Topic: CCminer(SP-MOD) Modded NVIDIA Maxwell / Pascal kernels. - page 1159. (Read 2347599 times)
Use release 39 and without a multiplier. should work. mine at yaamp.com (Hamsterpool is broken)
What multiplier do we need for Qubit?
Im only getting half my hash report at pool...750ti. Thx
Im only getting half my hash report at pool...750ti. Thx
Quote
I can't believed it stayed cpu only for 2 months...
*lol* ...but to late... amd is already there.edit: and whirlpool seems not that slow on amd - 50mhash per r9 270
nobody has been looking into that vanilla coin ? 
because it is nothing to code: you just need to xor the first 256bit of the whirlpool hash to the last 256bit
I can't believed it stayed cpu only for 2 months (it takes mostly 2 min, and 10 if you create a new algo in ccminer framework)
because it is nothing to code: you just need to xor the first 256bit of the whirlpool hash to the last 256bit
I can't believed it stayed cpu only for 2 months
(it takes mostly 2 min, and 10 if you create a new algo in ccminer framework) sp - is pluck part of this fork? ...
if not - will it be? ...
#crysx
if not - will it be? ...
#crysx
Not yet but I hope soon. For the time being sgminer is much faster than ccminer for nvidia cards which pretty much screams for optimizations.
To be honest I would be curious to look at the ptx generated by opencl (if there are a command to obtain it)...
Actually the main difference between the two, is that the cards on nvidia runs at 40% tdp while on sgminer it runs at 100%tdp...
ps: I think it would be interesting (but lengthy) to transpose the cuda neoscrypt to opencl and check how it does on nvidia
There's no pluck there. DJM34 made a fork which runs at around 2.3kh/s per 750 Ti while sgminer does ~3.7kh/s.
kool ...
will get it all sorted tomorrow ...
i think the last time i tried to compile i was gettign errors - and djm34 pointed out that i needed to use the latest cuda 6.5 ...
if thats the case for the sgminer / ccminer compiles - i will have a bit of work to do to build another linux machine thats more up to date than the fedora 19 x64 that i have ...
:|
#crysx
yes there is:
https://github.com/KlausT/ccminer/commit/fa44f730b875489e7f1c5f7e72179dc3c960e86f
But I think the code is made by DJM34
https://github.com/KlausT/ccminer/commit/fa44f730b875489e7f1c5f7e72179dc3c960e86f
But I think the code is made by DJM34
Oh right, branches.
There's no pluck there. DJM34 made a fork which runs at around 2.3kh/s per 750 Ti while sgminer does ~3.7kh/s.
kool ...
will get it all sorted tomorrow ...
i think the last time i tried to compile i was gettign errors - and djm34 pointed out that i needed to use the latest cuda 6.5 ...
if thats the case for the sgminer / ccminer compiles - i will have a bit of work to do to build another linux machine thats more up to date than the fedora 19 x64 that i have ...
:|
#crysx
yes there is:
https://github.com/KlausT/ccminer/commit/fa44f730b875489e7f1c5f7e72179dc3c960e86f
But I think the code is made by DJM34
https://github.com/KlausT/ccminer/commit/fa44f730b875489e7f1c5f7e72179dc3c960e86f
But I think the code is made by DJM34
There's no pluck there. DJM34 made a fork which runs at around 2.3kh/s per 750 Ti while sgminer does ~3.7kh/s.
sp - is pluck part of this fork? ...
if not - will it be? ...
#crysx
if not - will it be? ...
#crysx
Not yet but I hope soon. For the time being sgminer is much faster than ccminer for nvidia cards which pretty much screams for optimizations.
sp - is pluck part of this fork? ...
if not - will it be? ...
#crysx
if not - will it be? ...
#crysx
Must be a linux issue, or perhaps an issue with the 960 card.
