Need over 830 core results too. im using CGMiner experimenting a lot, let me calculate my results and update here
with 3x5970's at 850mhz core, i got 2272 average mhash on cgminer after 12 hours that's 757.3mhash per card, 378.6 per core
memory i always have 1/3 of core, but i will test the 150, and perhaps 1/6th of core.
one thing is when i set memory clock to 284 it registers 282.8mhz in GPU-Z so im figuring out if i really need to force it to something else to get the REAL speed closer to 1/3
I may have lost my optimal settings in software that got me 2272 earlier, figuring them out again (juggling sapphire trixx with cgminer's OC settings, CGMiner doesn't seem to set the vcore)
also unaware if this had helped me:
open taskmgr as Admin and set cgminer priority to Realtime
best results was pulling about 990 watts from the wall. only 850 when i was getting 2135Mhash total
12.6 driver with builtin SDK win7 x64
Topic: 5970 energy efficiency (Read 1787 times)
Optimum memory speed is highly dependant on the actual core speed. I'm working with blinky and chsx3 to find the optimum memory speeds for a cypress XT GPU (Hemlock, or 5970, uses the same GPU). So far I've gotten promising results from blinky's perl script. Basically, I'm brute forcing all of the normal combinations of memory and core speeds in multipels of 1 and 5 respectively. There is no set memory speed that is better than all the rest across ALL core speeds.
Chances are your miner for your 5xxx card is using phatk or phatk2 kernel, 128 worksize, vectors, and bfi_int, and hopefully yo uare using 2.1 SDK (not 2.6/2.7 which comes with the latest drivers. 2.4 and 2.5 is also suspected to be slightly slower). Here is some incomplete, but very GOOD information on what memory speeds you SHOULD be using. I have 4 5970 in 1 rig and was running 150 memory across all the cores. Using this spreadsheet, I was able to increase hashrate by 15 mhash. It's not much, but it WAS an increase. It's now doing exactly 2.7 Ghash with 910 watts, or 2.967 MHash/joule, compared to 2.685 earlier. Note that I can only recommend these speeds if you are using the above settings.
Chances are your miner for your 5xxx card is using phatk or phatk2 kernel, 128 worksize, vectors, and bfi_int, and hopefully yo uare using 2.1 SDK (not 2.6/2.7 which comes with the latest drivers. 2.4 and 2.5 is also suspected to be slightly slower). Here is some incomplete, but very GOOD information on what memory speeds you SHOULD be using. I have 4 5970 in 1 rig and was running 150 memory across all the cores. Using this spreadsheet, I was able to increase hashrate by 15 mhash. It's not much, but it WAS an increase. It's now doing exactly 2.7 Ghash with 910 watts, or 2.967 MHash/joule, compared to 2.685 earlier. Note that I can only recommend these speeds if you are using the above settings.
for me, 5970 gets best performance w/ phatk -w 256
As for improving effeciency further, if you can, drop the voltage on cpu and its clock speed to bout 1ghz, should be easily able to do 1v @ 1ghz, if not less voltage then the thing barely uses any power.
Also, run the RAM speed low as possible.
Disable any onboard components not needed(USB3, firewire, serial, parellel controller, ide controller, etc...)
My 5970's can do 810mhz @ 1.034v, give that a shot.
Also have one that does 760mhz @ .95v, pretty effecient.
Then of course the more effecient ur PSU...the better.
As for improving effeciency further, if you can, drop the voltage on cpu and its clock speed to bout 1ghz, should be easily able to do 1v @ 1ghz, if not less voltage then the thing barely uses any power.
Also, run the RAM speed low as possible.
Disable any onboard components not needed(USB3, firewire, serial, parellel controller, ide controller, etc...)
My 5970's can do 810mhz @ 1.034v, give that a shot.
Also have one that does 760mhz @ .95v, pretty effecient.
Then of course the more effecient ur PSU...the better.
Yep. I got a slight speed improvement when I dropped memclock to the min value instead 1/3 of the GPU. Freeing myself from that 'requirements let me push the GPU a bit faster. YMMV, but it works for me and runs cooler.
Try 167MHz for memory. Same power but slightly better MH/s than 150. Probably the OP wants to make memclock 150.
Yep. I got a slight speed improvement when I dropped memclock to the min value instead 1/3 of the GPU. Freeing myself from that 'requirements let me push the GPU a bit faster. YMMV, but it works for me and runs cooler.
Probably the OP wants to make memclock 150.
Hi all,
Just thought I'd share my findings on power vs hashrate for a single 5970, 2.8ghz core 2 cpu, cgminer-2.6.3-win32, OpenCL 1.2 AMD-APP (923.1) :
Max hashes
-I 9 -v 2 -w 128 -k diablo --gpu-engine 825 --gpu-memclock 150 --gpu-vddc 1.051
726MH/s, 363w = 2MH/watt
Max efficiency
-I 9 -v 2 -w 128 -k diablo --gpu-engine 680 --gpu-memclock 150 --gpu-vddc 0.950
601MH/s, 252w = 2.39MH/watt
The cost of electricity in the UK is quite expensive, so I'm looking to move to FPGA where hopefully I'll be getting around 10MH/watt
Just thought I'd share my findings on power vs hashrate for a single 5970, 2.8ghz core 2 cpu, cgminer-2.6.3-win32, OpenCL 1.2 AMD-APP (923.1) :
Max hashes
-I 9 -v 2 -w 128 -k diablo --gpu-engine 825 --gpu-memclock 150 --gpu-vddc 1.051
726MH/s, 363w = 2MH/watt
Max efficiency
-I 9 -v 2 -w 128 -k diablo --gpu-engine 680 --gpu-memclock 150 --gpu-vddc 0.950
601MH/s, 252w = 2.39MH/watt
The cost of electricity in the UK is quite expensive, so I'm looking to move to FPGA where hopefully I'll be getting around 10MH/watt
Jump to: