CCminer(SP-MOD) Modded GPU kernels.

[Dotcommie]

-x 32 is good on the 980 card, but not so good on the 970 card.  (100khash lower than with -x 12)

x 22 seems to be best for my G1 970s and I get about the same as jpouza.  Haven't tried x 32 but I doubt that would run on mine.  Everything is already really laggy.

[antonio8]

I am running factory stock OC 750ti and I get the best with -x 15. I could probably get more but I am also cpu mining with the cpu set to start/low with 3 of 6 cores.

[chrysophylax]

I am running factory stock OC 750ti and I get the best with -x 15. I could probably get more but I am also cpu mining with the cpu set to start/low with 3 of 6 cores.

i cant get this windows version to run under linux - no matter what i do ...

im still going to donate more as i appreciate the work that you have done anyway - but i would really appreciate if i could have one that will work under linux sp ...

#crysx

[antonio8]

Sorry don't know how to embed pics but anyone find blocks yet?

I am 0 for 3. http://s3.postimg.org/nrdjpvxkj/Rejected_blocks.png

EDIT: My first rejected block shows no yay or boo but the wallet did not accept it.

[Dotcommie]

Sorry don't know how to embed pics but anyone find blocks yet?

I am 0 for 3. http://s3.postimg.org/nrdjpvxkj/Rejected_blocks.png

EDIT: My first rejected block shows no yay or boo but the wallet did not accept it.

Yes, I was about to say the same thing.  I had to wait about 6hrs for the first block, but it was also a booo.  Don't think the current implementation is working. (at least for me)

It also seems my rig with 970s are able to use -x 22 on the original spreadminer and they're only down maybe 50-100khs compared to the first beta.  Patiently waiting for the next release, but in the meantime...anyone get an excepted share with SP's release yet? Curious to find out if it works for some but not all.

[bathrobehero]

I haven't got the email yet. Btw, both binaries and the source is being distributed?

[sp_]

I have found a bug in the hash. The problem is that tsivs code is missing verfication by the cpu before sending the results. Build 2 comming soon

[tsiv]

I have found a bug in the hash. The problem is that tsivs code is missing verfication by the cpu before sending the results. Build 2 comming soon

Care to elaborate on the bug part? Might see if I CBA to fix it in the official release.

[tsiv]

Something I noticed about the throughput values and it kinda makes sense too: Multiples of SMM count seem to work best for the -x parameter. 5 for 750 Ti, 13 for 970, 16 for 980. Only have 750s and a 970 to test with but 20 or 15 for 750 Ti and 13 or 26 for 970 seem to work well.

[Epsylon3]

I have found a bug in the hash. The problem is that tsivs code is missing verfication by the cpu before sending the results. Build 2 comming soon

Care to elaborate on the bug part? Might see if I CBA to fix it in the official release.

tsiv: I started to merge your spread code in ccminer, but without testnet or a pool, its hard to finalize it...

I added the cpu part

https://github.com/tpruvot/ccminer/tree/spreadx11

i will rebase it on my final 1.5.2

[tsiv]

Something I noticed about the throughput values and it kinda makes sense too: Multiples of SMM count seem to work best for the -x parameter. 5 for 750 Ti, 13 for 970, 16 for 980. Only have 750s and a 970 to test with but 20 or 15 for 750 Ti and 13 or 26 for 970 seem to work well.

Isn't that usually the case?

Makes sense to me, just never really crossed my mind before. Still most if not every bit of ccminer code seems to just use a more or less arbitrary numbers for throughput. Probably wouldn't be a bad idea to always base it on SMM count one way or another.

[tsiv]

I have found a bug in the hash. The problem is that tsivs code is missing verfication by the cpu before sending the results. Build 2 comming soon

Care to elaborate on the bug part? Might see if I CBA to fix it in the official release.

tsiv: I started to merge your spread code in ccminer, but without testnet or a pool, its hard to finalize it...

I added the cpu part

https://github.com/tpruvot/ccminer/tree/spreadx11

i will rebase it on my final 1.5.2

That CPU code seems to be just the X11 part of the hash though, completely ignoring the miner signature and whole block hash? And yea, I know exactly how annoying it is to test when your only method of verification is actually solving a block. I ended up hard coding it to work on a static block of input data that I knew the correct hash for until I got the damn thing working Well, more or less working apparently.

[sp_]

I will merge the original c code into the spreadcoin fork. the gpu finds a solution, it will be verified by the cpu. The current build have no verification done by the cpu.

[tsiv]

I will merge the original c code into the spreadcoin fork. the gpu finds a solution, it will be verified by the cpu. The current build have no verification done by the cpu.

I'm confused. You said you found a bug in the hash? Or are you considering the lack of verification to be a bug? I left it out kinda intentionally since submitting "bad" solutions doesn't really hurt and I just didn't bother with the CPU verification. But yea it would be nice and very helpful in debugging. Running in benchmark mode with a low target would be a good tool for checking if the GPU hashing works or not, instead of having to wait until you actually solve a block.

[sp_]

I'm confused. You said you found a bug in the hash? Or are you considering the lack of verification to be a bug? I left it out kinda intentionally since submitting "bad" solutions doesn't really hurt and I just didn't bother with the CPU verification. But yea it would be nice and very helpful in debugging. Running in benchmark mode with a low target would be a good tool for checking if the GPU hashing works or not, instead of having to wait until you actually solve a block.

It is a bug in my version. I have replaced some of the kernals in the miner to give a small boost. When changing kernals, things can go wrong, and it would be helpful to have the cpu to doublecheck the result.

[chrysophylax]

wow ...

is it just me - or is it that you guys seem to be working together ANYWAY? ...

why dont you merge all the work that you have been doing ( yes i know - easier said than done ) and just work off one fork? ...

seems that each of you awesome devs can bring their own thing to the table - and improve ccminer to an unprecedented level ...

just my observation here guys ...

i know if i could employ you all under the one roof - thats exactly what would happen ...

#crysx

[tsiv]

On a related note, I am I bit concerned about the miner on non-750Ti cards. I'm getting pretty much zero rejects and exactly as many blocks as you could expect against the current network hash rate on my 750 rig, but a fair amount of rejects (but not 100%) on the 970. Can't see any particular reason why the same code would work on 750 and not on 970 though. It is a mystery. Just seems bit odd that near all the rejects I get are on the 970. Could be coincidence but I can't help but wonder...

Code:

[2015-01-23 01:11:53] accepted: 1/1 (100.00%), 10038 khash/s (yay!!!)
[2015-01-23 04:56:03] accepted: 2/2 (100.00%), 10036 khash/s (yay!!!)
[2015-01-23 16:48:20] accepted: 3/3 (100.00%), 10040 khash/s (yay!!!)
[2015-01-24 07:28:52] accepted: 4/4 (100.00%), 10039 khash/s (yay!!!)
[2015-01-24 07:53:30] accepted: 5/5 (100.00%), 10034 khash/s (yay!!!)
[2015-01-24 09:22:16] accepted: 6/6 (100.00%), 10034 khash/s (yay!!!)
[2015-01-25 05:06:02] accepted: 7/7 (100.00%), 10035 khash/s (yay!!!)
[2015-01-25 16:22:12] accepted: 8/8 (100.00%), 10047 khash/s (yay!!!)
[2015-01-25 21:06:44] accepted: 9/9 (100.00%), 10027 khash/s (yay!!!)
[2015-01-26 04:59:57] accepted: 10/10 (100.00%), 10034 khash/s (yay!!!)
[2015-01-26 13:35:12] accepted: 11/11 (100.00%), 10038 khash/s (yay!!!)

[tsiv]

I'm confused. You said you found a bug in the hash? Or are you considering the lack of verification to be a bug? I left it out kinda intentionally since submitting "bad" solutions doesn't really hurt and I just didn't bother with the CPU verification. But yea it would be nice and very helpful in debugging. Running in benchmark mode with a low target would be a good tool for checking if the GPU hashing works or not, instead of having to wait until you actually solve a block.