51% of the qubit hashrate is now with my latest version of the mod @yaamp
ccminer/1.5.39-git(SP-MOD) 14 51% 161.1 mh/s 1%
I will be testing on the 960 this weekend.
The latest version on github has a bit high intensity. I get out of memory sometimes on my testrig, but the hashrate is +30-50KHASH. The GPU has plenty of memory left, but I think this is an issue with memory on the mainboard. I try to allocate a total of 1.2 gig on the graphic card, and it fails. 700mb is ok. (2GB 750ti's)
51% of the qubit hashrate is now with my latest version of the mod @yaamp
ccminer/1.5.39-git(SP-MOD) 14 51% 161.1 mh/s 1%
I will be testing on the 960 this weekend.
The latest version on github has a bit high intensity. I get out of memory sometimes on my testrig, but the hashrate is +30-50KHASH. The GPU has plenty of memory left, but I think this is an issue with memory on the mainboard. I try to allocate a total of 1.2 gig on the graphic card, and it fails. 700mb is ok. (2GB 750ti's)
Probably a 960 issue... works fine Linux (2) 750 Ti.
QUBIT ISSUE:
I have 2 6x750ti rigs, 1 is Linux, the other is Win 8 (now). I also have a GTX 960 on Win 7. Qubit does not behave well, recently. Both my 750ti cards and my GTX 960 card have periods where no shares are accepted or submitted, although hashing goes on. No rejects, just hashing speed and new block notifications. Essentially, they burn electricity and do no useful work.
Also, I get strange "doubled-up" behavior, where every other share is rejected. This happens with tpruvot's ccminer v1.5.3. Apparently, it happens about the same time as trouble with SP_ 's ccminer.
Later, they may run qubit properly. I don't know why. --scryptr
EDIT: I suppose the reason I am posting about qubit is that qubit is a top-ranking earner just now. If I let my Windows rigs mine on the round-robin batch file, they get stuck on qubit and actually earn nothing. --scryptr
Commit "Bether default throughput qubit(+30khash 750ti)" gives error
"Cuda error in func 'x11_simd512_cpu_init' at line 634 : out of memory."
on my poor gtx750 non-ti with only 1gb memory
Will try to lower "int intensity = 256 * 256 * 14;"
I tried 256 * 256 * 12 - it crashes driver when monitor attached and works but slow without monitor.
256 * 256 * 10 gives me 2-4 khs benefit compares to default 1U << 19 (256 * 256 * 8) from previous version
So I went to try -i command line parameter and figured out that for my card -i 19.3 is max for qubit (681472 cuda threads)
"Cuda error in func 'x11_simd512_cpu_init' at line 634 : out of memory."
on my poor gtx750 non-ti with only 1gb memory
Will try to lower "int intensity = 256 * 256 * 14;"
I tried 256 * 256 * 12 - it crashes driver when monitor attached and works but slow without monitor.
256 * 256 * 10 gives me 2-4 khs benefit compares to default 1U << 19 (256 * 256 * 8) from previous version
So I went to try -i command line parameter and figured out that for my card -i 19.3 is max for qubit (681472 cuda threads)
Must be a linux issue, or perhaps an issue with the 960 card.
51% of the qubit hashrate is now with my latest version of the mod @yaamp
ccminer/1.5.39-git(SP-MOD) 14 51% 161.1 mh/s 1%
I will be testing on the 960 this weekend.
The latest version on github has a bit high intensity. I get out of memory sometimes on my testrig, but the hashrate is +30-50KHASH. The GPU has plenty of memory left, but I think this is an issue with memory on the mainboard. I try to allocate a total of 1.2 gig on the graphic card, and it fails. 700mb is ok. (2GB 750ti's)
51% of the qubit hashrate is now with my latest version of the mod @yaamp
ccminer/1.5.39-git(SP-MOD) 14 51% 161.1 mh/s 1%
I will be testing on the 960 this weekend.
