Overclock.net › Forums › Overclockers Care › Overclock.net BOINC Team › GPU performance vs power consumption
New Posts  All Forums:Forum Nav:

GPU performance vs power consumption - Page 2

post #11 of 60
6950 does great for dnetc and milkyway. 560i will do good in prime. to me the 6950 if it can be unlocked to a 6970 will give you a higher ppd if thats what you are interested in.
    
CPUMotherboardGraphicsRAM
Intel I7-5820k@4.2ghz Gigabyte X99-UD3P AMD 295x2 16GB G.Skill DDR4-2400 
Hard DriveOptical DriveCoolingOS
Corsair Force GT 120gb (OS), 2x 1tb WD Blacks R... LG Bluray CoolerMaster MasterLiquid Pro 120 Win 10 x64 
MonitorKeyboardPowerCase
26" Emprex Logitech G15 CoolerMaster SPH-1300 Coolermaster HAF922 
MouseMouse Pad
CM Sentinel II CoolerMaster HSM Battle Pad SSK 
  hide details  
Reply
    
CPUMotherboardGraphicsRAM
Intel I7-5820k@4.2ghz Gigabyte X99-UD3P AMD 295x2 16GB G.Skill DDR4-2400 
Hard DriveOptical DriveCoolingOS
Corsair Force GT 120gb (OS), 2x 1tb WD Blacks R... LG Bluray CoolerMaster MasterLiquid Pro 120 Win 10 x64 
MonitorKeyboardPowerCase
26" Emprex Logitech G15 CoolerMaster SPH-1300 Coolermaster HAF922 
MouseMouse Pad
CM Sentinel II CoolerMaster HSM Battle Pad SSK 
  hide details  
Reply
post #12 of 60
Thread Starter 
How about Collatz Conjecture and Milkyway@home.

Quote:
Originally Posted by Tex1954 View Post
I would be happy to provide some task times to help you compare if others could do the same. Just pick some GPU tasks from my profile:

http://boincstats.com/search/all_pro...ff2d483c31de84


Edited by magic8192 - 6/3/11 at 10:14am
Precious
(23 items)
 
Intel 4P Rig
(16 items)
 
AMD 4P Rig
(9 items)
 
CPUCPUCPUCPU
Xeon E5-4650 (ES) C0 Xeon E5-4650 (ES) C0 Xeon E5-4650 (ES) C0 Xeon E5-4650 (ES) C0 
MotherboardGraphicsRAMHard Drive
SuperMicro X9QRi-F+ G200 on board  4GB PC3-12800R x16 OCZ Deneva  
Optical DriveCoolingOSMonitor
Samsung CD/DVD burner 4 x Cooler Master Hyper 212 Server 2012 Standard None 
PowerCaseMouse
OCZ ZX 1250 Rosewill Blackhawk Ultra None 
CPUCPUCPUCPU
Opteron 6376 Opteron 6376 Opteron 6376 Opteron 6376 
MotherboardRAMHard DriveCooling
H8QGi+-F Hynix 16 x HMT151R7BFR4C-H9 4GB 2RX4 PC3-10600R Silicon Power Slim S55 2.5" 480GB SATA III SSD ... Cooler Master Hyper 212 
OS
Ubuntu Server 14.04 
  hide details  
Reply
Precious
(23 items)
 
Intel 4P Rig
(16 items)
 
AMD 4P Rig
(9 items)
 
CPUCPUCPUCPU
Xeon E5-4650 (ES) C0 Xeon E5-4650 (ES) C0 Xeon E5-4650 (ES) C0 Xeon E5-4650 (ES) C0 
MotherboardGraphicsRAMHard Drive
SuperMicro X9QRi-F+ G200 on board  4GB PC3-12800R x16 OCZ Deneva  
Optical DriveCoolingOSMonitor
Samsung CD/DVD burner 4 x Cooler Master Hyper 212 Server 2012 Standard None 
PowerCaseMouse
OCZ ZX 1250 Rosewill Blackhawk Ultra None 
CPUCPUCPUCPU
Opteron 6376 Opteron 6376 Opteron 6376 Opteron 6376 
MotherboardRAMHard DriveCooling
H8QGi+-F Hynix 16 x HMT151R7BFR4C-H9 4GB 2RX4 PC3-10600R Silicon Power Slim S55 2.5" 480GB SATA III SSD ... Cooler Master Hyper 212 
OS
Ubuntu Server 14.04 
  hide details  
Reply
post #13 of 60
Thread Starter 
Quote:
Originally Posted by DarkRyder View Post
6950 does great for dnetc and milkyway. 560i will do good in prime. to me the 6950 if it can be unlocked to a 6970 will give you a higher ppd if thats what you are interested in.
I am certainly interested in ppd, but also in efficiency. I have been browsing the older threads and noticed where some people were cutting back on their contributions because of the power bill.
Precious
(23 items)
 
