CCminer(SP-MOD) Modded GPU kernels.

[sp_]

I see no reason why it shouldn't work.

Code:

accepted: 32/32 (100.00%), 709.96 khash/s yay!!!

What card and what intensity?

card gtx 750, intensity not set, default, yaamp

Gtx 750 has only got 1GB of memory. I will lower the default intensity for lyra2 on compute 5.0

[Handofmidas]

Have been testing DRK mining at coinotron for a while now with release34, and the pool is underreporting hashrate. Haven't tested previous releases there in detail. Unfortunately, coinotron does not draw nice graphs AFAIK, but the hashrate reported is always an average of the last 60 minutes, so it should be quite accurate. I checked several time today and it was always reported between 14.7 and 15.2 MH, while it should be at around 17.5 MH according to the miner.

[mirny]

Yeah. 200mhash  on the 750ti. You can solomine and wait for the jackpot. One block is 50 btc. Good luck:)

precisely, BTC block is awarded with 25 BTC's

[rednoW]

Gtx 750 has only got 1GB of memory. I will lower the default intensity for lyra2 on compute 5.0

Thanx for the reply. I tried to play with intensity -i parameter and found that Release 34 works for my GTX750 up to 20.3 value.
I think it's not a memory issue, gpu-z shows very small memory load but the fact that simple 750 has less CUDA cores - only 512 instead of 640 750ti has ...

[sp_]

Have been testing DRK mining at coinotron for a while now with release34, and the pool is underreporting hashrate. Haven't tested previous releases there in detail. Unfortunately, coinotron does not draw nice graphs AFAIK, but the hashrate reported is always an average of the last 60 minutes, so it should be quite accurate. I checked several time today and it was always reported between 14.7 and 15.2 MH, while it should be at around 17.5 MH according to the miner.

Seems to be a bug in the stats on the low difficulties. If a hash is found it exits quicker, but 

*hashes_done = pdata[19] - first_nonce + throughput;

counts the whole list of nounces.

Will fix it

[sp_]

Rolled back 2 nounce simd changes. (high rates on low diff, but lower on the pool)
Added bether launch config in simd +6.6% faster.

- Fixed lyra. (intensity was wrong in the merge)
- x11 +25KHASH (750ti) (+faster x13,x14,x15,x17)
- fresh + 50KHASH (750ti)

etc..

building release 35 now

[sp_]

1.5.35(sp-MOD) is available here: (5-feb-2015)

https://github.com/sp-hash/ccminer/releases/tag/1.5.35

The sourcecode is available here:

https://github.com/sp-hash/ccminer

[ryen123]

X11 on Release35 doesn't seem to work on my 750ti. Looks like its running but no shares submitted to pool. Tested with Coinking, Nicehash & YAAMP.

[rednoW]

X11 on Release35 doesn't seem to work on my 750ti. Looks like its running but no shares submitted to pool. Tested with Coinking, Nicehash & YAAMP.

+1 on gtx750 with yaamp. crunching but no accepted shares (yays or boos)

[sp_]

Strange, it works here.

Make sure you have the latest driver.

I might have copied the wrong exe file though. If you build it from github it should work.

[rednoW]

Strange, it works here.

Make sure you have the latest driver.

I might have copied the wrong exe file though. If you build it from github it should work.

just recompiled. The same. No accepts. No boos. Driver is from CUDA 7.0 pack as always 347.12
Previous releases run x11 normally

[sp_]

You need cuda 6.5

[sp_]

Seems to work for others. (yaamp.com)

Wich algos are broken? Did you try fresh or quark? Might be an issue with cuda 7.0 (wich is only released in beta version)

[rednoW]

Seems to work for others. (yaamp.com)

Wich algos are broken? Did you try fresh or quark? Might be an issue with cuda 7.0 (wich is only released in beta version)

only x11 is broken. All others work fine.
I compiled with cuda 6.5 preset (as default). Cuda 7 installed is not an issue.
You should recheck a code. I didn't see any working Release 35 clients on yaamp's x11 stats ...

[sp_]

Seems to work for others. (yaamp.com)

Wich algos are broken? Did you try fresh or quark? Might be an issue with cuda 7.0 (wich is only released in beta version)

only x11 is broken. All others work fine.
I compiled with cuda 6.5 preset (as default). Cuda 7 installed is not an issue.
You should recheck a code. I didn't see any working Release 35 clients on yaamp's x11 stats ...

Ahh, I found the problem now

remove the line

h_found[thr_id][0] = 0xffffffff;

in x11.cu

[rednoW]

Ahh, I found the problem now

remove the line

h_found[thr_id][0] = 0xffffffff;

in x11.cu

yeah, the next line "if (h_found[thr_id][0] != 0xffffffff)" is slightly not compatible with it

upd: now it works. ~2850 kh (the same speed as in r31 & r32)

[sp_]

Ahh, I found the problem now
remove the line
h_found[thr_id][0] = 0xffffffff;
in x11.cu

yeah, the next line "if (h_found[thr_id][0] != 0xffffffff)" is slightly not compatible with it

I added the line in order to speed test the simd kernal alone, but forgot to remove it. :/ I am away from my dev computer for a while, so will fix it on github later tonight.

[sp_]

upd: now it works. ~2850 kh (the same speed as in r31 & r32)

With vardiff I get around 25khash bether in 35 than in 33/34 , but I lost around 25 KHASH after merging the fork with klaus_t (from  32 ->  33).

Will take another look tonight.


But it could be this commit:

https://github.com/sp-hash/ccminer/commit/704e609b9003620dd29a74b2f43924209e193d24

Might work good on the 960-980 cards, but not on the 750ti.

Not many changes though.

[rednoW]

With vardiff I get around 25khash bether in 35 than in 33/34 , but I lost around 25 KHASH after merging the fork with klaus_t (from  32 ->  33).

Will take another look tonight.


But it could be this commit:

https://github.com/sp-hash/ccminer/commit/704e609b9003620dd29a74b2f43924209e193d24

Might work good on the 960-980 cards, but not on the 750ti.

Not many changes though.

Yes, i put back those "__global__ __launch_bounds__(256, 4)" etc and got some near measurement error gain of ~20kh

[sp_]

With vardiff I get around 25khash bether in 35 than in 33/34 , but I lost around 25 KHASH after merging the fork with klaus_t (from  32 ->  33).
Will take another look tonight.
But it could be this commit:
https://github.com/sp-hash/ccminer/commit/704e609b9003620dd29a74b2f43924209e193d24
Might work good on the 960-980 cards, but not on the 750ti.
Not many changes though.

Yes, i put back those "__global__ __launch_bounds__(256, 4)" etc and got some near measurement error gain of ~20kh

This code will force the kernal to use 64 registers. bether on the 750ti worse on the 970/980.
What I usually do is to case on the compute version and run different configurations for different compute version.
I think many of the kernals haven't been tweaked for a while, and hash can be gained.
If you see my Bitcoin change, all I did was to change the kernal launch configutration for a 17% speedup.

here is the commit:

https://github.com/sp-hash/ccminer/commit/c79f622969393f52f3462e2c3e967777cae1d7d3