Jump to content

CUDA-Z Info+Bench (Nvidia only) - updated Oct 2012


  • Please log in to reply
74 replies to this topic

#41
Flashe

Flashe

    Flashy ~ Flasheu

  • Members
  • PipPipPipPipPip
  • 298 posts
  • Location:93 Carats
Hello too all,

The application asks me to do an update when I have the latest driver 4.0.17 with Mac OS X 10.7 (11A494a).

Posted Image

This problem since the update 11A480e to 11A494a.Why ?

Someone one knows the reason for this problem.


Thanks

#42
mitch_de

mitch_de

    InsanelyMacaholic

  • Local Moderators
  • 2,884 posts
  • Gender:Male
  • Location:Stuttgart / Germany
Nope, i think Nvidia CUDA Prefpane simple didnt now the newest LION build and therefore it says "update".
For my knowledge there are no newer CUDA drivers.

#43
myrorym

myrorym

    InsanelyMac Protégé

  • Members
  • PipPip
  • 79 posts
  • Gender:Male

This problem since the update 11A480e to 11A494a.Why ?

Someone one knows the reason for this problem.

Following n et k a s, I think there is a solution for you.
Check this

This update brings opencl support for fermi to Snow Leopard and breaks cuda
To get CUDA back, execute this command in terminal:

cp /Library/Frameworks/CUDA.framework/Versions/A/Libraries/libcuda_256.01.10.dylib /Library/Frameworks/CUDA.framework/Versions/A/Libraries/libcuda_256.02.05.dylib


CUDA needs a libcuda_270.05.05.dylib in this directory.

try to copy cudaframework for previous Lion gfx driver... look in directory to see what CUDA put there. sub that for the 270.xx.xx. I not sure, not at CUDA driven pc atm, but this should work

cp /Library/Frameworks/CUDA.framework/Versions/A/Libraries/libcuda_270.xx.xx.dylib /Library/Frameworks/CUDA.framework/Versions/A/Libraries/libcuda_270.05.05.dylib

I haven't done LION yet, but I imagine this is what CUDA is looking for.

#44
Flashe

Flashe

    Flashy ~ Flasheu

  • Members
  • PipPipPipPipPip
  • 298 posts
  • Location:93 Carats
Hi myrorym,

I did not understand your explain to me.I made a copy / paste in terminal ?

cp /Library/Frameworks/CUDA.framework/Versions/A/Libraries/libcuda_256.01.10.dylib /Library/Frameworks/CUDA.framework/Versions/A/Libraries/libcuda_256.02.05.dylib

Posted Image

Can you give me 270.05.05.dylib please so with this file removes the message.Because I have no file in this folder.

Look in my picture above.

#45
mamadou

mamadou

    InsanelyMac Geek

  • Members
  • PipPipPip
  • 118 posts
Salut Flashe
to disable naggy startup cuda message you can delete file /Library/LaunchAgents/com.nvidia.CUDASoftwareUpdate.plist
Voilà voilou.

#46
sir_bazz

sir_bazz

    InsanelyMac Protégé

  • Members
  • Pip
  • 16 posts
  • Gender:Male
  • Location:Melb, Australia

Can you give me 270.05.05.dylib please so with this file removes the message.Because I have no file in this folder.

Look in my picture above.


Try this cmd

cp /Library/Frameworks/CUDA.framework/Versions/A/Libraries/libcuda_270.05.00.dylib /Library/Frameworks/CUDA.framework/Versions/A/Libraries/libcuda_270.05.05.dylib

I just had a similar problem with the cuda-z app not finding cuda and that fixed it for me.

bazz.

#47
myrorym

myrorym

    InsanelyMac Protégé

  • Members
  • PipPip
  • 79 posts
  • Gender:Male
cp /Library/Frameworks/CUDA.framework/Versions/A/Libraries/libcuda_270.05.00.dylib /Library/Frameworks/CUDA.framework/Versions/A/Libraries/libcuda_270.05.05.dylib
Yes that is it. You are just copying a previous version of the file to create a new file, one that matches your current video driver.
CUDA update 4.0.17 was before latest LION DP update. CUDA looks for dylib file matching version of your video driver. This creates that file by copying previous version.
I put xx.xx because I wasn't sure what would be in your folder.

#48
Flashe

Flashe

    Flashy ~ Flasheu

  • Members
  • PipPipPipPipPip
  • 298 posts
  • Location:93 Carats
Ok,

The problem is still the same with that file, this message still appears.

Posted Image

As the photo top .

Salut Flasheto disable naggy startup cuda message you can delete file /Library/LaunchAgents/com.nvidia.CUDASoftwareUpdate.plistVoilà voilou

Salut mamadou,
Le problème reste le meme avec ta méthode aucun changement ainsi que les autres à la suite de mon méssage.

Posted Image

#49
myrorym

myrorym

    InsanelyMac Protégé

  • Members
  • PipPip
  • 79 posts
  • Gender:Male

Ok,

The problem is still the same with that file, this message still appears.

Not sure then. Can you use CUDA successfully? If so, the warning would appear cosmetic. I still 10.6.7 so I can't really test, but the command should make CUDA usable if it hadn't worked previously

#50
mitch_de

mitch_de

    InsanelyMacaholic

  • Local Moderators
  • 2,884 posts
  • Gender:Male
  • Location:Stuttgart / Germany
Yep - or simple remove the prefpane CUDA - its useless (mostly).
The update function often fails and this is the only usage of that prefpane.
CUDA works complete independed of that prefpane !
To remove that right click on that CUDA prefpane (system settings) and then click on the remove...

You can always get here the newest CUDA (win/os x / linux)
http://developer.nvi.../cuda-downloads

#51
Carstiman

Carstiman

    InsanelyMac Geek

  • Members
  • PipPipPip
  • 116 posts
  • Gender:Male
hello, what is the solution this time for the new driver 256.02.25f1v1 in 10.6.8 ?
i used:
cp /Library/Frameworks/CUDA.framework/Versions/A/Libraries/libcuda_256.01.10.dylib /Library/Frameworks/CUDA.framework/Versions/A/Libraries/libcuda_256.02.25.dylib
but it wont work.

for the last 256.02.05f01 driver:
cp /Library/Frameworks/CUDA.framework/Versions/A/Libraries/libcuda_256.01.10.dylib /Library/Frameworks/CUDA.framework/Versions/A/Libraries/libcuda_256.02.05.dylib
worked.

thx in advane

ps: could i add a cuda 3.2 lib too that i could use boinc collatz cuda 3.2 apps ?

#52
mitch_de

mitch_de

    InsanelyMacaholic

  • Local Moderators
  • 2,884 posts
  • Gender:Male
  • Location:Stuttgart / Germany
Netkas found out:
http://netkas.org/?p=846
In short: That’s trash, not drivers!

#53
Gringo Vermelho

Gringo Vermelho

    The Jan Bird fix

  • Supervisors
  • 6,111 posts
  • Gender:Male
  • Location:Brazil
CUDA works on Fermi again on 10.6.8 with the latest CUDA 4.0.19 drivers + June 24 Quadro 4000 drivers. No modifications needed.
http://www.insanelym...p...t&p=1705325
CUDA-Z works, but the CUDA preference pane didn't tell me about the 4.0.19 drivers.

Haven't tried anything OpenCL yet.

#54
Carstiman

Carstiman

    InsanelyMac Geek

  • Members
  • PipPipPip
  • 116 posts
  • Gender:Male

the CUDA preference pane didn't tell me about the 4.0.19 drivers.


Hi, CUDA preference works for me with 4.0.19

#55
Graebags

Graebags

    InsanelyMac Geek

  • Members
  • PipPipPip
  • 129 posts
  • Gender:Male
  • Location:Canberra
Worked for me with 4.0.19 till the first reboot. now wants update! update! update! damn!



Think its about the same with 511 as DP4, despite new Nvidia GPU and Cuda drivers.

Attached Files



#56
meroy

meroy

    InsanelyMac Protégé

  • Members
  • Pip
  • 46 posts
Hi all. My GTX 295 Co-op edition is up and running in Snow Leopard 10.6.8 using Apple's native drivers in 10.6.8 and the OpenGL framework. There are 2 GPUs on this card, therefore x2 the results.

I added a post to a thread on insanelymac to give solution for GTX 295 owners who want to update to 10.6.8:

http://www.insanelym...p;#entry1716804

Basically, there are changes made in 10.6.8 which render the existing EFI string useless unless 2 extensions are installed in /Extra/Extensions.

The following is the result of combining both GPUs on the GTX 295 card for the N-Body demo in CUDA 4.0.

./nbody -numdevices=2 -n=61440 -benchmark

[nbody] starting...
Run "nbody -benchmark [-n=<numBodies>]" to measure perfomance.
	-fullscreen (run n-body simulation in fullscreen mode)
	-fp64	   (use double precision floating point values for simulation)
	-numdevices=N (use first N CUDA devices for simulation)

> Windowed mode
> Simulation data stored in system memory
> Single precision floating point simulation
> 2 Devices used for simulation
> Compute 1.3 CUDA device: [GeForce GTX 295]
> Compute 1.3 CUDA device: [GeForce GTX 295]

61440 bodies, total time for 10 iterations: 697.011 ms
= 54.158 billion interactions per second
= 1083.161 single-precision GFLOP/s at 20 flops per interaction

[nbody] test results...
PASSED

Attached Files



#57
mitch_de

mitch_de

    InsanelyMacaholic

  • Local Moderators
  • 2,884 posts
  • Gender:Male
  • Location:Stuttgart / Germany
Great nbody result!

