can we get someone with nvidia gfx to try this?
Printable View
can we get someone with nvidia gfx to try this?
I'd like to see it on an Nvidia GFX card myself. Unfortunately I believe it only works with Catalyst 9.3 and newer drivers.Quote:
Originally Posted by i found nemo
Thanks man, I'll start from scratch now :DQuote:
Originally Posted by EniGmA1987
@nsegative : Your GPU-Z says your'e using a PCIe x16 2.0 link, but those speeds are no where close! Is that truly a chipset bottleneck? Check your BIOS or something....wow....just wow.
Well the results are the same as the guy with core i7. Shrug.Quote:
Originally Posted by Mechromancer
Oh I see why. His CPU to GPU is ~4GB/s, but he has QPI. You have FSB so yours is not as high. Yes, that is normal for your setup it would seem...Quote:
Originally Posted by nsegative
Hold up, I take that back. The Core 2 guys over here: Intel Forum PCIe Speed Test don't have that issue. Something looks quirky about your results.
That is kinda weird. Both Intel systems with an X2 get the GPU -> CPU speed WAY lower than AMD systems with X2s, yet higher than AMD single card setups. jcool had normal CPU -> GPU bandwidth, but nsegative has half the bandwidth of AMD users. The difference between jcool and nsegative is jcool has an i7 with QPI, and nsegative has older fsb tech. the fsb cant really be limiting his bandwidth that much could it? And either way, why do both Intel based computers with 4870X2s have such low GPU -> CPU bandwidth compared to AMD?
EDIT: ya if you look at the guys with fsb based systems in the Intel section, they have more bandwidth than the AMD guys here with their HTT. So FSB definitely isnt a limiting factor. And since HTT has more bandwidth than FSB, something else is limiting the AMD system bandwidth.
Some people in that thread don't then the guy with the i7 does. Odd.Quote:
Originally Posted by Mechromancer
Ran the test again here are the resutls. Also I don't have any problems running the test in Crossfire. Can run it over and over in Crossfire without crashes but disable Crossfire and it will crash every once in a while.
http://img5.imageshack.us/img5/5568/testfour.png
By knuckles6 at 2009-06-03
Oh that's my new toy on the desktop
Mechromancer said it takes 10-15 minutes somewhere in one of his first posts...?Quote:
Originally Posted by EniGmA1987
Anyway, it's either that program being messed up, or really a driver issue, since we tested so many combinations it's gotta be one of those.
Gonna re-run with raised QPI and PCIe, hold on.
NM, it only takes about 5 minutes to run, if you system works right. I had some probs I had to fix when my system was running it in 10+ minutes.
Run #2...
4870X2, Server 08 x64 SP2, Cat. 9,4 - high QPI (7,68GT/s) and 110Mhz PCIe
http://database.he-computer.de/Bilde...PCIebandw2.jpg
Helped a little, but GPU--> CPU is still way too slow..
Oh and by the way, the app doesn't run with an NV card.. just tried on another i7 rig with a GTX260, just says ati something .dll missing...
I'm in the same boat as you jcool. Seems intel quads are affected by this.
@jcool: Your system is running like most single GPU systems. Other i7 and 4870x2 users are getting full GPU to CPU bandwidth, but not you. If Crossfire is properly enabled, we have another wierd result to figure out.Quote:
Originally Posted by jcool
With an X2, there is no option to enable/disable crossfire.. it's hardwired really, and can't be disabled.
I was just about to try it with my 8800gtx:(Quote:
Originally Posted by jcool
I want CF:mad:
Wow, Jcool's falls on its nose for GPU -> CPU for Peak Bandwidth.
I pushed my overclock up to 4.01Ghz and ran the test again and it made no difference in my results.
Actually disabling catalys AI forces it off into single core mode. :up:Quote:
Originally Posted by jcool
Going back to the weird options in the bios, I enabled the option to split the ht link and I got an extra 75ish mhz with the same vcore but I had to increase the nb volts slightly because its now running above 3.1ghz. But no extra performance for gpu to cpu with that for me.
Yeah, stupid AMD card eh ;)Quote:
Originally Posted by Stoner133
I just ran the tool on my "old" MSI K9A2-Neo with AMD770 chipset and single PCIe slot - Sapphire HD4870 and GPUz says PCIe 2.0 @ x16 - Cat 9.3
All runs crash with GPU Recover when starting the last level of the test but up to there it shows : CPU->GPU 4.1GHZ & GPU->CPU 880Mhz .... I am guessing that is not good?
>>> delete <<<
IT IS NORMAL. Please read my orginal post and check out the link to the AMD Developer Forums. It usually crashes, especially on Vista because stresses the hell out of the system. Run it from the CMD prompt to make things easier too. Take all screen shots before it completely crashes out, maybe around the 256 to 512mb range. I say this because all of us see our max bandwidths well before that point in the test.Quote:
Originally Posted by Russ_64
Mechromancer I did as you said and ran it this time directly from the Command Prompt and low and behold the screen doesn't close when the test completes as can be seen here:
http://img40.imageshack.us/img40/3928/testfive.png
By knuckles6 at 2009-06-04
The only thing this screen shot doesn't display correctly is GPU activity in Catalyst Control Center but thats only because the screen shot is taken after the test completes. That actual activity hits 98% during the test.
You have some huge bandwidth on that rig. Way to go! I've decided it would be more than worth it to go with an X2 card the next time I upgrade. I'll have to see what these DX11 cards can do when they appear in a few months. I love my single GPU, but theres something to be said for single-slot Crossfire configurations. The performance rocks and you have the flexibility of using one or both GPUs as needed.
BTW, Beyond3D Update: http://forum.beyond3d.com/showthread.php?t=54321. They have the luxury of getting a bit more specific, technical information from people in-the-know about these things.
My next cards will still be singles, just because I can use the waterblocks I currently have on them.
Try make BAT file to save result to notepad. (sory ,i dont read from page 1 to end) so i dont know if anyone also post
http://img145.imageshack.us/img145/5494/batulk.jpg
http://img145.imageshack.us/img145/9619/36444339.jpg
Another real world benchmark
http://graphics.stanford.edu/projects/gpubench/
(must make BAT too) but i like this program, because this can work on NVDIOA and ATI, and can find out your max GDDR dan Gigaflops.
Interisting when i tes 9600Gt on X58 and P35, bandwith x58 more bigger , so i think QPI bandwith is true working , with this program you can find out your max bandwith chipset.
remenber to make BAT file (or i uploaded to rapidshare gpu bench modifiend later)
sory if my english not good
just to report:
single 3870, boosting the htt/pcie clocks ~10% each resulted in very similar changes as aGeoM already graphed: cpu->gpu increased pretymuch 10% gpu->cpu didnt move atall (actual changes cpugpu +9% (from 5.197 to 5.628, gpucpu from 5.322 to 5.317. cpu/ram/nb held at 3250/1000/2250 htt/pcie from 2000/100 to 2250/110)
pcie 110 = 5.628/5.317
pcie 120 = 5.798/5.316
pcie 125 = 5.858/5.311(see below)
pcie 130 = 5.924/5.313
pcie 140 = 6.042/5.315
pcie 150 = 6.109/5.318
pcie 160 = 6.105/5.312
pcie 170 = 6.105/5.313
pcie 180 = 6.107/5.316
pcie 190 = 6.099/5.312
pcie 200 = 6.106/5.313 (is this a record for amd 770 nb pcie clock?)
so. it apears neither PCIE clock or HTT clock limits the GPU->cpu trafic. but Both, at the same time, have an effect on cpu->gpu traffic (perhaps during the buffered stage, in nb, that allows for unsynced clocks?) on my machine 125mhz pcie would be an exact ratio of htt clock, so...
nope. its in line.. next theory?
This is what I got with 2 3870 Toxic editions OC 860 both cores and 1265 mem on primary and 1230 mem on secondary. AMD Athlon 64 X2 6000+ OC to 3.4ghz with a multiplier of 13 and HTT Clock @ 260.9mhz and a HTT Speed 1304mhz. This was the first time that I ran the benchmark and got over 3 gb/s every other test was just under 3; I got above 3 when I raised the clock on the HTT Speed to 1304mhz which really didn't increase performance but everyso slightly but that is what it is about though getting the most that you can out of what you have right?
:rofl:http://img520.imageshack.us/img520/785/38382933.th.jpg
http://img520.imageshack.us/img520/633/30wlid.th.jpg
http://img520.imageshack.us/img520/9568/30w.th.jpg
[img=http://img520.imageshack.us/img520/785/38382933.th.jpg]
[img=http://img520.imageshack.us/img520/633/30wlid.th.jpg]
[img=http://img520.imageshack.us/img520/9568/30w.th.jpg]
ok after upgrading to a 5870 and latest 1.7B5 bios for the MSI GD70 i got the following...
way better than before and no crashes anymore.
could you post link for that BIOS DL-oad?Quote:
Originally Posted by xPliziT
http://www.ocztechnologyforum.com/fo...ad.php?t=59825Quote:
Originally Posted by Nedjo
Hmmm I guess its ok right?
http://i271.photobucket.com/albums/j...1/pcietest.jpg