Author

Topic: [ANN] ccminer 2.3 - opensource - GPL (tpruvot) - page 152. (Read 500112 times)

sr. member
Activity: 346
Merit: 250
I've always gone with what was stated in the wiki. I'm pretty sure the same info is stated on the nVidia web site as well. I will gladly try compiling for 3.0 and see if it works. Probably not till tomorrow though. Will report back my results. thanks all for chipping in to help.  Smiley

Z
Just reporting back that yes indeed, building for SM3.0 for my card that nVidia and the Wiki both state is 3.5 did the trick. Looking through the makefile of ccminer 1.2 I see that it was building for both 3.0 and 3.5 so that's why it worked. Some bitquark (quark algo) hashrate comparisons from that machine:

ccminer 1.2 = ~1200KH/s
ccminer 1.6.6 = ~950KH/s

A little disappointed with that. Hoping the share rates are higher to compensate but that's harder to gauge. I'll be trying 1.7 tonight.
legendary
Activity: 1484
Merit: 1082
ccminer/cpuminer developer
No, its maybe why the 950 is there, to replace the 750ti which seems different about software power limits. There is a setting in bios, but i dont think its tunable on the fly
legendary
Activity: 2870
Merit: 1091
--- ChainWorks Industries ---
Same on windows, admin rights and some unlock to do...

do you know whether nvidia have included the 'lower' class cards ( like our gigabyte 750ti oc lp cards ) are included in the list of cards supported for oc in the newer nvidia update? ...

i cant seem to find this info anywhere - as nvidia did not support oc with these cards with cuda 6.5 ...

#crysx
legendary
Activity: 1484
Merit: 1082
ccminer/cpuminer developer
Same on windows, admin rights and some unlock to do...
legendary
Activity: 2870
Merit: 1091
--- ChainWorks Industries ---
Yes, if you have doubts, use the x64 one which is a bit slower but use the exact same code/upx packer :p it also allow nvml api use (pstate and power limit)

i dont use windows - so i dont have this issue Wink ...

though there are limitations with oc parameters in linux ...

#crysx
legendary
Activity: 1484
Merit: 1082
ccminer/cpuminer developer
Yes, if you have doubts, use the x64 one which is a bit slower but use the exact same code/upx packer :p it also allow nvml api use (pstate and power limit)
legendary
Activity: 2870
Merit: 1091
--- ChainWorks Industries ---
I dont know, both are compiled the same, on the same system. I seriously begin to hate these Anti virus

Maybe they need to test it before, or are waiting user feedback, i dont know and i dont care about that.

You said to download ? from github ?

heuristics used in all current antivirus applications will ALWAYS have false positive responses to those apps that have similar type signatures to the viral signature ...

this does NOT mean its a virus - but of a pain in the backside when releasing new code that the heuristics engine does not know ...

so the machine that you are using to compile these windows packages is probably clean epsylon3 - its just the anti-virus packages need to 'learn' that you package is not a virus ( or contains one ) ...

#crysx
sr. member
Activity: 444
Merit: 250
Yes.. from github, windows version ...  Embarrassed
legendary
Activity: 1484
Merit: 1082
ccminer/cpuminer developer
I dont know, both are compiled the same, on the same system. I seriously begin to hate these Anti virus

Maybe they need to test it before, or are waiting user feedback, i dont know and i dont care about that.

You said to download ? from github ?
sr. member
Activity: 444
Merit: 250
My McAfee won't allow me to download the latest version 1.7.0 where as it doesn't complain with 1.6.6 ...anything its detecting false positive with the latest version?
legendary
Activity: 1484
Merit: 1082
ccminer/cpuminer developer
reason is simple, i cant test on SM 2.0 :p only have an old GTX 460 which is 2.1

but from what ive read, its only fine tuning like -mtune, both should be handled/compatible
sr. member
Activity: 457
Merit: 273
So, you added support for compute capability 2.1, but not 2.0? Any reason for that? Both are Fermi architecture...

Keep up the good work!
legendary
Activity: 1484
Merit: 1082
ccminer/cpuminer developer
no, this algo is too slow even on recent cards

maybe whirlpoolx
full member
Activity: 140
Merit: 100
hello dev do you planning to add neoscrypt sm 2.1 support ?
legendary
Activity: 2870
Merit: 1091
--- ChainWorks Industries ---
Released the 1.7 today with quark and lyra2v2 SM2.1 Compatibily

Ready for the winter with that...

testing when i get in to the office today ...

tanx ...

#crysx
legendary
Activity: 2716
Merit: 1094
Black Belt Developer
Thanks very much Sir!
legendary
Activity: 1484
Merit: 1082
ccminer/cpuminer developer
Released the 1.7 today with quark and lyra2v2 SM2.1 Compatibily

Ready for the winter with that...
legendary
Activity: 2870
Merit: 1091
--- ChainWorks Industries ---
Awesome release Epsy! I'm out for a few more days and can't reliably reach my rig for the time being, but can't wait to test this out.
The pool (& algo) switching will be super handy, great work indeed!

same ...

when i get into the office tomorrow - ill certainly be testing ...

and if it works - will be fully testing on epsylons donation link on all the algos i can Wink ...

#crysx
legendary
Activity: 1154
Merit: 1001
Awesome release Epsy! I'm out for a few more days and can't reliably reach my rig for the time being, but can't wait to test this out.
The pool (& algo) switching will be super handy, great work indeed!
legendary
Activity: 2870
Merit: 1091
--- ChainWorks Industries ---
Sigh.. He should stop building binaries 2 days before my releases...

Will double check / delay the release a bit more... On which card ?

I added sm 2.1 support only a few days ago : https://github.com/tpruvot/ccminer/commits/windows/x11/cuda_x11_simd512_sm2.cuh

better a small delay with a solid release epsylon3 ...

you are doing one hell of a job ...

#crysx
Jump to: