LTT Official Folding Month VI

[GOTSpectrum]

1 minute ago, cbigfoot said:

Points look good @Baha you have been warned Quit Speeding the speed limit is  23,663,074 PPD 

Hmm... why does that number look familiar 

[GOTSpectrum]

6 minutes ago, cbigfoot said:

OK, Just trying to make @GOTSpectrum Happy. but got to try.

[leadeater]

12m PPD CPU only, niceee

[cbigfoot]

2 minutes ago, leadeater said:

12m PPD CPU only, niceee

I would hate to see the power bill for that...

[GOTSpectrum]

 

[Jacob_1999]

42 minutes ago, GOTSpectrum said:

OMG A Warhammer reference lol my people 

[GOTSpectrum]

Just now, Jacob_1999 said:

OMG A Warhammer reference lol my people 

There  have been at least two others in the chat so far 

[CWP]

49 minutes ago, cbigfoot said:

Points look good @Baha you have been warned Quit Speeding the speed limit is  23,663,074 PPD 

 

 

OK, Just trying to make @GOTSpectrum Happy. but got to try.

Points Police Department. Pull over!

[GOTSpectrum]

2 minutes ago, CWP said:

Points Police Department. Pull over!

Don't worry my zipties come tomorrow, which means a higher speed limit!  

[dogwitch]

3 minutes ago, GOTSpectrum said:

There  have been at least two others in the chat so far 

Avenge Me Brother!!!!!!!!!!!!!!!

[Baha]

1 hour ago, cbigfoot said:

Points look good @Baha you have been warned Quit Speeding the speed limit is  23,663,074 PPD 

 

 

OK, Just trying to make @GOTSpectrum Happy. but got to try.

What was that you say? More gas???? Okay you got it friend!!!

[GOTSpectrum]

38 minutes ago, dogwitch said:

Avenge Me Brother!!!!!!!!!!!!!!!

I will destroy their minds and burn their bodies.

[Gorgon]

9 hours ago, Alex Atkin UK said:

I have it set in /etc/sddm/Xsetup mainly because I couldn't figure out how else to run nvidia-settings with the right display access to set the fan controls.  Drawback there is it requires you to use Xorg not Wayland, well its a drawback if Fedora kill off Xorg.

 

nvidia-settings -a "[gpu:0]/GPUFanControlState=1" -a "[fan:0]/GPUTargetFanSpeed=65"
nvidia-smi -pm 1
nvidia-smi -gtt 70 # Aim for this target operating temperature
nvidia-smi -pl 200 # Power Limit

 

Target temperature doesn't seem to really work though which is why I have the power limit there too, as this is the 2080 config where the thermal pads are not in great condition so hot spot runs about 20-25C higher than core temp.  As you can't monitor hot spot in Linux, I just have to try to keep the core temp down and hope hot spot tracks linearly.

 

Both 4090s are regrettably on Windows as I'm really curious if they score higher on Linux, but I'm not completely insane, both those boxes have actual uses that require Windows.  I'm not about to spend 4090 money and ONLY use it for Folding.

 

The 3080, INNO3D 4070 Ti and 2080 are dedicated Foldboxes, the rest are in PCs where I actually need them for stuff.

Assuming you already have CoolBits enabled ...

 

Find the XAuthority file gdm is using:

sudo ps -ef | grep auth

gdm         1129    1127  0 12:14 tty1     00:00:01 /usr/lib/xorg/Xorg vt1 -displayfd 3 -auth /run/user/128/gdm/Xauthority -nolisten tcp -background none -noreset -keeptty -novtswitch -verbose 3

and use it in the command:

DISPLAY=:0 XAUTHORITY=/run/user/128/gdm/Xauthority nvidia-settings \
 -a [gpu:0]/GPUPowerMizerMode=1 \
 -a [gpu:0]/GPUGraphicsClockOffsetAllPerformanceLevels=50
 -a [gpu:0]/GPUFanControlState=1 \
 -a [fan:0]/GPUTargetFanSpeed=75 \
 -a [fan:1]/GPUTargetFanSpeed=75 \

 

[TheLANguy]

Day 4-5 summary:

We are burning down the house now!
My point production has begun to level out and become steady, except for yesterday when I bumped the power strip and everything turned off.

It seems the team is beginning to do the same - minus the turning off part. Speaking of my point production... I wonder what hardware my "rivals" @dahoff and @jnkl are running.

 

Our teams "rival" Curecoin currently has 115,484,209,955 more points than us.

