Printable View
http://davibu.interfree.it/opencl/sm...mallptGPU.html
download the .tgz at the bottom of the page and extract.
note: this is not cinebench, the render will never finish!
http://img189.imageshack.us/img189/7351/oclrt.png
you dont have to do 5,000 passes.:p:
sample/sec: 1,123K
Thanks for the post! :)
Looks good... Searching for an extractor app now.
This worked:
http://download.cnet.com/IZArc/3000-...ml?tag=lst-0-1
Looks like the app uses 1 GPU, and at least 4 cores on a CPU...
http://img130.imageshack.us/img130/9163/smallix.th.jpg
4890 seems to do well in this one.
Indeed!!
And this one: http://www.xtremesystems.org/forums/...d.php?t=241794
I think ATI is rocking on OpenCL right now...
Could you verify that you are also just using 1 GPU, with GPU-z?
Actually, I only have one. I forgot to update my sig when I sold the second one :-)Quote:
Originally Posted by Talonman
--Matt
I can't get this thing to run right... it's all fuzzy and runs really slow. I extracted with winrar and ran the app.
It does start out fuzzy...
It is still rendering getting better with each pass.
This is after letting it run about 10 minutes:
http://img43.imageshack.us/img43/1645/longtime.jpg
Wow... :)Quote:
Originally Posted by mattkosem
Thanks!
I hope it supports multi-GPU's soon.
I could use the extra speed. :up:
For trivia, the 1/2 of my air cooled 295 that is working, heated up to 64C max. (Room currently 72.1F)
Even though my GPU's utilization was 99%, it's temp never got excessive.
I think for what it's doing, the app runs rather cool.
I hit 36c at the end with a room temp of 23c. I'm under water though.Quote:
Originally Posted by Talonman
--Matt
http://www.evga.com/forums/tm.aspx?m=91863
We are currently exploring the idea that ATI is running OpenCL apps much faster, but the Nvidia boys seem to have higher CPU utilization in more app than 1.
If it helping us, or hurting us, I can't say, but would love to know...
How does your CPU utilization look?
Note that my Q6600@3.81GHz is at 26% utilization in post #2.
I see 0%.
--Matt
Thanks again Matt...
So where is an expert when you need one? :)
Could that CPU work be hurting our performance, and better suited for the GPU?
I think our OpenCL driver could be miss-routing work, or maybe some funky global memory accessing, also causing higher CPU utilization?
I really have no idea. :shrug:
Nvidia doesn't support OpenCL on the CPU, only the GPU.
yeah, according to gpu-z and ccc, my 2nd gpu is doing nothing. also not seeing any load on the cpu as it runs
Thanks for the post. :)
4870 @ 512mb, 790mhz.
http://img341.imageshack.us/img341/1856/13195137.png
http://img710.imageshack.us/img710/2631/25247002.jpg
http://img710.imageshack.us/img710/6707/gpuzwe.jpg
>> Q9450 @ 3608MHz
>> CPU utilization: ~0%
Code:OpenCL Platform 0: Advanced Micro Devices, Inc.
OpenCL Device 0: Type = TYPE_GPU
OpenCL Device 0: Name = Cypress
OpenCL Device 0: Compute units = 20
OpenCL Device 0: Max. work group size = 256
Reading file 'rendering_kernel.cl' (size 2997 bytes)
OpenCL Device 0: kernel work group size = 256
I was board... and I'm a list guy.
DosDuoNo ---------- GTX 260 --- Sample/sec --- 1093.2K
Chumbucket843 ---- GTX 260 --- Sample/sec --- 1123.2K
Talonman --------- 1/2 a 295 --- Sample/sec --- 1159.2K
NovoRei ------------ 4870 ------ Sample/sec --- 5616.1K
PyrO ----------- 1/2 a 4870X2 -- Sample/sec --- 5796.2K
mattkosem --------- 4890 ------ Sample/sec --- 6751.6K
fellix bg ------------ 5870 ------ Sample/sec -- 13719.6K
Another outstanding ATI Score... :)
DosDuoNo ---------- GTX 260 --- Sample/sec --- 1093.2K
Chumbucket843 ---- GTX 260 --- Sample/sec --- 1123.2K
Talonman --------- 1/2 a 295 --- Sample/sec --- 1159.2K
NovoRei ------------ 4870 ------ Sample/sec --- 5616.1K
PyrO ----------- 1/2 a 4870X2 -- Sample/sec --- 5796.2K
mattkosem --------- 4890 ------ Sample/sec --- 6751.6K
fellix bg ------------ 5870 ------ Sample/sec -- 13719.6K
Toysoldier ---------- 5870 ------ Sample/sec -- 13719.6K
FYI - The programmer is looking to fix us Nvidia guys up:
http://forum.beyond3d.com/showthread.php?t=55913
"Hi, I'm David, the author of SmallptGPU. I think I can clarify few points:
- About the poor performances on Nvidia, I have developed SmallptGPU on an ATI HD4870. Both ATI and NVIDIA OpenCL drivers are in a early stage of the development and have their fair amount of problems/bugs/etc. I have avoided problematic paths on ATI because is my card while I have never tested SmallptGPU on NVIDIA. I assume I'm doing something that the NVIDIA OpenCL driver doesn't like at all. The high CPU usage is a good hint of this problem.
- The sources are available on the web site, so if anyone has a fix for NVIDA cards, I will be happy to apply it.
- SmallptGPU uses the first GPU device available (there is a command line option to run on CPU device). About all the load should be on GPU, CPU is nearly unused. It is not able to use multiple devices at the same time so any SLI/CrossFire configuration will be used only at 50% of its capabilities.
- 5870 is horrible fast ... I' trying to not buy one ;)"
a single 5970
Sweet! :)
DosDuoNo ---------- GTX 260 --- Sample/sec --- 1,093.2K
Chumbucket843 ---- GTX 260 --- Sample/sec --- 1,123.2K
Talonman --------- 1/2 a 295 --- Sample/sec --- 1,159.2K
NovoRei ------------ 4870 ------ Sample/sec --- 5,616.1K
PyrO ----------- 1/2 a 4870X2 -- Sample/sec --- 5,796.2K
mattkosem --------- 4890 ------ Sample/sec --- 6,751.6K
safan80 ------------ 5970 ------ Sample/sec -- 11,012.8K
fellix bg ------------ 5870 ------ Sample/sec -- 13,719.6K
Toysoldier ---------- 5870 ------ Sample/sec -- 13,719.6K
I'd like to update my score.
--Matt
I don't know if we are officially recording them... :)
But glad to!
DosDuoNo ---------- GTX 260 --- Sample/sec --- 1,093.2K
Chumbucket843 ---- GTX 260 --- Sample/sec --- 1,123.2K
Talonman --------- 1/2 a 295 --- Sample/sec --- 1,159.2K
NovoRei ------------ 4870 ------ Sample/sec --- 5,616.1K
PyrO ----------- 1/2 a 4870X2 -- Sample/sec --- 5,796.2K
mattkosem --------- 4890 ------ Sample/sec --- 7,520.9K
safan80 ------------ 5970 ------ Sample/sec -- 11,012.8K
fellix bg ------------ 5870 ------ Sample/sec -- 13,719.6K
Toysoldier ---------- 5870 ------ Sample/sec -- 13,719.6K
i'f you're going to have the scores in a list, i think it might be best to have clock speeds listed too