The latest version on github has a bit high intensity. I get out of memory sometimes on my testrig, but the hashrate is +30-50KHASH. The GPU has plenty of memory left, but I think this is an issue with memory on the mainboard. I try to allocate a total of 1.2 gig on the graphic card, and it fails. 700mb is ok. (2GB 750ti's)
Probably a 960 issue... works fine Linux (2) 750 Ti.
Interesting to see bitcoin (sha256) support added. No chance of profit but fun to have most of the algos supported in one miner. Of note, this doesn't seem to be a very optimal implementation. With cgminer (opencl) I was getting about 405,000 kh/s on my 2 x 750 Ti setup. With release 34 ccminer I'm down to about 345,000 kh/s. Still, it sits at the very bottom of the profitability list with either hash rate so I'm not too concerned. It burns many more times in power than it earns.
Klaus_t implementation was pretty good, but the launchconfiguration of the kernal was way off. I submitted a 17.5% speed increase to github. (750ti)
Hah! 17.5% increase puts it right at the 405 mh/s I had with the OpenCL miner.
If all the ASIC miners out there suddenly brick themselves we'll have a nice GPU Bitcoin miner here.
Must be a linux issue, or perhaps an issue with the 960 card.
51% of the qubit hashrate is now with my latest version of the mod @yaamp
ccminer/1.5.39-git(SP-MOD) 14 51% 161.1 mh/s 1%
I will be testing on the 960 this weekend.
The latest version on github has a bit high intensity. I get out of memory sometimes on my testrig, but the hashrate is +30-50KHASH. The GPU has plenty of memory left, but I think this is an issue with memory on the mainboard. I try to allocate a total of 1.2 gig on the graphic card, and it fails. 700mb is ok. (2GB 750ti's)
51% of the qubit hashrate is now with my latest version of the mod @yaamp
ccminer/1.5.39-git(SP-MOD) 14 51% 161.1 mh/s 1%
I will be testing on the 960 this weekend.
The latest version on github has a bit high intensity. I get out of memory sometimes on my testrig, but the hashrate is +30-50KHASH. The GPU has plenty of memory left, but I think this is an issue with memory on the mainboard. I try to allocate a total of 1.2 gig on the graphic card, and it fails. 700mb is ok. (2GB 750ti's)
Thanks for the quick reply! This card is running on Win7 x64. I also posted on tpruvot's thread. --scryptr
More free hash is comming in x13/x14/x15 (fugue).
I can precalc most of the first round of the FUGUE512_3 function since it is starting with alot of constant inputs. It seems to give another 20-30KHASH(750ti), but more work is needed before I can know for sure. .
I can precalc most of the first round of the FUGUE512_3 function since it is starting with alot of constant inputs. It seems to give another 20-30KHASH(750ti), but more work is needed before I can know for sure. .
Must be a linux issue, or perhaps an issue with the 960 card.
51% of the qubit hashrate is now with my latest version of the mod @yaamp
ccminer/1.5.39-git(SP-MOD) 14 51% 161.1 mh/s 1%
I will be testing on the 960 this weekend.
The latest version on github has a bit high intensity. I get out of memory sometimes on my testrig, but the hashrate is +30-50KHASH. The GPU has plenty of memory left, but I think this is an issue with memory on the mainboard. I try to allocate a total of 1.2 gig on the graphic card, and it fails. 700mb is ok. (2GB 750ti's)
51% of the qubit hashrate is now with my latest version of the mod @yaamp
ccminer/1.5.39-git(SP-MOD) 14 51% 161.1 mh/s 1%
I will be testing on the 960 this weekend.
The latest version on github has a bit high intensity. I get out of memory sometimes on my testrig, but the hashrate is +30-50KHASH. The GPU has plenty of memory left, but I think this is an issue with memory on the mainboard. I try to allocate a total of 1.2 gig on the graphic card, and it fails. 700mb is ok. (2GB 750ti's)
Jump to: