Bitcoin Forum

Alternate cryptocurrencies => Altcoin Discussion => Topic started by: primedigger on July 19, 2013, 12:40:42 PM



Title: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: primedigger on July 19, 2013, 12:40:42 PM
The amount of greed surrounding a possible GPU primecoin miner seems to be phenomenal.
Too many people seem to fall for "send me bitcoin and get early access". Sorry, but soliciting payment for early access is just dirty.

Let stop that bullshit - I'm going to share my code, even though my miner isn't fully functional right now (but I think I've already taken it quite far).

Github link: http://github.com/primedigger/primecoin

I repeat, THE CODE DOESN'T RUN CORRECTLY CURRENTLY AND CRASHES, this is just for developers who want to join and help and to show that there is a transparent development.

My general plan for porting this to CUDA:

I couldn't get getblocktemplate to work, so I added CUDA directly to the qt-client.  
Let the CPU handle the candidate search, candidates get send to the GPU and the GPU acts as a co-processor that only does "ProbablePrimeChainTest" very fast. This makes it easier to have a proof-of-concept GPU miner soon.

What I did so far :

I started with the latest high performance client (hp4).

- Ported the code path in "ProbablePrimeChainTest" so that it runs with pure mpz_t like functions and minimised the number of functions that are needed. I compiled this successfully against the big integer library in https://github.com/dmatlack/cuda-rsa.

- Changed code in "MineProbablePrimeChain" so that candidates are collected for ProbablePrimeChainTest. I made sure to measure that candidate collection is much faster than testing them in "ProbablePrimeChainTest" and this will put a theoretical limit on the speed up. If I didn't do a mistake while measuring it, the upper limit is somwhere in the 100x-1000x range. So this should be a viable approach.

- Candidates are transfered to the GPU as hex char* strings. The big integer library in cuda-rsa has "mpz_set_str", but unfortunately that doesn't work on the GPU as of now. It might be better to produce the mpz format that the GPU needs directly on the CPU instead of parsing strings. Note: Later on, transfers to the GPU can be made async, so that CPU and GPU mine in tandem.

(I also changed the sieve of Erastothenes to sieve of Atkin - I had that code flying around anyway - , but that has nothing to do with the GPU.)

The hardest part is having a reliable big integer library for CUDA. Thats why there is no working GPU miner yet. The one in https://github.com/dmatlack/cuda-rsa needs more testing and someone could work on this independently from this project. It was the best library I could find for big integer+modulo arithmetic. If you know a better one let me know.

What I like: One way or another we will end up with a highly optimised big integer library for GPUs. That is something big on its own!

 Stop sending your money to someone claiming to give you early access - my guess, the first functional GPU miners won't be very fast anyway (e.g. single digit speedup compared to hp4). An unoptimised big integer library won't outperform GMP by much.
 
P.s. You should call the binary with "-printtoconsole -printmining -testnet" to debug. Also I'm developing this under linux 64bit with standard paths. I only updated the qtcreator project file, so if you have CUDA working under a 64bit linux you should be able to build the project with qtcreator.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: bidji29 on July 19, 2013, 12:50:36 PM
Thanks for sharing


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: tomtommy on July 19, 2013, 12:51:22 PM
+1


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: junglist.massive on July 19, 2013, 12:58:47 PM
lets try it


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: Chemisist on July 19, 2013, 01:06:35 PM
Thanks for sharing!  Looking through your Weave() algo, might want to replace the "break" in line 841 with a "return false" since if a new block is detected, I'm pretty sure you'd want to terminate the Weave() algo immediately.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: Vorksholk on July 19, 2013, 01:19:17 PM
Awesome :D Send 50 Primecoins you way.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: TheSwede75 on July 19, 2013, 01:21:03 PM
Accepting donations for "early access" is about as close to an outright scam as you can get, and I NEVER expected the person behind reaper etc. to pull something as cheap.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: Vorksholk on July 19, 2013, 01:25:40 PM
Accepting donations for "early access" is about as close to an outright scam as you can get, and I NEVER expected the person behind reaper etc. to pull something as cheap.

It's essentially buying software. People give him money, he gives them a product. Anyone is perfectly entitled to buy, the only prerequisite is one Bitcoin, or less than $100. People in many fields purchase software and make money from the product of the software, think CAD, think Graphics Design, think electrical engineering, think multimedia...

Everyone freaks out when someone doesn't do something for free, it's a bit sad. The project will be released free, but for those who want early access, they can buy the software. He's selling a product. And taking pre-orders.

EDIT: Also, a 'scam' is generally when the person who pays money doesn't get a product. Sure we can't prove this is the true reaper developer, but if it is (and chances are high!), then it won't be a 'scam,' but rather not giving the software away to everyone. He wants to be paid for his work, but he still wants to be able to release the software as open source and support it as freeware. Mrtlt has earned like 21 Bitcoins, or around $2,000 for his efforts. If it takes him just 40 hours to develop the software, he's getting paid less than most top-notch programmers.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: Stalast on July 19, 2013, 01:29:48 PM
This guy. This guy is a good person.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: mikaelh on July 19, 2013, 01:35:09 PM
Oh, wow. You integrated CUDA code into the Primecoin client. I'll take a closer look later. Maybe I can even get it working. ;)


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: ig0tik3d on July 19, 2013, 01:38:41 PM
Oh, wow. You integrated CUDA code into the Primecoin client. I'll take a closer look later. Maybe I can even get it working. ;)
"In mikaelh We Trust" ))


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: vingaard on July 19, 2013, 02:03:01 PM
Thanks  ;)


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: mustyoshi on July 19, 2013, 02:04:09 PM
Using GPU for primality testing is the logical first step.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: billotronic on July 19, 2013, 02:12:24 PM
+1


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: Vorksholk on July 19, 2013, 02:19:22 PM
I'm sure OP is already aware, but I have a bounty for 6 BTC for a primecoin CUDA miner, which this is well on the way to earning. Bounty of course requires a public release of the code and binaries, which it looks like OP is going to provide no problem.   


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: Remember remember the 5th of November on July 19, 2013, 02:21:54 PM
Why all the attention to CUDA?


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: blastbob on July 19, 2013, 02:22:24 PM
Good post to get prices unstable :)


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: 6strings on July 19, 2013, 02:25:01 PM
Awesome.
So does that mean I have to order an Nvidia card now rather than the 7950 I was going to buy with the BTC I made last night off XPM? lol


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: Vorksholk on July 19, 2013, 02:27:11 PM
Why all the attention to CUDA?

Based on previous prime-number-based-research projects, CUDA has outperformed OpenCL.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: svirus on July 19, 2013, 02:43:12 PM
anyone try this and find block ?


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: mikaelh on July 19, 2013, 02:59:32 PM
Awesome.
So does that mean I have to order an Nvidia card now rather than the 7950 I was going to buy with the BTC I made last night off XPM? lol

Before anyone actually starts ordering new GPUs, I'd like to say that I see no reason why CUDA should outperform OpenCL. The current mining code is mostly doing integer math. As most people should know from Bitcoin/Litecoin mining, GPUs from AMD are very good at integer math. I would advise waiting until properly optimized CUDA and OpenCL miners are available until you decide which GPUs you will be buying in the future.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: TheSwede75 on July 19, 2013, 03:04:34 PM
Accepting donations for "early access" is about as close to an outright scam as you can get, and I NEVER expected the person behind reaper etc. to pull something as cheap.

It's essentially buying software. People give him money, he gives them a product. Anyone is perfectly entitled to buy, the only prerequisite is one Bitcoin, or less than $100. People in many fields purchase software and make money from the product of the software, think CAD, think Graphics Design, think electrical engineering, think multimedia...

Everyone freaks out when someone doesn't do something for free, it's a bit sad. The project will be released free, but for those who want early access, they can buy the software. He's selling a product. And taking pre-orders.

EDIT: Also, a 'scam' is generally when the person who pays money doesn't get a product. Sure we can't prove this is the true reaper developer, but if it is (and chances are high!), then it won't be a 'scam,' but rather not giving the software away to everyone. He wants to be paid for his work, but he still wants to be able to release the software as open source and support it as freeware. Mrtlt has earned like 21 Bitcoins, or around $2,000 for his efforts. If it takes him just 40 hours to develop the software, he's getting paid less than most top-notch programmers.


I have absolutely nothing against someone charging for software. What is shady is the idea of paying for priority access of a software with unknown specs, exclusive for people who are willing to 'risk' money on a gamble.

Just does not smell right. The idea of this project, a bounty and open development is far more open and honest in my opinion.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: Tamis on July 19, 2013, 03:13:36 PM
Are you planning to release windows binaries ?


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: oroqen on July 19, 2013, 03:21:45 PM
Awesome.
So does that mean I have to order an Nvidia card now rather than the 7950 I was going to buy with the BTC I made last night off XPM? lol

Before anyone actually starts ordering new GPUs, I'd like to say that I see no reason why CUDA should outperform OpenCL. The current mining code is mostly doing integer math. As most people should know from Bitcoin/Litecoin mining, GPUs from AMD are very good at integer math. I would advise waiting until properly optimized CUDA and OpenCL miners are available until you decide which GPUs you will be buying in the future.
I refer you too http://bitcoin.stackexchange.com/questions/9854/why-do-amds-gpus-mine-faster-than-nvidias

if the theory there is correct then trying to compair opencl and cuda prime searching is moot until we have some working code to test and optimize


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: eCoinomist on July 19, 2013, 03:23:30 PM
The latest client is hp5


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: primedigger on July 19, 2013, 03:28:56 PM
Right who knows! I'm only doing this in CUDA because: I have a Nvidia graphics card and I know CUDA fairly well.

It's not even sure that GPU mining primecoins will be significantly faster. And as I said, getting big integer math on GPU right isn't exactly easy to do. If ATI or NVIDIA is faster with this, pure speculation.

Thanks for all the support!!!  I hope to have something functional soon, but progress over the weekend will be slower.

Mikaelh, since my code is a fork of your client... whats the status of your github page?
https://github.com/mikaelh2/primecoin

Or did you move permanently to bitbucket / or sourceforge with this?


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: mikaelh on July 19, 2013, 03:33:55 PM
Mikaelh, since my code is a fork of your client... whats the status of your github page?
https://github.com/mikaelh2/primecoin

Or did you move permanently to bitbucket / or sourceforge with this?

Github support hasn't responded to me. Right now I'm forced to use Bitbucket as a replacement.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: gudmunsn on July 19, 2013, 04:09:30 PM
you sir, are awesome. I will click the trust in your name to give you some :)


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: kderaymond on July 19, 2013, 04:38:39 PM
Very excited to see this. Thank you for all your hard work, it is very much appreciated.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: Eli0t on July 19, 2013, 04:49:30 PM
awesome work! cant wait  ;D


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: xelite on July 19, 2013, 05:04:09 PM
Please compile source code for windows.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: TheSwede75 on July 19, 2013, 05:08:46 PM
FYI I believe there are already large prime libraries and Wuite a bit of CUDA code on Mersenne a prime solving that "should" be able to plug into a CUDA miner.

I have a few nVidia cards I would be happy to test early betas on.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: ReCat on July 19, 2013, 06:43:58 PM
Wow. Good job on publicizing that. Too bad I don't own any CUDA cards. :( Can buy one, though. Been wanting to for a while just for the sake of using blender cycles.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: lemons on July 19, 2013, 06:48:11 PM
+1


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: kimosan on July 19, 2013, 06:58:45 PM
I have a few nVidia cards I would be happy to test early betas on.

Ditto.

Thank you for taking on this project.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: TheSwede75 on July 19, 2013, 07:08:10 PM
I have a few nVidia cards I would be happy to test early betas on.

Ditto.

Thank you for taking on this project.

Have to mention that I prefer a Win 7 x64 build over Linux since I run Win 7 native and VM's tend to be quite a bit slower.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: nhminer on July 19, 2013, 07:38:50 PM
I have a few nVidia cards I would be happy to test early betas on.

Ditto.

Thank you for taking on this project.

DoubleDitto


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: 6strings on July 19, 2013, 08:26:23 PM
I also have several nvidia cards around to test on. Just 2x 240's but hey its testing isn't it? lol


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: nhminer on July 19, 2013, 08:27:53 PM
I also have several nvidia cards around to test on. Just 2x 240's but hey its testing isn't it? lol

I have a 550Ti I can run it on


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: Stinky_Pete on July 19, 2013, 08:30:17 PM
I also have several nvidia cards around to test on. Just 2x 240's but hey its testing isn't it? lol

I have a 550Ti I can run it on
590 and 680 ready and waiting


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: gateway on July 19, 2013, 08:42:41 PM
have you seen the cuda gmp pdf?
http://individual.utoronto.ca/haojunliu/courses/ECE1724_Report.pdf

cheers on not being greedy, btw i have many nvidia cards sitting idle here in our office :)



Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: TheSwede75 on July 19, 2013, 08:46:59 PM
have you seen the cuda gmp pdf?
http://individual.utoronto.ca/haojunliu/courses/ECE1724_Report.pdf

cheers on not being greedy, btw i have many nvidia cards sitting idle here in our office :)



Very much agree. A public release to a larger group of people is the best for Primecoin.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: loloman on July 19, 2013, 09:45:22 PM
what about the perfs of this client?


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: Kouye on July 19, 2013, 09:49:50 PM
what about the perfs of this client?
Insane perfs. As soon as you gracefully push your BigInt lightning fast and reliable cuda library.
Any ETA on that ?


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: gateway on July 19, 2013, 10:10:26 PM
anyone here working with him? .. seems like he wants devs to step up and work on a client with him, asking if its done or something is prob not the right thing here..


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: Sunny King on July 19, 2013, 10:21:15 PM
Great initiative and looking forward  :)


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: sumantso on July 19, 2013, 10:27:16 PM
Nice work OP. I will just wait for the nice folks to provide a one-click Windows 7 miner  8)

Meanwhile I have to find my nVIDIA GPU and see whether its alive. I think it was a 8400.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: primedigger on July 20, 2013, 12:37:07 PM
Github-link: http://github.com/primedigger/primecoin

update: added a stroxl function from http://www.jbox.dk/sanos/

The code still needs a lot of debugging and the mpz cuda library + string functions a lot of testing, if any CUDA developer wants to give it a chance, go for it!

Weather is too good outside to continue this now...


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: alexxy on July 20, 2013, 01:11:28 PM
Github-link: http://github.com/primedigger/primecoin

update: added a stroxl function from http://www.jbox.dk/sanos/

The code still needs a lot of debugging and the mpz cuda library + string functions a lot of testing, if any CUDA developer wants to give it a chance, go for it!

Weather is too good outside to continue this now...

Code compiles with some makefiles modifications but fails near imidiently with memory errors

Code:

Cuda finished
Have 250 candidates after main loop
Cuda finished
Have 344 candidates after main loop
Cuda finished
Have 312 candidates after main loop
Cuda finished
Have 1235 candidates after main loop
Cuda error: cudaMemcpy: cudaMemcpyDeviceToHost, unspecified launch failure


Also current code will only works with sm_20 and higher devices


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: alexxy on July 20, 2013, 01:25:20 PM
Aha

Seems like it starts failing if more then 1000 prime chain candidates found

Code:
Cuda finished
Added 361 addresses from 199.83.89.217: 180 tried, 12837 new
Added 199 addresses from 199.83.89.217: 180 tried, 12827 new
Have 1314 candidates after main loop
[0] start
[0] candidate is 1edf6ebf2abf2b6396b2f78eacddc21a2f9c0b963356d861e2035aa210ed01e094668
[0] mpz init done!