Intel 4P Rig
(16 items)
 
AMD 4P Rig
(9 items)
 
CPUCPUCPUCPU
Xeon E5-4650 (ES) C0 Xeon E5-4650 (ES) C0 Xeon E5-4650 (ES) C0 Xeon E5-4650 (ES) C0 
MotherboardGraphicsRAMHard Drive
SuperMicro X9QRi-F+ G200 on board  4GB PC3-12800R x16 OCZ Deneva  
Optical DriveCoolingOSMonitor
Samsung CD/DVD burner 4 x Cooler Master Hyper 212 Server 2012 Standard None 
PowerCaseMouse
OCZ ZX 1250 Rosewill Blackhawk Ultra None 
CPUCPUCPUCPU
Opteron 6376 Opteron 6376 Opteron 6376 Opteron 6376 
MotherboardRAMHard DriveCooling
H8QGi+-F Hynix 16 x HMT151R7BFR4C-H9 4GB 2RX4 PC3-10600R Silicon Power Slim S55 2.5" 480GB SATA III SSD ... Cooler Master Hyper 212 
OS
Ubuntu Server 14.04 
  hide details  
Reply
Precious
(23 items)
 
Intel 4P Rig
(16 items)
 
AMD 4P Rig
(9 items)
 
CPUCPUCPUCPU
Xeon E5-4650 (ES) C0 Xeon E5-4650 (ES) C0 Xeon E5-4650 (ES) C0 Xeon E5-4650 (ES) C0 
MotherboardGraphicsRAMHard Drive
SuperMicro X9QRi-F+ G200 on board  4GB PC3-12800R x16 OCZ Deneva  
Optical DriveCoolingOSMonitor
Samsung CD/DVD burner 4 x Cooler Master Hyper 212 Server 2012 Standard None 
PowerCaseMouse
OCZ ZX 1250 Rosewill Blackhawk Ultra None 
CPUCPUCPUCPU
Opteron 6376 Opteron 6376 Opteron 6376 Opteron 6376 
MotherboardRAMHard DriveCooling
H8QGi+-F Hynix 16 x HMT151R7BFR4C-H9 4GB 2RX4 PC3-10600R Silicon Power Slim S55 2.5" 480GB SATA III SSD ... Cooler Master Hyper 212 
OS
Ubuntu Server 14.04 
  hide details  
Reply
post #14 of 60
Yes Darkryder, there is a lot of talk about certain Projects that run faster on the ATI cards... but I have yet to find a place where real numbers are posted in general for many cards vs. tasks. GPUGrid has a forum sticky that puts their "relative processing power" numbers up against different cards, but that is not applicable to all projects.

Could you or gamer11200 start a sticky thread, post the known GPU projects OCN BOINC Team members run (like maybe a google spreadsheet or something) and we could fill in the blanks? I'm sure a lot of folks would appreciate that... I have no idea how to do such a thing myself. We could have tasks listed on one side and various cards/clockspeeds listed on the other and times in the pigeon holes...


Edited by Tex1954 - 6/3/11 at 10:31am
Blue Beast
(13 items)
 
  
CPUMotherboardGraphicsRAM
W3670 4.0 GHz (HT On) 1.345v 24/7 Asus P6X58D Premium DUAL EVGA GTX 560 Ti SC 1G in SLI 12G Corsair Dominator GT2000MHz 
Hard DriveOptical DriveOSMonitor
Kingston V300 120G SSD, 2x 1TB Barracuda HD's LG BlueRay/LightScribe Burner Windows 7 Pro 64Bit 3x24", 1x22" LCD 
KeyboardPowerCaseMouse
Wireless Logitech K320 ULTRA X4 1200W Custom Danger Den LDR-29 Wireless Logitech M310 
Mouse Pad
Custom 
  hide details  
Reply
Blue Beast
(13 items)
 
  
CPUMotherboardGraphicsRAM
W3670 4.0 GHz (HT On) 1.345v 24/7 Asus P6X58D Premium DUAL EVGA GTX 560 Ti SC 1G in SLI 12G Corsair Dominator GT2000MHz 
Hard DriveOptical DriveOSMonitor
Kingston V300 120G SSD, 2x 1TB Barracuda HD's LG BlueRay/LightScribe Burner Windows 7 Pro 64Bit 3x24", 1x22" LCD 
KeyboardPowerCaseMouse
Wireless Logitech K320 ULTRA X4 1200W Custom Danger Den LDR-29 Wireless Logitech M310 
Mouse Pad
Custom 
  hide details  
Reply
post #15 of 60
Thread Starter 
Quote:
Originally Posted by Tex1954 View Post
Yes Darkryder, there is a lot of talk about certain Projects that run faster on the ATI cards... but I have yet to find a place where real numbers are posted in general for many cards vs. tasks. GPUGrid has a forum sticky that puts their "relative processing power" numbers up against different cards, but that is applicable to all projects.

Could you or gamer11200 start a sticky thread, post the known GPU projects OCN BOINC Team members run (like maybe a google spreadsheet or something) and we could fill in the blanks? I'm sure a lot of folks would appreciate that... I have no idea how to do such a thing myself. We could have tasks listed on one side and various cards/clockspeeds listed on the other and times in the pigeon holes...

I was looking at this
http://www.overclock.net/overclock-n...-database.html

The problem with it is that the Nvidia and Ati/AMD users don't have stats in the same projects.
Precious
(23 items)
 
Intel 4P Rig
(16 items)
 
AMD 4P Rig
(9 items)
 
CPUCPUCPUCPU
Xeon E5-4650 (ES) C0 Xeon E5-4650 (ES) C0 Xeon E5-4650 (ES) C0 Xeon E5-4650 (ES) C0 
MotherboardGraphicsRAMHard Drive
SuperMicro X9QRi-F+ G200 on board  4GB PC3-12800R x16 OCZ Deneva  
Optical DriveCoolingOSMonitor
Samsung CD/DVD burner 4 x Cooler Master Hyper 212 Server 2012 Standard None 
PowerCaseMouse
OCZ ZX 1250 Rosewill Blackhawk Ultra None 
CPUCPUCPUCPU
Opteron 6376 Opteron 6376 Opteron 6376 Opteron 6376 
MotherboardRAMHard DriveCooling
H8QGi+-F Hynix 16 x HMT151R7BFR4C-H9 4GB 2RX4 PC3-10600R Silicon Power Slim S55 2.5" 480GB SATA III SSD ... Cooler Master Hyper 212 
OS
Ubuntu Server 14.04 
  hide details  
Reply
Precious
(23 items)
 
Intel 4P Rig
(16 items)
 
AMD 4P Rig
(9 items)
 
CPUCPUCPUCPU
Xeon E5-4650 (ES) C0 Xeon E5-4650 (ES) C0 Xeon E5-4650 (ES) C0 Xeon E5-4650 (ES) C0 
MotherboardGraphicsRAMHard Drive
SuperMicro X9QRi-F+ G200 on board  4GB PC3-12800R x16 OCZ Deneva  
Optical DriveCoolingOSMonitor
Samsung CD/DVD burner 4 x Cooler Master Hyper 212 Server 2012 Standard None 
PowerCaseMouse
OCZ ZX 1250 Rosewill Blackhawk Ultra None 
CPUCPUCPUCPU
Opteron 6376 Opteron 6376 Opteron 6376 Opteron 6376 
MotherboardRAMHard DriveCooling
H8QGi+-F Hynix 16 x HMT151R7BFR4C-H9 4GB 2RX4 PC3-10600R Silicon Power Slim S55 2.5" 480GB SATA III SSD ... Cooler Master Hyper 212 
OS
Ubuntu Server 14.04 
  hide details  
Reply
post #16 of 60
Well, we don't need points for performance comparisons, only times vs. card/GPU type/speed data. I think it would be a lot more useful in general to just have the task name and time it took to complete on a specific card. Points really don't enter into it. Also, we could post what BOINC says our cards do... for instance, on my 2 running systems, I get this:

Corsair-800D (460's running 763MHz)

6/3/2011 10:36:21 AMNVIDIA GPU 0: GeForce GTX 460 (driver version 27527, CUDA version 4000, compute capability 2.1, 962MB, 702 GFLOPS peak)
6/3/2011 10:36:21 AMNVIDIA GPU 1: GeForce GTX 460 (driver version 27527, CUDA version 4000, compute capability 2.1, 962MB, 702 GFLOPS peak)

Phenom-1090T (560 Ti's running 952MHz)

6/3/2011 11:28:23 AMNVIDIA GPU 0: GeForce GTX 560 Ti (driver version 27527, CUDA version 4000, compute capability 2.1, 962MB, 979 GFLOPS peak)
6/3/2011 11:28:23 AMNVIDIA GPU 1: GeForce GTX 560 Ti (driver version 27527, CUDA version 4000, compute capability 2.1, 962MB, 979 GFLOPS peak)



PS: I put my times in other post for the 560... when MilkyWay gets back online, I will update it.

PPS: IN FACT DarkRyder, for the reasons you mentioned, perhaps the ONLY and best thing we could put on a spreadsheet would be what BOINC reports the GFLOPS Peak is.... That would about cover everything since we know the tasks are written differently for the 2 different vendors and thus the speeds would naturally be different too. I can see in THAT case where a PPD comparison would be wanted and more useful... But I run many different tasks daily, so would have to spend 24 hrs on ONE task to get any usable PPD info...
Edited by Tex1954 - 6/3/11 at 10:55am
Blue Beast
(13 items)
 
  
CPUMotherboardGraphicsRAM
W3670 4.0 GHz (HT On) 1.345v 24/7 Asus P6X58D Premium DUAL EVGA GTX 560 Ti SC 1G in SLI 12G Corsair Dominator GT2000MHz 
Hard DriveOptical DriveOSMonitor
Kingston V300 120G SSD, 2x 1TB Barracuda HD's LG BlueRay/LightScribe Burner Windows 7 Pro 64Bit 3x24", 1x22" LCD 
KeyboardPowerCaseMouse
Wireless Logitech K320 ULTRA X4 1200W Custom Danger Den LDR-29 Wireless Logitech M310 
Mouse Pad
Custom 
  hide details  
Reply
Blue Beast
(13 items)
 
  
CPUMotherboardGraphicsRAM
W3670 4.0 GHz (HT On) 1.345v 24/7 Asus P6X58D Premium DUAL EVGA GTX 560 Ti SC 1G in SLI 12G Corsair Dominator GT2000MHz 
Hard DriveOptical DriveOSMonitor
Kingston V300 120G SSD, 2x 1TB Barracuda HD's LG BlueRay/LightScribe Burner Windows 7 Pro 64Bit 3x24", 1x22" LCD 
KeyboardPowerCaseMouse
Wireless Logitech K320 ULTRA X4 1200W Custom Danger Den LDR-29 Wireless Logitech M310 
Mouse Pad
Custom 
  hide details  
Reply
post #17 of 60
gflops wont make a difference in boinc. Cuda and openci are 2 way different technologies. your 979glops from your 560ti's dont compare to my 3370gflops per gpu in my 6990. my stock 5870 said something like 2800gflops and could get like 300-330k in dnetc.
comparing nvidia to amd/ati is like comparing apples to oranges. They dont process data the same way. Each project we compute data for has to have a compatible application written for each architecture. my ati5550 got like 770gflops in boinc and you know as well as i do its no where near even close to performance wise as a 560ti. my old 5770 got like 1700gflops. my 5850 gets like 2350 at stock. You see where I am going with this ?
    
CPUMotherboardGraphicsRAM
Intel I7-5820k@4.2ghz Gigabyte X99-UD3P AMD 295x2 16GB G.Skill DDR4-2400 
Hard DriveOptical DriveCoolingOS
Corsair Force GT 120gb (OS), 2x 1tb WD Blacks R... LG Bluray CoolerMaster MasterLiquid Pro 120 Win 10 x64 
MonitorKeyboardPowerCase
26" Emprex Logitech G15 CoolerMaster SPH-1300 Coolermaster HAF922 
MouseMouse Pad
CM Sentinel II CoolerMaster HSM Battle Pad SSK 
  hide details  
Reply
    
CPUMotherboardGraphicsRAM
Intel I7-5820k@4.2ghz Gigabyte X99-UD3P AMD 295x2 16GB G.Skill DDR4-2400 
Hard DriveOptical DriveCoolingOS
Corsair Force GT 120gb (OS), 2x 1tb WD Blacks R... LG Bluray CoolerMaster MasterLiquid Pro 120 Win 10 x64 
MonitorKeyboardPowerCase
26" Emprex Logitech G15 CoolerMaster SPH-1300 Coolermaster HAF922 
MouseMouse Pad
CM Sentinel II CoolerMaster HSM Battle Pad SSK 
  hide details  
Reply
post #18 of 60
Well, then we are stuck using what does work and that clarifies the "PPD" method the others are using. It would seem the only thing that could be considered "common" would be PPD provided the projects generally award like points for like tasks...

Too many variables, need to figure out a way, method, something that is common and gives a realistic idea of what can be expected. GPU-Z does do what we need either...

Sigh, got no more idea's at the moment...

Blue Beast
(13 items)
 
  
CPUMotherboardGraphicsRAM
W3670 4.0 GHz (HT On) 1.345v 24/7 Asus P6X58D Premium DUAL EVGA GTX 560 Ti SC 1G in SLI 12G Corsair Dominator GT2000MHz 
Hard DriveOptical DriveOSMonitor
Kingston V300 120G SSD, 2x 1TB Barracuda HD's LG BlueRay/LightScribe Burner Windows 7 Pro 64Bit 3x24", 1x22" LCD 
KeyboardPowerCaseMouse
Wireless Logitech K320 ULTRA X4 1200W Custom Danger Den LDR-29 Wireless Logitech M310 
Mouse Pad
Custom 
  hide details  
Reply
Blue Beast
(13 items)
 
  
CPUMotherboardGraphicsRAM
W3670 4.0 GHz (HT On) 1.345v 24/7 Asus P6X58D Premium DUAL EVGA GTX 560 Ti SC 1G in SLI 12G Corsair Dominator GT2000MHz 
Hard DriveOptical DriveOSMonitor
Kingston V300 120G SSD, 2x 1TB Barracuda HD's LG BlueRay/LightScribe Burner Windows 7 Pro 64Bit 3x24", 1x22" LCD 
KeyboardPowerCaseMouse
Wireless Logitech K320 ULTRA X4 1200W Custom Danger Den LDR-29 Wireless Logitech M310 
Mouse Pad
Custom 
  hide details  
Reply
post #19 of 60
the points per task in dnetc seem to be the same per task. difference is if you have a more powerful card or you overclock you get them done quicker which means more points accumulated in a days time since you will be getting more work units processed.
    
CPUMotherboardGraphicsRAM
Intel I7-5820k@4.2ghz Gigabyte X99-UD3P AMD 295x2 16GB G.Skill DDR4-2400 
Hard DriveOptical DriveCoolingOS
Corsair Force GT 120gb (OS), 2x 1tb WD Blacks R... LG Bluray CoolerMaster MasterLiquid Pro 120 Win 10 x64 
MonitorKeyboardPowerCase
26" Emprex Logitech G15 CoolerMaster SPH-1300 Coolermaster HAF922 
MouseMouse Pad
CM Sentinel II CoolerMaster HSM Battle Pad SSK 
  hide details  
Reply
    
CPUMotherboardGraphicsRAM
Intel I7-5820k@4.2ghz Gigabyte X99-UD3P AMD 295x2 16GB G.Skill DDR4-2400 
Hard DriveOptical DriveCoolingOS
Corsair Force GT 120gb (OS), 2x 1tb WD Blacks R... LG Bluray CoolerMaster MasterLiquid Pro 120 Win 10 x64 
MonitorKeyboardPowerCase
26" Emprex Logitech G15 CoolerMaster SPH-1300 Coolermaster HAF922 
MouseMouse Pad
CM Sentinel II CoolerMaster HSM Battle Pad SSK 
  hide details  
Reply
post #20 of 60
Yup, makes sense to me. I would want to see at least a 48 hr or at least an average of 30 tasks to form a useful number since the tasks do vary a bit anyways...

How about Collatz Conjecture? They run ATI, you run that?

Blue Beast
(13 items)
 
  
CPUMotherboardGraphicsRAM
W3670 4.0 GHz (HT On) 1.345v 24/7 Asus P6X58D Premium DUAL EVGA GTX 560 Ti SC 1G in SLI 12G Corsair Dominator GT2000MHz 
Hard DriveOptical DriveOSMonitor
Kingston V300 120G SSD, 2x 1TB Barracuda HD's LG BlueRay/LightScribe Burner Windows 7 Pro 64Bit 3x24", 1x22" LCD 
KeyboardPowerCaseMouse
Wireless Logitech K320 ULTRA X4 1200W Custom Danger Den LDR-29 Wireless Logitech M310 
Mouse Pad
Custom 
  hide details  
Reply
Blue Beast
(13 items)
 
  
CPUMotherboardGraphicsRAM
W3670 4.0 GHz (HT On) 1.345v 24/7 Asus P6X58D Premium DUAL EVGA GTX 560 Ti SC 1G in SLI 12G Corsair Dominator GT2000MHz 
Hard DriveOptical DriveOSMonitor
Kingston V300 120G SSD, 2x 1TB Barracuda HD's LG BlueRay/LightScribe Burner Windows 7 Pro 64Bit 3x24", 1x22" LCD 
KeyboardPowerCaseMouse
Wireless Logitech K320 ULTRA X4 1200W Custom Danger Den LDR-29 Wireless Logitech M310 
Mouse Pad
Custom 
  hide details  
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Overclock.net BOINC Team
Overclock.net › Forums › Overclockers Care › Overclock.net BOINC Team › GPU performance vs power consumption