4070ti pdd?

[finder]

The average PDD for the 18717 task of the collapsed database update is 8,780,000

Why can't my 4070TI computing capacity increase, can only maintain at about 6,000,000？

Power consumption is maintained at 130W-150W and cannot be increased to 200W

[JimboPalmer]

Welcome to Folding@Home!

viewtopic.php?t=26036
Items 6 to 8 show how to post a log of your folding setup, usually to first 150 lines have your configuration info and are the most interesting.

[finder]

Welcome to Folding@Home!

viewtopic.php?t=26036
Items 6 to 8 show how to post a lot of your folding setup, usually to first 150 lines have your configuration info and are the most interesting.

Lines 6-8 only publish logs and do not address the problem of reaching average speed

[JimboPalmer]

Ok, sorry to attempt to help. Carry on.

[BobWilliams757]

Welcome to Folding@Home!

viewtopic.php?t=26036
Items 6 to 8 show how to post a lot of your folding setup, usually to first 150 lines have your configuration info and are the most interesting.

Lines 6-8 only publish logs and do not address the problem of reaching average speed

The problem of reaching average speed depends on quite a few factors. Often some of the most important are shown in the logs which tell people if the proper drivers and configuration are in place.

There are plenty of other things impacting average speed, including being aware that it's just an average and far from an absolute. The more information you provide the more chance someone can point you in the right direction.

[finder]

my computer
CPU I9 10900kf
RAM 32gb 3200
gpu 4070TI

[Joe_H]

my computer
CPU I9 10900kf
RAM 32gb 3200
gpu 4070TI

So that tells us the absolute bare minimum information about your system. It provides no information about how the client software sees the hardware and software of the system it is running on, and also provides no information on how the client itself is configured.

If you are looking for actual advice and assistance in figuring out how your system is functioning on the F@h WUs, I would suggest providing the actual information requested instead of dribbling out one piece of information at a time

[finder]

my computer
CPU I9 10900kf
RAM 32gb 3200
gpu 4070TI

So that tells us the absolute bare minimum information about your system. It provides no information about how the client software sees the hardware and software of the system it is running on, and also provides no information on how the client itself is configured.

If you are looking for actual advice and assistance in figuring out how your system is functioning on the F@h WUs, I would suggest providing the actual information requested instead of dribbling out one piece of information at a time

Does the underlying hardware not provide accurate PDD numbers？

[Lazvon]

Does the underlying hardware not provide accurate PDD numbers？

No, just knowing what kind of hardware you have does nothing for anyone to answer your question.

Folks have asked you to paste the logs as outlined in a linked post in order to be able to help answer your question, but you have NOT posted them multiple times. The log messages could help those who know what to look for in the logs guide you a bit. Is there an old driver, is something not being recognized, are all of your CPU cores folding and none are available for your GPU, have you somehow misconfigured your folding slots... or the like.

In other words, if you want help - post your logs as requested. If you don't understand that, then I'm not sure there is anything anyone can do to help you.

Here is the tiniest bit of useless help I can give you. In your first question, you mentioned you are only seeing the card draw 130-150W and while the words you use "cannot be increased to" make it sound like you are probably using a translator - or you really thought you could go and set the wattage to 200W somewhere. I'll assume a translator and what you were trying to say was that "you were expecting it to draw 200W and not the 130-150W" you are seeing. If that's the case, then perhaps your cooling is inadequate and it is having to throttle your graphics card.

Good luck, but remember - post your logs as requested if you want someone to really help you.

[Joe_H]

Does the underlying hardware not provide accurate PDD numbers？

By itself, no.

[finder]

Does the underlying hardware not provide accurate PDD numbers？

No, just knowing what kind of hardware you have does nothing for anyone to answer your question.

Folks have asked you to paste the logs as outlined in a linked post in order to be able to help answer your question, but you have NOT posted them multiple times. The log messages could help those who know what to look for in the logs guide you a bit. Is there an old driver, is something not being recognized, are all of your CPU cores folding and none are available for your GPU, have you somehow misconfigured your folding slots... or the like.

In other words, if you want help - post your logs as requested. If you don't understand that, then I'm not sure there is anything anyone can do to help you.

Here is the tiniest bit of useless help I can give you. In your first question, you mentioned you are only seeing the card draw 130-150W and while the words you use "cannot be increased to" make it sound like you are probably using a translator - or you really thought you could go and set the wattage to 200W somewhere. I'll assume a translator and what you were trying to say was that "you were expecting it to draw 200W and not the 130-150W" you are seeing. If that's the case, then perhaps your cooling is inadequate and it is having to throttle your graphics card.

Good luck, but remember - post your logs as requested if you want someone to really help you.

Yeah, I expect him to operate on high power

[Lazvon]

Yeah, I expect him to operate on high power

Without logs, no one will be able to help you.

