Bitcoin Forum

So, for the fun of it, and because I don't have any newer graphics cards, I've modified m0mchil's poclbm into something that works on R600 video cards. These are the Radeon HD 2000 and 3000 series. I never bothered to rename it.

64-bit Windows binary http://dl.dropbox.com/u/38079179/poclbm-ati-brook-x64.zip (http://dl.dropbox.com/u/38079179/poclbm-ati-brook-x64.zip)

Source code http://dl.dropbox.com/u/38079179/poclbm-ati-brook-src.zip (http://dl.dropbox.com/u/38079179/poclbm-ati-brook-src.zip)

Some of the hints on how to build the source code are in the comments on the bottom of the C++ files. To use the Python source, you need some current version of Python 2, Numpy, and Boost::Python. To use the Brook+ source, you need ATI Stream SDK 1.4 beta and some C++ compiler, and some common sense.

Let's discuss the reasons not to mine on an R600.

• By now, they are fairly old. Not all of them were that fast, either. I think my HP notebook is thermally throttling the video card.
• They don't support the gather/scatter I/O operations that the OpenCL miners use to minimize memory and bandwidth use.
• They don't have the 1-instruction rotate operation, that makes the modern Radeons so much faster than the modern GeForces. Nor do they have the 1-instruction bitselect instruction that AMD strangely doesn't expose to any APIs.
• Shifting is extremely inefficient, only 1/5 the speed of normal operations. That's because the ATI design has 4 simple ALUs and 1 "Transcendental" ALU in each processor. In the R600, only the T-unit does integer multiplies and logical shifts. (All of them do floating-point multiplies.) With the number of shifts in SHA-256, most of the time the simple ALUs sit around waiting for the T-unit.
• As a result, my video card (Mobility Radeon HD 3410) crunches through fewer hashes than my CPU (AMD Turion Neo X2 L625). If I didn't mangle poclbm too badly, it estimates that it crunches at a rate of roughly 0.850 MH/s.
• This miner doesn't adjust its work size. That's because of point 2 above, so adjusting work size involves destroying and allocating buffers. For some reason, it's really slow on my computer, taking up to several seconds. This is not something that can be done several times per second.

The AMD KernelAnalyzer says this SHA-256 kernel should operate at a rate of 5M threads/sec on a Radeon HD 2900, or 57M threads/sec on a Radeon HD 6970. Clearly, this is less efficient than the OpenGL versions on the GPUs that can run OpenGL. I may also have made horrible mistakes in modifying the source code, especially in BitcoinMiner.py.

In retrospect, when I saw that Stream SDK was 64-bit, I should have installed the 32-bit Stream SDK, instead of installing the 64-bit Python. And, when I saw that Klöckner had used Boost::Python for PyOpenCL, I should have ignored that and used the raw Python C API.

Also, while destroying and creating buffers is extremely slow, I don't see why it won't work if I have it allocate several buffers and switch between them according to load. I expect that a dozen buffers should take only a few MB, and my video card has 512MB total. Something to do maybe later, if it didn't have such low payoff.

Anyway, I think I can be convinced to produce a 32-bit binary for the low, low price of 5 BTC. :)

You can send me tips if you feel like it, too.
19gShNE2sdo9NP7N3kTYPjkqQ6ukPCP8jH

So the old radeons are slower than current nvidia cards? I sold my radeon a long time ago, so I can't test it.

Question decade. Can you make it work for an IGP? I.e an integrated HD4200(which people say is actually an HD3k igp)

Well...

- I have 2 ATI cards in my known technology collection (amongst about 7-10 nVidias, about 6 of which are capable of mining - at 2-5 Mhash/sec).
-- One of which is the 6770 that I use for mining 24/7.
-- The other is a 3700, if I remember correctly.
- I'd really like some freedom to play with other pools.
- I've only ever made a grand total of about 0.6 Bitcoin in the entire ~3 weeks of 24/7 mining on the 6770.
-- Low, low price of 5 BTC my ass? ;) I think it'd take an eternity to mine that back...

If the miner goes faster on the 3700, than the ~20 Mhash/sec that my horrifyingly over-powered (and now blown-out, evidently) 8800GTS produces, I'd love to play around with it... even if it's stupidly inefficient, you'd really just need to step back and look at how bad nVidias are that people still try to mine with ;)

Oops, I rearranged and forgot to renumber. Fixed.