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: tonyback on July 20, 2013, 01:48:20 PM
thanks for this topic


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: primedigger on July 20, 2013, 01:52:31 PM
The 1000 is an arbitrary limit I put in so that enough candidates are available, thats when the GPU is used the first time.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: alexxy on July 20, 2013, 01:54:40 PM
The 1000 is an arbitrary limit I put in so that enough candidates are available, thats when the GPU is used the first time.

Are you interesting in makefiles for cuda enabled version?

PS is it intened to allow it to work only for sm_20+ devices?


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: TheSwede75 on July 20, 2013, 01:57:47 PM
Very interesting. Would love to test out once Windows binaries are available.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: Vorksholk on July 20, 2013, 02:02:39 PM
I have a 780 I can test on, even if it's just builds that aim to test for performance, not actually mine blocks. Sadly my limited CUDA knowledge doesn't serve me very well here. :(


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: hak8or on July 20, 2013, 02:06:33 PM
This is absolutely awesome! I was wondering why so many were falling for that 1 BTC early access thing, and even without any escrow! I know I should have started to learn cuda months ago, but I chose Ruby for RoR instead, heh.

I will see where I can help out with the non cuda stuff, or maybe throwing in some comments here and there so other dev's can catch up easier, all while learning some cuda on the way. As others have said, you have quite a few people willing to test things out for you. If interested, I might be able to provide you root access to a machine running a 570 GTX.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: TheSwede75 on July 20, 2013, 02:23:54 PM
This is absolutely awesome! I was wondering why so many were falling for that 1 BTC early access thing, and even without any escrow! I know I should have started to learn cuda months ago, but I chose Ruby for RoR instead, heh.

I will see where I can help out with the non cuda stuff, or maybe throwing in some comments here and there so other dev's can catch up easier, all while learning some cuda on the way. As others have said, you have quite a few people willing to test things out for you. If interested, I might be able to provide you root access to a machine running a 570 GTX.

+1

I have a GTX 475 SLI 2x setup I can provide root access to as well.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: primedigger on July 20, 2013, 02:26:04 PM
If someone wants to help, we need test cases for the cuda mpz functions to be sure that the cuda mpz library works (it's not mine). Also debugging this with gdb (the one from nvidia that can step through gpu execution) would be insightful, to see at which code line it crashes.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: alexxy on July 20, 2013, 02:37:00 PM
If someone wants to help, we need test cases for the cuda mpz functions to be sure that the cuda mpz library works (it's not mine). Also debugging this with gdb (the one from nvidia that can step through gpu execution) would be insightful, to see at which code line it crashes.


I can try debuggin it on linux with cuda 5.5

Btw why didnt you use cump library? it seems to be gmp (mpz) compatible


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: hak8or on July 20, 2013, 02:57:09 PM
If someone wants to help, we need test cases for the cuda mpz functions to be sure that the cuda mpz library works (it's not mine). Also debugging this with gdb (the one from nvidia that can step through gpu execution) would be insightful, to see at which code line it crashes.

Sadly I haven't the faintest how to work with GDB. Also, adding on to my root access offer, I just remembered that Amazon offers GPU instances! For the life of me I cannot find the GPU option in the instance list though. If anyone gets back to me about this, I can happily offer you a donation to pay for a bit of that instance or I can try to set it up for you.

https://aws.amazon.com/ec2/instance-types/


Quote
CG1 Instances are backed by 2 x Intel Xeon X5570, quad-core with hyperthread plus 2 NVIDIA Tesla M2050 GPUs


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: Lauda on July 20, 2013, 02:58:52 PM
How is this project going? I like it   8)


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: x3maniac on July 20, 2013, 06:51:42 PM
If someone wants to help, we need test cases for the cuda mpz functions to be sure that the cuda mpz library works (it's not mine). Also debugging this with gdb (the one from nvidia that can step through gpu execution) would be insightful, to see at which code line it crashes.

Thanks for sharing your code I am also using your base and coding in vs2010 for windows. I will have a git  later but I'm on a different approach with a stand alone miner. Not integrated to primecoind. I will keep u posted. Main thing that different is I don't use gmp rather I am staying with bignums for now


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: primedigger on July 20, 2013, 11:20:54 PM

Code compiles with some makefiles modifications but fails near imidiently with memory errors


Ah sorry my bad, I forgot to commit the makefile changes... but they should be up now


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: hasle2 on July 21, 2013, 03:35:58 AM
I posted a run down of some will and will not change when gpu miners are released in the other miner thread:

https://bitcointalk.org/index.php?topic=258540.msg2771239#msg2771239

Hope it helps some people.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: dadaguo on July 21, 2013, 03:50:17 AM
If someone wants to help, we need test cases for the cuda mpz functions to be sure that the cuda mpz library works (it's not mine). Also debugging this with gdb (the one from nvidia that can step through gpu execution) would be insightful, to see at which code line it crashes.

Sadly I haven't the faintest how to work with GDB. Also, adding on to my root access offer, I just remembered that Amazon offers GPU instances! For the life of me I cannot find the GPU option in the instance list though. If anyone gets back to me about this, I can happily offer you a donation to pay for a bit of that instance or I can try to set it up for you.

https://aws.amazon.com/ec2/instance-types/


Quote
CG1 Instances are backed by 2 x Intel Xeon X5570, quad-core with hyperthread plus 2 NVIDIA Tesla M2050 GPUs


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: Entz on July 21, 2013, 03:58:33 AM
Btw why didnt you use cump library? it seems to be gmp (mpz) compatible

CUMP is mpf not mpz, unfortunately (if it was a miner would already be done)


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: primedigger on July 21, 2013, 12:01:57 PM
Btw why didnt you use cump library? it seems to be gmp (mpz) compatible

CUMP is mpf not mpz, unfortunately (if it was a miner would already be done)

Exactly! It's floating point only and that doesn't help for big int+modulo math. GPU mining primecoins will create an optimized big int GPU library as a nice side effect.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: LZ on July 22, 2013, 01:29:18 AM
Looks interesting. I'll try to compile it.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: 1l1l11ll1l on July 22, 2013, 03:44:41 AM
Excellent, 47 K10's and 14 K20's get bored and lonely evenings and weekends.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: jaakkop on July 22, 2013, 06:53:57 AM
Looking forward to windows binaries. I'll make sure to tip some once they go live :)


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: primedigger on July 23, 2013, 09:23:39 AM
Recent update with some fixes for the GPU :

- No more crashes on the GPU
- Fermat test seems to run fine
- Numbers are still sometimes truncated when parsed from strings

But we're getting there!

Help me to stay motivated with some XPM: AeB7voX4imYfA7T3V1PJgB3vV5jU526tz1

And happy hacking if you're trying this out.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: crendore on July 23, 2013, 09:57:05 AM
- Numbers are still sometimes truncated when parsed from strings

Sounds like you need a better library for int. something that handles really really big ints.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: primedigger on July 23, 2013, 10:14:07 AM
- Numbers are still sometimes truncated when parsed from strings

Sounds like you need a better library for int. something that handles really really big ints.

Imho there is no better library out there for GPUs. I'm hacking on the mpz library from a cuda-rsa student project, which is in my repo (https://github.com/primedigger/primecoin/blob/master/src/cuda/mpz.h). It uses its own structures for big ints and that's why you can't exactly load gmp mpz_t into the GPU - I'm using hex strings for that. The library had a function for loading numbers from hex strings, but unfortunately that function is a bit buggy. I guess in the long run, it would be wiser to assemble the GPU's big int data structures on the CPU.



Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: dreamwatcher on July 23, 2013, 10:31:49 AM
I specifically bought a Nvidia card for my main computer so I would not worry about mining with it when not using it.

Now this...dammit  :D


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: maco on July 23, 2013, 10:40:27 AM
I may want to try this.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: Joe13 on July 23, 2013, 10:50:20 AM
hi
if you guys dont mind i would also try it on a nvidia card ...
thx


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: primedigger on July 23, 2013, 11:03:58 AM
hi
if you guys dont mind i would also try it on a nvidia card ...
thx

Of course, if you want to hack on this - go ahead! Just to clarify: The project isn't in a stage were you can start mining with this, but we're not far away from that either.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: karsy on July 23, 2013, 04:28:13 PM
Precompiled windows version please! All of us don't have compilers handy you know.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: DiThi on July 23, 2013, 04:40:21 PM
Precompiled windows version please! All of us don't have compilers handy you know.

Why do you want a miner that can't mine yet?


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: karsy on July 23, 2013, 04:42:52 PM
Precompiled windows version please! All of us don't have compilers handy you know.

Why do you want a miner that can't mine yet?
Oh yeah, didn't see that until now.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: azwccc on July 23, 2013, 11:10:43 PM
I would definitely donate if the CUDA miner can outperform the AMD OCL miner (on GPU in same price range).

Best of luck to you.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: jaakkop on July 24, 2013, 05:52:16 AM
Donated some BTC just now, keep up the good work :)

BTW, are there any guides how to compile this for Windows (or how to cross-compile with Linux?)
or should Bitcoin-QT building guides apply to this as well?


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: wetroof on July 24, 2013, 06:03:48 AM
if I do mine with this with this, I'll pledge to give 10% of the total XPM I mine for the first 5 days to primedigger. I have two 2.0 cuda compute capability cards and one 3.0.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: teknohog on July 24, 2013, 09:21:44 AM
Based on previous prime-number-based-research projects, CUDA has outperformed OpenCL.

AFAIK, there are two reasons why the Lucas-Lehmer test for Mersenne primes has been done in CUDA rather than OpenCL. It uses floating point math, and there are better FFT libraries available for CUDA. (For the huge numbers involved, multiplication is more efficient via Fourier transform and convolution.)

On the other hand, trial division (another important part of Mersenne prime search) seems to be more efficient with OpenCL on AMD, as it uses integer math. I assume Primecoin would work fine with just integer math.

Sources: http://mersenneforum.org/forumdisplay.php?f=92

Of course, there is a more ideological point about using a language that works across many platforms (CPUs and GPUs from several vendors) than tying yourself to Nvidia, especially in an open source project.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: maco on July 24, 2013, 09:35:40 AM
How is progress so far? CUDA is a must :)


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: primedigger on July 24, 2013, 02:55:20 PM
Progress update:

- GPU does Fermats tests now and CPU does the rest. Fermat tests seem to work fine now.
- I can find prime chains, but couldn't find a block on testnet in a timely manner

Todo:
- Transferring mpz types with strings is slow so I will transform gmp mpz's directly to my CUDA format on the CPU.
- A lot happenend with the high performance client, I will update my codebase to hp7
- The changes in hp6 could also be useful on the GPU ( -> fast divisibility tests before doing the expensive Fermat's test)
- Interleave CPU+GPU computations and async memory copys. Without this, my client won't be very fast.

I don't like to put an ETA on this, it's done when it's done. But I hope to have something by next week which outperforms my old intel core2 quad core.

My 2 cents: mining entirely on GPU wont be easy and is impractical, but tandem mining with interleaved CPU+GPU computations may very well give good speed ups.

Edit: And thanks for the donations / pledges guys!


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: hasle2 on July 24, 2013, 07:30:02 PM
Progress update:

- GPU does Fermats tests now and CPU does the rest. Fermat tests seem to work fine now.
- I can find prime chains, but couldn't find a block on testnet in a timely manner

Todo:
- Transferring mpz types with strings is slow so I will transform gmp mpz's directly to my CUDA format on the CPU.
- A lot happenend with the high performance client, I will update my codebase to hp7
- The changes in hp6 could also be useful on the GPU ( -> fast divisibility tests before doing the expensive Fermat's test)
- Interleave CPU+GPU computations and async memory copys. Without this, my client won't be very fast.

I don't like to put an ETA on this, it's done when it's done. But I hope to have something by next week which outperforms my old intel core2 quad core.

My 2 cents: mining entirely on GPU wont be easy and is impractical, but tandem mining with interleaved CPU+GPU computations may very well give good speed ups.

Edit: And thanks for the donations / pledges guys!

The fact that you are finding blocks on testnet is quite an achievement. Keep up the good work :)


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: wetroof on July 24, 2013, 09:13:52 PM
Progress update:

- GPU does Fermats tests now and CPU does the rest. Fermat tests seem to work fine now.
- I can find prime chains, but couldn't find a block on testnet in a timely manner

Todo:
- Transferring mpz types with strings is slow so I will transform gmp mpz's directly to my CUDA format on the CPU.
- A lot happenend with the high performance client, I will update my codebase to hp7
- The changes in hp6 could also be useful on the GPU ( -> fast divisibility tests before doing the expensive Fermat's test)
- Interleave CPU+GPU computations and async memory copys. Without this, my client won't be very fast.

I don't like to put an ETA on this, it's done when it's done. But I hope to have something by next week which outperforms my old intel core2 quad core.

My 2 cents: mining entirely on GPU wont be easy and is impractical, but tandem mining with interleaved CPU+GPU computations may very well give good speed ups.

Edit: And thanks for the donations / pledges guys!

I donated .2 BTC . thanks for the update.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: jhxhlj on July 25, 2013, 01:59:10 AM
Any update? ;D ;D ;D ;D ;D ;D ;D ;D


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: Sunny King on July 25, 2013, 02:12:02 AM
My 2 cents: mining entirely on GPU wont be easy and is impractical, but tandem mining with interleaved CPU+GPU computations may very well give good speed ups.

Some feedback from knowledgeable people indicates that mod_exp probably would not speed up as well on gpu. However I think if gpu can do the sieve much more efficiently it could generate a lot less candidates for the Fermat test, which could speed things up quite a bit.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: primedigger on July 25, 2013, 01:06:02 PM
My 2 cents: mining entirely on GPU wont be easy and is impractical, but tandem mining with interleaved CPU+GPU computations may very well give good speed ups.

Some feedback from knowledgeable people indicates that mod_exp probably would not speed up as well on gpu. However I think if gpu can do the sieve much more efficiently it could generate a lot less candidates for the Fermat test, which could speed things up quite a bit.

There is indeed a problem with the speeds of Fermat tests on the GPU. GNU GMP uses the most sophisticated algorithms available, the student library I found and which I started to extend uses the most basic algorithms.

Mpz_powmod needs fast multiplications of big ints, GMP's algorithm is most likely in O(log(n)*n), school book multiplication which the GPU now uses is O(n^2). I hoped that for the ~400 bit numbers involved it wouldn't make such a difference. Currently, the new version in my repo does Fermat tests on the GPU (rate is 10 per second), but my CPU is still faster due to better algorithms and a better big num implementation.

But don't worry, I won't give up so fast! The current situation is that I either need to look into porting better algorithms to the GPU or to do something else than Fermat tests on the GPU to sieve candidates (e.g. trial division with most common primes).

Anybody with a better GPU than the Geforce 570 TI I own, please test this! My stats (base version is still hp4):

2013-07-24 21:53:38 primemeter     24303 prime/h    490729 test/h        47 5-chains/h

prime/h  and test/h seem to fluctuate enormously and seem to be rather meaningless. As most tests are on the GPU, I have no idea if this is even measuring the tests right. 5-chains is accurate though.

You have to use setgenerate true 1, i.e. one CPU thread for mining. 


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: paulthetafy on July 25, 2013, 01:13:24 PM
Good job primedigger!  Just looking at your 5-chains score, I think that means that this is slower than most CPU's though right?  My crappy 4-core AMD does 400-800 5-chains/ hour (around 3000PPS).


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: primedigger on July 25, 2013, 01:20:57 PM
Good job primedigger!  Just looking at your 5-chains score, I think that means that this is slower than most CPU's though right?  My crappy 4-core AMD does 400-800 5-chains/ hour (around 3000PPS).

Yes, it's an extremely crappy score, but it's a start. The high performance client does a good job of squeezing out as much as possible out of your CPU though!


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: K1773R on July 25, 2013, 01:31:53 PM
My 2 cents: mining entirely on GPU wont be easy and is impractical, but tandem mining with interleaved CPU+GPU computations may very well give good speed ups.

Some feedback from knowledgeable people indicates that mod_exp probably would not speed up as well on gpu. However I think if gpu can do the sieve much more efficiently it could generate a lot less candidates for the Fermat test, which could speed things up quite a bit.

There is indeed a problem with the speeds of Fermat tests on the GPU. GNU GMP uses the most sophisticated algorithms available, the student library I found and which I started to extend uses the most basic algorithms.

Mpz_powmod needs fast multiplications of big ints, GMP's algorithm is most likely in O(log(n)*n), school book multiplication which the GPU now uses is O(n^2). I hoped that for the ~400 bit numbers involved it wouldn't make such a difference. Currently, the new version in my repo does Fermat tests on the GPU (rate is 10 per second), but my CPU is still faster due to better algorithms and a better big num implementation.

But don't worry, I won't give up so fast! The current situation is that I either need to look into porting better algorithms to the GPU or to do something else than Fermat tests on the GPU to sieve candidates (e.g. trial division with most common primes).

Anybody with a better GPU than the Geforce 570 TI I own, please test this! My stats (base version is still hp4):

2013-07-24 21:53:38 primemeter     24303 prime/h    490729 test/h        47 5-chains/h

prime/h  and test/h seem to fluctuate enormously and seem to be rather meaningless. As most tests are on the GPU, I have no idea if this is even measuring the tests right. 5-chains is accurate though.

You have to use setgenerate true 1, i.e. one CPU thread for mining.  
running current git (b0c7062f3925482935f5eb352b17737d21b95c5b) and i cant see any usage of my GPU, no heat nor used mem increases when using the QT. anything special to activate so it mines with the GPU? i got a powerfull GPU to test with ;)

EDIT:
Code:
2013-07-25 13:35:43 primemeter         0 prime/h   34261932 test/h         0 5-chains/h
seems the miner thread which should launch the CUDA is borked?

EDIT2:
Code:
Have 2400 candidates after main loop
Cuda start!
Cuda error: cudaMemcpy: cudaMemcpyDeviceToHost, the launch timed out and was terminated
from debug.log
after the message it segfaults, going to debug with gdb ;)


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: primedigger on July 25, 2013, 01:45:44 PM
My 2 cents: mining entirely on GPU wont be easy and is impractical, but tandem mining with interleaved CPU+GPU computations may very well give good speed ups.

Some feedback from knowledgeable people indicates that mod_exp probably would not speed up as well on gpu. However I think if gpu can do the sieve much more efficiently it could generate a lot less candidates for the Fermat test, which could speed things up quite a bit.

There is indeed a problem with the speeds of Fermat tests on the GPU. GNU GMP uses the most sophisticated algorithms available, the student library I found and which I started to extend uses the most basic algorithms.

Mpz_powmod needs fast multiplications of big ints, GMP's algorithm is most likely in O(log(n)*n), school book multiplication which the GPU now uses is O(n^2). I hoped that for the ~400 bit numbers involved it wouldn't make such a difference. Currently, the new version in my repo does Fermat tests on the GPU (rate is 10 per second), but my CPU is still faster due to better algorithms and a better big num implementation.

But don't worry, I won't give up so fast! The current situation is that I either need to look into porting better algorithms to the GPU or to do something else than Fermat tests on the GPU to sieve candidates (e.g. trial division with most common primes).

Anybody with a better GPU than the Geforce 570 TI I own, please test this! My stats (base version is still hp4):

2013-07-24 21:53:38 primemeter     24303 prime/h    490729 test/h        47 5-chains/h

prime/h  and test/h seem to fluctuate enormously and seem to be rather meaningless. As most tests are on the GPU, I have no idea if this is even measuring the tests right. 5-chains is accurate though.

You have to use setgenerate true 1, i.e. one CPU thread for mining.  
running current git (b0c7062f3925482935f5eb352b17737d21b95c5b) and i cant see any usage of my GPU, no heat nor used mem increases when using the QT. anything special to activate so it mines with the GPU? i got a powerfull GPU to test with ;)

EDIT:
Code:
2013-07-25 13:35:43 primemeter         0 prime/h   34261932 test/h         0 5-chains/h
seems the miner thread which should launch the CUDA is borked?

EDIT2:
Code:
Have 2400 candidates after main loop
Cuda start!
Cuda error: cudaMemcpy: cudaMemcpyDeviceToHost, the launch timed out and was terminated
from debug log

You can also run it with -printmining -printtoconsole to see that output directly. Could you compile the cuda portion with -G -g (change the qt project file where it invokes nvcc) and give me the output of cuda-memcheck?

You can also #define CUDA_DEBUG in the cu file, to see the GPU printfs from the console


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: Vorksholk on July 25, 2013, 01:47:59 PM
If someone could give me some specific compilation directions (or a windows binary!) I can test on a 780. :)


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: K1773R on July 25, 2013, 01:48:56 PM
My 2 cents: mining entirely on GPU wont be easy and is impractical, but tandem mining with interleaved CPU+GPU computations may very well give good speed ups.

Some feedback from knowledgeable people indicates that mod_exp probably would not speed up as well on gpu. However I think if gpu can do the sieve much more efficiently it could generate a lot less candidates for the Fermat test, which could speed things up quite a bit.

There is indeed a problem with the speeds of Fermat tests on the GPU. GNU GMP uses the most sophisticated algorithms available, the student library I found and which I started to extend uses the most basic algorithms.

Mpz_powmod needs fast multiplications of big ints, GMP's algorithm is most likely in O(log(n)*n), school book multiplication which the GPU now uses is O(n^2). I hoped that for the ~400 bit numbers involved it wouldn't make such a difference. Currently, the new version in my repo does Fermat tests on the GPU (rate is 10 per second), but my CPU is still faster due to better algorithms and a better big num implementation.

But don't worry, I won't give up so fast! The current situation is that I either need to look into porting better algorithms to the GPU or to do something else than Fermat tests on the GPU to sieve candidates (e.g. trial division with most common primes).

Anybody with a better GPU than the Geforce 570 TI I own, please test this! My stats (base version is still hp4):

2013-07-24 21:53:38 primemeter     24303 prime/h    490729 test/h        47 5-chains/h

prime/h  and test/h seem to fluctuate enormously and seem to be rather meaningless. As most tests are on the GPU, I have no idea if this is even measuring the tests right. 5-chains is accurate though.

You have to use setgenerate true 1, i.e. one CPU thread for mining.  
running current git (b0c7062f3925482935f5eb352b17737d21b95c5b) and i cant see any usage of my GPU, no heat nor used mem increases when using the QT. anything special to activate so it mines with the GPU? i got a powerfull GPU to test with ;)

EDIT:
Code:
2013-07-25 13:35:43 primemeter         0 prime/h   34261932 test/h         0 5-chains/h
seems the miner thread which should launch the CUDA is borked?

EDIT2:
Code:
Have 2400 candidates after main loop
Cuda start!
Cuda error: cudaMemcpy: cudaMemcpyDeviceToHost, the launch timed out and was terminated
from debug log

You can also run it with -printmining -printtoconsole to see that output directly. Could you compile the cuda portion with -G -g (change the qt project file where it invokes nvcc) and give me the output of cuda-memcheck?

You can also #define CUDA_DEBUG in the cu file, to see the GPU printfs from the console
was already running with -g just waiting for the "cuda start message", stoped it now and recompile with -D CUDA_DEBUG
EDIT: its up and running, waiting for the cuda init + crash ;)
EDIT2: why does it take so long until the miner starts the cuda thread? that seems stupid :S
EDIT3: here we go, it crashed :)
debug.log
Code:
Have 2400 candidates after main loop
Cuda start!
Cuda error: cudaMemcpy: cudaMemcpyDeviceToHost, unspecified launch failure
stdout
Code:
[0] start! 
sizeof(struct) = 400
mpz_print:mpz_capacity: 0
[0] string candidate is 
[0] N is: mpz_capacity: 30 ffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffff
[0] E is: mpz_capacity: 30 fffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffe
gdb: dont want to spam, sending per PM and message too big -.-


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: TheSwede75 on July 25, 2013, 02:11:30 PM
More than wiling to help perform tests as instructed if a windows binary is posted. Got an old GTX 475 rattling around that I could out to work..


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: primedigger on July 25, 2013, 03:27:25 PM
Please check that you're using the latest SDK. I also encountered memory problems with cuda 5.0 and I'm using 5.5 now which works for me.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: Schleicher on July 25, 2013, 04:14:24 PM
Would it make any difference if we use __restricted__ pointers in the CUDA code?


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: K1773R on July 25, 2013, 04:56:10 PM
Please check that you're using the latest SDK. I also encountered memory problems with cuda 5.0 and I'm using 5.5 now which works for me.
ACK, will do later and report back ;)


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: Entz on July 25, 2013, 04:57:51 PM
Just got this compiled  (Talk about a mess, when my cuda sdk was installed the paths were completely different then they should of /nvidia-304 vs /nvidia-current etc, then some fun Qt conflicts).

Anyone have a working node for testnet they can post? Not having any luck connecting.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: bcp19 on July 25, 2013, 07:03:20 PM
Please check that you're using the latest SDK. I also encountered memory problems with cuda 5.0 and I'm using 5.5 now which works for me.
Just curious, have you looked at the Mfaktc source code at all?  While it is used for trial factoring Mersenne Primes, which may not be helpful, the writer did get it to sieve completely on the GPU, which may.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: Entz on July 25, 2013, 07:46:31 PM
Having the same problem as K1773R  (GTX670 using CUDA 5.5 and the driver it includes). Tried it on mainnet as I still cannot connect to testnet for some reason.

Code:
Have 101 candidates after main loop
Cuda start!
{... some block messages i.e. getblocks -1 to blah, accept etc}
Have -1 candidates after main loop
Cuda+host test round finished with -1 candidates (0 host chain tests)
Cuda error: cudaMemcpy: cudaMemcpyDeviceToHost, unspecified launch failure
ERROR: PrimecoinMiner() : primorial minimum overflow
ERROR: PrimecoinMiner() : primorial minimum overflow
ERROR: PrimecoinMiner() : primorial minimum overflow
ERROR: PrimecoinMiner() : primorial minimum overflow
ERROR: PrimecoinMiner() : primorial minimum overflow
ERROR: PrimecoinMiner() : primorial minimum overflow
ERROR: PrimecoinMiner() : primorial minimum overflow
ERROR: PrimecoinMiner() : primorial minimum overflow

from GDB
Code:
[0] start! 
sizeof(struct) = 400
mpz_print:mpz_capacity: 0
[0] string candidate is 
[0] N is: mpz_capacity: 30 ffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffff
[0] E is: mpz_capacity: 30 fffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffe

Edit: This may just be PEBKAC / RTFM issue on my part. Just saw your note about running with 1 cpu only.

Still crashed, managed to get a few rounds. Used to crash right away
Code:
2013-07-25 19:51:56 primemeter         0 prime/h    498885 test/h         0 5-chains/h
2013-07-25 19:52:56 primemeter         0 prime/h   8404040 test/h         0 5-chains/h
2013-07-25 19:53:56 primemeter         0 prime/h   4184750 test/h         0 5-chains/h


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: ReCat on July 26, 2013, 12:24:41 AM
Fascinating. This CUDA miner is already vaguely functional now? Now that's some community effort. I wonder what will be the eventual result of this. Will fast CPU's and GPU's working together be the new mining rigs?


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: liteuser on July 26, 2013, 10:09:24 AM
I've updated to cuda-5.5 (and driver 319.21)

Running with cuda-gdb I get the following error:

Code:
Have 2400 candidates after main loop
Cuda start!
[New Thread 0x7fffacc38700 (LWP 14248)]
[Context Create of context 0x7fff700234f0 on Device 0]
[Launch of CUDA Kernel 0 (runPrimeCandidateSearch<<<(25,1,1),(192,1,1)>>>) on Device 0]

Program received signal CUDA_EXCEPTION_10, Device Illegal Address.
[Switching focus to CUDA kernel 0, grid 1, block (15,0,0), thread (0,0,0), device 0, sm 3, warp 0, lane 0]
0x00007fff7091b760 in long_multiplication(unsigned int * @generic, unsigned int * @generic, unsigned int * @generic, unsigned int, unsigned int) (
    product=0x3fff6b4, op1=0x3fff734, op2=0x3fff634, num_digits=17,
    prod_capacity=1073741824)
    at primecoin/src/cuda/digit.h:406
406     product[i] = 0;


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: primedigger on July 26, 2013, 12:20:14 PM
Please check that you're using the latest SDK. I also encountered memory problems with cuda 5.0 and I'm using 5.5 now which works for me.
Just curious, have you looked at the Mfaktc source code at all?  While it is used for trial factoring Mersenne Primes, which may not be helpful, the writer did get it to sieve completely on the GPU, which may.

I looked into it, yes. Code is not very understandable though...


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: hasle2 on July 30, 2013, 03:55:19 PM
Is this project still active or has it been abandoned?


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: ReCat on July 30, 2013, 09:55:06 PM
It's abandoned. Lol. Probably everyone figured out that this is too difficult. Heck even mlmrt was having trouble.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: Kouye on July 30, 2013, 10:26:10 PM
It's abandoned. Lol. Probably everyone figured out that this is too difficult. Heck even mlmrt was having trouble.

LIES! He's managed to have the same efficiency as an AMD multi-core.

With an AMD multi-core + a HD6990.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: bcp19 on July 30, 2013, 10:40:30 PM
It's abandoned. Lol. Probably everyone figured out that this is too difficult. Heck even mlmrt was having trouble.

LIES! He's managed to have the same efficiency as an AMD multi-core.

With an AMD multi-core + a HD6990.

WOW, we can spend ~194 watts running an AMD multi-core or ~525 watts running the AMD and a GPU and get the same results!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
I WANT I WANT I WANT I WANT!


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: maco on July 30, 2013, 11:56:54 PM
has anyone tested this yet? is it working?


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: primedigger on July 31, 2013, 09:52:58 AM
I'm still on it - with a different idea. As it turns out, doing Fermat tests on the GPU is not a no brainer and getting that fast requires too much effort for now, so I'll try to port something else to the GPU.

I'm still sure a GPU miner is possible, but right now I would say it's a lot harder than for the other coins. The other OpenCL miner project is (amusingly!) also having problems.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: refer_2_me on July 31, 2013, 11:21:23 AM
I'm still on it - with a different idea. As it turns out, doing Fermat tests on the GPU is not a no brainer and getting that fast requires too much effort for now, so I'll try to port something else to the GPU.

I'm still sure a GPU miner is possible, but right now I would say it's a lot harder than for the other coins. The other OpenCL miner project is (amusingly!) also having problems.

As i'm sure you are already aware, mlmrt ported the sieve to the GPU. Is that what you are going after?



Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: ReCat on July 31, 2013, 03:39:23 PM
It's abandoned. Lol. Probably everyone figured out that this is too difficult. Heck even mlmrt was having trouble.

LIES! He's managed to have the same efficiency as an AMD multi-core.

With an AMD multi-core + a HD6990.


Wait... that's better?


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: jaakkop on August 06, 2013, 05:55:31 AM
What's the progress so far?


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: Lauda on August 06, 2013, 06:55:59 AM
It's abandoned. Lol. Probably everyone figured out that this is too difficult. Heck even mlmrt was having trouble.

LIES! He's managed to have the same efficiency as an AMD multi-core.

With an AMD multi-core + a HD6990.


Wait... that's better?
It's not..


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: ReCat on August 07, 2013, 09:39:41 PM
Forget about it guys, this miner was never gonna happen. Obviously the people who started and promoted this thread had no idea what they were getting into.

The reaper guy's miner is probably the only GPU miner we will EVER be seeing.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: refer_2_me on August 07, 2013, 09:43:06 PM
Forget about it guys, this miner was never gonna happen. Obviously the people who started and promoted this thread had no idea what they were getting into.

The reaper guy's miner is probably the only GPU miner we will EVER be seeing.

So it would seem, sadly. I hope once mrtlt's is open sourced we can get some real community development going for the good of the coin. Hopefully by that point, I will have finish the crunch at work and I can try to dive in as well.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: primedigger on August 13, 2013, 03:00:41 PM
I was away for the past week and will look into it again this week. Yes, it's just a hobby project and it got bigger than I expected. Currently, I'm the only one working on this, so if someone wants to chip in and help (programming), send me a PM.

Status:

I will push my lastest changes soon, I have updated my code basis to hp-9 and I implemented a fast big num small prime trial division for the GPU. Depending on the settings, this can filter out 10-90% of all candidates. The CPU than computes the fermat tests on the remaining candidates. I was under the impression that the sieve would already filter out all chains versus small primes, but apparently the high performance client still filters out some candidates with trial divisions and does this before doing fermat tests.

If a fast fermat test for the GPU surfaces, than filtering+fermat tests could be chained directly on the GPU to give a better speed up.

To clarify: I didn't push my changes because I still have a silly bug somewhere, so that apparently not all prime divisors are found. But doing more prime division tests than what the high performance client does by default yields already better speed ups directly on the CPU.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: ncr1pt0r on August 13, 2013, 03:25:52 PM
good news , glad to see your still working on it


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: Spoetnik on August 13, 2013, 05:13:21 PM
i am watching you ;)

CUDA !!!!!!


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: Lauda on August 13, 2013, 08:37:27 PM
good news , glad to see your still working on it
I wondered if he quit it, so we got some good news  :)


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: jaakkop on August 14, 2013, 05:46:11 AM
I was away for the past week and will look into it again this week. Yes, it's just a hobby project and it got bigger than I expected. Currently, I'm the only one working on this, so if someone wants to chip in and help (programming), send me a PM.

Status:

I will push my lastest changes soon, I have updated my code basis to hp-9 and I implemented a fast big num small prime trial division for the GPU. Depending on the settings, this can filter out 10-90% of all candidates. The CPU than computes the fermat tests on the remaining candidates. I was under the impression that the sieve would already filter out all chains versus small primes, but apparently the high performance client still filters out some candidates with trial divisions and does this before doing fermat tests.

If a fast fermat test for the GPU surfaces, than filtering+fermat tests could be chained directly on the GPU to give a better speed up.

To clarify: I didn't push my changes because I still have a silly bug somewhere, so that apparently not all prime divisors are found. But doing more prime division tests than what the high performance client does by default yields already better speed ups directly on the CPU.

Thanks for the update and keep up the good work :)


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: SynergyCores on August 14, 2013, 04:42:03 PM

I was away for the past week and will look into it again this week. Yes, it's just a hobby project and it got bigger than I expected. Currently, I'm the only one working on this, so if someone wants to chip in and help (programming), send me a PM.

Status:

I will push my lastest changes soon, I have updated my code basis to hp-9 and I implemented a fast big num small prime trial division for the GPU. Depending on the settings, this can filter out 10-90% of all candidates. The CPU than computes the fermat tests on the remaining candidates. I was under the impression that the sieve would already filter out all chains versus small primes, but apparently the high performance client still filters out some candidates with trial divisions and does this before doing fermat tests.

If a fast fermat test for the GPU surfaces, than filtering+fermat tests could be chained directly on the GPU to give a better speed up.

To clarify: I didn't push my changes because I still have a silly bug somewhere, so that apparently not all prime divisors are found. But doing more prime division tests than what the high performance client does by default yields already better speed ups directly on the CPU.

I only wish that I knew enough to help, but as it is, I know nothing. Thanks for the update!


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: primedigger on August 15, 2013, 05:10:59 PM
So, sad news:

I think there might be a bug in hp-9 somewhere, so that the trial division doesn't work quite right in that version and sorts out wrong candidates. I will need to confirm this with the hp-9 sources without my changes, so that I'm sure I didn't introduce that bug. The CUDA trial division seems to be doing the right think and it doesn't find any candidates to discard, because if I understand it right, the sieve already did that. To put it in different words: this idea is likely a dead end.

I pushed my changes for anyone who that wants to play with it. There is also still a very slow ported version of the Fermat test, which is easily outperformed by GMP's implementation on the CPU. I think there is no easy way to avoid doing Fermat tests on the GPU. So for now, there is sadly nothing for the GPU which is faster than hp-9 on the CPU.

I will have a close look at Mtrlt's project, but as it seems, he might have similar problems. It would be a major achievement if gets a GPU Fermat test working with a good speed-up. This means that a fast GPU "exponentiation by squaring" algorithm is available to the research community and prime research would benefit from that in general, as most prime tests (not only Fermat's test) need that. There are also a couple of papers on Fermat tests on the GPU (e.g. http://www.gpgpgpu.com/gecco2009/6.pdf), however these implementations are usually assuming that n is smaller than 32 or 64bits, which makes the test much easier.

Also if Mtrlt succeds, he really deserves his price money... it's really not an easy task and I doubt other GPU implementations are in the wild. I would also then port over his method to my CUDA project.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: lemons on August 15, 2013, 05:36:20 PM
CUDA ++1


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: gigawatt on August 15, 2013, 05:58:29 PM
I take it CUMP (http://www.hpcs.cs.tsukuba.ac.jp/~nakayama/cump/) didn't have what you needed?


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: gigawatt on August 15, 2013, 06:25:27 PM
There are also a couple of papers on Fermat tests on the GPU (e.g. http://www.gpgpgpu.com/gecco2009/6.pdf), however these implementations are usually assuming that n is smaller than 32 or 64bits, which makes the test much easier.

I just skimmed over that paper.  Their results are novel, but almost useless in application.
If you're doing Fermat tests, there's a good chance the numbers you want to analyze are greater than 2^64.   :(


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: primedigger on August 15, 2013, 07:20:27 PM
There are also a couple of papers on Fermat tests on the GPU (e.g. http://www.gpgpgpu.com/gecco2009/6.pdf), however these implementations are usually assuming that n is smaller than 32 or 64bits, which makes the test much easier.

I just skimmed over that paper.  Their results are novel, but almost useless in application.
If you're doing Fermat tests, there's a good chance the numbers you want to analyze are greater than 2^64.   :(

Exactly, and they are much greater than 2^64 in primcoin.

As for CUMP, it is completely useless for primecoin, as it only implements floating point arithmetic and then only addition, multiplication and subtraction.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: refer_2_me on August 16, 2013, 04:30:33 AM
Keep on keeping on, I guess.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: mikaelh on August 16, 2013, 06:51:20 AM
I think there might be a bug in hp-9 somewhere, so that the trial division doesn't work quite right in that version and sorts out wrong candidates. I will need to confirm this with the hp-9 sources without my changes, so that I'm sure I didn't introduce that bug. The CUDA trial division seems to be doing the right think and it doesn't find any candidates to discard, because if I understand it right, the sieve already did that. To put it in different words: this idea is likely a dead end.

I can confirm that fast division is buggy in hp9. It's also not needed in hp9 anymore because of Sunny's optimization. I'll remove the code in my next release. Thanks for spotting the bug.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: crendore on August 16, 2013, 09:37:42 AM
I think there might be a bug in hp-9 somewhere, so that the trial division doesn't work quite right in that version and sorts out wrong candidates. I will need to confirm this with the hp-9 sources without my changes, so that I'm sure I didn't introduce that bug. The CUDA trial division seems to be doing the right think and it doesn't find any candidates to discard, because if I understand it right, the sieve already did that. To put it in different words: this idea is likely a dead end.

I can confirm that fast division is buggy in hp9. It's also not needed in hp9 anymore because of Sunny's optimization. I'll remove the code in my next release. Thanks for spotting the bug.

So uhh... should we not be using HP9 then?


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: Lauda on August 16, 2013, 09:50:08 AM
CUDA ++1
We all love CUDA because of XPM  :D


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: mikaelh on August 16, 2013, 11:16:26 AM
I think there might be a bug in hp-9 somewhere, so that the trial division doesn't work quite right in that version and sorts out wrong candidates. I will need to confirm this with the hp-9 sources without my changes, so that I'm sure I didn't introduce that bug. The CUDA trial division seems to be doing the right think and it doesn't find any candidates to discard, because if I understand it right, the sieve already did that. To put it in different words: this idea is likely a dead end.

I can confirm that fast division is buggy in hp9. It's also not needed in hp9 anymore because of Sunny's optimization. I'll remove the code in my next release. Thanks for spotting the bug.

So uhh... should we not be using HP9 then?

The fast-div test seems to be filtering out about 6% of candidates. HP9 should still be faster than HP8 despite that.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: primedigger on August 16, 2013, 11:48:06 AM
If you can't wait for Mikaelh's fix: The fast div can be disabled easily, open prime.cpp in an editor and comment out the whole "if (fFastDiv) {...}" block in the function FermatProbablePrimalityTestFast and comment out the block after the comment "// Compute parameters for fast div test" in MineProbablePrimeChain.

The number of candidates it filtered out was depended on "nFastDivPrimes", which by default was chosen fairly low, thus only ~6% of candidates are (wrongly) discarded. But hp-9 is still doing unnecessary divisions on all candidates, so you can speed up hp-9 by commenting out those divisions.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: crendore on August 16, 2013, 12:03:42 PM
If you can't wait for Mikaelh's fix: The fast div can be disabled easily, open prime.cpp in an editor and comment out the whole "if (fFastDiv) {...}" block in the function FermatProbablePrimalityTestFast and comment out the block after the comment "// Compute parameters for fast div test" in MineProbablePrimeChain.

The number of candidates it filtered out was depended on "nFastDivPrimes", which by default was chosen fairly low, thus only ~6% of candidates are (wrongly) discarded. But hp-9 is still doing unnecessary divisions on all candidates, so you can speed up hp-9 by commenting out those divisions.

Thanks!  Just as an aside, in c++ what does it mean to wrap a block of code in { } when there isn't a function definition, if, else, while, for, etc. statement? (in regards to the second block of code you mentioned to comment out, it seems wrapped in {} with no initial statement.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: paulthetafy on August 16, 2013, 12:13:47 PM
Thanks!  Just as an aside, in c++ what does it mean to wrap a block of code in { } when there isn't a function definition, if, else, while, for, etc. statement? (in regards to the second block of code you mentioned to comment out, it seems wrapped in {} with no initial statement.
It's a nice simple way to limit the scope of any variables declared inside the {} to only be "visible" inside those brackets


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: crendore on August 16, 2013, 12:17:57 PM
Thanks!  Just as an aside, in c++ what does it mean to wrap a block of code in { } when there isn't a function definition, if, else, while, for, etc. statement? (in regards to the second block of code you mentioned to comment out, it seems wrapped in {} with no initial statement.
It's a nice simple way to limit the scope of any variables declared inside the {} to only be "visible" inside those brackets

Ahhh. sweet, good to know.  I don't write much c++, just learning it now.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: crendore on August 16, 2013, 12:21:07 PM
I can confirm this gave me a speed up on my 4 machines as follows:
1: 6.88%
2: 4.52%
3: 4.69%
4: 8.42%


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: paulthetafy on August 16, 2013, 12:31:25 PM
I can confirm this gave me a speed up on my 4 machines as follows:
1: 6.88%
2: 4.52%
3: 4.69%
4: 8.42%


Interesting - I saw a 9% decrease in Chains/day???


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: crendore on August 16, 2013, 12:40:45 PM
How long did you leave it running for before checking? (and yeah my numbers were based on change in chains per day)


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: paulthetafy on August 16, 2013, 12:52:11 PM
How long did you leave it running for before checking? (and yeah my numbers were based on change in chains per day)
About 10 minutes, although I have just checked again and it is still 9% slower.    I can also see the marked drop in the chains/day stats in debug.log.  Weird.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: crendore on August 16, 2013, 01:04:46 PM
How long did you leave it running for before checking? (and yeah my numbers were based on change in chains per day)
About 10 minutes, although I have just checked again and it is still 9% slower.    I can also see the marked drop in the chains/day stats in debug.log.  Weird.

Upon further inspection the chainsperday number seems to jump around all over the place.  anywhere from 0.58 to 0.89.  Currently it seems to have stabilized at 0.77, which is still markedly higher than it was before the changes at 0.69.

edit: are you using other settings (or default settings)?

Also, sorry for thread jacking primedigger.  We really should move this conversation back to the correct thread.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: paulthetafy on August 16, 2013, 01:09:46 PM
How long did you leave it running for before checking? (and yeah my numbers were based on change in chains per day)
About 10 minutes, although I have just checked again and it is still 9% slower.    I can also see the marked drop in the chains/day stats in debug.log.  Weird.

Upon further inspection the chainsperday number seems to jump around all over the place.  anywhere from 0.58 to 0.89.  Currently it seems to have stabilized at 0.77, which is still markedly higher than it was before the changes at 0.69.

edit: are you using other settings?

Also, sorry for thread jacking primedigger.  We really should move this conversation back to the correct thread.
My i7 was running stable at 1.10, it is now running stable at 1.00, which is actually 10% slower.  I'll revert the code back later and see if it makes a difference.
Anyway we're hijacking this thread so suggest starting a new one if you want to continue to figuring this out.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: mechs on August 31, 2013, 08:06:29 PM
Is this project dead?


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: Lauda on September 01, 2013, 10:04:01 AM
Is this project dead?
Looks like it.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: dadaguo on September 01, 2013, 11:26:46 AM
yes,it is~~ :-[


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: Lauda on September 01, 2013, 12:08:12 PM
Damn.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: mechs on September 01, 2013, 04:27:18 PM
Too bad, I have an awesome nvidia based gaming rig.  I use the cpu for xpm mining but the gpus just sit idle


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: ReCat on September 01, 2013, 06:12:58 PM
Community projects always die. It needs to have a motivation. ie. bitcoins.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: mechs on September 01, 2013, 06:14:37 PM
Community projects always die. It needs to have a motivation. ie. bitcoins.
I would donate to a developer willing to take this on.


Title: Re: [XPM] CUDA enabled qt client miner for primecoins. Source code inside. WIP
Post by: ReCat on September 01, 2013, 06:38:52 PM
I can take this on. Donate 75 bitcoins to me and you will receive an early beta. :P

(joking, of course)