[finder]

Code: Select all

06:56:23:WU00:FS01:0x22:       Core: Core22
06:56:23:WU00:FS01:0x22:       Type: 0x22
06:56:23:WU00:FS01:0x22:    Version: 0.0.20
06:56:23:WU00:FS01:0x22:     Author: Joseph Coffland <joseph@cauldrondevelopment.com>
06:56:23:WU00:FS01:0x22:  Copyright: 2020 foldingathome.org
06:56:23:WU00:FS01:0x22:   Homepage: https://foldingathome.org/
06:56:23:WU00:FS01:0x22:       Date: Jan 20 2022
06:56:23:WU00:FS01:0x22:       Time: 01:15:36
06:56:23:WU00:FS01:0x22:   Revision: 3f211b8a4346514edbff34e3cb1c0e0ec951373c
06:56:23:WU00:FS01:0x22:     Branch: HEAD
06:56:23:WU00:FS01:0x22:   Compiler: Visual C++
06:56:23:WU00:FS01:0x22:    Options: /TP /nologo /EHa /wd4297 /wd4103 /O2 /Zc:throwingNew /MT
06:56:23:WU00:FS01:0x22:             -DOPENMM_VERSION="\"7.7.0\""
06:56:23:WU00:FS01:0x22:   Platform: win32 10
06:56:23:WU00:FS01:0x22:       Bits: 64
06:56:23:WU00:FS01:0x22:       Mode: Release
06:56:23:WU00:FS01:0x22:Maintainers: John Chodera <john.chodera@choderalab.org> and Peter Eastman
06:56:23:WU00:FS01:0x22:             <peastman@stanford.edu>
06:56:23:WU00:FS01:0x22:       Args: -dir 00 -suffix 01 -version 706 -lifeline 2988 -checkpoint 15
06:56:23:WU00:FS01:0x22:             -opencl-platform 0 -opencl-device 0 -cuda-device 0 -gpu-vendor
06:56:23:WU00:FS01:0x22:             nvidia -gpu 0 -gpu-usage 100
06:56:23:WU00:FS01:0x22:************************************ libFAH ************************************
06:56:23:WU00:FS01:0x22:       Date: Jan 20 2022
06:56:23:WU00:FS01:0x22:       Time: 01:14:17
06:56:23:WU00:FS01:0x22:   Revision: 9f4ad694e75c2350d4bb6b8b5b769ba27e483a2f
06:56:23:WU00:FS01:0x22:     Branch: HEAD
06:56:23:WU00:FS01:0x22:   Compiler: Visual C++
06:56:23:WU00:FS01:0x22:    Options: /TP /nologo /EHa /wd4297 /wd4103 /O2 /Zc:throwingNew /MT
06:56:23:WU00:FS01:0x22:   Platform: win32 10
06:56:23:WU00:FS01:0x22:       Bits: 64
06:56:23:WU00:FS01:0x22:       Mode: Release
06:56:23:WU00:FS01:0x22:************************************ CBang *************************************
06:56:23:WU00:FS01:0x22:       Date: Jan 20 2022
06:56:23:WU00:FS01:0x22:       Time: 01:13:20
06:56:23:WU00:FS01:0x22:   Revision: ab023d155b446906d55b0f6c9a1eedeea04f7a1a
06:56:23:WU00:FS01:0x22:     Branch: HEAD
06:56:23:WU00:FS01:0x22:   Compiler: Visual C++
06:56:23:WU00:FS01:0x22:    Options: /TP /nologo /EHa /wd4297 /wd4103 /O2 /Zc:throwingNew /MT
06:56:23:WU00:FS01:0x22:   Platform: win32 10
06:56:23:WU00:FS01:0x22:       Bits: 64
06:56:23:WU00:FS01:0x22:       Mode: Release
06:56:23:WU00:FS01:0x22:************************************ System ************************************
06:56:23:WU00:FS01:0x22:        CPU: Intel(R) Core(TM) i9-10900KF CPU @ 3.70GHz
06:56:23:WU00:FS01:0x22:     CPU ID: GenuineIntel Family 6 Model 165 Stepping 5
06:56:23:WU00:FS01:0x22:       CPUs: 20
06:56:23:WU00:FS01:0x22:     Memory: 31.86GiB
06:56:23:WU00:FS01:0x22:Free Memory: 29.49GiB
06:56:23:WU00:FS01:0x22:    Threads: WINDOWS_THREADS
06:56:23:WU00:FS01:0x22: OS Version: 6.2
06:56:23:WU00:FS01:0x22:Has Battery: false
06:56:23:WU00:FS01:0x22: On Battery: false
06:56:23:WU00:FS01:0x22: UTC Offset: 8
06:56:23:WU00:FS01:0x22:        PID: 13200
06:56:23:WU00:FS01:0x22:        CWD: C:\ProgramData\FAHClient\work
06:56:23:WU00:FS01:0x22:************************************ OpenMM ************************************
06:56:23:WU00:FS01:0x22:    Version: 7.7.0
06:56:23:WU00:FS01:0x22:********************************************************************************
06:56:23:WU00:FS01:0x22:Project: 18717 (Run 45, Clone 38, Gen 163)
06:56:23:WU00:FS01:0x22:Reading tar file core.xml
06:56:23:WU00:FS01:0x22:Reading tar file integrator.xml.bz2
06:56:23:WU00:FS01:0x22:Reading tar file state.xml.bz2
06:56:23:WU00:FS01:0x22:Reading tar file system.xml.bz2
06:56:23:WU00:FS01:0x22:Digital signatures verified
06:56:23:WU00:FS01:0x22:Folding@home GPU Core22 Folding@home Core
06:56:23:WU00:FS01:0x22:Version 0.0.20
06:56:23:WU00:FS01:0x22:  Checkpoint write interval: 250000 steps (5%) [20 total]
06:56:23:WU00:FS01:0x22:  JSON viewer frame write interval: 50000 steps (1%) [100 total]
06:56:23:WU00:FS01:0x22:  XTC frame write interval: 50000 steps (1%) [100 total]
06:56:23:WU00:FS01:0x22:  Global context and integrator variables write interval: disabled
06:56:23:WU00:FS01:0x22:There are 4 platforms available.
06:56:23:WU00:FS01:0x22:Platform 0: Reference
06:56:23:WU00:FS01:0x22:Platform 1: CPU
06:56:23:WU00:FS01:0x22:Platform 2: OpenCL
06:56:23:WU00:FS01:0x22:  opencl-device 0 specified
06:56:23:WU00:FS01:0x22:Platform 3: CUDA
06:56:23:WU00:FS01:0x22:  cuda-device 0 specified
06:56:28:WU00:FS01:0x22:Attempting to create CUDA context:
06:56:28:WU00:FS01:0x22:  Configuring platform CUDA
06:56:29:WU00:FS01:0x22:  Using CUDA and gpu 0
06:56:29:WU00:FS01:0x22:Completed 0 out of 5000000 steps (0%)
06:56:29:WU00:FS01:0x22:Checkpoint completed at step 0
06:56:30:WU01:FS01:Upload complete
06:56:30:WU01:FS01:Server responded WORK_ACK (400)
06:56:30:WU01:FS01:Final credit estimate, 271902.00 points
06:56:30:WU01:FS01:Cleaning up
06:57:41:WU00:FS01:0x22:Completed 50000 out of 5000000 steps (1%)
06:58:54:WU00:FS01:0x22:Completed 100000 out of 5000000 steps (2%)
07:00:07:WU00:FS01:0x22:Completed 150000 out of 5000000 steps (3%)
07:01:19:WU00:FS01:0x22:Completed 200000 out of 5000000 steps (4%)
07:02:32:WU00:FS01:0x22:Completed 250000 out of 5000000 steps (5%)
07:02:32:WU00:FS01:0x22:Checkpoint completed at step 250000
07:03:45:WU00:FS01:0x22:Completed 300000 out of 5000000 steps (6%)
07:04:57:WU00:FS01:0x22:Completed 350000 out of 5000000 steps (7%)
07:06:09:WU00:FS01:0x22:Completed 400000 out of 5000000 steps (8%)
07:07:22:WU00:FS01:0x22:Completed 450000 out of 5000000 steps (9%)
07:08:34:WU00:FS01:0x22:Completed 500000 out of 5000000 steps (10%)
07:08:35:WU00:FS01:0x22:Checkpoint completed at step 500000

