Does it only work on DX 11 GPUs?

not sure what it actually compares but here are my results

Attached Thumbnails

 

why score is N/A ...

From readme file:
2) You DON'T need DX11 hardware, DirectCompute is supported on DX10 class devices if only drivers support it (forceware >= 190.62, Catalyst currently doesn't support DirectCompute on DX10/10.1 GPUs)

Help me out here. I read the Readme.txt file and it states that this can be run on DX10 hardware as long as DX11 is installed. I also have Windows 7 installed.

But, when I try to start the Benchmark, I get a "No DirectCompute Support" screen pop-up.

EDIT: OK....Catalyst/AMD Does NOT support this at this point on non-DX 11 hardware.

Attached Thumbnails

 

interesting tool. I will give it a try.

Multithreaded, wow my cpu got higher score than the gpu?? We propably need better drivers for directcompute from ati?



Singlethreaded...

EDIT: OH BTW, thx for the maker of this little bench. Keep em coming!! Its impressive how little cpu is used when the gpu is used. Very different from earlier DX10 benchies and games?

Hello,
with Catalyst 9.10 on Win7 x64 and HD5870 (900/1300)

Got M11849 and S11847

Why is there no difference between the multi an single bench ?
HaZe303 got the same results.

Is this a catalyst related issue ?

I´d guess so, think about it, we have only had one official driver since release. And Win7 is new as well, so it could be something MS needs to fix. But dont worry about it, it will get sorted out eventually! Im sure of it.

on my 8800 GTS 512 doesent work, (forceware191.07, vista 32-bit sp2, dx11 trough win update) WHY?????

Can someone having an Nvidia card run this tool? So we get som numbers to compare to?

DirectCompute not working!

Freature level
D3D_FEATURE_LEVEL_10_0

DirectCompute support
NO

Vendor / Device / Subsystem
10DE / 5E2 / 58510DE


NVIDIA GeForce GTX 260
Vista x64
DirectX Redist (March 2009)
DX 11 Update KB971644
DxDiag shows the right version (DX11)

try 195.39 wow it work

Soooo... now we've got to wait just a year or two before somebody puts that into real app, right? :P

thanksssss

You also need DirectCompute enabled driver, get forceware 190 or newer.

How is it my GTX280 scores nigh on 70k whereas my 5870 scores barely 40k lol? Is this supposed to be a comparetive score?

Oh, just noticed people with 5770's are scoring the same as 5870's as well, seems odd considering they've only got half the SP's? xD

Because some drivers aren't mature enough.

I want to clear up any doubts about my benchmark:

1) I don't favour any vendor, the codepath is one and the same for ATI/NV any any other DirectCompute enabled device that could be istalled in the system. It looks like this:

benchmark(compute shader code) -> DirectX11 API (compiler) -> DirectX11 API (run) -> drivers -> GPU.

I personally own HD4890 Toxic as a primary rendering device and GeForce 9800gt as a dedicated PhysX card.

2) Rendering capability is not the same as computing capability - look at FluidMark scores, you can find GTX260/275/295 scores lower or comparable to 9800gt.

3) NVidia has enabled DirectCompute feature in 190.62 WHQL driver (21.08.2009) so we can assume it's more stable and mature than ATI implementation (note that NVidia had CUDA/PhysX for long time so we can assume it was easier for them to enable DirectCompte)

4) Benchmark is still in 0.xx version which means it CAN have some bugs.

And last word about my drivers expectations. I really wait for a DirectCompute enabled drivers for HD4xxx. I know ATI wants to sell DX11 cards but it's not fair that you can have DirectCompute on 9400gt and not on HD4890...

What are the MScore of GPUs?

It's relative performance index, GeForce 9800gt OC (700/1800/1000MHz) reach ~M15000.

Thanks Pat, works nice with my 2x GTX285 i will get GPU Score arround M79000!

But one thing is confuse:
Why does a Q6600 get more CPU Points than a Q9550 with ver.0.25?
Must be the Q9550 not a bit faster than the Q6600?

Best Regards

Well, it depends but generally Q9550 should be faster. What are the exact scores you get? (on v0.25) Look for any backgroud tasks like antivir scans or some other processes that could slowdown. What is the operating system? Same for both CPUs? Finally the memory modules/FSB - same for both platforms?

The CPU-Scores are both from ver.0.25
Q6600 - Vista x32: M2716
Q9550 - Win7 x64: M1066

Ok I must ask the guys, if the use the same Mem-Moduls and FSB, also the Background tasks.

Thanks for info & you fast reply!

Best Regards

Pat,

What are the single- and multi-threaded aspects of this benchmark and how do they apply to GPU's?

What development environment are you using for creating this benchmark, just curious.

Pat,
Great tool! Any chance we could get the source code? I'm working my own DC project and it looks like you've cracked the code on a number of things (e.g. device enumeration, running shaders that take >5 sec, etc).

Thanks!

I've just uploaded OpenCL enabled benchmark. Now you can select the GPGPU API. The tool is here.

Please try it, especially OpenCL part. Please also read the short readme if you're not sure about OpenCL support in your system.

Hi,
I try the new 195.62 ForceWare with my GTX285 and a 955 BE CPU all @ stock:

I must switch the PhysX off in cause of openCL Bench will crash before it start running
Also my values look a little bit low, compared to your 9800GT.

I´m all ready reading the Thread about the PhysX mod driver.
GenL is working on 195.62, can this be help to get better Score?
Because you writing also the directComputing and the opneCL Score are the same with the mod driver.

So far, I hope reading from you...

I've just created a dedicated thread for DirectCompute benchmark

It's here so please post any comments/questions/bugs there.

Here is my score, on a C2D E6300 1.86@3.04Ghz on version 0.43b

ATI Radeon HD 4800 Series @ 750 MHz (1002 / 9440 / 851174B)
Intel(R) Core(TM)2 CPU 6300 @ 1.86GHz (2 logical CPUs)

DirectCompute: D2364.9
OpenCL: C1882.3
CPU: M17.7

Windows 7 x64 Ultimate Edition (build 7600)

Question is what does that mean ??? is it good? bad ?

AMD Athlon(tm) 64 X2 Dual Core Processor 6000+ (2 logical CPUs)
ATI Radeon HD 4800 Series @ 750 MHz (1002 / 9440 / 851174B)
ATI Radeon Kernel Mode Driver
atikmdag 8.01.01.984
Windows 7 x64 Home Premium Edition (build 7600)

Version 0.35
DirectCompute: D137968
OpenCL: N/A
CPU: M656

Version 0.43b
DirectCompute: D2389.9
OpenCL: N/A
CPU: M13.3

Version 0.44b
DirectCompute: D2382.6
OpenCL: N/A
CPU: M16.9

NVIDIA GeForce GTX 285 @ 1512 MHz (10DE / 5E3 / C9119F1)
Intel(R) Core(TM) i7 CPU 920 @ 2.67GHz (8 logical CPUs) (3.2GHz)

Version 0.44b
DirectCompute: D804.1
OpenCL: C1907.9
CPU: M62.1

Windows 7 x64 (build 7600)

Can you confirm the GPU Load factor (GPU-Z sensor tab) during those tests?

Please, reply here. We've moved the thread.

This is 4870 result I guess? If so, that's good 4890 OC'ed can reach D3200/C2500.

It means your Radeon floating point performance is about 133x higher than CPU.

Please, write any comments/questions/results here. We've already moved the thread.

Here is MY score, on a AMD Phenom BlackEdition version 0.35

Driver Version:
ATI Radeon Kernel Mode Driver
atikmdag 8.01.01.984
Windows 7 x64 Ultimate Edition (build 7600)

Feature Level:
DX10.1

Vendor/Device/Subsystem:
1002 / 9440 / 22671787

DirectCompute Score: D150746
CPU Score: M946

Xfire: simonlange