Generating Bitcoins with your video card (OpenCL/CUDA)

[laszlo]

I have a working prototype of Bitcoin generation with OpenCL.  OpenCL is similar to OpenGL but it's for doing computation, not graphics.  Other similar technologies are CUDA (NVIDIA) and DirectCompute (Microsoft).  OpenCL is Apple's version of it but it is available for Linux and Windows as well.

I posted a Mac OS X binary package in my other thread.  I will create a patch and write up instructions for each platform this week as I have time but I wanted to find out who if anyone is interested in this even.

Please reply if you're interested and what OS/Video card you have.

Supported video cards:
NVIDIA: http://www.nvidia.com/object/cuda_gpus.html
AMD (ATI): http://developer.amd.com/gpu/ATIStreamSDK/Pages/default.aspx#two

Just to give people an idea of the benefit..
In my development computer I have an Intel E8600 CPU which runs at 3.3ghz normally.. I have it clocked at 4.1ghz
With 2 threads running the normal way I get about 1800k iterations per second in the bitcoin miner.
With an NVIDIA 8800 GTS (G80 GPU) video card I get around 3300-3800k iterations per second (varies with tuning of the code).

I find the best total performance comes from running one CPU thread and one OpenCL thread for the GPU (which does a bunch of work in parallel).  I get about 5000k/sec total that way on my development machine.  If I use 2 CPU threads it slows down the overall results to about 3600k/sec because the CPU is tied up computing and not pumping data to the GPU fast enough.

[1714303538]

1714303538

[1714303538]

1714303538

[fergalish]

Ubuntu 10.04, nvidia geforce 8300, AMD Athlon II 64 X2.  The GPU has only 8 cores, so it'll not do much.

[SmokeTooMuch]

Win7 Pro 64-Bit
Radeon HD4870 512MB

[m0mchil]

Win7 Pro 64-Bit
Radeon HD4350

[laszlo]

Thanks for the replies.  I have it working somewhat but it's really inefficient at the moment.  I will try to get a windows and linux version built this weekend so other people can play with it.

[allinvain]

Windows 7 Ultimate 64-bit
Nvidia Geforce 7900GS

But I'm soon going to get a ATI Radeon HD 5850

Btw, I think this is an excellent idea!

[SmokeTooMuch]

is the gpu-computing-client compatible with the normal client so the gpu-generated bitcoins will emmit into the same pool as the cpu-generated ones ?

[Xunie]

is the gpu-computing-client compatible with the normal client so the gpu-generated bitcoins will emmit into the same pool as the cpu-generated ones ?

Yes, totally.
It's just your GPU doing the work of your CPU.

[laszlo]

I ran into some technical problems while experimenting with this and didn't get much farther than the Mac OS version.  It seems that the NVIDIA CUDA 3.0 SDK for linux at least is very buggy when it comes to OpenCL.  It never quite worked right on Mac OS either, it just leaks a lot of memory (the OpenCL implementation, not bitcoin).  I will post an update if I come up with some solutions but I think the issues in the CUDA SDK will just have to be addressed by the vendor.. this stuff is all very beta right now, but I will keep my eye open for any new developments with the SDKs.  I do not have an AMD GPU so I have not been able to develop for that at all, but their support is pretty spotty right now too, almost all of the current cards are listed as experimental support.

I think this tech has a lot of potential but I think most people who are using it for anything real are rolling their own vendor specific solutions.  I would rather not pass out something that only partially works and only works for a few people.

[lachesis]

While this is definitely really cool, it won't benefit anyone if you release it. After two weeks, the difficulty system will adjust and we'll be back to generating bitcoins at the same rate but using much more computer power. Although I suppose this will benefit those with fast video cards instead of just fast CPUs.

I'm one of those people, so full steam ahead!

[alystair]

Windows 7 x64 / Debian VM
GTX 260

Extremely interested to see CUDA work considering I hardly strain this card anymore

[fflaguna]

Windows 7 x64
nVidia GTX 280M

I'm computer-savvy and ready to try out the CUDA binary.

[asdfman]

Unfortunately I dont have a compatible GPU on the machine Im running, but I think this is an awesome idea!!

[StinkiePhish]

OS X 10.6.4

Geforce GT 220.

Thank you!

[ultrasonicsite]

Windows XP
Geforce 8400 GS 512MB

[Ground Loop]

Windows XP SP3 32-bit
GeForce GT 240
(96 CUDA cores, 1GB)

Happy to try experimental code.

[letharion]

I have a couple of GTX480 and some other nvidia cards I could try this with on Gentoo Linux.
I've also done a bit of Cuda programming previously, so maybe I could lend a hand.

[EricJ2190]

Windows Vista SP2 64-bit
nVidia 8800 GTS (Overclocked)

[SmokeTooMuch]

I think posting all of that data is a bit useless now, afaik the threadstarter has canceled paused his project.

[sgtstein]

As far as I am concerned, if I can get the source that he has I can branch off of it and continue working. Otherwise I am starting work on my own CUDA build using C/C++.

Just a heads up...

[WARlrus]

Windows 7 64-bit
Dual NVIDIA GeForce 9800 GT

Can't wait to see this in action!

[StinkiePhish]

I didn't realize that the Mac dmg was already posted to a different thread.

http://heliacal.net/~solar/bitcoin/Bitcoin-MacOSX-Intel-svn-75-opencl-2010-05-10.dmg

Looks like it is a little outdated.

Any chance of the opencl code being patched to the current SVN version?

[sgtstein]

I didn't realize that the Mac dmg was already posted to a different thread.

http://heliacal.net/~solar/bitcoin/Bitcoin-MacOSX-Intel-svn-75-opencl-2010-05-10.dmg

Looks like it is a little outdated.

Any chance of the opencl code being patched to the current SVN version?

I can't patch it to the current as I have no way of testing it. Though, if I can determine how to disassemble his I might be able to get a leg up on working off the SVN and making a new one.

[Ground Loop]

I didn't realize that the Mac dmg was already posted to a different thread.

Thanks for the link StinkiePhish!  I installed it on my Macbook Pro (GeForce GT 330M), and it runs/connects.
It says "Generating", but I suspect it's not.  The graphics card is still on the Intel integrated chip.

I forced it over to the NVidia, but it still just says "Generating", with no khash/s readout.
Activity Monitor shows Bitcoin using 400% of the CPU, so I suspect nothing is going on the GPU.

Anyone else?

[Limb]

I didn't realize that the Mac dmg was already posted to a different thread.

Thanks for the link StinkiePhish!  I installed it on my Macbook Pro (GeForce GT 330M), and it runs/connects.
It says "Generating", but I suspect it's not.  The graphics card is still on the Intel integrated chip.

I forced it over to the NVidia, but it still just says "Generating", with no khash/s readout.
Activity Monitor shows Bitcoin using 400% of the CPU, so I suspect nothing is going on the GPU.

Anyone else?

Run

Code:

tail -f ~/Library/Application\ Support/Bitcoin/debug.log

You'll see if its using your card as an extra thread, will also show it initializing the OpenCL program.

[adavid]

Mac OS X 10.6.4
NVIDIA GeForce GT 330M

I tried the version posted and I was wondering if there would be the possibility to do generation only with the GPU (0 CPU threads and 1 GPU thread).

If the code would be checked-in somewhere, I'd be happy to check out and make other tweaks.

[Kalidratorma]

Windows XP SP3 32-bit
GeForce 9600 GT 512Mb

Happy to try experimental code.

[Ground Loop]

Run

Code:

tail -f ~/Library/Application\ Support/Bitcoin/debug.log

You'll see if its using your card as an extra thread, will also show it initializing the OpenCL program.

Excellent info, Limb!  It seems the OpenCL doesn't automatically 'kick in' the NVidia graphics core. It stays on Intel.  I'm not sure how the OS auto-switches, but whatever it's looking for, bitcoin doesn't trip it.

When Intel is selected, the relevant portion seems to be:

Code:

GetAvailableDevices():  Found 1 OpenCL platform(s)
ThreadOpenConnections started
ThreadMessageHandler started
GetAvailableDevices():  Checking OpenCL platform 'Apple' (index 0)
proxy connecting 
proxy connected 
GetAvailableDevices():  Found 0 devices on platform 'Apple' (index 0)
4 processors

When I force the NVidia, we have ignition!

Code:

GetAvailableDevices():  Found 1 devices on platform 'Apple' (index 0)
Printing info for device index 0  Vendor: NVIDIA  Device: GeForce GT 330M
...
Device 'GeForce GT 330M (NVIDIA)' : CL_KERNEL_WORK_GROUP_SIZE : 192
Device successfully initialized!  (GeForce GT 330M (NVIDIA))

It runs one bitminer thread on OpenCL, and three on the CPU.

Code:

**Perf - thread 1 : 1128k iter/sec
**Perf - thread 2 : 296k iter/sec
**Perf - thread 3 : 293k iter/sec
**Perf - thread 4 : 296k iter/sec
**Perf - total : 2013k iter/sec (4 threads)

I've never seen the MBP fan this excited.

[laszlo]

Yea, I tried that too, after seeing it at 90C temps for a while I decided to not generate with my MBP

It's great that it's working for you - I'm not sure if it's possible to ask the OS to switch to the high performance video automatically, as far as I know you have to go to the power settings and tweak it manually, like you did.

[mcdett]

The cooling fans are working hard on my 13 MBP

OSX 10.6.4 - NVIDIA GeForce 9400M

**Perf - thread 1 : 519k iter/sec
**Perf - thread 2 : 317k iter/sec
**Perf - thread 3 : 315k iter/sec
**Perf - total : 1151k iter/sec (3 threads)

Running the original bitcoin I was getting around 6-800k iter/sec

[Ground Loop]

OpenCL bitcoin crashes with a Bus Error on a 27" iMac (ATI Radeon).

Anyone get it to work on a current-gen iMac?

[Leonassan]

I'm interested
Windows 7 Ultimate 64
NVidia GTX295

[em3rgentOrdr]

I would very much like OpenCL/Cuda code to generate bitcoins with my GPU:

GF9800gx2 (http://www.nvidia.com/object/product_geforce_9800_gx2_us.html)
Ubuntu 10.04

[Olipro]

szia lászló

Intel Core 2 Quad Q9550 8GB DDR3-1333 RAM
Nvidia Geforce 9800GT Golden Sample
Server 2008 R2 (x64)


erdekelne a programod, de sokkal jobban erdekelne a forraskod.
I'm interested, I'd be twice as interested in the source code though.

[TryBitCoin]

del

[aaaxxx]

Windows 7 Ultimate 64Bit

CPU: AMD Athlon 64 X2 5200+ 2.71GHz
DDR2 3GB 245MHz
GPU: Palit 8600GT DDR2 1GB

[dougztr]

I have Windows Vista 64 and an ATI Radeon HD 4830

Also, I have a working 0.3.2 Gentoo Linux version which I compiled from source. I have ATI Stream SDK installed so I have a fully working OpenCL developer kit, the only thing I'm lacking is some sources to add in OpenCL support.

I'd be happy to work on the Linux version.

[em3rgentOrdr]

Anyone know when this OpenCL/CUDA code will be released?  Are there any development versions of this OpenCL/CUDA source code out yet?

[knightmb]

As far as I can tell, nothing else has been done. Seems to only work on a few select Mac at the moment and even then appears to be a little unstable.

[ledskof]

Intel Core i7 965
12 GB DDR3
Nvidia GTX480
Windows 7 64bit
Funtoo Linux 64bit

[Ground Loop]

Runs fine on the i5 Macbook Pro.  Generates a lot of heat, but it's more stable than the speedy Windows x64 version by far.

[omegadraconis]

I finally got around to testing this on my hackintosh pc and it seems to be running just fine, although it is definitely not ready for the masses. The speed keep jumping around (although this maybe related to my hackintosh nvidia driver), The debug log shows it is coming in @
**Perf - thread 1 : 1636k iter/sec
**Perf - thread 4 : 1616k iter/sec
**Perf - thread 5 : 1295k iter/sec
**Perf - total : 4547k iter/sec (3 threads)

**Perf - thread 1 : 3363k iter/sec
**Perf - thread 3 : 2109k iter/sec
**Perf - total : 5472k iter/sec (2 threads)

**Perf - thread 1 : 2630k iter/sec
**Perf - thread 3 : 2069k iter/sec
**Perf - total : 4699k iter/sec (2 threads)

Seems like the speed is dropping over time after running for a short bit of time, anyone else having this issue?

At it's peak, this is about 2200khash better then in windows with the x64bit optimized version. The normal mac client is only coming in at 858khash/s for some reason with two cores. I think is is real promise with using opencl to add khashs.

My hardware is an intel Dual-core celeron @3.8ghz, My GPU is an Geforce 8800GTX, 2GB ram, and I am running OSX 10.6.3.

[nimnul]

While this is definitely really cool, it won't benefit anyone if you release it. After two weeks, the difficulty system will adjust and we'll be back to generating bitcoins at the same rate but using much more computer power. Although I suppose this will benefit those with fast video cards instead of just fast CPUs.

On the contrary, it will benefit everyone. Remember, bitcoin is not about generating free bitcoins - it's about creating a market. And more complexity means more stable market as it becomes harder to rent a farm of servers and put generated coins to market to destabilize it.

[kwukduck]

Yes, it brings more power to the desktop user instead of the server people that are just farming coins in big numbers, servers don't have strong GPU's so it just pulls the desktop machines up to level.

Win 7 32bit + Ubuntu 10.04LTS
nVidia 8800GT

[LZ]

Fedora 13, Mac OS 10.5, Windows 7
nVidia GeForce GTS 250

[puddinpop]

I've created a Windows 32 bit CUDA enabled client.  Consider it a beta right now.  You'll need the Visual Studio 2005 runtime to run it.  I spent a considerable amount of effort to create this, so I'm asking that if you use it, you send 5 me BTC for every block you generate while using it.  I suggest you try it out first with 2 clients connected to each other and an empty block database, so you can see that it works correctly for you relatively quickly.

See my signature for the latest download.

[BitLex]

.. so I'm asking that if you use it, you send 5 me BTC for every block you generate while using it. ..

your askin'?
would be cool, but seems like you hardcoded your 10%-share to auto-send you coins,
as much as i'd like to reward your efforts, i really don't like that.

besides that and the fact, that generating totally slows down my system (not even surfing is fun),
it seems to work, it shows (!) me  ~33.000khs on a gtx260, not bad (if it's true) compared to the regular ~3.800khs the triple-core-cpu gets.
can't do anything else besides running it though, so i shut it down again after a short test.

[Gespenster]

Not sure if I trust this, might be a scam to rip off your coins or something.

[puddinpop]

Yes, the transaction is automatic if you have an available balance.  It is detailed in the readme.txt file.  Please don't use it if you feel uncomfortable with it.  I just want to give a better chance of generating blocks to those with CUDA enabled hardware.  There is already 1 person with 20% of the hashing capability, so I thought this would even the field better.

besides that and the fact, that generating totally slows down my system (not even surfing is fun),

While the kernel is running, the GPU is 100% dedicated to running the kernel only, not updating the screen.  This is why you will see sluggishness.  Actually, you could squeeze out some more performance at the cost of more sluggishness.  This is perfectly normal.  If you have 2 graphics cards, you could use the slow one for your display and the fast one just for CUDA.

Reflector says it is not a .NET assembly.

That is quite correct.  Bitcoin is a c++ application, so obviously there won't be any .NET code.

how do we know you won't take more or do other things? Your intentions seem to be muddy.

Please test it in a sandbox environment if you are uneasy.  The only thing it does differently form the standard client is generate using CUDA, and send 5 BTC for every block you generate while using the CUDA miner.  Everything else is standard.

[Gespenster]

There is already 1 person with 20% of the hashing capability, so I thought this would even the field better.

Could you elaborate on that?

[puddinpop]

There is already 1 person with 20% of the hashing capability, so I thought this would even the field better.

Could you elaborate on that?

ArtForz, on IRC, has like 12 or so 5770s running his own OpenCL client.  The entire network is doing something around 5-6Ghash/s, and he alone has stated that he has over 1Ghash/s

[MoonShadow]

While the kernel is running, the GPU is 100% dedicated to running the kernel only, not updating the screen.  This is why you will see sluggishness.  Actually, you could squeeze out some more performance at the cost of more sluggishness.  This is perfectly normal.  If you have 2 graphics cards, you could use the slow one for your display and the fast one just for CUDA.

Is there some way to tie the GPU portion of generation to a screensaver-like start/stop?

[puddinpop]

Is there some way to tie the GPU portion of generation to a screensaver-like start/stop?

Sure, if someone wanted to code a screensaver they could do that.

[Ground Loop]

Everything above is true here too.

My GT240 is pumping 16,339 khash/sec.  Yow!
And yes, the machine is damn near unuseable.

By my math, I'm generating coins at 3x the previous rate, so if there's a 10% author tax, I can live with that.

[omegadraconis]

My 8800GTX is currently crunching at a blistering 25000-24000 khashs with one cpu core pegged @ 50%. The hardware is the exact same as above where I posted my results from the mac OSX Cuda Client but, running windows 7 ultimate 64bit. My system is still usable (browsed the web with no noticeable difference) but, windows menus and other aero enabled windows are laggy. Not trying to be "that guy" but, I am a little un-trusting of a closed-source client that sends bitcoins to a random address without knowing what else is under the hood. Then again if it nets me coins and only cost 5 thats not bad as long as that is all it does. Anyone using it generate coins yet? Also how many coins would you want to opensource the client?

[EDIT]
Wanted to add a note here that I have been running the client for about 8 hrs now and it is stable no crashes. Seems that puddinpop did a good job with the code. I think I can live with the 10% tax for now once I see the coins start rolling in.

[mizerydearia]

Although I don't use windows, I'll donate some bitcoins to puddinpop if they release the client source code as open source.  Otherwise I'll donate some bitcoins to the first person to release a hacked/modified version of the client that disables the 10% autopayment on block generation, whichever comes first.

puddinpop would ultimately establish far more profits if the client remains closed, at least until using the client is no longer beneficial as the difficulty will have increased far too much for it to be as beneficial as it was compared to noncuda clients.  However, reverting back to a noncuda client would be worse than continuing to use the cuda client and therefore unless an alternative is established, puddinpop will continue to receive plenty of profits from usage of the client.  While it will be nice to donate a nice amount to puddinpop for their efforts, I personally feel 10% for all block generations from everyone is too much, and I would like to encourage open source or otherwise rewarding those that defeat the proprietary profiting addition.

Keep in mind while khash/sec is increased providing higher chance to generate a block, more electricity is used to do so.  I don't believe puddinpop is willing to pay 10% of electricity bill or cost of extra electricities used from usage of this software.

Also, Satoshi and sirius-m Have put forth a lot of hard work into the open source Bitcoin client and do not force donations.  I would like to donate to Satoshi for his efforts. ^_^

[omegadraconis]

Although I don't use windows, I'll donate some bitcoins to puddinpop if they release the client source code as open source.  Otherwise I'll donate some bitcoins to the first person to release a hacked/modified version of the client that disables the 10% autopayment on block generation, whichever comes first.

I will donate for an opensource client.

[TTBit]

I'm not a programmer or expert user, so go easy on me please.

I have an old athlon machine that can generate ~1850 khash/s. I have a geforce 8500 gt on that machine, and used this to up production to ~2150 khash/s (says GPU 2150, so I know I'm using the -gpu). How can I get these 20,000 khash/sec that everyone else is? Is my athlon the bottleneck here, or can I just buy a nicer video card?

[omegadraconis]

I'm not a programmer or expert user, so go easy on me please.

I have an old athlon machine that can generate ~1850 khash/s. I have a geforce 8500 gt on that machine, and used this to up production to ~2150 khash/s (says GPU 2150, so I know I'm using the -gpu). How can I get these 20,000 khash/sec that everyone else is? Is my athlon the bottleneck here, or can I just buy a nicer video card?

For Cuda I think (emphasis on think, not know) the best thing to look for would be the number of Stream Processors, which is like the number of cores. If you are using an old Athlon and your stuck with agp I think this will slow you down abit vs a pci-e bus but, I really don't know. Also you would want to make sure you cpu can keep up with pumping data to your gpu. Like I said in my post one of my cpu's cores is pegged @ 100% usage while the gpu crunches away.

[puddinpop]

If I were to receive a sizable donation, I might be willing to open source the code.  However, any talk about hacking or cracking the current client makes me less likely to do this, and I will not accept donations from anyone who has already advocated for such measures.

Here's a little story to think about.

There is an apple tree growing in some public place where everyone is free to take apples whenever they wish.  Everyone is happy and content eating the free apples directly from the tree.  Now some start to think about all the possibilities that the apples can provide.  You can make apple juice, apple cider, apple pie, etc.  Most only think about the possibilities and don't take action.  Either because they can't cook, or they don't have the time or resources.  Now someone comes along and decides to make apple juice.  He sets up a stand and sells the juice to the people.  Some people are fine with this, and happily pay.  They understand the value added to the apple by processing it into juice.  It took time, effort, and resources to do so.  Others are upset that the apples they had been getting for free have been processed and that they now have to pay for the refined product.  They would rather take the apple juice by force rather than paying for it or making their own.

I understand that some of you are uncomfortable with the client.  If that is the case the obvious solution is not to use the client.

[omegadraconis]

Just wanted to send another update here that the client is running just fine for the most part of about 14hrs or so now. I had to stop it once to add a fan to my gpu because it was overheating, Otherwise generation is stable.

Also something that I noticed, when I use the -server and -daemon switch the client exits after the command has been sent. For example when I do a getinfo, the info box comes up but, after that the client closes, no crash just exit. The log shows the getdata command and then some operations afterwards, so it doesn't seem like it's bailing out right at the command being sent. Anyone else having this issue?

[Ground Loop]

I set up a new wallet for this client.  Starting balance is zero, so it can only be a net win regardless of what the code wants to do.

That said, my passively-cooled Zotac GT240 is not up to the task.  After five minutes, the generation rate has dropped down to nearly nothing.  I suspect the card has throttled itself down as it bakes under the passive heatsink.  Hmm.

[jgarzik]

I set up a new wallet for this client.  Starting balance is zero, so it can only be a net win regardless of what the code wants to do.

Stealing your BTC isn't the only thing a closed-source client might do.  Make sure you don't have any personal information stored on the computer running the client, no bank account numbers etc.

And monitor the network communication to make sure it only communicates with other bitcoin P2P clients, and not other botnets as well.  It would be too easy to hide a key-generation botnet client inside something appearing to be a bitcoin client.

[MoonShadow]

I set up a new wallet for this client.  Starting balance is zero, so it can only be a net win regardless of what the code wants to do.

Stealing your BTC isn't the only thing a closed-source client might do.  Make sure you don't have any personal information stored on the computer running the client, no bank account numbers etc.

And monitor the network communication to make sure it only communicates with other bitcoin P2P clients, and not other botnets as well.  It would be too easy to hide a key-generation botnet client inside something appearing to be a bitcoin client.

Wow, that's paranoid, and rude.  Dude, we are not talking about some shady third party, you can ask the programmer the how and why right here.  Nor do you have to trust him or use his code.  You could do it, and open source it if you like, if you have the skills.  For the time being, however, the code belongs to he who wrote it, and he can dictate the conditions.  The client is open, but for now, the gpu code is not.  If you want to help to make it so, someone could make an offer of a number; after which the code is open sourced by the author, whether that number comes from the gpu client or from regular donations.

How about 50K bitcoins from all sources or one year after the release date, whichever comes first?

[BitLex]

paranoid?
who do you tell?
those that want to stay completely anonymous, those using Tor and similar stuffs to hide theyr identities? 
those are paranoid anyway and they know it.

believe it or not, in a closed source client anything might be possible.

i already made the mistake to trust someone i know nothing about and lost a few coins,
i don't think it's paranoid to be sceptical.


i'm with mizery and in for another donation on a first open source cuda-client.

[neptop]

Windows 7 x64
HD4650
ATI stream SDK has been installed for a while now to play a bit.

I'd also donate for the source code.

[MoonShadow]

believe it or not, in a closed source client anything might be possible.

i already made the mistake to trust someone i know nothing about and lost a few coins,
i don't think it's paranoid to be sceptical.

I am aware that a closed application could do just about anything.  But there are sound reasons for preferring anonimity beyond paranoia.  This list is pretty anonymous anyway, since we really don't know who each other are, but you can still converse with the author.  If you don't trust his code, don't use it.  Collective bounties for opening code is a valid method.  I would wager that there are a number of ways that the total amount sent to the author's encoded address could be tallied.  Just tracking the transfers to that address, either donations or generation commissions, can be seen within the blockchain by anyone willing to write a program to do so.  If he agrees to a price, and refuses to comply, then you can be dick.

[BitLex]

If you don't trust his code, don't use it.

i didn't say anything else, did i? 
i'm not asking people to hack into and modify puddin's client,
i'm not even asking puddin to open his code, it's his choice alone what he wants to release,
just offering some coins to another programmer, that might be willing to show us the code to proof it does what it's supposed to.

I would wager that there are a number of ways that the total amount sent to the author's encoded address could be tallied.  Just tracking the transfers to that address, either donations or generation commissions, can be seen within the blockchain by anyone willing to write a program to do so.  If he agrees to a price, and refuses to comply, then you can be dick.

do whatever you want, track coins, workaround the auto-send somehow (nothing's easier than that), or even hack/modify the client to your needs, if your able, willing and bored enough to, that again is your choice.


not a goal of mine to be dick (yuck) at all,
i'd just like to use a cuda-client cuz my gfx-card is kinda bored most of the time and could do ~10times the work, my cpu is able to do,
so if anyone releases one, i'm willing to donate some of my hard earned, bought and generated coins.

[GeorgeH]

Anyone have this running on windows?

[omegadraconis]

Anyone have this running on windows?

The client posted by puddinpop is a windows client. Probably at this point could use it's own thread and perhaps a mod should split the topic at the point that puddinpop posted the client to avoid further confusion. I have been running the client on windows 7 64bit for a few days on and off.

I am currently doing some testing on the client on a closed network of clients. Thus far the client has not done anything outside of generate coins and send some to the authors address. Netstat does not show any extra ports being listened to. I am fairly confident that the client is not doing anything out of the norm for a bitclient (i.e. virus, malware, etc.) and it does pass online anti-virus tests. The client may have a bug that I am trying to verify tonight. I don't want to cause a stir until I am sure and would like very much for some one else verify they are getting the same results as me. I will post more once or if I see the client do the same thing again as I deleted the data files and started over clean. I will know in about another 75blocks (sometime in the middle of the night). In the mean time has anyone else linked a client with this one and let the run from block 0 on a closed test network?

[Update] OK no bug here just a matter of the way the client works. Everything a block is generated the "debit" is incremented by 5btc. This debit is sent right then if coins are available. What I was seeing on my test network was that clocks were being generated and taking a long time to mature so that once I have 50btcs generated and matured they would all be debited out of the client because the blocks had already been generated. I would say my test of the client makes me trust it with a fair level of confidence.

[Immanuel]

I use a 9500GT and use Debian 5.0 64-bit and Windows 7 64-bit. I am very interested in this. Thanks!

[Cdecker]

This is probably one of the best reasons we should switch the main client to a more restrictive license, that wouldn't allow to redistribute a closed source client based on the original client.

Anyway am I the only one who's wondering if the reported performance is actually what is going on under the hood? It doesn't take a lot of effort to just multiply the real number of khash/s by a constant factor, which would urge people to use the modified client which in turn sends part of the coins to the author, without providing actually any benefits?

Has anyone verfied the odds of generating coins with the theoretical number [1]?

[1] http://www.alloscomp.com/bitcoin/calculator.php

[omegadraconis]

This is probably one of the best reasons we should switch the main client to a more restrictive license, that wouldn't allow to redistribute a closed source client based on the original client.

Anyway am I the only one who's wondering if the reported performance is actually what is going on under the hood? It doesn't take a lot of effort to just multiply the real number of khash/s by a constant factor, which would urge people to use the modified client which in turn sends part of the coins to the author, without providing actually any benefits?

Has anyone verfied the odds of generating coins with the theoretical number [1]?

[1] http://www.alloscomp.com/bitcoin/calculator.php

In my closed test network experiment with the client the speeds did seem to be on par with what should have been seen. The difficulty was 1 and I khashed @ a rate of ~28500KHPS. I found that I should be seeing a block roughly 7 minutes with a 95% probability. I was seeing them within this time frame regularly. It seems to me to be working but, I have not gotten to put it to much use on the public network as of yet. I am going to start running it full tilt on the public network now to see what happens.

[puddinpop]

Here's a build with the latest svn, 149.

See my signature for the latest download.

[true]

Fuck your ransomware, I'll work on cracking this.

[aceat64]

Fuck your ransomware, I'll work on cracking this.

That's not really helpful to anyone, a more productive use of your time is to work on an opensource CUDA/OpenCL client. Assuming you have the required programming expertise, which I don't.

[true]

Fuck your ransomware, I'll work on cracking this.

That's not really helpful to anyone, a more productive use of your time is to work on an opensource CUDA/OpenCL client. Assuming you have the required programming expertise, which I don't.

It's helpful to at least two people, and considering it's already done, the time/payoff seems pretty high to me. Seemed to be a pretty productive use of my time.

[torservers]

For some reason, the CUDA client doesn't use my GPU (Win7 64bit, 8800GT). I do like that it shows khashes/s though.

[LZ]

With -gpu option?

[torservers]

I have reinstalled the NVIDIA driver and now it works. Thanks. Would be nice to have it use CPU cores and GPU.

[kwukduck]

great idea, take an open source project, make a closed source mod that sends you cash, profit.

no thanks i'll wait for an open version, who knows what happens to my wallet using this thing..

[nelisky]

great idea, take an open source project, make a closed source mod that sends you cash, profit.

no thanks i'll wait for an open version, who knows what happens to my wallet using this thing..

I have nothing against closed source by itself, and I surely understand the profit taken after hard work has been done, me being a coder working on the comercial arm of an open source project. But that's where my positive reasoning ends.

Charging for work is important, but not so much taking a ransom. And are these 5 BTCs per generated block only from blocks generated using the GPU? And are 4 of these 5 given to satoshi and all the others working on the open source?

Anyway, still crude and already outdated but look here -> http://bitcointalk.org/index.php?topic=1009.msg12264#msg12264
I didn't understand what all the fuss was about in this cuda development so I decided to take a stab at it. I only have very limited time, but I am now testing a version doing 6Mh/s as opposed to teh 1.4MHs I get from the CPU. And it has only been mildly optimized, so there's still much room for improvement.

My hardware is limited too, so I only tested in my OSX cli client, but the idea is there for other to work on. I can be convinced to work harder on it, but this was just proving a point for me.

[puddinpop]

There is no ransom involved.  Use of the client is at the will of the user, and the cost is explained and known about up front.  Would you feel better if there was a single large up front cost for each user?  The way it is now, the cost is proportional to the number of blocks you generate, which I think is much more fair.  Otherwise it will be severely disproportionate for those who use it very little or have a slow GPU and those who use it a lot on a fast GPU.

Now I have seen a few say they would be willing to donate for an open source client.  I would be willing to release the source provided a large enough donation is made.

[BitLex]

And are 4 of these 5 given to satoshi and all the others working on the open source?

would be nice, i'd then even think about using it.

did a few more tests on puppin's client, more to check results of gpu-OC'ing though.
so far i was able to squeeze another ~10Mhash outof my GTX260, get ~42Mhash now
(default clock 575/1240/1000, OC'ed 700/1500/1200).

i also tested my older 8600GT, which gets about 5200khash stock, 6500khash slightly OC'ed,
on a dual-core amd be-2350 (cpu gets <2000).
not sure yet, what extra-power the card consumes, system without card is/was around 60Watt @100% cpu load,
the card is powered by PCIe only, so it's max.+75W, my guess is <120W for the system.
gpu-mining would even here be "more effective" than cpu-mining, but prolly still non-profitable.

anyone tested (or able to) an ION platform yet?
those are pretty cheap and using a dual-core ULV cpu they won't need much power.


both testings run quite stable (OCed for about 20hrs) but didnt generate anything.
guess i'm done for now, as soon as there's another client, i'll get back testing.  

[nelisky]

There is no ransom involved.  Use of the client is at the will of the user, and the cost is explained and known about up front.  Would you feel better if there was a single large up front cost for each user?  The way it is now, the cost is proportional to the number of blocks you generate, which I think is much more fair.  Otherwise it will be severely disproportionate for those who use it very little or have a slow GPU and those who use it a lot on a fast GPU.

Now I have seen a few say they would be willing to donate for an open source client.  I would be willing to release the source provided a large enough donation is made.