[BobWilliams757]

I don't see any glaring issues in the logs, and it looks like the configuration is correct and running on CUDA. I don't know of any recent releases of drivers that have hurt performance to any real extent on NVIDIA cards, so that might not be any real concern.

This would lead me to the many specific questions that might impact how your GPU performs.

Power limits if any? Cooling ability of GPU + case and fan curves.... what are your GPU temps? Do you use the system for other work while folding? What is the specific model of the GPU, clocks, memory, etc? Do you run the GPU bone stock, or do you use software to control power or temp limits, overclock/undervolt, alter memory speeds, etc?


Keep in mind that the info on LARS is subject to flaws and bugs as well, and may not really be representative of the "average" return in the sense most of us might consider it. If the majority using LARS use higher end GPU's and cooling solutions, overclock their gear to within and inch of it's life, and never touch the system while folding in hopes of raising PPD, then it can't be compared to the out of the box GPU in another persons system, who might also use the computer for other things while folding.

Here and there people have mentioned that there are methods to fool the database as well. On top of that, now and then a assignment glitch will skew numbers. I noticed that for that GPU the highest PPD return was over a billion points. Though we know those are fast graphics cards, we also know they don't go THAT fast. If just a few of those crazy high returns show on that project, it would skew the averages quite a bit. Recently there were some work units that went out for another project that were configured wrong and were only about 1/5 the desired length. Due to that I had a return of over 12 million points per day..... on a 1660 Super. So in short, LARS isn't even close to error free. If you compare numbers of others with known equipment it might be a better indication or real world expectations.


I did notice that the project mentioned shows as among the lowest returns for the 4070ti GPU. Does your GPU perform more closely to the averages on most projects, or does it seem to be on the slow side compared to the other GPU's of the same type?

[Mxyzptlk]

I'll get roughly the same PPD and power draw as the OP with my 4070ti on 18717. It's a tough WU for 40XX Cards.