Probably. Bullet points 3 and 4. Bullet point 3 is a reference to the Bitcoin wiki. https://en.bitcoin.it/wiki/Why_a_GPU_mines_faster_than_a_CPU#Why_are_AMD_GPUs_faster_than_Nvidia_GPUs.3F (https://en.bitcoin.it/wiki/Why_a_GPU_mines_faster_than_a_CPU#Why_are_AMD_GPUs_faster_than_Nvidia_GPUs.3F)

Essentially, by the same measure that a modern Radeon has a 1.7x advantage over a modern GeForce, the older Radeon does not have that advantage, and in fact has a sloppily estimated 3.5x disadvantage. And that doesn't account for manufacturing process or runtime differences.

I don't know. I don't have a Radeon IGP. AMD's spec thing says the Radeon 4200 has the 40 unified shaders supporting the ATI Stream Technology, so it should already work with 64-bit Windows. For the low, low price of 5 BTC, I might be convinced to try to make it work on other platforms, too. I doubt you could recover even the costs of electricity with that chip.

I'm trying to consider the value of my time, here. And while the difficulty of mining increases, the exchange rate decreases, so I think even 5 BTC is only worth it for playing around. Instead, I give you my source code, which is everything you need to try it on your own, except for the links to the SDKs, and however long it takes it learn how to work those tools.

KernelAnalyzer says the SHA-256 kernel should do 6M threads/sec on a Radeon HD 3870, so I doubt that it will do better than a GeForce 8800GTS. Unless I also made horrible mistakes in writing the kernel.

Oh, yeah, for reference, I'm using ATI Stream SDK 1.4.0 beta, Visual Studio 2008, Python 2.7.2, NumPy 1.6.1, and Boost 1.47.0.

Can you convert 6 mill threads to mh/s?

Also, these 5 bitcoins...5 bitcoins per person, or it could be small donations from everyone to make 5 bitcoins?

No. I don't have any of the GPUs in KernelAnalyzer's group of simulated GPUs. I have only this Radeon HD 3410. I assume it's roughly 4 hashes per thread, so 6 Mthreads/s approximates 24 MH/s. Probably less because of runtime overhead.

Let's say 5 BTC per platform.

Okay, fine. If you can do it, count me in for 3 BTC. Anyone else want to cough up the missing 2?

I could really use this at the office. Best card I've got in my desk PC is an nVidia 8600GT that bakes 83 C crunching out 3-4 Mhash/sec, or an nVidia Ion with an Atom that, if overclocked, crunches 7 Mhash/sec. And I've got a couple cards laying around gathering dust that could be crunching 20? Sheez. It's worth it!

Well i can donate 0.5BTC if i can get miner for my RV620 (Ati HD 3450) :)
Im just curious to see if i can use my server for mining in its idletimes :D

So what i need to have/do/adjust :) so guide me up here :D

@rTech: I think we just need to wait for the remaining 1.5 BTC to be volunteered :)

Yep, I'm in for 3, so if you're in for 2, there's our 5 Bitcoin :)

Yo, Decade! :D

edit: Just checked on the GPU I'll be looking at using this on, and it's a Radeon HD 3600. Just FYI.

I actually managed to get 64 version to work with my ASUS EAH 3450 256MB.

1.1 Mhash is the neat result, ill post pictures later.

Wait, wait. I don't think you're mining with your GPU there. What's your CPU usage? What's the miner? If you have AMD-APP installed, it'll emulate GPU-processing using the CPU if you don't select the right device...

Brook+ is something deprecated in ATI Stream (now AMD APP) SDK a long time ago. You can consider it as some kind of an OpenCL predecessor. That's was the framework that I used when I got into GPGPU development. It looks like a very very limited C++, rather brain-damaging. Common things like arrays were not allowed in kernels. No local memory, no barriers, limited scatter-gather that was performed into reduction steps and stuff like that, it's like a nightmare from the past. Lucky for me I quickly switched to OpenCL which was something new back then and was also very badly supported. And no, Brook+ is not a heterogenous compute environment, it does not support CPUs. It also has nothing to do with OpenCL.

Actually I doubt recent AMD APP SDKs ship with even brcc. Headers are gone a long time ago. Well, even CAL/IL is being deprecated nowadays.

Did you read the 1st post?

http://img64.imageshack.us/img64/7195/3450.png

irrelevant info:
..i use this pc as gameserver and it has intel E2140 oc'd to 2.4ghz.
I have tested ufasoft cpu miner and it gives 5.7 mhash @ 73c

http://smile.it.cx/confused53.gif (http://smile.it.cx)
That's worse than nVidia mining...

yeah its already mentioned in op's title :)  Its still fun to test new things.

..anyway there is no point to use R6xx serie in mining.

Max i got out from my HD 3450 was 1.226 mhash and that was oc core 700mhz stable 48c.

Before you ask, my 3450 is low profile passive model.

You keep saying things to the effect of, "hey dumbass, can't you read?"... and to answer that, yes, I'm not a fucking moron, unlike most of the twats around here, I actually read the posts before I reply. Most of the time. Sometimes I jump in with an "ooh shiney!" effect, but typically I'm the guy slapping everyone straight that didn't read the whole topic.

Case in point... hey dumbass, can't you read?

And for the attention-impaired...

Far fucking cry from the CPU- or nVidia-like 1-point-something hashes a sec.

Additionally, these hashing functions used in kernels like phatk and poclbm seem to be centered around a small handful of custom functions. So to port the mining operation to a new language with a stripped down function set, it should STILL be about as simple as writing those "core" functions (rotate, math, etc) in an optimized way in the new language, and adapting all the function calls of the original code to work with the new language. And before you get started, yes, I DO know what I'm speaking of here (http://falconfour.com) (that is, I "program" in several languages, but I'm not a professional "programmer").

Plus, I would REALLY, FUCKING REALLY love to know what you're doing to even be able to write statements like "Max I got out of my HD 3450 was 1.226 mhash"... when Decade - the guy WRITING the first-and-only miner for these older GPUs - hasn't even fucking REPLIED to this thread yet since our offers. So there's no conceivable way (unless I'm getting raped here) that you could have your hands on the software we're talking about paying to develop here...

edit: wate. what the hell? i swear to christ that link wasn't up there in OP before. Now there's a binary and source? I don't see how I could've skimmed past that - only thing I plan on using this on is Win7 x64. Shit. Well, I was under the impression it hadn't even been made usable yet, and I guess it hadn't... with all the optimizations being considered so far, and the potential of >=20MHPS speeds on junker GPUs, I still think it's worth the investment in optimizing!

Where i've been hostile against you? Im not native english speaker so i have to use small sentences. My offer for 2 btc is now out of case cause you being incredibly hostile against me. My results were based on that binary 64 version and i runned several times to see what it can produce. This is same case like in other miner topics, everyone posted max result they got with that version what they had. So my max results were those what i informed. Im not coder so i cant use source. It also seems that coder of this product isnt very active here so this miner thread is pretty mutch forget and move on.

I'l hope you re happy when you haven't even run that app to check what it offers for you and post your results. I atleast runned it several times and show'd my results for all who may check this topic.

It was yours + PLaci1982's replies that had this whole sense of "this is pointless, what are you noobs bothering for" about it, and at the time I hadn't realized there was even anything to test to begin with. Dead serious, I was 100% under the impression nothing had been posted yet and we were paying to have the program written.

Since neither of us had paid for the program to be written yet, obviously I was a bit ruffled that there were people in this thread (and with a huge screenshot) acting like something already "exists and sucks"... so yeah. Sorry 'bout that. Take your 2 BTC away if you like, but I don't even know if Decade is going to be back anyway :/

OMG m8 don't be like a 8yo n00b.... MH/s will be different for different cards... HD 2400, HD 3100-3400, HD 4200 GPUs have ~1/3 of compute power compared to a HD 2600, HD 3600, ~1/6 compared to a HD 2900 GT and ~1/8 compared to a HD 2900 Pro/XT, HD 3800...

http://www.planetcalypsoforum.com/gallery/files/3/8/6/7/pot-kettle.jpg

Anywho, I gave the current build a whirl on the 3600 computer. No love at all. It gives me a brief message stating "brook_cal DLL Not Found!", followed by two messages of "Failed to initialize CAL. Falling back to CPU". I'm running from the command line in the extracted ("dist") folder, with Catalyst 11.6 on Win7 x64.

Think I'll go try smacking it with Process Monitor, but I really do hate that thing... useful, but a pain in the ass to use :/

edit: also, I get a kick out of your sig explicitly calling yourself an "expert". I'm not saying you're not - that line in your sig did it for me. Honestly, I don't have to go around waving such self-proclaimed labels in my posts because it's the post itself (specifically, attention to detail in spelling and grammar, goes a LONG way to establishing credibility, like it or not) that shows I know what I'm talking about. Just friendly advice here, and this is completely unrelated to the topic/discussion/argument at hand, but: you really do seem like a bit of a douche putting that in your sig, IMO.  :P

My sole intention of posting that pic was just proof that i got it work and in what rate. Your own imaginaion ruined everything else. I have not been smartass anywhere in my posts. You just imagine and blame me for things your own brains farted. Still my test is currently only non author result here. So maybe you need to take break and think before you act. BTW it was you asking about cpu usage so i added it too for my pic with gpu info also. Anyway you re pretty hostile dude. Also re-read my posts to figure out i had compiled testing version what i used. You just blamed me things i didnt do (as i had tool and results). My post said "its fun to test new things" and after my own results i said its not good thing to use R620 for mining.