Right, I may simply be dense, sorry. The way I see it, you did an amazing job which I, unfortunately, can't make use of. If I could, maybe I'd gladly give the 5 coins to you, who knows? But the fact remains that while I feel that this has a niche to it, and while I'd love to have a farm of multi tesla computers to generate coins to my heart's contempt, I don't like the idea of closed source, binary distributed software for security reasons. And on top of that, you are forcing the (fairly distributed) payment of your work, but fail to move that income down the line, to the base of your own work.

But that's more philosophical  than anything else, and you did nothing wrong, you disclosed your intentions from the start, etc. I just feel that the community would be more forthcoming with sharing their earnings if given the chance, that's all. Having said that, I don't expect a bitcent for the  work I did, but then again I'm not doing it for profit.

[BitLex]

Would be nice to have it use CPU cores and GPU.

you could just use a VM to run a CPU node on the same machine.

[puddinpop]

And are 4 of these 5 given to satoshi and all the others working on the open source?

Satoshi and the others wrote the code that made it possible to write the client, but they did not write this code that uses the GPU to generate the blocks.  I would be more than happy to add an extra 5, 10, 20, or whatever payment for satoshi and any others who worked on the bitcoin client if you like.  Anyway, are you really donating to them now after every block you generate?  Are you going to donate to them after every block you generate on your CUDA client?  After all, they made it possible for you to write it.

And on top of that, you are forcing the (fairly distributed) payment of your work, but fail to move that income down the line, to the base of your own work.

You are perfectly capable to make that payment yourself when you generate a block.

[nelisky]

And are 4 of these 5 given to satoshi and all the others working on the open source?

Satoshi and the others wrote the code that made it possible to write the client, but they did not write this code that uses the GPU to generate the blocks.  I would be more than happy to add an extra 5, 10, 20, or whatever payment for satoshi and any others who worked on the bitcoin client if you like.  Anyway, are you really donating to them now after every block you generate?  Are you going to donate to them after every block you generate on your CUDA client?  After all, they made it possible for you to write it.

And on top of that, you are forcing the (fairly distributed) payment of your work, but fail to move that income down the line, to the base of your own work.

You are perfectly capable to make that payment yourself when you generate a block.

This is starting to be another nitpick thread, but I'll try my best to keep it civil

I'm not arguing everyone should pay everytime  they generate a thread. That is the purpose of the infrastructure and people get paid for the work they are doing suporting that infrastructure. In fact, if noone generated blocks, we wouldn't be here in the first place. So, yes, I'm doing my best to help satoshi and everyone else by generating more blocks and making the network harder to defeat. I don't have to pay others for the work I do, that is just nonsense!

What you are doing is slightly different. You create a better hashing mechanism/algorithm and you can capitalize on it by generating more blocks. Go for it, it's what we are all doing anyway. But then you decide the effort you put in this isn't going to pay off anytime soon by just generating, moreso because regardless of how fast you hash, there's always a component of luck.

So what do you do? You sell your work. Actually, you rent your work. So now you are not suporting the network, you are capitalizing on top of it. You have every right to do so, mind you, I'm not arguing with that. If you had a way to hash 100MH's on a P3, I know you could sell that on biddingpond for a lot of coins. It's not the idea that bugs me, it's the process. By providing a closed source system you depend on other's trust, which is fine if they trust you. I don't trust you any more than you trust me, insofar as we don't know each other.

I run very little closed source software, and yes, I run OSX and linux and no, I haven't read the whole source for the kernel, but still a group of other people have, and I trust numbers.

Again, I'm not pointing a finger or stating you are wrong in your intentions. I'm just saying I don't see what the big deal for this to be so secretive and closed, and I have put my money where my mouth is and went ahead and developed something myself. It's crude, gives me ~7MHs at this moment on my macbook, which is a 4x speedup over the CPU version, I'm happy I did this in two partial days without any previous cuda experience. Yeah, I'm a coder by trade, but still this gave me a warm feeling...

I will continue development until someone else picks it up, at the pace I can afford. It  may end up amounting to nothing, but hey, at the end of the day, I feel very good about myself, as I know you do to, so two thumbs up, my friend.

[fawkes]

Just wanted to share my experience, I decided to try this version of Bitcoin, more specifically version 0.3.12.1 beta CUDA enabled and it really works as it increased my generating ability 10x, but it is useless as it is now because after finally having generated my first 50 coins so that I finally had a balance other then 0.00 they were all sent to 1HZNsUqQxKVLmfPfCAzLwrnVDzx8CxwxnM leaving me with a balance of 0.00 again.

I thought only 5 coins would be sent not 50, if 50 are sent then what's the point of generating coins if they are all going to someone else? At least with the normal Bitcoin I get to the coins even though I will be generating far less.

Did anyone else experience something similar?

[puddinpop]

The 5 per block are cumulative, so If you had a 0 balance and generated multiple blocks it waits until a balance is available and sends 5 * generated blocks.  Did you really generate 10+ blocks so fast?

[fawkes]

The 5 per block are cumulative, so If you had a 0 balance and generated multiple blocks it waits until a balance is available and sends 5 * generated blocks.  Did you really generate 10+ blocks so fast?

Thats the thing, I only generated 1 block not 10 so that is should have been a 5 coin payment not 50, what it says here is balance + 50 and then after it was confirmed it said - 50 with a transaction to 1HZNsUqQxKVLmfPfCAzLwrnVDzx8CxwxnM so there seems to be a problem here, anyone that was just taxed 5 coins and not 50?

[puddinpop]

I just triple checked the amount by connecting 2 clients together and the payment is definitely only 5.  Did you use the same wallet to do testing and then go on the live network?  The wallet will accumulate the payment.  If you post an address I'll send back your coins, but first delete your wallet or it will probably happen again.

[mizerydearia]

The 5 per block are cumulative, so If you had a 0 balance and generated multiple blocks it waits until a balance is available and sends 5 * generated blocks.  Did you really generate 10+ blocks so fast?

Does this mean:

Generate 1 block, +50btc, 5btc * 1 == 5btc sent to puddinpop, 0+50-5 == 45btc total balance
Generate 2 blocks, +50btc, 5btc * 2 == 10btc sent to puddinpop, 45+50-10 == 85btc total balance
Generate 3 blocks, +50btc, 5btc * 3 == 15btc sent to puddinpop, 85+50-15 == 120btc total balance
Generate 4 blocks, +50btc, 5btc * 4 == 20btc sent to puddinpop, 120+50-20 == 150btc total balance
Generate 5 blocks, +50btc, 5btc * 5 == 25btc sent to puddinpop, 150+50-25 == 175btc total balance
Generate 6 blocks, +50btc, 5btc * 6 == 30btc sent to puddinpop, 175+50-30 == 195btc total balance
Generate 7 blocks, +50btc, 5btc * 7 == 35btc sent to puddinpop, 195+50-35 == 210btc total balance
Generate 8 blocks, +50btc, 5btc * 8 == 40btc sent to puddinpop, 210+50-40 == 220btc total balance
Generate 9 blocks, +50btc, 5btc * 9 == 45btc sent to puddinpop, 220+50-45 == 225btc total balance
Generate 10 blocks, +50btc, 5btc * 10 == 50btc sent to puddinpop, 225+50-50 == 225btc total balance
Generate 11 blocks, +50btc, 5btc * 11 == 55btc sent to puddinpop, 225+50-55 == 220btc total balance
Generate 12 blocks, +50btc, 5btc * 12 == 60btc sent to puddinpop, 220+50-60 == 210btc total balance
Generate 13 blocks, +50btc, 5btc * 13 == 65btc sent to puddinpop, 210+50-65 == 195btc total balance
Generate 14 blocks, +50btc, 5btc * 14 == 70btc sent to puddinpop, 195+50-70 == 175btc total balance
Generate 15 blocks, +50btc, 5btc * 15 == 75btc sent to puddinpop, 175+50-75 == 150btc total balance
Generate 16 blocks, +50btc, 5btc * 16 == 80btc sent to puddinpop, 150+50-80 == 120btc total balance
Generate 17 blocks, +50btc, 5btc * 17 == 85btc sent to puddinpop, 120+50-85 == 85btc total balance
Generate 18 blocks, +50btc, 5btc * 18 == 90btc sent to puddinpop, 85+50-90 == 45btc total balance
Generate 19 blocks, +50btc, 5btc * 19 == 95btc sent to puddinpop, 45+50-95 == 0btc total balance
Generate 20 blocks, +50btc, 5btc * 20 == 100btc sent to puddinpop, 0+50-100 == 0btc total balance
Generate 21 blocks, +50btc, 5btc * 21 == 105btc sent to puddinpop, 0+50-105 == 0btc total balance
Generate 22 blocks, +50btc, 5btc * 22 == 110btc sent to puddinpop, 0+50-110 == 0btc total balance
Generate 23 blocks, +50btc, 5btc * 23 == 115btc sent to puddinpop, 0+50-115 == 0btc total balance
Generate 24 blocks, +50btc, 5btc * 24 == 120btc sent to puddinpop, 0+50-120 == 0btc total balance
Generate 25 blocks, +50btc, 5btc * 25 == 125btc sent to puddinpop, 0+50-125 == 0btc total balance
Generate 26 blocks, +50btc, 5btc * 26 == 130btc sent to puddinpop, 0+50-130 == 0btc total balance
Generate 27 blocks, +50btc, 5btc * 27 == 135btc sent to puddinpop, 0+50-135 == 0btc total balance
Generate 28 blocks, +50btc, 5btc * 28 == 140btc sent to puddinpop, 0+50-140 == 0btc total balance
Generate 29 blocks, +50btc, 5btc * 29 == 145btc sent to puddinpop, 0+50-145 == 0btc total balance

?

[puddinpop]

No, it's 5 coins per block.  It doesn't matter how many blocks you've already generated or how many you're going to generate, it's always 5.  If you have 0 balance when you generate a block, the payment will wait and accumulate if you generate more blocks until you have a balance.

[true]

Your ransomware has bugs and is ripping people off.

I was testing mine on the live network but I think I'll test with a test local network just to make sure it works and release it so you stop ripping people off.

[nelisky]

Your ransomware has bugs and is ripping people off.

I was testing mine on the live network but I think I'll test with a test local network just to make sure it works and release it so you stop ripping people off.

Yes, please do release yours! I loved the experience of thinking massive parallel for coding mine (and I also need to release my 2x speedup update). I'm really interested is understanding your approach to the problem!

[true]

Read previous posts to understand what I mean, nelisky.

[nelisky]

Read previous posts to understand what I mean, nelisky.

Heh, got confused there, sorry. But if by 'cracking this' you mean reverse engineer it, from experience it's easier to just code it from scratch! If, on the other hand, you mean remove the ransom part, then yeah, I guess it can be done with little effort.

[LobsterMan]

Here's a little story to think about.

There is an apple tree growing in some public place where everyone is free to take apples whenever they wish.  Everyone is happy and content eating the free apples directly from the tree.  Now some start to think about all the possibilities that the apples can provide.  You can make apple juice, apple cider, apple pie, etc.  Most only think about the possibilities and don't take action.  Either because they can't cook, or they don't have the time or resources.  Now someone comes along and decides to make apple juice.  He sets up a stand and sells the juice to the people.  Some people are fine with this, and happily pay.  They understand the value added to the apple by processing it into juice.  It took time, effort, and resources to do so.  Others are upset that the apples they had been getting for free have been processed and that they now have to pay for the refined product.  They would rather take the apple juice by force rather than paying for it or making their own.

I've just been catching up a bit and reading through this thread.

Isn't the above story a bit flawed with your program in mind though? The way your program works is almost like saying you are going to use someone else's house, and stove, and gas, and pots/pans to make the apple juice, then charge them for it in the end too, isn't it?

[nelisky]

Here's a little story to think about.

There is an apple tree growing in some public place where everyone is free to take apples whenever they wish.  Everyone is happy and content eating the free apples directly from the tree.  Now some start to think about all the possibilities that the apples can provide.  You can make apple juice, apple cider, apple pie, etc.  Most only think about the possibilities and don't take action.  Either because they can't cook, or they don't have the time or resources.  Now someone comes along and decides to make apple juice.  He sets up a stand and sells the juice to the people.  Some people are fine with this, and happily pay.  They understand the value added to the apple by processing it into juice.  It took time, effort, and resources to do so.  Others are upset that the apples they had been getting for free have been processed and that they now have to pay for the refined product.  They would rather take the apple juice by force rather than paying for it or making their own.

I've just been catching up a bit and reading through this thread.

Isn't the above story a bit flawed with your program in mind though? The way your program works is almost like saying you are going to use someone else's house, and stove, and gas, and pots/pans to make the apple juice, then charge them for it in the end too, isn't it?

Almost right. You do your own cooking, not him, but by using his special spoon, you'll get your meal quicker You will, however, loose 10% of your meal in the end.

[puddinpop]

Here's a little story to think about.

There is an apple tree growing in some public place where everyone is free to take apples whenever they wish.  Everyone is happy and content eating the free apples directly from the tree.  Now some start to think about all the possibilities that the apples can provide.  You can make apple juice, apple cider, apple pie, etc.  Most only think about the possibilities and don't take action.  Either because they can't cook, or they don't have the time or resources.  Now someone comes along and decides to make apple juice.  He sets up a stand and sells the juice to the people.  Some people are fine with this, and happily pay.  They understand the value added to the apple by processing it into juice.  It took time, effort, and resources to do so.  Others are upset that the apples they had been getting for free have been processed and that they now have to pay for the refined product.  They would rather take the apple juice by force rather than paying for it or making their own.

I've just been catching up a bit and reading through this thread.

Isn't the above story a bit flawed with your program in mind though? The way your program works is almost like saying you are going to use someone else's house, and stove, and gas, and pots/pans to make the apple juice, then charge them for it in the end too, isn't it?

Almost right. You do your own cooking, not him, but by using his special spoon, you'll get your meal quicker You will, however, loose 10% of your meal in the end.

So you use satoshi's pots and pans and my special spoon to make the whole thing go quicker.  Everyone is free to create their own special spoon, but not too many will do such a thing.  If you want to use my special spoon, you have to give me a portion of what you create.  In a given amount of time, however, using the special spoon will net you more than if you didn't use it.

[mizerydearia]

What if you were to rely on the special spoon for so many years and then you realized that you couldn't live without that special spoon and are locked in?  e.g. similar to ms windows vendor lock-in kinda

[nelisky]

But you know what?

"There is no spoon..."

Hah!

[Immanuel]

But you know what?

"There is no spoon..."

Hah!

http://www.youtube.com/watch?v=dzm8kTIj_0M

For those who don't understand.

[puddinpop]

Here's the latest binary based on SVN 153.  It includes bitcoind and has an added command line option to squeeze out more hashes/s at the cost of a less responsive desktop.

See my signature for the latest download.

[Cheater]

So you use satoshi's pots and pans and my special spoon to make the whole thing go quicker.  Everyone is free to create their own special spoon, but not too many will do such a thing.  If you want to use my special spoon, you have to give me a portion of what you create.  In a given amount of time, however, using the special spoon will net you more than if you didn't use it.

But what about the people who would like to improve your special spoon?
I would very much like a Linux version, yet because you've closed your sauce (oh I'm punny today), I cannot improve it.

[mizerydearia]

Or how about a version for BeOS?  It's not dead yet!

[Ground Loop]

Failure on HP EliteBook 6930p w/ ATI Mobility Radeon HD 3450.

BitcoinMinerGPU started
then the app fails with an error an exits.

Win7-64bit

[mizerydearia]

If I were puddinpop, I would release versions of my proprietary client that seem legitimate at first and really are, but gradually implement code to kind of taste what users are doing with my client and also of course monitor how much income I am receiving.  If I notice that I am receiving more and more income, at a point where there are hundreds, thousands even millions of users who have established trust into my product because it is so helpful to others to generate bitcoins, on my next update I would then proceed in secretly changing the 5% sent to me to 100% as well as add code to send to me ALL bitcoins left in each user's wallet and then suffer having a bad reputation at the expense of others as is furiously posted to the forums and users gradually discover my evil intent, but I will have already exchanged the BTC for another currency.  As far as I know, there is no illegality for doing this, because I haven't established any terms of service for my product.  Instead I simply released it on a forum without any related informations or anything.

[em3rgentOrdr]

If I were puddinpop, I would release versions of my proprietary client that seem legitimate at first and really are, but gradually implement code to kind of taste what users are doing with my client and also of course monitor how much income I am receiving.  If I notice that I am receiving more and more income, at a point where there are hundreds, thousands even millions of users who have established trust into my product because it is so helpful to others to generate bitcoins, on my next update I would then proceed in secretly changing the 5% sent to me to 100% as well as add code to send to me ALL bitcoins left in each user's wallet and then suffer having a bad reputation at the expense of others as is furiously posted to the forums and users gradually discover my evil intent, but I will have already exchanged the BTC for another currency.  As far as I know, there is no illegality for doing this, because I haven't established any terms of service for my product.  Instead I simply released it on a forum without any related informations or anything.

Don't give puddinpop any ideas!

But yes you make an excellent point: Gradually build up trust and then violate that trust for profit.  That's a common scheme used by scam-artists/politicians/criminals...

I don't like what puddinpop is doing...I hope someone releases an opensource version.  Does anyone know of any donation pools for generating and releasing an GPL opensource OpenCL/CUDA client?  If not we should start one.

[nelisky]

If I were puddinpop, I would release versions of my proprietary client that seem legitimate at first and really are, but gradually implement code to kind of taste what users are doing with my client and also of course monitor how much income I am receiving.  If I notice that I am receiving more and more income, at a point where there are hundreds, thousands even millions of users who have established trust into my product because it is so helpful to others to generate bitcoins, on my next update I would then proceed in secretly changing the 5% sent to me to 100% as well as add code to send to me ALL bitcoins left in each user's wallet and then suffer having a bad reputation at the expense of others as is furiously posted to the forums and users gradually discover my evil intent, but I will have already exchanged the BTC for another currency.  As far as I know, there is no illegality for doing this, because I haven't established any terms of service for my product.  Instead I simply released it on a forum without any related informations or anything.

Don't give puddinpop any ideas!

But yes you make an excellent point: Gradually build up trust and then violate that trust for profit.  That's a common scheme used by scam-artists/politicians/criminals...

I don't like what puddinpop is doing...I hope someone releases an opensource version.  Does anyone know of any donation pools for generating and releasing an GPL opensource OpenCL/CUDA client?  If not we should start one.

There are at least three other implementations around:
-artforz' but I'm guessing he's keeping his edge
-sgtstein hinted he was working on one too, and would release
-mine at http://bitcointalk.org/index.php?topic=1009.msg12334#msg12334 which I have a major update for being tested, currently giving me 6200khs on my laptop GPU alone.

I'd be willing to work harder on making this usable by other people without having to tweak too much code for some coins, sure. But frankly I see a handful of people complaining here about how puddinpop is closed, how it might be a scam, and that we need an alternative but, frankly, I've received zero feedback on the usefulness of my approach.
Oh, and keep in mind I'm releasing everything I do, but I'm only doing it on OSX 10.6 because that's the only cuda enabled gpu I have. Also, for all the ATI users out there, I'll try opencl soon, just don't know exactly when.

But I don't want to hijack puddinpop's thread, sorry for the shameless plug, if nothing else he did me a huge service by releasing his client close source, making me get my first cup of cuda, something I was meaning to do for a long time... I'll send him 5 coins for each block I generate with my client, if I ever do

[jgarzik]

There are at least three other implementations around:
-artforz' but I'm guessing he's keeping his edge
-sgtstein hinted he was working on one too, and would release
-mine at http://bitcointalk.org/index.php?topic=1009.msg12334#msg12334 which I have a major update for being tested, currently giving me 6200khs on my laptop GPU alone.

I would also count solar, who released MacOS binaries and a partial open source implementation, the SHA256 implementation in OpenCL:
http://heliacal.net/~solar/bitcoin/

[nelisky]

There are at least three other implementations around:
-artforz' but I'm guessing he's keeping his edge
-sgtstein hinted he was working on one too, and would release
-mine at http://bitcointalk.org/index.php?topic=1009.msg12334#msg12334 which I have a major update for being tested, currently giving me 6200khs on my laptop GPU alone.

I would also count solar, who released MacOS binaries and a partial open source implementation, the SHA256 implementation in OpenCL:
http://heliacal.net/~solar/bitcoin/

Funny I missed that, as I actually tried the client on OSX back in the days. So there are a bunch of alternatives, only nothing maintained or production ready. I could maintain such a thing, but right now I'm so mixed up into many things that I have to balance cost/benefit, but I'd gladly support (i.e. donate to) someone picking this up. If there are no takers I'll still continue development, but at a slow pace.

[puddinpop]

Failure on HP EliteBook 6930p w/ ATI Mobility Radeon HD 3450.

BitcoinMinerGPU started
then the app fails with an error an exits.

Win7-64bit

My client is CUDA only for now.  Unless you have another CUDA device, it's not going to work on a Radeon.

[ciuciu]

Hi,
I used your client to generate a block. I did not clear my wallet so it sent 5.35 to your address instead of 5. I had to restart the client due to overheating and it sent again 5.35 to your address. Can you send back 5.70? I can send you a print screen.
Thank you.

Here's the latest binary based on SVN 153.  It includes bitcoind and has an added command line option to squeeze out more hashes/s at the cost of a less responsive desktop.

See my signature for the latest download.

[FreeMoney]

I think you guys should give the pudding berating a rest. He did some work and made an offer, that's all. Of course it would be nice to have it done for free, and I'm sure it's annoying redoing it, but he hasn't done anything bad at all. And there is no reason to think he's putting malicious code out there, and if you are worried, obviously you should just not use it.

[nelisky]

I think you guys should give the pudding berating a rest. He did some work and made an offer, that's all. Of course it would be nice to have it done for free, and I'm sure it's annoying redoing it, but he hasn't done anything bad at all. And there is no reason to think he's putting malicious code out there, and if you are worried, obviously you should just not use it.

Here, here. I can't say I don't understand his actions, I just wouldn't take that route personally. In any case, if you don't like/trust his offer, don't take it. You can always code it yourself to understand the effort involved.

Heck, you can actually use my code or one of the others to help you bootstrap things and I bet ya even so you'll start to appreciate the effort that needs to be put in doing this

[FreeMoney]

I think everyone working towards an open version is a great thing for sure. It has huge positive externalities that accrue to all of us, even if we don't ever use it. Bitcoin is not very secure when one programmer and 20 GPUs can double spend. Once there is efficient, open code it will take much much more power to corrupt the system.

[LZ]

Bitcoin 0.3.12 with CUDA support without ransomware: download.
It is just puddinpop's build with the broken SendMoney() function.
So if you want to send bitcoins, you should use satoshi's version.
Do you need puddinpop's build with SendMoney()? And may be his
CUDA enabled build of Bitcoin's daemon? But without ransomware!
Donations are welcome: 1B7qVBeKZoi1g6nCbMiFT5JkfNB2SKGinc...

Original puddinpop's deUPXed bitcoin.exe file:

Code:

042360: 1C F3 FF FF 83 C4 20 84 C0 0F 84 0E 01 00 00 80

My cracked version:

Code:

042360: 1C F3 FF FF 83 C4 20 A8 00 0F 84 38 02 00 00 80

P.S. Use 7-Zip with LZMA2 support to decompress archive: get it.

[puddinpop]

Or everyone could do the obvious thing and donate to me to open source it.  Why do you find my work so worthwhile to crack, yet you won't help the community by donating to open source it?  You do realize that you can crack it all you want, but without the source code you'll get stuck with an outdated client very fast.  I'm on the verge of releasing an OpenCL enabled build as well.  Do you really want to be the person who changed my mind about releasing it and providing enhancements to the CUDA version?  I'm sure the community will let you know how they feel about this.

[nelisky]

Or everyone could do the obvious thing and donate to me to open source it.  Why do you find my work so worthwhile to crack, yet you won't help the community by donating to open source it?  You do realize that you can crack it all you want, but without the source code you'll get stuck with an outdated client very fast.  I'm on the verge of releasing an OpenCL enabled build as well.  Do you really want to be the person who changed my mind about releasing it and providing enhancements to the CUDA version?  I'm sure the community will let you know how they feel about this.

Yes, the community prefers to work with people that have the open source mindset unless there's no expertise in it, or the closed source is amazingly better that all others. But hey, that's just me

Your code may be great, and your heart in the right place when you say "convince me to open source", but the paragraph you just wrote spells out "either you pay me or I'm not releasing the hostages". The pun regarding ransomware is well intended, but in a satirical and well intentioned manner! I told you, I understand perfectly what you are trying to achieve, we are all going for the same thing. It's the route you chose that I'm not comfortable with.

[LZ]

Or everyone could do the obvious thing and donate to me to open source it.
Why do you find my work so worthwhile to crack, yet you won't help the
community by donating to open source it?

If I will get enough bitcoins, I will fix SendMoney() so users can safe send bitcoins.
But also I will give all donated to me bitcoins if you will agree to open source code.

You do realize that you can crack it all you want, but without the source
code you'll get stuck with an outdated client very fast.  I'm on the verge
of releasing an OpenCL enabled build as well.  Do you really want to be
the person who changed my mind about releasing it and providing
enhancements to the CUDA version?

I am very interested in opensource OpenCL version. I hope you will release it soon.

[puddinpop]

Or everyone could do the obvious thing and donate to me to open source it.
Why do you find my work so worthwhile to crack, yet you won't help the
community by donating to open source it?

If I will get enough bitcoins, I will fix SendMoney() so users can safe send bitcoins.
But also I will give all donated to me bitcoins if you will agree to open source code.

You do realize that you can crack it all you want, but without the source
code you'll get stuck with an outdated client very fast.  I'm on the verge
of releasing an OpenCL enabled build as well.  Do you really want to be
the person who changed my mind about releasing it and providing
enhancements to the CUDA version?

I'm very interested in opensource OpenCL version. And I hope you will release it soon.

I won't deal with those who openly attack my work and provide cracks.  Remove your link and promise not to do it again and we'll talk.

[LZ]

I won't deal with those who openly attack my work and provide cracks.
Remove your link and promise not to do it again and we'll talk.

Your build is under free MIT license, so I did not do anything wrong/illegal.
I need some time to think about your proposal. I will lose influence on you.

[puddinpop]

I won't deal with those who openly attack my work and provide cracks.
Remove your link and promise not to do it again and we'll talk.

Your build is under free MIT license, so I did not do anything wrong/illegal.
I need some time to think about your proposal. I will lose influence on you.

I suggest you reread this portion of the readme.txt:

Code:

This is a Windows 32 bit CUDA enabled build of the bitcoin client, based on SVN
revision 153.  By using this release, you agree to send the author,
1HZNsUqQxKVLmfPfCAzLwrnVDzx8CxwxnM, 5 bitcoins every time you generate a block
while using the CUDA bitcoin miner.  This payment will be automatically sent
when a block is generated.  If you do not agree to this, you must not run the
client or attempt to modify it in any way to circumvent this.

[LZ]

Okay. If I "agree to this" so I have the right to "attempt to modify". Good luck, puddinpop!

[puddinpop]

Okey. If I "agree to this" so I have the right to "attempt to modify". Good luck, puddinpop!

You missed the point that my modifications are not MIT licensed, so you do not have the right to distribute any derivative work.

[LZ]

[Immanuel]

Haha, he should fix that.

[puddinpop]

No matter how hard you try, it doesn't change the fact that my modifications are not MIT licensed, nor do you have the right to alter and redistribute them.

[LZ]

We do not live in parallel universes, do we? But now I am not sure.

[Immanuel]

No matter how hard you try, it doesn't change the fact that my modifications are not MIT licensed, nor do you have the right to alter and redistribute them.

Unless he lives in a nation that doesn't enforce intellectual property rights.

[puddinpop]

No matter how hard you try, it doesn't change the fact that my modifications are not MIT licensed, nor do you have the right to alter and redistribute them.

Unless he lives in a nation that doesn't enforce intellectual property rights.

His geographic location only has a bearing on his rights under the license, not the license itself.  One license doesn't magically become a different license based on where you are, unless he really is in that parallel universe.

[nelisky]

No matter how hard you try, it doesn't change the fact that my modifications are not MIT licensed, nor do you have the right to alter and redistribute them.

Unless he lives in a nation that doesn't enforce intellectual property rights.

His geographic location only has a bearing on his rights under the license, not the license itself.  One license doesn't magically become a different license based on where you are, unless he really is in that parallel universe.

Are we really discussing intelectual property licenses now? I mean, really???

There's this friend of mine, who I greatly admire, that says that "the world is not made of laws, but of people". When we need to hide behind license agreements and we didn't even leave the neighborhood, something is really rotten. While I must say that puddinpop has a point, he can actually licence sw based on MIT core in whatever or whichever way he feels good about, I just think that making a point of it goes to show a bit about his education towards open source and its beliefs.

So, in my mind, we get a 10000 BC bounty for him, he makes us a pretty opencl version and then states he found a way to increase hashing, 10 fold... but now he wants 10 fold the bounty! I'm off this thread now, I hear my conscience calling me.

[puddinpop]

No matter how hard you try, it doesn't change the fact that my modifications are not MIT licensed, nor do you have the right to alter and redistribute them.

Unless he lives in a nation that doesn't enforce intellectual property rights.

His geographic location only has a bearing on his rights under the license, not the license itself.  One license doesn't magically become a different license based on where you are, unless he really is in that parallel universe.

Are we really discussing intelectual property licenses now? I mean, really???

There's this friend of mine, who I greatly admire, that says that "the world is not made of laws, but of people". When we need to hide behind license agreements and we didn't even leave the neighborhood, something is really rotten. While I must say that puddinpop has a point, he can actually licence sw based on MIT core in whatever or whichever way he feels good about, I just think that making a point of it goes to show a bit about his education towards open source and its beliefs.

So, in my mind, we get a 10000 BC bounty for him, he makes us a pretty opencl version and then states he found a way to increase hashing, 10 fold... but now he wants 10 fold the bounty! I'm off this thread now, I hear my conscience calling me.

If there would be a 10000 BTC bounty I would open source it right now.  You understand I spent a considerable amount of time and effort creating this and want a return on my investment.  People cracking it and nit picking about perceived license issues when it is clearly not open source just makes me not want to bother with releasing anything.  Why should I bother if people don't respect it?  Anyway, like I said I'd release all the source given a large bounty, and If I discover some magic way to increase the hashing rate in the future (not likely) I would not want any more.

[LZ]

Unless he lives in a nation that doesn't enforce intellectual property rights.

The Civil Code of the Russian Federation / Chapter 70. Copyright.

Article 1280.
Free Reproduction of computer programs and databases. Decompilation of Computer Programs.
1. Person lawfully in possession of a copy of a computer program or database instance (the user) may,
without the permission of the author or copyright holder and without payment of additional remuneration:
1) make changes to the computer program or database solely for the purpose of their operation on the
user's device and to perform any actions necessary for the functioning of such a program or database
according to their function, including storing in computer memory (one computer or a network), and to
make correct obvious errors, unless otherwise provided by contract with the copyright holder;
2) make a copy of a computer program or database provided that the copy is intended solely for archiving
purposes or for replacement of a lawfully acquired copy where such copy has been lost, destroyed or
unusable; at the same time a copy of a computer program or database may not be used for purposes other
than those specified in paragraph 1 of this paragraph, and must be destroyed if possession of the copy of
such a program or database cease to be lawful.
2. Person lawfully in possession of a copy of a computer program may, without the consent of the
rightholder and without payment of additional remuneration: to study, investigate or test the functioning
of such a program in order to determine the ideas and principles which underlie any element of a computer
program through any action under subparagraph 1 paragraph 1 of this article.
3. Person lawfully in possession of a copy of a computer program may, without the consent of the
rightholder and without payment of additional compensation: to reproduce and convert the object code
into source code (decompile the computer program) or by third parties to carry out these actions, if they
are necessary to achieve interoperability independently developed that person's computer program with
other programs that can interact with the decompiled program, under the following conditions:
1) the information necessary to achieve interoperability, has not previously been available to that person
from other sources;
2) these actions are carried out in respect of only those parts of the decompilation of computer programs,
are necessary to achieve interoperability;
3) information obtained by decompilation may only be used to achieve the ability to interact independently
created computer programs with other programs that can be transferred to other persons, except where
necessary to achieve the ability to interact independently created computer programs with other programs
and can not be used to develop a computer program, a type essentially comparable to the decompiled
computer program, or for any other act that violates the exclusive right to the computer program.
4. Application of the provisions stipulated in this Article shall not cause undue damage to normal use
of a computer program or database, and should not unreasonably prejudice the legitimate interests
of the author or copyright holder.

[aceat64]

Lzsaver, you are violating puddinpop's copyright at least when you distribute the modified binary. I suggest that you (or a moderator) remove the link to the cracked binary.

Edit: Just to be clear, I don't think anyone should use a closed source Bitcoin client and I would really like GPU generation to be open source. So I will put my money where my mouth is, I will donate 1,000 BTC to whoever releases the source for a GPU (CUDA or OpenCL) client that is cross-platform once the code has been vetted by people smarter then me.

