Page 16 of 17

Re: PPD Database - HFM exports needed

Posted: Wed May 27, 2020 11:35 am
by skydivingcatfan
foldinghomealone wrote:The database is currently under maintanance.
There is an issue that that WU name and path don't match.
I don't know what messed this up, probably it was me.

The PPD and HW stats seem to be ok, though.

Currently I don't have time to set up everything again.
There are too many uploaded single files I have to process manually.

Thanks for your work on this.

Re: PPD Database - HFM exports needed

Posted: Wed May 27, 2020 1:03 pm
by NoMoreQuarantine
skydivingcatfan wrote:That is interesting. The opposite of what was reported on GamersNexus and the database on TechPowerup.

How did you determine that you had a TU106?
I can tell you from my end that GPUs.txt does not differentiate between TU106 2060s and TU104 versions and reports them all as TU106-200A. I suspect that's what markdotgooley is looking at. GPU-Z should be used to check the actual GPU.

Re: PPD Database - HFM exports needed

Posted: Wed May 27, 2020 1:06 pm
by NoMoreQuarantine
NormalDiffusion wrote:
NoMoreQuarantine wrote: I have a backup I made on the 18th. It looks like the name and path lined up then.

https://docs.google.com/spreadsheets/d/ ... sp=sharing
It's also showing the problem. Check the 16425 projects, you can see it (reported in path as 13xxx).
Ah, you are correct. I only checked the first couple, but there are mismatches interspersed throughout.

Re: PPD Database - HFM exports needed

Posted: Wed May 27, 2020 5:37 pm
by markdotgooley
NoMoreQuarantine wrote:
skydivingcatfan wrote:That is interesting. The opposite of what was reported on GamersNexus and the database on TechPowerup.

How did you determine that you had a TU106?
I can tell you from my end that GPUs.txt does not differentiate between TU106 2060s and TU104 versions and reports them all as TU106-200A. I suspect that's what markdotgooley is looking at. GPU-Z should be used to check the actual GPU.
All I can find on this Ubuntu Linux machine is what's on the FAHControl application, and the 2060 KO is called a TU106-200A and the vanilla 2060 is called a TU-106. I don't know where else to look. The NVIDIA X Server Settings calls both merely RTX 2060. Clueless about where else to look.

Re: PPD Database - HFM exports needed

Posted: Sun May 31, 2020 4:13 am
by bruce
I'd think that both GPU-Z and TechPowerUp would differentiate based on their database. The real key is the device code

What does GPU-Z say about both versions ... the TU106 2060s and TU104 versions. Specifically, the reported device codes.

Re: PPD Database - HFM exports needed

Posted: Mon Jun 01, 2020 8:11 pm
by NoMoreQuarantine
bruce wrote:I'd think that both GPU-Z and TechPowerUp would differentiate based on their database. The real key is the device code

What does GPU-Z say about both versions ... the TU106 2060s and TU104 versions. Specifically, the reported device codes.
That's a good point. GPU-Z gives the device ID 10DE 1E89 for my 2060 KO. A 2060 binned TU106 should report 10DE 1F08 I believe.

markdotgooley, you can get this info in Linux using

Code: Select all

lspci -nn

Re: PPD Database - HFM exports needed

Posted: Mon Jun 01, 2020 10:52 pm
by bruce
What corrections would you suggest?
NVidia 0x1e81 TU104 [GeForce RTX 2080 Super]
NVidia 0x1e82 TU104 [GeForce RTX 2080]
NVidia 0x1e84 TU104 [GeForce RTX 2070 Super] 8218
NVidia 0x1e87 TU104 [GeForce RTX 2080 Rev. A] 10068
NVidia 0x1e89 TU106-200A [GeForce 2060 KO]
NVidia 0x1e90 TU104M [GeForce RTX 2080 Mobile]
NVidia 0x1e91 TU104M [GeForce RTX 2070 Super Mobile / Max-Q]
NVidia 0x1e93 TU104M [GeForce RTX 2080 Super Mobile / Max-Q]
NVidia 0x1eab TU104M
NVidia 0x1eae TU104M
NVidia 0x1eb0 TU104GL [Quadro RTX 5000]
NVidia 0x1eb1 TU104GL [Quadro RTX 4000]
NVidia 0x1eb5 TU104GLM [Quadro RTX 5000 Mobile / Max-Q]
NVidia 0x1eb6 TU104GLM [Quadro RTX 4000 Mobile / Max-Q]
NVidia 0x1eb8 TU104GL [Tesla T4] 8141
NVidia 0x1eb9 TU104GL
NVidia 0x1ebe TU104GL
NVidia 0x1ec2 TU104 [GeForce RTX 2070 SUPER]
NVidia 0x1ed0 TU104BM [GeForce RTX 2080 Mobile]
NVidia 0x1f06 TU106 [GeForce RTX 2060 Super]
NVidia 0x1f08 TU106 [Geforce RTX 2060]
NVidia 0x1f11 TU106 [Geforce RTX 2060 Mobile]
NVidia 0x1f12 TU106M [GeForce RTX 2060 Mobile / Max-Q] 4550
NVidia 0x1f15 TU106M [GeForce RTX 2060 Mobile]
NVidia 0x1f42 TU106 [GeForce RTX 2060 SUPER] 7181
NVidia 0x1f47 TU106 [GeForce RTX 2060 Super]
NVidia 0x1f51 TU106M [GeForce RTX 2060 Mobile] 4608

Re: PPD Database - HFM exports needed

Posted: Tue Jun 02, 2020 12:17 am
by NoMoreQuarantine
bruce wrote:What corrections would you suggest?
I would keep it simple and just label it 2060, but change it to TU104. I can tell the current format is the same as that used by devicehunt.com. The format is pretty good, but I would love to know what "GL" & "GLM" stands for if you know. Also, my OCD is being triggered by the inconsistent use of upper and lower case Super/SUPER (the brand uses all upper case FYI), and that some entries don't have any names at all. I notice you are adding the GFLOPS to the end of some of the entries; would you like me to find that info for the rest?
NVidia 0x1e81 TU104 [GeForce RTX 2080 Super]
NVidia 0x1e82 TU104 [GeForce RTX 2080]
NVidia 0x1e84 TU104 [GeForce RTX 2070 Super] 8218
NVidia 0x1e87 TU104 [GeForce RTX 2080 Rev. A] 10068
NVidia 0x1e89 TU104 [GeForce RTX 2060]
NVidia 0x1e90 TU104M [GeForce RTX 2080 Mobile]
NVidia 0x1e91 TU104M [GeForce RTX 2070 Super Mobile / Max-Q]
NVidia 0x1e93 TU104M [GeForce RTX 2080 Super Mobile / Max-Q]
NVidia 0x1eab TU104M
NVidia 0x1eae TU104M
NVidia 0x1eb0 TU104GL [Quadro RTX 5000]
NVidia 0x1eb1 TU104GL [Quadro RTX 4000]
NVidia 0x1eb5 TU104GLM [Quadro RTX 5000 Mobile / Max-Q]
NVidia 0x1eb6 TU104GLM [Quadro RTX 4000 Mobile / Max-Q]
NVidia 0x1eb8 TU104GL [Tesla T4] 8141
NVidia 0x1eb9 TU104GL
NVidia 0x1ebe TU104GL
NVidia 0x1ec2 TU104 [GeForce RTX 2070 SUPER]
NVidia 0x1ed0 TU104BM [GeForce RTX 2080 Mobile]
NVidia 0x1f06 TU106 [GeForce RTX 2060 Super]
NVidia 0x1f08 TU106 [Geforce RTX 2060]
NVidia 0x1f11 TU106 [Geforce RTX 2060 Mobile]
NVidia 0x1f12 TU106M [GeForce RTX 2060 Mobile / Max-Q] 4550
NVidia 0x1f15 TU106M [GeForce RTX 2060 Mobile]
NVidia 0x1f42 TU106 [GeForce RTX 2060 SUPER] 7181
NVidia 0x1f47 TU106 [GeForce RTX 2060 Super]
NVidia 0x1f51 TU106M [GeForce RTX 2060 Mobile] 4608

Re: PPD Database - HFM exports needed

Posted: Tue Jun 02, 2020 5:28 pm
by toTOW
I made the change, my source agrees with yours on :
1e89 TU104 [GeForce RTX 2060]

All Super are now SUPER since my source also label them like this. I also corrected some missing designations I saw in the process.