It is a bug in my version. I have replaced some of the kernals in the miner to give a small boost. When changing kernals, things can go wrong, and it would be helpful to have the cpu to doublecheck the result.

Aah, got it. Thought you found something wrong in the original.

[Dotcommie]

On a related note, I am I bit concerned about the miner on non-750Ti cards. I'm getting pretty much zero rejects and exactly as many blocks as you could expect against the current network hash rate on my 750 rig, but a fair amount of rejects (but not 100%) on the 970. Can't see any particular reason why the same code would work on 750 and not on 970 though. It is a mystery. Just seems bit odd that near all the rejects I get are on the 970. Could be coincidence but I can't help but wonder...

Code:

[2015-01-23 01:11:53] accepted: 1/1 (100.00%), 10038 khash/s (yay!!!)
[2015-01-23 04:56:03] accepted: 2/2 (100.00%), 10036 khash/s (yay!!!)
[2015-01-23 16:48:20] accepted: 3/3 (100.00%), 10040 khash/s (yay!!!)
[2015-01-24 07:28:52] accepted: 4/4 (100.00%), 10039 khash/s (yay!!!)
[2015-01-24 07:53:30] accepted: 5/5 (100.00%), 10034 khash/s (yay!!!)
[2015-01-24 09:22:16] accepted: 6/6 (100.00%), 10034 khash/s (yay!!!)
[2015-01-25 05:06:02] accepted: 7/7 (100.00%), 10035 khash/s (yay!!!)
[2015-01-25 16:22:12] accepted: 8/8 (100.00%), 10047 khash/s (yay!!!)
[2015-01-25 21:06:44] accepted: 9/9 (100.00%), 10027 khash/s (yay!!!)
[2015-01-26 04:59:57] accepted: 10/10 (100.00%), 10034 khash/s (yay!!!)
[2015-01-26 13:35:12] accepted: 11/11 (100.00%), 10038 khash/s (yay!!!)

I don't believe I had any rejects on your original spreadminer on my 970s.  Seemed to work with the same consistency as 750s for me, unless this is newer code you're talking about.

[tsiv]

On a related note, I am I bit concerned about the miner on non-750Ti cards. I'm getting pretty much zero rejects and exactly as many blocks as you could expect against the current network hash rate on my 750 rig, but a fair amount of rejects (but not 100%) on the 970. Can't see any particular reason why the same code would work on 750 and not on 970 though. It is a mystery. Just seems bit odd that near all the rejects I get are on the 970. Could be coincidence but I can't help but wonder...

Code:

[2015-01-23 01:11:53] accepted: 1/1 (100.00%), 10038 khash/s (yay!!!)
[2015-01-23 04:56:03] accepted: 2/2 (100.00%), 10036 khash/s (yay!!!)
[2015-01-23 16:48:20] accepted: 3/3 (100.00%), 10040 khash/s (yay!!!)
[2015-01-24 07:28:52] accepted: 4/4 (100.00%), 10039 khash/s (yay!!!)
[2015-01-24 07:53:30] accepted: 5/5 (100.00%), 10034 khash/s (yay!!!)
[2015-01-24 09:22:16] accepted: 6/6 (100.00%), 10034 khash/s (yay!!!)
[2015-01-25 05:06:02] accepted: 7/7 (100.00%), 10035 khash/s (yay!!!)
[2015-01-25 16:22:12] accepted: 8/8 (100.00%), 10047 khash/s (yay!!!)
[2015-01-25 21:06:44] accepted: 9/9 (100.00%), 10027 khash/s (yay!!!)
[2015-01-26 04:59:57] accepted: 10/10 (100.00%), 10034 khash/s (yay!!!)
[2015-01-26 13:35:12] accepted: 11/11 (100.00%), 10038 khash/s (yay!!!)

I don't believe I had any rejects on your original spreadminer on my 970s.  Seemed to work with the same consistency as 750s for me, unless this is newer code you're talking about.

Talking about the original, yep. Oh well, maybe my 970 is just jinxed