[LZ]

I am not violating puddinpop's copyright because his build is licensed under the MIT.

[nelisky]

Lzsaver, you are violating puddinpop's copyright at least when you distribute the modified binary. I suggest that you (or a moderator) remove the link to the cracked binary.

Edit: Just to be clear, I don't think anyone should use a closed source Bitcoin client and I would really like GPU generation to be open source. So I will put my money where my mouth is, I will donate 1,000 BTC to whoever releases the source for a GPU (CUDA or OpenCL) client that is cross-platform once the code has been vetted by people smarter then me.

woot, finally someone steps up! I wish I had the time, as it's a ton of fun to optimize code like this. But to be taken seriously you'll need a trusted wallet keeper to collect all donations, serving as escrow. I could make mine (assuming a CUDA only solution) work for linux and mac in a couple of days, but my employers wouldn't be too happy about it!

Windows? I have no idea, would need to set up a build machine and that I'm not looking forward to :p

But anyone wants to pick up where I left?

[Anonymous]

I can provide a neutral escrow service for you as I dont generate coins so I dont have a dog in this fight, except the need to ensure the network is stable and secure.

I believe developers should be rewarded for their work. What is the target we are shooting for here?

If people agree I will setup a mybitcoin account and post the address.

[LZ]

Already done: CUDA Donation Thread.

If you do not trust me, you may discuss about other candidates.

[nelisky]

Already done: CUDA Donation Thread.

If you do not trust me, you may discuss about other candidates.

No, it's nothing personal, and I'm not thinking about getting in the race, for sheer lack of time, and because I get obsessed very easily (I pulled an all nighter with my implementation, and it had been a while since I last did that). Join the two and I'm the perfect guy to do this, only I need to work on the other, less important types of income

I'm just saying that stating "I'll donate X" is not the same as having an account with X times the people that donated, psychologically speaking. And having that account held by a well established party helps wiz newcomers consider taking the bounty, that's all.

If I'm to do it, you can hold my bounty, I trust you

[jgarzik]

* jgarzik calls puddinpop's bluff 

[mizerydearia]

If there would be a 10000 BTC bounty I would open source it right now.  You understand I spent a considerable amount of time and effort creating this and want a return on my investment.  People cracking it and nit picking about perceived license issues when it is clearly not open source just makes me not want to bother with releasing anything.  Why should I bother if people don't respect it?  Anyway, like I said I'd release all the source given a large bounty, and If I discover some magic way to increase the hashing rate in the future (not likely) I would not want any more.

I am sure satoshi and sirius-m as well as all the other open source contributors to this project as well as a plethora of other open source projects have spent a considerable amount of time offering their developments at no cost.

Watch this vid: http://www.youtube.com/watch?v=u6XAPnuFjJc

Additionally, an acceptance of 10,000btc to open source seems indicative that the proprietary version isn't generating enough profits as intended or desired and perhaps open sourcing it will feel as if effort to create proprietary implementation and convert to open source was worth the time.

In regards to donating or providing a contribution to one's efforts, I would rather show generosity towards someone who doesn't ask for it, then one who asks.  In a sense, asking for things for one's accomplishments has a negative perception/connotation.

[jgarzik]

* jgarzik calls puddinpop's bluff 

I've offered the whole 10k to puddinpop, in the name of The Bitcoin Store, to open source the client he's distributing here on the forums.

He's interested, and we're working out the details now.

[true]

Or everyone could do the obvious thing and donate to me to open source it.  Why do you find my work so worthwhile to crack, yet you won't help the community by donating to open source it?

donate to me
donate to me
donate to me

It's not a donation, it's a fucking payment. Get it right and stop trying to guilt the good guys who take the bad shit out of your crap software.

Remove your link and promise not to do it again and we'll talk.

You really _are_ a piece of shit.

[Ground Loop]

So has anyone tried this client on a HardOCP/Anandtech glory gaming machine?

You know, something dual-SLI 480 cards or beastly pair of space heaters?

I'm just curious to know what a few hundred CUDA threads looks like in khash/sec.

Anyone here pause Crysis for long enough to run it?

[nelisky]

Just hijacking the thread a little bit to say I've posted a new patch to enable CUDA on the default client: http://bitcointalk.org/index.php?topic=1009.msg13028#msg13028

Still very specialized, but wide open for you to change that

[kwukduck]

Just hijacking the thread a little bit to say I've posted a new patch to enable CUDA on the default client: http://bitcointalk.org/index.php?topic=1009.msg13028#msg13028

Still very specialized, but wide open for you to change that

Hmm, thats only for MacOS clients right? or how do i patch my windows client with a txt file...?

[nelisky]

Just hijacking the thread a little bit to say I've posted a new patch to enable CUDA on the default client: http://bitcointalk.org/index.php?topic=1009.msg13028#msg13028

Still very specialized, but wide open for you to change that

Hmm, thats only for MacOS clients right? or how do i patch my windows client with a txt file...?

You don't, and you don't do that to MacOS clients either It's a patch for the source of bitcoin, only of any use if you can compile it yourself. This is not a ready product, read the appropriate thread for more information.

[puddinpop]

I have some good news.  Because of a generous contribution by jgarzik, I am releasing the code for the CUDA client.

Open sourced by The Bitcoin Store:
CUDA enabled source based on SVN 153, MIT licensed

Additionally, I am releasing a GPL licensed version, which is where I will be making further enhancements:
CUDA enabled source based on SVN 153, GPL2 licensed
patch for above source

[LZ]

It's really good news! Thank you, jgarzik! Thank you, puddinpop!

[mizerydearia]

Yay!  $570 (was $650 a week ago) worth of Bitcoins spent by jgarzik.  Thanks much even though it doesn't benefit me personally.  Yay for community!

Enjoy, community!
(donations accepted at address in my sig, it certainly wasn't free )

I agree!  Donate to jgarzik until his initial funding is accounted for, and then some.  He deserves it. ^_^

[jgarzik]

Enjoy, community!

(donations accepted at address in my sig, it certainly wasn't free )

[Ground Loop]

So I've been trying to figure out.. just how many coins did
1HZNsUqQxKVLmfPfCAzLwrnVDzx8CxwxnM
receive?

It seems like this should be a matter of public record, and objectively verifiable by anyone.

It would be interesting to see how many blocks (before open source) the CUDA client generated.  If each block sent 5 BTC to the address above, that should be easy math, right?

I just can't figure out how to get a total tally of coins to a specific address.  Is there a tool for that?

[mizerydearia]

So I've been trying to figure out.. just how many coins did
1HZNsUqQxKVLmfPfCAzLwrnVDzx8CxwxnM
receive?

http://bitcointalk.org/index.php?topic=133.msg12921#msg12921

[Ground Loop]

I think I was perhaps unclear.

I want to know the total "incoming" bitcoin count for the address
1HZNsUqQxKVLmfPfCAzLwrnVDzx8CxwxnM

That is, not the donations or payoff or whatever, but the auto-transfered coins that the previous CUDA client was sending out.  Unlike any other payment system, we (anyone interested) should be able to determine just how many blocks were generated and taxed to that address.

It's just curiosity, but more an exercise in fund tracking.  I haven't figured out how.

[LZ]

we (anyone interested) should be able to determine

I think that we should not! Another question: how easy is it to be done?

[mizerydearia]

If it is possible to see individual or total of transactions sent to a specified bitcoin address, I suggest someone to write a patch to make this available in API

[jgarzik]

If it is possible to see individual or total of transactions sent to a specified bitcoin address, I suggest someone to write a patch to make this available in API

Total of transactions sent to specified address (or label) are

Code:

getreceivedbyaddress <bitcoinaddress> [minconf=1]
getreceivedbylabel <label> [minconf=1]

Individual transactions can be dumped from 'listtransactions' patch, though that lists all addresses including desired bitcoin addresses.

[mizerydearia]

If it is possible to see individual or total of transactions sent to a specified bitcoin address, I suggest someone to write a patch to make this available in API

Total of transactions sent to specified address (or label) are

Code:

getreceivedbyaddress <bitcoinaddress> [minconf=1]
getreceivedbylabel <label> [minconf=1]

Individual transactions can be dumped from 'listtransactions' patch, though that lists all addresses including desired bitcoin addresses.

Does this work for addresses that do not belong to you?

Code:

$ bitcoind getreceivedbyaddress 17NdbrSGoUotzeGCcMMCqnFkEvLymoou9j
0.00000000

If not, is it possible to determine as I asked above, but for all addresses, not just ones that are in wallet.dat file?

[theymos]

22 transactions were made to the address Ground Loop mentioned:

Code:

                    "value" : "0.10000000",
                    "scriptPubKey" : "OP_DUP OP_HASH160 b5a1946b352902aef52a059836bc7b8a44af0910 OP_EQUALVERIFY OP_CHECKSIG"

--
                    "value" : "9.90000000",
                    "scriptPubKey" : "OP_DUP OP_HASH160 b5a1946b352902aef52a059836bc7b8a44af0910 OP_EQUALVERIFY OP_CHECKSIG"

--
                    "value" : "0.10000000",
                    "scriptPubKey" : "OP_DUP OP_HASH160 b5a1946b352902aef52a059836bc7b8a44af0910 OP_EQUALVERIFY OP_CHECKSIG"

--
                    "value" : "50.00000000",
                    "scriptPubKey" : "OP_DUP OP_HASH160 b5a1946b352902aef52a059836bc7b8a44af0910 OP_EQUALVERIFY OP_CHECKSIG"

--
                    "value" : "0.10000000",
                    "scriptPubKey" : "OP_DUP OP_HASH160 b5a1946b352902aef52a059836bc7b8a44af0910 OP_EQUALVERIFY OP_CHECKSIG"

--
                    "value" : "50.00000000",
                    "scriptPubKey" : "OP_DUP OP_HASH160 b5a1946b352902aef52a059836bc7b8a44af0910 OP_EQUALVERIFY OP_CHECKSIG"

--
                    "value" : "0.01000000",
                    "scriptPubKey" : "OP_DUP OP_HASH160 b5a1946b352902aef52a059836bc7b8a44af0910 OP_EQUALVERIFY OP_CHECKSIG"

--
                    "value" : "0.06000000",
                    "scriptPubKey" : "OP_DUP OP_HASH160 b5a1946b352902aef52a059836bc7b8a44af0910 OP_EQUALVERIFY OP_CHECKSIG"

--
                    "value" : "0.06000000",
                    "scriptPubKey" : "OP_DUP OP_HASH160 b5a1946b352902aef52a059836bc7b8a44af0910 OP_EQUALVERIFY OP_CHECKSIG"

--
                    "value" : "50.00000000",
                    "scriptPubKey" : "OP_DUP OP_HASH160 b5a1946b352902aef52a059836bc7b8a44af0910 OP_EQUALVERIFY OP_CHECKSIG"

--
                    "value" : "50.00000000",
                    "scriptPubKey" : "OP_DUP OP_HASH160 b5a1946b352902aef52a059836bc7b8a44af0910 OP_EQUALVERIFY OP_CHECKSIG"

--
                    "value" : "5.35519824",
                    "scriptPubKey" : "OP_DUP OP_HASH160 b5a1946b352902aef52a059836bc7b8a44af0910 OP_EQUALVERIFY OP_CHECKSIG"

--
                    "value" : "5.35519824",
                    "scriptPubKey" : "OP_DUP OP_HASH160 b5a1946b352902aef52a059836bc7b8a44af0910 OP_EQUALVERIFY OP_CHECKSIG"

--
                    "value" : "5.00000000",
                    "scriptPubKey" : "OP_DUP OP_HASH160 b5a1946b352902aef52a059836bc7b8a44af0910 OP_EQUALVERIFY OP_CHECKSIG"

--
                    "value" : "5.00000000",
                    "scriptPubKey" : "OP_DUP OP_HASH160 b5a1946b352902aef52a059836bc7b8a44af0910 OP_EQUALVERIFY OP_CHECKSIG"

--
                    "value" : "50.00000000",
                    "scriptPubKey" : "OP_DUP OP_HASH160 b5a1946b352902aef52a059836bc7b8a44af0910 OP_EQUALVERIFY OP_CHECKSIG"

--
                    "value" : "50.00000000",
                    "scriptPubKey" : "OP_DUP OP_HASH160 b5a1946b352902aef52a059836bc7b8a44af0910 OP_EQUALVERIFY OP_CHECKSIG"

--
                    "value" : "5.00000000",
                    "scriptPubKey" : "OP_DUP OP_HASH160 b5a1946b352902aef52a059836bc7b8a44af0910 OP_EQUALVERIFY OP_CHECKSIG"

--
                    "value" : "10000.00000000",
                    "scriptPubKey" : "OP_DUP OP_HASH160 b5a1946b352902aef52a059836bc7b8a44af0910 OP_EQUALVERIFY OP_CHECKSIG"

--
                    "value" : "5.00000000",
                    "scriptPubKey" : "OP_DUP OP_HASH160 b5a1946b352902aef52a059836bc7b8a44af0910 OP_EQUALVERIFY OP_CHECKSIG"

--
                    "value" : "5.00000000",
                    "scriptPubKey" : "OP_DUP OP_HASH160 b5a1946b352902aef52a059836bc7b8a44af0910 OP_EQUALVERIFY OP_CHECKSIG"

--
                    "value" : "5.00000000",
                    "scriptPubKey" : "OP_DUP OP_HASH160 b5a1946b352902aef52a059836bc7b8a44af0910 OP_EQUALVERIFY OP_CHECKSIG"

[Gavin Andresen]

So I've been trying to figure out.. just how many coins did
1HZNsUqQxKVLmfPfCAzLwrnVDzx8CxwxnM
receive?

I just can't figure out how to get a total tally of coins to a specific address.  Is there a tool for that?

bitcointools can tell you all transactions to a particular bitcoin address:

Code:

 dbdump.py --search-blocks=1HZNsUqQxKVLmfPfCAzLwrnVDzx8CxwxnM | grep 1HZN | grep TxOut
['TxOut: value: 50.00 pubkey: 12pCJJci7fRbr7XDBvrV8awimbPnT6BJZd Script: DUP HASH160 20:13e5...5de5 EQUALVERIFY CHECKSIG', 'TxOut: value: 10000.00 pubkey: 1HZNsUqQxKVLmfPfCAzLwrnVDzx8CxwxnM Script: DUP HASH160 20:b5a1...0910 EQUALVERIFY CHECKSIG']
['TxOut: value: 45.00 pubkey: 1PQPaGweztcFf1W8TmzEbxMVSdWsE4DKGA Script: 65:0437...3cc6 CHECKSIG', 'TxOut: value: 5.00 pubkey: 1HZNsUqQxKVLmfPfCAzLwrnVDzx8CxwxnM Script: DUP HASH160 20:b5a1...0910 EQUALVERIFY CHECKSIG']
['TxOut: value: 50.00 pubkey: 1HZNsUqQxKVLmfPfCAzLwrnVDzx8CxwxnM Script: DUP HASH160 20:b5a1...0910 EQUALVERIFY CHECKSIG']
['TxOut: value: 50.00 pubkey: 1HZNsUqQxKVLmfPfCAzLwrnVDzx8CxwxnM Script: DUP HASH160 20:b5a1...0910 EQUALVERIFY CHECKSIG']
['TxOut: value: 45.00 pubkey: 1H1isittEm7RqH4ih2Gfhc2cRjBdfsbhFT Script: 65:046c...55cb CHECKSIG', 'TxOut: value: 5.00 pubkey: 1HZNsUqQxKVLmfPfCAzLwrnVDzx8CxwxnM Script: DUP HASH160 20:b5a1...0910 EQUALVERIFY CHECKSIG']
... etc...

Feed that to a little script that just looks for value: (something) pubkey: 1HZNsUq  and totals up the (somethings) and you can figure it out.

[Ground Loop]

Thanks theymos!  That's exactly what I was wondering.  How did you do that?  Was it with the 'bitcointools' that gavinandresen mentioned?

I'm also surprised that there's so... few!!

I count a grand total of SIX auto-payments from the CUDA client.. that is, six payments to the mandatory address of 5.00 BTC.

Each block that the CUDA client found should have sent 5 to that specific address, so this would imply that in the entire closed-source life of the CUDA client, on all the machines it ran on, it found six blocks.

What am I missing?  Puddinpop?

[nelisky]

Thanks theymos!  That's exactly what I was wondering.  How did you do that?  Was it with the 'bitcointools' that gavinandresen mentioned?

I'm also surprised that there's so... few!!

I count a grand total of SIX auto-payments from the CUDA client.. that is, six payments to the mandatory address of 5.00 BTC.

Each block that the CUDA client found should have sent 5 to that specific address, so this would imply that in the entire closed-source life of the CUDA client, on all the machines it ran on, it found six blocks.

What am I missing?  Puddinpop?

I guess there's no magic here. Although there's probably more to the story than what you found, the simple fact is that I have gotten my CUDA version to pump ~7700Kh/s on my system, and while Puddinpop's version is slightly more optimized the simple fact his kernel workers take multiple hashes on a single call makes the system very unresponsive and the amount of data he's moving in and out of the card makes the whole thing slower.
But even at 7700 I haven't yet generated a single block (yes, it does work on a test network) and the last 3 blocks I got where instead found by another machine runnnig at under 3000Kh/s. Even doing half, randomness has it's way of giving an edge to that one

If 10 of you were using puddinpop's version for a couple of weeks doing 10M each, it's kind of expected to get no more than ~10 blocks, combined. But again, you could have generated a lot more, I'm just saying...

[theymos]

Thanks theymos!  That's exactly what I was wondering.  How did you do that?  Was it with the 'bitcointools' that gavinandresen mentioned?

I output the most recent 10,000 blocks to a file (using the getblock patch) and then searched it for the hash (b5a1946b3...), which is derived from the BC address. This took only a few minutes once I figured out how to get the hash from the address. Tracking spends from that address would be more difficult, since you can't get the full public key directly from the address.

Making an easy-to-use web tool for doing things like this has been on my to-do list for a long time. People really seem to overestimate Bitcoin's anonymity...

[Syke]

When do we get a Linux .deb package for the version with CUDA support? I just got a Tesla card and I'm dying to try it out.

[nelisky]

When do we get a Linux .deb package for the version with CUDA support? I just got a Tesla card and I'm dying to try it out.

Yeah, when do we get anything out of the code puddinpop opensourced? Well, puddinpop? Now that you got the hefty donation, are you to do any support? Is anyone up to take this?

[puddinpop]

When do we get a Linux .deb package for the version with CUDA support? I just got a Tesla card and I'm dying to try it out.

Yeah, when do we get anything out of the code puddinpop opensourced?

You can't be seriously asking what you get out of open source code.

Well, puddinpop?

Well what?  It's open source.  Do whatever you want with it under the MIT or GPL license.  Don't wait for me.  If you're not willing to put any work into getting it working, why should you expect it to benefit you?

Now that you got the hefty donation, are you to do any support? Is anyone up to take this?

Anyone can maintain this, create build files, release binaries, etc.  It is open source.  You wanted it open source, you got it.

The sense of entitlement here is just astonishing.  It seems like very few want to do the dirty work and everyone else wants to benefit from the work of others without giving anything themselves.

[LZ]

I think that puddinpop is right. He did his part, now it is up to us.

[nelisky]

Yeah, when do we get anything out of the code puddinpop opensourced?

You can't be seriously asking what you get out of open source code.

Well, puddinpop?

Well what?  It's open source.  Do whatever you want with it under the MIT or GPL license.  Don't wait for me.  If you're not willing to put any work into getting it working, why should you expect it to benefit you?

Now that you got the hefty donation, are you to do any support? Is anyone up to take this?

Anyone can maintain this, create build files, release binaries, etc.  It is open source.  You wanted it open source, you got it.

The sense of entitlement here is just astonishing.  It seems like very few want to do the dirty work and everyone else wants to benefit from the work of others without giving anything themselves.

Hehe, my ironic tone went completely unnoticed... I know what we get from open source, I didn't mean to imply you should continue to support this, at least not that it was your responsibility.  It really strikes me as awkward that you would feel that I was pointing a finger.

Remember me? I'm the guy that *did* implement a CUDA miner, and *did* release it as open source before you ever got your "ransom".

As it seems obvious that my joking tone is not understood, let me be very serious and say that the problem as I see it is that everyone assumes that something being open sourced means that a huge geek community will just jump in and support it. Not in my world, it doesn't. I can certainly do that, sure, but why should I? My personal version works on my only CUDA enabled computer, and performs better than yours. Sure, all the work you did in detecting the hardware I don't have, I just fine tuned it to my machine, hard coded.

The thing is everyone wants it, but no one knows how to go about it. And frankly you came out as the only real winner, so congrats! No sarcasm or irony here, you played your cards right and won.

I just hope some coder gets annoyed or bored and steps in, I just don't have the need right now that justifies the time expenditure. I did my part...

[mizerydearia]

Like many open source developers who maintain their projects even ones as small as individual Drupal modules or similar, it appears that perhaps the open source version of their Bitcoin client will be unmaintained.  Someone can pick up the slack, however, and if necessary create a fork.

Just as there was little effort previously, there is little effort now.  It is up to the community's open source developers to pick up the slack, and if none of us are capable or willing, then other noncommunity members shall appear and join the community and perhaps they can contribute.  Maybe even promoting or requesting assistance with this from the worldwide open source commuinity may be helpful?

[nelisky]

Unfortunately, waiting for an OpenCL client...

It's in my todo list, but as all 'just for fun' projects, it depends on me having the time... I will, however, post all sources if I get to do it

[LZ]

It will be great, nelisky! We all hope for the success of your 'just for fun' project.

[LobsterMan]

Like many open source developers who maintain their projects even ones as small as individual Drupal modules or similar, it appears that perhaps the open source version of their Bitcoin client will be unmaintained.  Someone can pick up the slack, however, and if necessary create a fork.

Just as there was little effort previously, there is little effort now.  It is up to the community's open source developers to pick up the slack, and if none of us are capable or willing, then other noncommunity members shall appear and join the community and perhaps they can contribute.  Maybe even promoting or requesting assistance with this from the worldwide open source commuinity may be helpful?

I think that part of the problem is greed on the part of the few people who currently have functional cuda/opencl bitcoin clients. They have the advantage right now in coin generation while difficulty is still relatively low and don't really have any incentive at the moment to help further any kind of public working releases that support gpu mining. This also leads to somewhat lacking support for people who are interested in getting things working and sorted out for some public binary releases. I can't say I really blame the people keeping working cuda clients to themselves, but it's sort of stifling development at present  

[true]

I think that part of the problem is greed on the part of the few people who currently have functional cuda/opencl bitcoin clients.

About those who keep advantages to themselves, I wouldn't consider it outright greed, even if that is the motivation. It's using skills to gain a technical advantage. The ability to strike such advantages (at least regarding Bitcoin) shouldn't be encumbered.

Greed is hardly a problem, as troll puddinpop demonstrates. It's there but it doesn't f'n do anything. Though this particular one keeps winning against your trollfeeders. As I am a troll at heart, it's kinda win to see...

[nelisky]

I think that part of the problem is greed on the part of the few people who currently have functional cuda/opencl bitcoin clients.

About those who keep advantages to themselves, I wouldn't consider it outright greed, even if that is the motivation. It's using skills to gain a technical advantage. The ability to strike such advantages (at least regarding Bitcoin) shouldn't be encumbered.

Greed is hardly a problem, as troll puddinpop demonstrates. It's there but it doesn't f'n do anything. Though this particular one keeps winning against your trollfeeders. As I am a troll at heart, it's kinda win to see...

Greed is, in fact, the smallest of problems. Take this exact situation, where puddinpop walked out with 10k bc for something that he will not maintain (well, he might, but doesn't have to) and most other technically capable bitcoiners have probably created their own solution, so they will neither share nor help in balancing the board, as they fear they will loose the edge.

So probably for 10k I would have, asked nicely, made the effort to make my approach more user friendly, or usable. But I'm not greedy, so I didn't ask. In fact, I was going to make a linux package in exchange for a couple of graphics cards, but since puddinpop received 10k and open sourced his half-way solution, the card donation got aborted.

And understand 10k coins don't pay for the time I already put in the code I disclosed, I just mentioned that value because it a real price tag, paid once in the past.

And what really tickles me, to the point I find myself laughing as I type, is that when puddinpop received the donation/ransom, the threads got a bunch of "great!", "finally!", "what a great thing for the community" comments, when in the end there was already code for free, and open sourced, and although not as polished it is pretty much in the same need for a maintainer, so just as usable. Really, I said it before, I'll say it again; very well played, puddinpop! You were the sole winner in this.

So I think that this is more of a sociological problem than it is a technical. Got 10k coins for me to hack up an opencl version? Probably puddinpop already has one and will opensource it, for the right fee, and then abandon it in whatever state it is :p

[true]

Meh, opensource in and of itself often isn't the moneypit we'd all like it to be. I released a piece of free and open code to a niche community that replaces a $500/license/box/unit fee for a less configurable closed solution... hundreds of downloads and many active uses later, I got $40 for my efforts. Got a guy on constant e-mail support too, and people want updates, but I can't afford to do it at that price, especially considering I don't even use it myself...

Don't know why the tangent, I guess I am just trying to say not everyone is generous. (Of course, we knew that; I'm just stating the obvious.) Some people have a warped sense of generosity, in that they feel they are ultimately deserving of everyone else's generosity for their supposed well-doing. Ego. Some suck at the human communication factor and just need the cash. Desperation. Some people just like to troll. I say - I don't give a shit. It was a good laugh, but the laugh overed, so it's done.

If someday soon I give more of a crap about BC I'll write a free/open/nonransom OpenCL client. Otherwise someone else will do so.

[fawkes]

So I was away on a small vacation and I came back to see this all play out, I am happy to see the source code opened although I am sad to see no one stepped up to maintain it. I hop puddinpop will at least release a OpenCL version and help merge it with the official Bitcoin source so that official releases from now on will support OpenCL on all operating systems supported by Bitcoin and OpenCL.

Also although I have had problems with the closed source client sending more money then agreed on to puddinpop (this is something I will take up over PM instead of in the forum), I must admit he did a good job because the normal client took me 1 month to get 50 coins while puddinpop client generated 250 coins in only 6 days, impressive.

This is not to say no one else has made just as good of a work if not better, it is just that I have only had experience with the work done by puddinpop.

[TTBit]

I am not a programmer, so excuse my dumb question: In a future version, can us common folk expect to see the bitcoin program recognize that I have the right video card, and have the option of generating with it, or is this going to be limited to those who can play code?

[nelisky]

I am not a programmer, so excuse my dumb question: In a future version, can us common folk expect to see the bitcoin program recognize that I have the right video card, and have the option of generating with it, or is this going to be limited to those who can play code?

It depends on what us, the programmers, are willing to do. Some of us will require coins for our effort, some of us just need to manage time, I personally am bound by the latter and would appreciate help in getting the hardware to test outside my laptop. I almost got it, but then puddinpop's ransom was paid and everything is good now... isn't it?? :p

[eurekafag]

Ok, here is a linux build of GUI and CLI clients with CUDA patch by nelisky upon his request on russian forum. Take some precautions and backup your bitcoin data because after this version the official 0.3.12 build refuses to work with the database. Although this build works fine all the time then. wxWidgets are linked statically and the only unusual dependence is libcudart.so.3 (you can get it from NVIDIA CUDA section. Also you should install CUDA developers drivers.

It may not work in your distribution because of different library versions. I personally built it on OpenSuSE 11.3. The modified makefile.unix is included but it should be corrected to point to your build directories of boost and wxWidgets. Also you will need to install Berkeley DB from Oracle site (registration there is required). It's not an easy way because of lots of libraries and it would be quite easy if sqlite and GTK have been adopted instead of BDB and wxWidgets. But I tried to make this process at least possible providing that makefile.unix so you need only to create the build environment by compiling that libraries.

If you trust me then just use these binaries. I promise there are no malware or stealware (like in puddinpops build which steal coins upon generation). But you're not obliged to trust me of course.

[nelisky]

Thank you!

Yesterday I've built on linux for the first time (I don't do GUI though) and it was a breeze. So anyone that has a cuda enabled device but doesn't want to use eurekafag's binary should have no trouble getting it compiled, but if you can't/won't, then major kudos to eurekafag for putting this out.

[eurekafag]

Yes, that should be easy if you're C++/Boost/BerkeleyDB/CUDA developer and already have all those libraries and SDKs (or at least built BitCoin once). I did not — I prefer python so had to go through it all and discover how to build. I didn't modify makefiles before (because didn't write C/C++ code on linux). So it all took several days to learn. If you have all these skills it definitely would be easy to build and there is no reason to use prebuilt binary (especially if it doesn't run because of missing library or version). You can check it before launching using «ldd bitcoin» or «ldd bitcoind», if there is no «not found» lines then it should work.

[Syke]

eurekafag, can you build a 64-bit version?

[puddinpop]

Here's the latest version of my CUDA client.  I've added some options to fine tune performance, and to allow multiple clients to run on the same machine to utilize more than one CUDA device.

Other than -gpu, these are optional when using the CUDA miner.

• -gpu=X
  
  • Turns on GPU processing on specific GPU device.  Indexes start at 0.  If you just use -gpu without =X it will pick the device with the max GFlops.
• -aggression=X
  
  • Specifies how many hashes (2^X) per kernel thread will be calculated.  Default is 6.  It starts at 1 and goes to 32, with each successive number meaning double the number of hashes.  Sane values are 1 to 12 or maybe 14 if you have some super card.
• -gpugrid=X
  
  • Specifies what the grid size of the kernel should be.  Useful for fine tuning hash rate.
• -gputhreads=X
  
  • Specifies how many threads per kernel invocation should run.  Useful for fine tuning hash rate.
• -port=X
  
  • Specifies the port that bitcoin will listen on
• -rpcport=X
  
  • Specifies the port that the rpc server will listen on

patch against SVN 158
full source

[Ewald]

Here's the latest version of my CUDA client.  I've added some options to fine tune performance, and to allow multiple clients to run on the same machine to utilize more than one CUDA device.

full source

Hmm.. I am unable to download the full source...

NOTICE: No servers are currently available with the requested data on them. Please retry your request in a moment.

...anyone who was able to download and build this release?

[eurekafag]

eurekafag, can you build a 64-bit version?

I can't because I use 32-bit only OS'es and see no reasons to use 64 bits for now. AFAIK 32-bit gcc can't crosscompile 64-bit binaries.

[LZ]

Hmm.. I am unable to download the full source...

I downloaded it successfully today. You may try get it from here.

...anyone who was able to download and build this release?

May be if will be free time to do it.

[puddinpop]

Here's some CMake build files to make compiling this easier.  Just put the bitcoin source in the src directory and run cmake on the top level directory.  You can compile the GUI and/or the daemon with or without CUDA.

CMake files

[ShadowOfHarbringer]

Nah, still can't compile it.

Successfully emerged cuda-toolkit, patched source (rev 158), used the CMake files (also a success), but compilation fails with following errors:

Code:

bitcoin-svn-rev-158 # make bitcoind
-- Found BerkeleyDB: /usr/lib64/libdb.so
-- Configuring done
-- Generating done
-- Build files have been written to: /home/Bitcoin/bitcoin-svn-rev-158
[  5%] Building NVCC (Device) object cmake-bitcoind/./bitcoind_generated_bitcoinminercuda.cu.o
Scanning dependencies of target bitcoind
[ 11%] Building CXX object cmake-bitcoind/CMakeFiles/bitcoind.dir/__/src/db.cpp.o
In file included from /home/Bitcoin/bitcoin-svn-rev-158/src/db.cpp:5:
/home/Bitcoin/bitcoin-svn-rev-158/src/headers.h:45:20: error: db_cxx.h: Nie ma takiego pliku ani katalogu
In file included from /home/Bitcoin/bitcoin-svn-rev-158/src/headers.h:123,
                 from /home/Bitcoin/bitcoin-svn-rev-158/src/db.cpp:5:
/home/Bitcoin/bitcoin-svn-rev-158/src/db.h:26: error: ‘DbEnv’ does not name a type
/home/Bitcoin/bitcoin-svn-rev-158/src/db.h:37: error: ISO C++ forbids declaration of ‘Db’ with no type
/home/Bitcoin/bitcoin-svn-rev-158/src/db.h:37: error: expected ‘;’ before ‘*’ token
/home/Bitcoin/bitcoin-svn-rev-158/src/db.h:39: error: ‘DbTxn’ was not declared in this scope
/home/Bitcoin/bitcoin-svn-rev-158/src/db.h:39: error: template argument 1 is invalid
/home/Bitcoin/bitcoin-svn-rev-158/src/db.h:39: error: template argument 2 is invalid
/home/Bitcoin/bitcoin-svn-rev-158/src/db.h:152: error: ISO C++ forbids declaration of ‘Dbc’ with no type
/home/Bitcoin/bitcoin-svn-rev-158/src/db.h:152: error: expected ‘;’ before ‘*’ token
/home/Bitcoin/bitcoin-svn-rev-158/src/db.h:163: error: expected `;' before ‘int’
/home/Bitcoin/bitcoin-svn-rev-158/src/db.h:163: error: ‘Dbc’ has not been declared
/home/Bitcoin/bitcoin-svn-rev-158/src/db.h:202: error: ISO C++ forbids declaration of ‘DbTxn’ with no type
/home/Bitcoin/bitcoin-svn-rev-158/src/db.h:202: error: expected ‘;’ before ‘*’ token
/home/Bitcoin/bitcoin-svn-rev-158/src/db.h:210: error: expected `;' before ‘public’
/home/Bitcoin/bitcoin-svn-rev-158/src/db.h:163: error: ‘DB_NEXT’ was not declared in this scope
/home/Bitcoin/bitcoin-svn-rev-158/src/db.h: In member function ‘bool CDB::Read(const K&, T&)’:
/home/Bitcoin/bitcoin-svn-rev-158/src/db.h:54: error: ‘pdb’ was not declared in this scope
/home/Bitcoin/bitcoin-svn-rev-158/src/db.h:61: error: ‘Dbt’ was not declared in this scope
(...)
(...)

Any ideas what's wrong ?

I run an AMD64 Gentoo.

Code:

Linux tuxhost 2.6.34-gentoo-r2 #2 SMP Fri Jul 23 11:42:01 CEST 2010 x86_64 AMD Phenom(tm) 9850 Quad-Core Processor AuthenticAMD GNU/Linux

[ShadowOfHarbringer]

UPDATE:

I compiled it successfully using a modified ebuild.
It launches and seems to generally work, but at the same speed as normal.

I have 195.36.24 version of NVIDIA drivers (too old perhaps ?).

I guess i will wait for the official CUDA build.

[LZ]

You should use the latest driver version. Did you use -gpu switch?

[ShadowOfHarbringer]

Yeah i tried using the -gpu switch, nothing happened. Well, actually perhaps something happened - i think the speed was cut in half - i had only 2200khash/sec instead of normal 4500khash/sec or 8500 khash/sec (-4way switch).

I guess then will try the new drivers tomorrow.

[eurekafag]

Do you use CUDA-enabled driver or regular one?

[ShadowOfHarbringer]

Do you use CUDA-enabled driver or regular one?

It seems I used the normal driver....
Everything is clear now.

[L29Ah]

UPDATE:

I compiled it successfully using a modified ebuild.
It launches and seems to generally work, but at the same speed as normal.

I have 195.36.24 version of NVIDIA drivers (too old perhaps ?).

I guess i will wait for the official CUDA build.

can you please post the ebuild?

[LobsterMan]

Any chance of anyone posting compiled binaries for windows? I'm still having a hell of a time trying to get anything to compile 

[joe]

Unless he lives in a nation that doesn't enforce intellectual property rights.

The Civil Code of the Russian Federation / Chapter 70. Copyright.

Article 1280.
Free Reproduction of computer programs and databases. Decompilation of Computer Programs.
1. Person lawfully in possession of a copy of a computer program or database instance (the user) may,
without the permission of the author or copyright holder and without payment of additional remuneration:
1) make changes to the computer program or database solely for the purpose of their operation on the
user's device and to perform any actions necessary for the functioning of such a program or database
according to their function, including storing in computer memory (one computer or a network), and to
make correct obvious errors, unless otherwise provided by contract with the copyright holder;
2) make a copy of a computer program or database provided that the copy is intended solely for archiving
purposes or for replacement of a lawfully acquired copy where such copy has been lost, destroyed or
unusable; at the same time a copy of a computer program or database may not be used for purposes other
than those specified in paragraph 1 of this paragraph, and must be destroyed if possession of the copy of
such a program or database cease to be lawful.
2. Person lawfully in possession of a copy of a computer program may, without the consent of the
rightholder and without payment of additional remuneration: to study, investigate or test the functioning
of such a program in order to determine the ideas and principles which underlie any element of a computer
program through any action under subparagraph 1 paragraph 1 of this article.
3. Person lawfully in possession of a copy of a computer program may, without the consent of the
rightholder and without payment of additional compensation: to reproduce and convert the object code
into source code (decompile the computer program) or by third parties to carry out these actions, if they
are necessary to achieve interoperability independently developed that person's computer program with
other programs that can interact with the decompiled program, under the following conditions:
1) the information necessary to achieve interoperability, has not previously been available to that person
from other sources;
2) these actions are carried out in respect of only those parts of the decompilation of computer programs,
are necessary to achieve interoperability;
3) information obtained by decompilation may only be used to achieve the ability to interact independently
created computer programs with other programs that can be transferred to other persons, except where
necessary to achieve the ability to interact independently created computer programs with other programs
and can not be used to develop a computer program, a type essentially comparable to the decompiled
computer program, or for any other act that violates the exclusive right to the computer program.
4. Application of the provisions stipulated in this Article shall not cause undue damage to normal use
of a computer program or database, and should not unreasonably prejudice the legitimate interests
of the author or copyright holder.

What has the world come to where laws now tell us what we can do (and everything else is illegal) instead of the old days where laws told us what we can't do (and everything else was legal)?

[ShadowOfHarbringer]

can you please post the ebuild?

I deleted it already and now i have to re-do it again.
It may take a day or 2.

[ShadowOfHarbringer]

OK, now I'm confused.

1. I installed the NVIDIA developer drivers
2. I installed NVIDIA cuda toolkit
3. I downloaded the svn rev. 158 source
4. I patched the source
5. I used CMAKEfiles on the source to make it buildable
6. I compiled / emerged everything

And the result is that on GUI version i get 1200khash/sec always, it does not matter if i used -gpu, -aggresion, -gputhreads, or any other switch.
It seems that the client is not working on graphics card, it is just using CPU, and single core only.

Also, the client in graphics mode keeps breaking wxGTK/wxWidgets (wxWidgets often popup with some debug messages). But that is bearable.

----
EDIT:
I guess i will try compiling only the daemon now and tell You how it went.

[ShadowOfHarbringer]

UPDATE:

I tried everything again with bitcoind. The same results.
1200khash/sec with -gpu switch, and 4700 khash/sec without (and 8500 khash/sec with -4way switch).

So the client seems not to be using GPU at all.

BTW,
I have NVIDIA Geforce 9800GT

EDIT:
It would be really useful if the GPU client displayed any debugging info, because i don't even know if my card was detected, if CUDA is enabled properly etc etc.

[nightrow]

I have a nvidia card, 8800gt, i can test under windows and why not under linux in virtualbox.
Don't hesitate to publish step by test it, or publish a windows binary package, i'm willing to help as i can.

[ShadowOfHarbringer]

Here it is:

SVN Rev 158, patched & cmaked.
However, still not working.

http://www.tinyurl.pl/?KMsRtkuR

Maybe somebody will figure out what i did wrong.

BTW,
Forget any hope that it will work under VirtualBox. No way.

[puddinpop]

EDIT:
It would be really useful if the GPU client displayed any debugging info, because i don't even know if my card was detected, if CUDA is enabled properly etc etc.

Why are you assuming it doesn't display any debug info?  If you look in the debug.log, you'll see lots of info on what is detected and what it is doing.  Why was this info not useful to you?

Anyway, it sounds like you either don't have a CUDA capable driver installed, or you somehow compiled with CPU emulation turned on.  It's probably the CPU emulation.  A 9800 should get 21 Mhash/s.

[ShadowOfHarbringer]

Well i debugged the compile process a little, and concluded that i have wrong makefile, which does not include gpucommon.h/cpp and bitcoinminercuda.h/cpp, and does not convert them to objects.

However, after adding them to makefile,

Code:

HEADERS=(...) cuda/bitcoinminercuda.h cuda/cudashared.h gpucommon/gpucommon.h

all: bitcoind

OBJS= \
 (...)
        obj/gpucommon.o \
        obj/bitcoinminercuda.o

, make complains that it "cannot find rules to make object. Stop."

EDIT:

Code:

make: *** No rule to make target `obj/gpucommon.o', needed by `bitcoin'.  Stop.

.

I know, I'm not a C++ programmer.

[puddinpop]

So that points to you not using the CMake build files.  So your first step should be to use cmake to generate appropriate makefiles with CUDA enabled and then use those to make bitcoin.

[kslaughter]

puddinpop,

You are in violation of the GPL since you CUDA is linked to the bitcoin source.

I will build a CUDA enabled build of the bitcoin client and post the source code
here.
 
Ken

[em3rgentOrdr]

puddinpop,

You are in violation of the GPL since you CUDA is linked to the bitcoin source.

I will build a CUDA enabled build of the bitcoin client and post the source code
here.
 
Ken

Good point!  We should report puddinpop to gpl-violations.org immediately!  Get ready to be sued by Richard Stallman and his lawyer buddies!  

Just Kidding!  Bitcoin uses The MIT License/X11 license, which is permissive...

[mizerydearia]

puddinpop,

You are in violation of the GPL since you CUDA is linked to the bitcoin source.

I will build a CUDA enabled build of the bitcoin client and post the source code
here.
 
Ken

This doesn't seem correct.  How does GPL come into play when Bitcoin is under MIT license?

[ShadowOfHarbringer]

So that points to you not using the CMake build files.  So your first step should be to use cmake to generate appropriate makefiles with CUDA enabled and then use those to make bitcoin.

I tried that at first, but build failed. This is the debug output i got:

Code:

bitcoin-svn-rev-158 # make bitcoind
-- Found BerkeleyDB: /usr/lib64/libdb.so
-- Configuring done
-- Generating done
-- Build files have been written to: /home/Bitcoin/bitcoin-svn-rev-158
[  5%] Building NVCC (Device) object cmake-bitcoind/./bitcoind_generated_bitcoinminercuda.cu.o
Scanning dependencies of target bitcoind
[ 11%] Building CXX object cmake-bitcoind/CMakeFiles/bitcoind.dir/__/src/db.cpp.o
In file included from /home/Bitcoin/bitcoin-svn-rev-158/src/db.cpp:5:
/home/Bitcoin/bitcoin-svn-rev-158/src/headers.h:45:20: error: db_cxx.h: Nie ma takiego pliku ani katalogu
In file included from /home/Bitcoin/bitcoin-svn-rev-158/src/headers.h:123,
                 from /home/Bitcoin/bitcoin-svn-rev-158/src/db.cpp:5:
/home/Bitcoin/bitcoin-svn-rev-158/src/db.h:26: error: ‘DbEnv’ does not name a type
/home/Bitcoin/bitcoin-svn-rev-158/src/db.h:37: error: ISO C++ forbids declaration of ‘Db’ with no type
/home/Bitcoin/bitcoin-svn-rev-158/src/db.h:37: error: expected ‘;’ before ‘*’ token
/home/Bitcoin/bitcoin-svn-rev-158/src/db.h:39: error: ‘DbTxn’ was not declared in this scope
/home/Bitcoin/bitcoin-svn-rev-158/src/db.h:39: error: template argument 1 is invalid
/home/Bitcoin/bitcoin-svn-rev-158/src/db.h:39: error: template argument 2 is invalid
/home/Bitcoin/bitcoin-svn-rev-158/src/db.h:152: error: ISO C++ forbids declaration of ‘Dbc’ with no type
/home/Bitcoin/bitcoin-svn-rev-158/src/db.h:152: error: expected ‘;’ before ‘*’ token
/home/Bitcoin/bitcoin-svn-rev-158/src/db.h:163: error: expected `;' before ‘int’
/home/Bitcoin/bitcoin-svn-rev-158/src/db.h:163: error: ‘Dbc’ has not been declared
/home/Bitcoin/bitcoin-svn-rev-158/src/db.h:202: error: ISO C++ forbids declaration of ‘DbTxn’ with no type
/home/Bitcoin/bitcoin-svn-rev-158/src/db.h:202: error: expected ‘;’ before ‘*’ token
/home/Bitcoin/bitcoin-svn-rev-158/src/db.h:210: error: expected `;' before ‘public’
/home/Bitcoin/bitcoin-svn-rev-158/src/db.h:163: error: ‘DB_NEXT’ was not declared in this scope
/home/Bitcoin/bitcoin-svn-rev-158/src/db.h: In member function ‘bool CDB::Read(const K&, T&)’:
/home/Bitcoin/bitcoin-svn-rev-158/src/db.h:54: error: ‘pdb’ was not declared in this scope
/home/Bitcoin/bitcoin-svn-rev-158/src/db.h:61: error: ‘Dbt’ was not declared in this scope
(...)
(...)

However that approach won't work with Gentoo ebuild/emerge system, so i abandoned it.


EDIT:

This doesn't seem correct.  How does GPL come into play when Bitcoin is under MIT license?

He included GPL sourcecode in the CUDA-patch.

[ShadowOfHarbringer]

Here is my Gentoo makefile, if anybody is interested.

http://pastebin.com/73aFW42p

Also, i grepped all the CMake makefiles for "gpucommon.h" and "bitcoinminercuda.h". No results.
So i'm not so sure the CMakefiles are going to work at all.

[mizerydearia]

I modified my makefile for bitcoin-9999.ebuild for svn Bitcoin and then

Code:

LC_ALL=C TZ=UTC0 diff -Naur makefile.unix makefile.gentoo > files/bitcoin-9999-Makefile.patch
ebuild bitcoin-9999.ebuild digest
emerge bitcoin

The emerge failed.  What else do I need to do besides update the makefile?

[ShadowOfHarbringer]

I modified my makefile for bitcoin-9999.ebuild for svn Bitcoin and then

Code:

LC_ALL=C TZ=UTC0 diff -Naur makefile.unix makefile.gentoo > files/bitcoin-9999-Makefile.patch
ebuild bitcoin-9999.ebuild digest
emerge bitcoin

The emerge failed.  What else do I need to do besides update the makefile?

The ebuild-9999 is bad, don't use it.
I'm using a modified 0.3.13 ebuild all the time - it works for SVN rev 158 source too.

Also, You need to create additional separate makefile for CUDA version, place it in the files/ directory and then put an digest entry for it in the Manifest. At least this is what i did.
For me, it emerges successfully, however the CUDA header & cpp files are omitted. So i need a better makefile.

[mizerydearia]

The ebuild-9999 is bad, don't use it.

How is it bad?  More importantly, though, without losing focus of my original question, what else do I need to do besides update the makefile?

[ShadowOfHarbringer]

The ebuild-9999 is bad, don't use it.

How is it bad?  More importantly, though, without losing focus of my original question, what else do I need to do besides update the makefile?

Why it is bad ? Well, i looked at the source and i didn't like it, so i took the 0.3.13 one.

I have no idea what do You need. I downloaded source, patched it and **it just worked** with 0.3.13 ebuild on my Gentoo AMD64.
I only changed the digests in Manifest to match added/modified files.

[kslaughter]

I would not have made a mistake by assuming it was licensed under
the GPL if puddinpop had included in his license the MIT license as
was required.  Of course if he had included the MIT license all of
the restrictions he placed in his license would have been moot.

Ken

[jgarzik]

I would not have made a mistake by assuming it was licensed under
the GPL if puddinpop had included in his license the MIT license as
was required.  Of course if he had included the MIT license all of
the restrictions he placed in his license would have been moot.

puddinpop's original open sourced client was MIT.

But he makes his changes GPL'd, which is disappointing, leading to confusing situations like this.

[lfm]

So I've been trying to figure out.. just how many coins did
1HZNsUqQxKVLmfPfCAzLwrnVDzx8CxwxnM
receive?

I now see 37 blocks with the 5.00 BTC going to this address. It seems there are still people running the old proprietary version. Still negligible tho.

That latest was Block #84274 at 2010-10-09 20:29:52

[nightrow]

I'm one of the people running the proprietary version, even if i yet generated no block.
This is the only ready to use binary i found for windows.

I can get 21Mh/s with a 8800gt, comparing to 2.2Mhz with a e6750 2.66ghz processor.
I would say *10 -10% is better than nothing, that still make a *9 speedup.
sure, a *10 speedup would be best :p

Any news of a windows version that can get such a speed without compilation ?

[BitLex]

It seems there are still people running the old proprietary version.

is there any other version available (besides the old cracked proprietary version) for people that dont compile themselfes?
i havent noticed any.

[dkaparis]

So that points to you not using the CMake build files.  So your first step should be to use cmake to generate appropriate makefiles with CUDA enabled and then use those to make bitcoin.

I tried that at first, but build failed. This is the debug output i got:

Code:

bitcoin-svn-rev-158 # make bitcoind
-- Found BerkeleyDB: /usr/lib64/libdb.so
-- Configuring done
-- Generating done
-- Build files have been written to: /home/Bitcoin/bitcoin-svn-rev-158
[  5%] Building NVCC (Device) object cmake-bitcoind/./bitcoind_generated_bitcoinminercuda.cu.o
Scanning dependencies of target bitcoind
[ 11%] Building CXX object cmake-bitcoind/CMakeFiles/bitcoind.dir/__/src/db.cpp.o
In file included from /home/Bitcoin/bitcoin-svn-rev-158/src/db.cpp:5:
/home/Bitcoin/bitcoin-svn-rev-158/src/headers.h:45:20: error: db_cxx.h: Nie ma takiego pliku ani katalogu
In file included from /home/Bitcoin/bitcoin-svn-rev-158/src/headers.h:123,
                 from /home/Bitcoin/bitcoin-svn-rev-158/src/db.cpp:5:
/home/Bitcoin/bitcoin-svn-rev-158/src/db.h:26: error: ‘DbEnv’ does not name a type
/home/Bitcoin/bitcoin-svn-rev-158/src/db.h:37: error: ISO C++ forbids declaration of ‘Db’ with no type
/home/Bitcoin/bitcoin-svn-rev-158/src/db.h:37: error: expected ‘;’ before ‘*’ token
/home/Bitcoin/bitcoin-svn-rev-158/src/db.h:39: error: ‘DbTxn’ was not declared in this scope
/home/Bitcoin/bitcoin-svn-rev-158/src/db.h:39: error: template argument 1 is invalid
/home/Bitcoin/bitcoin-svn-rev-158/src/db.h:39: error: template argument 2 is invalid
/home/Bitcoin/bitcoin-svn-rev-158/src/db.h:152: error: ISO C++ forbids declaration of ‘Dbc’ with no type
/home/Bitcoin/bitcoin-svn-rev-158/src/db.h:152: error: expected ‘;’ before ‘*’ token
/home/Bitcoin/bitcoin-svn-rev-158/src/db.h:163: error: expected `;' before ‘int’
/home/Bitcoin/bitcoin-svn-rev-158/src/db.h:163: error: ‘Dbc’ has not been declared
/home/Bitcoin/bitcoin-svn-rev-158/src/db.h:202: error: ISO C++ forbids declaration of ‘DbTxn’ with no type
/home/Bitcoin/bitcoin-svn-rev-158/src/db.h:202: error: expected ‘;’ before ‘*’ token
/home/Bitcoin/bitcoin-svn-rev-158/src/db.h:210: error: expected `;' before ‘public’
/home/Bitcoin/bitcoin-svn-rev-158/src/db.h:163: error: ‘DB_NEXT’ was not declared in this scope
/home/Bitcoin/bitcoin-svn-rev-158/src/db.h: In member function ‘bool CDB::Read(const K&, T&)’:
/home/Bitcoin/bitcoin-svn-rev-158/src/db.h:54: error: ‘pdb’ was not declared in this scope
/home/Bitcoin/bitcoin-svn-rev-158/src/db.h:61: error: ‘Dbt’ was not declared in this scope
(...)
(...)

However that approach won't work with Gentoo ebuild/emerge system, so i abandoned it.

Seems the the db include directory was not set correctly. Can you search your system for db_cxx.h and db.h and post the results? I'll try to look at it when I have time.

[ShadowOfHarbringer]

I continued with the Gentoo ebuild approach, and got around most of the library dependencies problem. I will probably not continue with CMake way, because it is very gentoo-incompatibile.

However, now i get this while trying to compile code:

Code:

obj/bitcoinminercuda.o: In function `CUDARunner::RunStep()':
bitcoinminercuda.cpp:(.text+0x301): undefined reference to `cudaMemcpy'
bitcoinminercuda.cpp:(.text+0x332): undefined reference to `cuda_process_helper(cuda_in*, cuda_out*, unsigned int, unsigned int, int, int)'
bitcoinminercuda.cpp:(.text+0x351): undefined reference to `cudaMemcpy'
bitcoinminercuda.cpp:(.text+0x3e5): undefined reference to `cudaGetErrorString'
bitcoinminercuda.cpp:(.text+0x41c): undefined reference to `cudaGetErrorString'
obj/bitcoinminercuda.o: In function `CUDARunner::AllocateResources(int, int)':
bitcoinminercuda.cpp:(.text+0x48c): undefined reference to `cudaMalloc'
bitcoinminercuda.cpp:(.text+0x49c): undefined reference to `cudaMalloc'
bitcoinminercuda.cpp:(.text+0x4d1): undefined reference to `cudaGetErrorString'
bitcoinminercuda.cpp:(.text+0x508): undefined reference to `cudaGetErrorString'
obj/bitcoinminercuda.o: In function `CUDARunner::DeallocateResources()':
bitcoinminercuda.cpp:(.text+0x544): undefined reference to `cudaFree'
bitcoinminercuda.cpp:(.text+0x574): undefined reference to `cudaFree'
bitcoinminercuda.cpp:(.text+0x591): undefined reference to `cudaGetErrorString'
bitcoinminercuda.cpp:(.text+0x5c8): undefined reference to `cudaGetErrorString'
obj/bitcoinminercuda.o: In function `CUDARunner::~CUDARunner()':
bitcoinminercuda.cpp:(.text+0x5f1): undefined reference to `cudaThreadExit'
bitcoinminercuda.cpp:(.text+0x60c): undefined reference to `cudaGetErrorString'
obj/bitcoinminercuda.o: In function `CUDARunner::~CUDARunner()':
bitcoinminercuda.cpp:(.text+0x661): undefined reference to `cudaThreadExit'
bitcoinminercuda.cpp:(.text+0x675): undefined reference to `cudaGetErrorString'
obj/bitcoinminercuda.o: In function `CUDARunner::~CUDARunner()':
bitcoinminercuda.cpp:(.text+0x6d1): undefined reference to `cudaThreadExit'
bitcoinminercuda.cpp:(.text+0x6e5): undefined reference to `cudaGetErrorString'
obj/bitcoinminercuda.o: In function `CUDARunner::CUDARunner()':
bitcoinminercuda.cpp:(.text+0x76a): undefined reference to `cudaGetDeviceCount'
bitcoinminercuda.cpp:(.text+0x79c): undefined reference to `cudaSetDevice'
bitcoinminercuda.cpp:(.text+0x7c0): undefined reference to `cudaSetDevice'
bitcoinminercuda.cpp:(.text+0x7cb): undefined reference to `cudaGetErrorString'
bitcoinminercuda.cpp:(.text+0x7f8): undefined reference to `cudaGetErrorString'
bitcoinminercuda.cpp:(.text+0x82f): undefined reference to `cudaGetErrorString'
obj/bitcoinminercuda.o: In function `CUDARunner::CUDARunner()':
bitcoinminercuda.cpp:(.text+0x88a): undefined reference to `cudaGetDeviceCount'
bitcoinminercuda.cpp:(.text+0x8bc): undefined reference to `cudaSetDevice'
bitcoinminercuda.cpp:(.text+0x8e0): undefined reference to `cudaSetDevice'
bitcoinminercuda.cpp:(.text+0x8eb): undefined reference to `cudaGetErrorString'
bitcoinminercuda.cpp:(.text+0x918): undefined reference to `cudaGetErrorString'
bitcoinminercuda.cpp:(.text+0x94f): undefined reference to `cudaGetErrorString'
obj/bitcoinminercuda.o: In function `CUDARunner::FindBestConfiguration()':
bitcoinminercuda.cpp:(.text+0xa7b): undefined reference to `cudaGetLastError'
bitcoinminercuda.cpp:(.text+0xaab): undefined reference to `cudaMemcpy'
bitcoinminercuda.cpp:(.text+0xad0): undefined reference to `cuda_process_helper(cuda_in*, cuda_out*, unsigned int, unsigned int, int, int)'
bitcoinminercuda.cpp:(.text+0xae5): undefined reference to `cudaMemcpy'
bitcoinminercuda.cpp:(.text+0xaf2): undefined reference to `cudaGetLastError'
bitcoinminercuda.cpp:(.text+0xc1e): undefined reference to `cudaGetErrorString'
bitcoinminercuda.cpp:(.text+0xc55): undefined reference to `cudaGetErrorString'
obj/bitcoinminercuda.o: In function `cutGetMaxGflopsDeviceId()':
bitcoinminercuda.cpp:(.text._Z23cutGetMaxGflopsDeviceIdv[cutGetMaxGflopsDeviceId()]+0x46): undefined reference to `cudaGetDeviceCount'
bitcoinminercuda.cpp:(.text._Z23cutGetMaxGflopsDeviceIdv[cutGetMaxGflopsDeviceId()]+0x66): undefined reference to `cudaGetDeviceProperties'
bitcoinminercuda.cpp:(.text._Z23cutGetMaxGflopsDeviceIdv[cutGetMaxGflopsDeviceId()]+0xca): undefined reference to `cudaGetDeviceProperties'
collect2: ld returned 1 exit status

It seems I am missing some library or something.

Also, my libraries paths & include paths in the makefile are as follows:

Code:

INCLUDEPATHS= \
         -I"/usr/include" \
         -I"/usr/include/crypto++" \
         -I"@@GENTOO_DB_INCLUDEDIR@@" \
         -I"/opt/cuda/sdk/C/common/inc/" \
         -I"/opt/cuda/include/"
LIBPATHS= \
         -L"/usr/lib" \
         -L"/opt/cuda/lib64" -lcufft -lcudart -Wl,-rpath,/opt/cuda/lib64
LIBS := ${LDFLAGS} -dead_strip \
 -ldb_cxx -lboost_system -lboost_filesystem -lboost_program_options -lboost_thread -lgthread-2.0 -lssl -lcrypto -lcrypto++ -lgmp -ldl -lz

Once again, I am not a C++ programmer (too bad).

EDIT:

To be clear, i have moved all of the *.cpp, *.h and *.cu files from the data/ & gpucommon/ directory, to the root directory to make it easier resolving dependencies.

You may also find this part of the makefile useful:

Code:

HEADERS=headers.h strlcpy.h serialize.h uint256.h util.h key.h bignum.h base58.h \
    script.h db.h net.h irc.h main.h rpc.h uibase.h ui.h noui.h init.h bitcoinminercuda.h \
    cudashared.h gpucommon.h

all: bitcoind

OBJS= \
        obj/util.o \
        obj/script.o \
        obj/db.o \
        obj/net.o \
        obj/irc.o \
        obj/main.o \
        obj/rpc.o \
        obj/init.o \
        obj/bitcoinminercuda.o \
        obj/gpucommon.o

#       cryptopp/obj/sha.o
#       cryptopp/obj/cpu.o

bitcoind: $(OBJS:obj/%=obj/nogui/%) obj/sha256.o
        g++ $(CFLAGS) -o $@ $^ $(LIBPATHS) $(LIBS)

bitcoin: $(OBJS) obj/ui.o obj/uibase.o obj/sha256.o
        g++ $(CFLAGS) -o $@ $^ $(WXLIBS) $(LIBS)

#cryptopp/obj/%.o: cryptopp/%.cpp
#       g++ -c $(CFLAGS) -O3 -o $@ $<

obj/%.o: %.cpp $(HEADERS)
        g++ -c $(CFLAGS) $(WXDEFS) -DGUI -o $@ $<

obj/sha256.o: sha256.cpp
        g++ -c $(CFLAGS) @@GENTOO_SHA256_SSE2@@ -o $@ $<

obj/nogui/%.o: %.cpp $(HEADERS)
        g++ -c $(CFLAGS) -o $@ $<

I promise to make the Gentoo cuda ebuild public after i make this work.

[ShadowOfHarbringer]

Seems the the db include directory was not set correctly. Can you search your system for db_cxx.h and db.h and post the results? I'll try to look at it when I have time.

Here You go:

db_cxx.h :

Code:

$ find /usr | grep db_cxx
/usr/include/db4.7/db_cxx.h
/usr/include/db4.8/db_cxx.h
/usr/include/db4.5/db_cxx.h
/usr/include/db4.6/db_cxx.h

db.h :

Code:

$ find /usr | fgrep '/db.h'
/usr/include/db4.7/db.h
/usr/include/htdig_db/db.h
/usr/include/db4.8/db.h
/usr/include/db4.5/db.h

[dkaparis]

Seems the the db include directory was not set correctly. Can you search your system for db_cxx.h and db.h and post the results? I'll try to look at it when I have time.

Here You go:

db_cxx.h :

Code:

$ find /usr | grep db_cxx
/usr/include/db4.7/db_cxx.h
/usr/include/db4.8/db_cxx.h
/usr/include/db4.5/db_cxx.h
/usr/include/db4.6/db_cxx.h

db.h :

Code:

$ find /usr | fgrep '/db.h'
/usr/include/db4.7/db.h
/usr/include/htdig_db/db.h
/usr/include/db4.8/db.h
/usr/include/db4.5/db.h

Interesting. You have your db headers installed in db4.7, while on my FreeBSD installation it's db47. I think it should work if you use my FindBerkeleyDB module and set DB_VERSION to 4.7, instead of 47. Anyway at least with the REQUIRED flag, the CMake configuration will bail out as it should, if it's unable to find the headers.

[mizerydearia]

SOH: I recommend you to create a new thread specific to OpenCL/CUDA gentoo ebuild developments; otherwise use my gentoo thread

[LZ]

It seems there are still people running the old proprietary version.

is there any other version available (besides the old cracked proprietary version) for people that dont compile themselfes?
i havent noticed any.

Yes, it is available for Windows now. You can get it here. Do not forget to use the -gpu key.

[ShadowOfHarbringer]

I have finally succeeded to create a working makefile.
Everything emerges properly, however dependencies are not automatically solved, so the ebuild will not work on any Gentoo out of the box. It needs improvements.
Also, my ebuild replaces normal bitcoin with its cuda version (as i don't know how to make separate binaries yet), so it's probably not the way it should be done anyway.

I will be posting the ebuild soon together with modified source, after i start my account at github.

[GeorgeH]

patch against SVN 158
full source

I'm attempting to compile the full source and I'm having a lot of dependency issues. Has anyone been able to build this for win32?

[LobsterMan]

patch against SVN 158
full source

I'm attempting to compile the full source and I'm having a lot of dependency issues. Has anyone been able to build this for win32?

Been trying for a long time with no success myself, perhaps take a look at this thread I made a while back to see if any of the problems you're having are similar to mine?

http://bitcointalk.org/index.php?topic=1034.0

[bagazaga]

It seems there are still people running the old proprietary version.

is there any other version available (besides the old cracked proprietary version) for people that dont compile themselfes?
i havent noticed any.

Yes, it is available for Windows now. You can get it here. Do not forget to use the -gpu key.

There is no CUDA support in this binary. So there is only old proprietary binary available(and cracked version of it).

[LobsterMan]

There is a windows binary for open cl, which I have just recently learned works on nvidia cards:

http://bitcointalk.org/index.php?topic=1334.0

[BEaSTFX]

Windows 7 64 bit
Nvidia Geforce GT 220

[film2240]

Is there a client that can generate BC with GPU and CPU working together to speed things up even more,that is available on the Mac as well as the PC?

[Silverpike]

Is there a client that can generate BC with GPU and CPU working together to speed things up even more,that is available on the Mac as well as the PC?

No, there is no advantage to these devices working together.  You can run them both at the same time though, although CPU mining is an utter waste of time and power at this point.

[lonewolf1996]

I am very interested

Windows XP Pro
ATI RADEON XPRESS 200M
1gb memory

[film2240]

Win 7 pro x64
ATI radeon HD 4670 (Oc'd to the max)

I'm only getting 37.7Mhash/s mas with this.ny help to boost mining performance?
Also How can I tell when Ive made BTC with GUIminer/because I want to send my work to my BTC wallet.

[Silverpike]

Win 7 pro x64
ATI radeon HD 4670 (Oc'd to the max)

I'm only getting 37.7Mhash/s mas with this.ny help to boost mining performance?
Also How can I tell when Ive made BTC with GUIminer/because I want to send my work to my BTC wallet.

Guys, STOP POSTING IN THIS THREAD.

We have a mining forum for specific issues.  This thread is old and useless now.

[Garr255]

Hey, just wondering if any of you that posted back in 2010 are still around...

If so, make a post

[DeepBit]

Wow, generating bitcoins with a video card should be interesting !