For your words "neither we have this tool"
I did said i used 64 version (you didnt bother check first topic by topic starter, where usually every software is by their author.. always 1st post in topic!). I think all others who read these post realises that :) But you re too horry to act before you even understand what others said. I assume you re Yankee... as you act like big moron one...

Shit, I'm hostile? Read your own post, dude, you're going on a world-burning rampage spewing all kinds of hate! I just hadn't realized the OP had already posted a preliminary version (and it would be very helpful if Decade would actually come back and give us some insight here...), so I was wondering why the heck you got a copy of it first! That's all.

And you all keep blindly overlooking the part where the guy said a theoretical speed on a mediocre card would be 20MH/s, which is hell and away from the 1MH/s you were reporting. Seeing as though I didn't know any software had been posted at the time, I had no idea what software you were even using. And at 1MH/s, it seemed more like a mis-configured Phoenix miner than anything (set to run AMD APP emulation on CPU, which happens with a bad DEVICE= entry).

It was a miscommunication, misunderstanding, mis-something.
Nobody's dying.
Nobody's throwing accusations around (except, well, you and the whole "you're pretty hostile dude" and "think before you act" thing, Mom... and PLaci1982 with the "lol m8 8yo n00b" thing).
We're on the same page now. Hopefully. At least in the same chapter. Which is good.
Now, can we MOVE ON?

I rely on my own results until there is another version (i didnt oversee anything here, i just posted my results). GPU usage is shown in my pic and CPU usage also. You just said that you didnt read before you acted :) so you didnt know about first post tool being posted and started to act hostile against me. Anyway case closed in my situation until next version comes out for testing.

Yep, I completely admit I screwed up not reading that there was already a version posted here. We both acted kinda douche. All sorted now. No hard feelings I hope.

Also keeping my fingers crossed for an update. I'm sure there's LOTS of tweaks that can be done to improve performance on these cards... I'm still convinced (knowing the technical details/limitations of how CUDA processing works versus AMD's solution being faster) that nVidia miners can be much more optimized as well... I think someone with the motivation and knowledge needs to dig into it. I might even take a crack at it... the only CUDA miner I know of out there is way old (rpcminer-cuda)! The architecture of the hash processing functions just needs to be adapted to match the architecture of those GPUs :)

Yeah no hard feelings :) Im really hoping that developer of this tool tryis to improve it bit :) As its always pleasure to test new things.

Makes me wish my 3850 was on a 64 bit capable system. As is my dual P3 board leaves me with no other option. Hope the developer comes back.

Really old post, but...

I've succeeded in building the R600 Bitcoin Miner on a 32-bit machine. It was kinda painful, as OP has said, but not too bad. :)

I'm willing to sell the 32-bit build of this app on a bounty for 3 BTC, 2 BTC less than OP's price (barring BTC value increases).

Please note that this miner doesn't have great performance on crappy GPUs, end of story.
(However, if you have a better old AMD ATI GPU, it might be useful...)

If anyone submits an improved Brook+ source file for mining, I'll deduct 0.5 BTC from the cost (up to 1.5 BTC).

I've also created a build system for the miner. I'll release it on GitHub for a bounty of 5 BTC, since it took me a while to get everything working.
Included is a list of all the programs needed, with direct or close-to-direct links to download the programs.

So in summary: 3 BTC for the 32-bit build (-0.5 BTC for every optimization), 5 BTC for the build source!

http://imageshack.us/a/img685/8015/screenshotminer.png (http://imageshack.us/photo/my-images/685/screenshotminer.png/)

(And yes, that's Windows 8... the Classic Shell and the Windows To Go feature are the only things keeping me here. :P Note that the build should also work on older OSes, like Windows XP/Vista/7.)

Build system screenshot:
http://imageshack.us/a/img40/7412/screenshotminerbuilder.png (http://imageshack.us/photo/my-images/40/screenshotminerbuilder.png/)

Built with Visual Studio 2010 Pro, Python 2.7.3, Cg 2.0, Boost C++ 1.51, Numpy 1.7.0, and Brook+ 1.4.1 beta on Windows 8 x86

Wow..


I have it compiled for Windows XP x86/x64, Windows Server 2003 x86/x64, Windows Vista x64, Windows 7 x64

If anyone need it I'll supply it for FREE!

Just pm me ; )

p.s.

I've also found a way to enable OpenCL on 38x00 series, if a1 interested - PM

The community would appreciate if you share your knowledge. Especially when it comes to OpenCL on HD3000 series.