Glimpse help would be of great value.Tutor wrote:Glimpse,
If you know about water cooling GPUs, I need your help/advice because that's what I plan to do to four or five Titans and an EVGA 8G GTX 980 Classified.
Tutor
How did you check your clocks? I have the same card and it reads boost 1253 MHz, Memory 3000 MHz in msi afterburner.pBarrelas wrote:My card: Asus GTX 980 4GB, GPU Clock 1127 MHz, Boost 1216 MHz, Memory 1753 MHz, Driver 344.60
I meant real clocks at which you render. There is no such thing as boost clock while rendering. You can read its value in MSI afterburner (or any other soft) while rendering.pBarrelas wrote: GPU Clock 1127 MHz, Boost 1216 MHz, Memory 1753 MHz...
Please give us descriptions (as I did) at which setting these speeds are obtained. In current layout these are hard to read. And there is no need to give '1000 samp/px and time' - this carries no useful information - give only ms/s.pBarrelas wrote:
PT 3.46 Ms/s, 1000 samp/px, 00:02:31
PT 3.55 Ms/s, 1000 samp/px, 00:02:28
PT 3.54 Ms/s, 1000 samp/px, 00:02:29
I am very skeptical about 900 series - I just simply don't believe that even hard code optimization will make them faster. If so, all cards will run faster and 980 will remain slow. I feel like we shall wait until 210 chip is released or next generation Titan, at 20nm process.pBarrelas wrote:I wish next release will be faster than 1.55.
I'm using Asus GPU Tweak and I'm not using priority.ratata wrote:How did you check your clocks? I have the same card and it reads boost 1253 MHz, Memory 3000 MHz in msi afterburner.
Are you maybe using priority?
All the speeds were obtained with the scene default settings. I've run them 3 times and mentioned the samples and completion times, as requested by Tutor.smicha wrote:Please give us descriptions (as I did) at which setting these speeds are obtained. In current layout these are hard to read. And there is no need to give '1000 samp/px and time' - this carries no useful information - give only ms/s.
Smicha, thanks for catching my careless error re "s/px" and providing the correction - "ms/s" (I'll correct my earlier post so as not to mislead anyone). Also, thanks for stepping in to lend your vast and excellent advice and observation. Finally, thanks for the reference to your H20 cooled build - it's truly amazing and worthy of emulation.smicha wrote: ... Please give us descriptions (as I did) at which setting these speeds are obtained. In current layout these are hard to read. And there is no need to give '1000 samp/px and time' - this carries no useful information - give only ms/s.
Rarely are questions asked stupid, but frequently the failure/refusal to ask is. So err on the side of always asking.pBarrelas wrote:I'll run the test but 1st I got a couple of questions, since I'm pretty new to Octane:
1-Where can I find Octane Demo scatter.ocs?
2-Why running theses tests using 1.2 and 2.06 demos and not the licensed versions?
I'm sure these are stupid questions but I've been using Octane from last Saturday, so a lot of this stuff is unfamiliar to me!
Sample/px and time is useful to compare different versions of Octane to test for convergence, but you have to compare the images to make it useful. There is a group out there that developed a method to compare noise levels in images but I forget who and where I saw that.Tutor wrote:Smicha, thanks for catching my careless error re "s/px" and providing the correction - "ms/s" (I'll correct my earlier post so as not to mislead anyone). Also, thanks for stepping in to lend your vast and excellent advice and observation. Finally, thanks for the reference to your H20 cooled build - it's truly amazing and worthy of emulation.smicha wrote: ... Please give us descriptions (as I did) at which setting these speeds are obtained. In current layout these are hard to read. And there is no need to give '1000 samp/px and time' - this carries no useful information - give only ms/s.