Here my 9600GT:

nbody -n=61440 -benchmark
[nbody] starting...
Run "nbody -benchmark [-n=<numBodies>]" to measure perfomance.
-fullscreen (run n-body simulation in fullscreen mode)
-fp64 (use double precision floating point values for simulation)
-numdevices=N (use first N CUDA devices for simulation)
> Windowed mode
> Simulation data stored in video memory
> Single precision floating point simulation
> 1 Devices used for simulation

> Compute 1.1 CUDA device: [GeForce 9600 GT]
61440 bodies, total time for 10 iterations: 4812.126 ms
= 7.845 billion interactions per second
= 156.890 single-precision GFLOP/s at 20 flops per interaction
[nbody] test results...
PASSED
GA_EP35:nbody ami$


For others: compiled nbody attatched!
use nbody -n=61440 -benchmark for comparing our results
without parameter -benchmark you will get an window which shows what happens (simulated)
I added the nbody resuls and nbody to first posting.
Happy nbody benching - which is much more usefull than the small (bandwithtest) CUDA-Z performance values.

Attached Files



#58
mm67

mm67

    InsanelyMac Legend

  • Members
  • PipPipPipPipPipPipPip
  • 673 posts
  • Gender:Male
  • Location:Finland
GTX 460 results :
[nbody] starting...
Run "nbody -benchmark [-n=<numBodies>]" to measure perfomance.
-fullscreen (run n-body simulation in fullscreen mode)
-fp64 (use double precision floating point values for simulation)
-numdevices=N (use first N CUDA devices for simulation)

> Windowed mode
> Simulation data stored in video memory
> Single precision floating point simulation
> 1 Devices used for simulation
> Compute 2.1 CUDA device: [GeForce GTX 460]
61440 bodies, total time for 10 iterations: 2008.823 ms
= 18.791 billion interactions per second
= 375.829 single-precision GFLOP/s at 20 flops per interaction
[nbody] test results...
PASSED

#59
meroy

meroy

    InsanelyMac Protégé

  • Members
  • Pip
  • 46 posts
My EVGA GTX 295 Co-op edition is over-clocked and has been running like that since several years now. The settings has allowed it to match a GTX 285 (times 2) in CUDA performance. I'm going to restore the 2 ROMs (one per GPU) back to the factory settings and report back the N-body results for the standard clocks.

Using the factory clocks, the N-body results against both GPUs running simultaneously are: (via -numdevices=2)

./nbody -n=61440 -numdevices=2 -benchmark
[nbody] starting...
Run "nbody -benchmark [-n=<numBodies>]" to measure perfomance.
	-fullscreen (run n-body simulation in fullscreen mode)
	-fp64	   (use double precision floating point values for simulation)
	-numdevices=N (use first N CUDA devices for simulation)

> Windowed mode
> Simulation data stored in system memory
> Single precision floating point simulation
> 2 Devices used for simulation
> Compute 1.3 CUDA device: [GeForce GTX 295]
> Compute 1.3 CUDA device: [GeForce GTX 295]
61440 bodies, total time for 10 iterations: 847.679 ms
= 44.532 billion interactions per second
= 890.638 single-precision GFLOP/s at 20 flops per interaction
[nbody] test results...
PASSED

And the N-body result against one GPU on the GTX 295:

./nbody -n=61440 -benchmark
[nbody] starting...
Run "nbody -benchmark [-n=<numBodies>]" to measure perfomance.
	-fullscreen (run n-body simulation in fullscreen mode)
	-fp64	   (use double precision floating point values for simulation)
	-numdevices=N (use first N CUDA devices for simulation)

> Windowed mode
> Simulation data stored in video memory
> Single precision floating point simulation
> 1 Devices used for simulation
> Compute 1.3 CUDA device: [GeForce GTX 295]
61440 bodies, total time for 10 iterations: 1687.764 ms
= 22.366 billion interactions per second
= 447.322 single-precision GFLOP/s at 20 flops per interaction
[nbody] test results...
PASSED


#60
mitch_de

mitch_de

    InsanelyMacaholic

  • Local Moderators
  • 2,884 posts
  • Gender:Male
  • Location:Stuttgart / Germany
I will collect all nbody results in first posting (bottom of first posting) for easy look+compare :worried_anim:





0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users

© 2014 InsanelyMac  |   News  |   Forum  |   Downloads  |   OSx86 Wiki  |   Mac Netbook  |   PHP hosting by CatN  |   Designed by Ed Gain  |   Logo by irfan  |   Privacy Policy