Re: PPD Database - HFM exports needed

Posted: Tue Jun 02, 2020 5:53 pm
by NoMoreQuarantine
toTOW wrote:I made the change, my source agrees with yours on :
1e89 TU104 [GeForce RTX 2060]

All Super are now SUPER since my source also label them like this. I also corrected some missing designations I saw in the process.
Thanks! I know it's a small change, but it could prevent future confusion. :)

Re: PPD Database - HFM exports needed

Posted: Tue Jun 02, 2020 10:43 pm
by markdotgooley
NoMoreQuarantine wrote:
bruce wrote:I'd think that both GPU-Z and TechPowerUp would differentiate based on their database. The real key is the device code

What does GPU-Z say about both versions ... the TU106 2060s and TU104 versions. Specifically, the reported device codes.
That's a good point. GPU-Z gives the device ID 10DE 1E89 for my 2060 KO. A 2060 binned TU106 should report 10DE 1F08 I believe.

markdotgooley, you can get this info in Linux using

Code: Select all

lspci -nn
Thanks! One card is 10de:1f08 or TU106 [GeForce RTX 2060 Rev. A]
and the other 10de:1e89 or TU104 [GeForce RTX 2060].

Which I guess makes sense. I hadn't known where to look.

Re: PPD Database - HFM exports needed

Posted: Sun Jun 14, 2020 4:56 pm
by ThWuensche
Radeon VII performance, as observed on my side and as mentioned by user NormalDiffussion, of around 1.2MPPD compared to 1.635MPPD mentioned in the table raises questions and makes me consider where the difference might come from. So it would be nice to have information how the 1.635MPPD have been achieved. That is about 35% higher than achieved here and even considering the (double effect of speed on) bonus points for faster return indicates a serious performance difference.

I wonder whether I could improve the performance of the four Radeon VII contributed to the project by me. Information might be helpful for others, too. My setup is stock clocks and voltages for the GPUs, Linux Debian Buster with CPU Ryzen 3700X on one, Ryzen 3900X on the other machine. Graphics driver is upstream kernel amdgpu (kernel 5.5/5.7) with ROCM opencl implementation.

What is performance on Linux with AMD proprietary driver, any experience? What is performance on Windows? Are the results in the table coming from overclocked Radeon VIIs or stock?

Would be appreciated if the contributors of data on Radeon VII for the table could share information about their systems. Thanks!

On another issue the bonus PPD are a decision by the project to increase incentive, but are misleading regarding the true performance and contribution of different GPUs to the project. Thus information on PPD for base points would be nice to have in the table - but that's probably not so easy to achieve?

Thanks for the great contribution with this work. It helps to judge whether the own system performs appropriate.

Re: PPD Database - HFM exports needed

Posted: Mon Jun 15, 2020 10:05 am
by Juggy
My 3950X should be arriving in the next 2 weeks, I will submit some CPU data for you then.

Re: PPD Database - HFM exports needed

Posted: Mon Jun 15, 2020 3:09 pm
by Ichbin3
ThWuensche wrote:My setup is ...
Four cards on how many computers?

Re: PPD Database - HFM exports needed

Posted: Mon Jun 15, 2020 4:07 pm
by ThWuensche
Ichbin3 wrote:
ThWuensche wrote:My setup is ...
Four cards on how many computers?
Two, one Ryzen 3700X, one Ryzen 3900X. The Ryzen 3900X received the second GPU only today, since a power supply upgrade was required to support two GPUs, so for the last days was working with one GPU only. But the PPD for each GPU is approximately the same, independant of how many GPUs are in the computer, so PPD is not limited by CPU overload.