On average Curecoin is producing 758,553,786 points a day. By the end of the Folding month they will have a total of about 2,286,497,241,992 points.

The LTT EOC stats from 11/4 show we are making 3,189,984,266 points a day. If we keep the same PPD we should have a total of 2,246,387,376,892 points.

That will leave only 40,109,865,100 points between us and second place!

 

That is really impressive!

Stay warm.

[AndrewHamOp]

7 hours ago, Heats with Nvidia said:

Thank you. I know this method and used it quite a lot, but somehow never thought of trying that on a GPU.

I tried it and found out, one of the VRM mosfets shorted out and the 12V now go directly into the GPU itself.

So this one is a no fix. But at least, now i know for sure.

 

Yikes! Sorry to hear that your GPU toasted itself, but at least you know for sure. Glad I could help even if the GPU is very dead. I really enjoy Necroware's videos myself, though the hardware he usually covers is a bit older than that GPU lol...

[Alex Atkin UK]

4 hours ago, GOTSpectrum said:

 

Darn you Gorgon and Firedfly, looks like I wont be able to hold my position much longer.

[Baha]

Decided to tinker around with getting the CPU of my linux "server" involved with folding -- I wasn't expecting a whole lot, because I have consumer grade hardware, not actual perf-per-watt efficient prosumer or enterprise stuff for multi-threaded, but at a roughly 200W limit my 13900 is kicking out a nice 700,000-1M PPD depending on WU. Still a fair bit less efficient than even my GTX 1000 series cards, but worth leaving on for the event. I'm pleasantly surprised!

[jnkl]

3 hours ago, TheLANguy said:

Day 4-5 summary:

We are burning down the house now!
My point production has begun to level out and become steady, except for yesterday when I bumped the power strip and everything turned off.

It seems the team is beginning to do the same - minus the turning off part. Speaking of my point production... I wonder what hardware my "rivals" @dahoff and @jnkl are running.

 

Our teams "rival" Curecoin currently has 115,484,209,955 more points than us.

On average Curecoin is producing 758,553,786 points a day. By the end of the Folding month they will have a total of about 2,286,497,241,992 points.

The LTT EOC stats from 11/4 show we are making 3,189,984,266 points a day. If we keep the same PPD we should have a total of 2,246,387,376,892 points.

That will leave only 40,109,865,100 points between us and second place!

 

That is really impressive!

Stay warm.

I am unsure whether my setup falls under the category of hardware or software.
It's a docker container running in Unraid, which is virtualized by Proxmox. My Unraid is given just 2 cores, no GPU, and is my backup server.
The Promox host is an Intel 11500T

[Alex Atkin UK]

1 minute ago, Baha said:

Decided to tinker around with getting the CPU of my linux "server" involved with folding -- I wasn't expecting a whole lot, because I have consumer grade hardware, not actual perf-per-watt efficient prosumer or enterprise stuff for multi-threaded, but at a roughly 200W limit my 13900 is kicking out a nice 700,000-1M PPD depending on WU. Still a fair bit less efficient than even my GTX 1000 series cards, but worth leaving on for the event. I'm pleasantly surprised!

Careful it doesn't bring the GPU performance down.

[Baha]

14 minutes ago, Alex Atkin UK said:

Careful it doesn't bring the GPU performance down.

Definitely! I'm being pretty conservative with it -- the CPU has 24 cores and 32 threads, I set it to fold on 20 cores (originally 22) to allow for the two GPUs to each have a dedicated physical core of overhead, and two extra on top for any duties it ends up needing to serve in the interim. Outside of folding right now it largely gets used for Plex (which uses primarily the iGPU), development work that won't require much compute any time soon, and a bunch of services in containers/VMs that are spiky which I've bound to the two free cores for duration of the event.

[Baha]

... Also, when I was doing some work on my server box earlier to get things ready to try out CPU folding, I took the GPUs out to shuffle some things around. When I put them back in, I changed the slots they were in, and this magical thing happened with the IDs.

[Lightwreather]

Somehow I managed to brick a cloud instance?

It flipping refused to even connect to ssh from within google's network…

Welp, time to start over

[dogwitch]

found a wu that loving my gpus and cpu atm.

good points and times

[Mxyzptlk]

3 hours ago, Baha said:

Definitely! I'm being pretty conservative with it -- the CPU has 24 cores and 32 threads, I set it to fold on 20 cores (originally 22) to allow for the two GPUs to each have a dedicated physical core of overhead, and two extra on top for any duties it ends up needing to serve in the interim. Outside of folding right now it largely gets used for Plex (which uses primarily the iGPU), development work that won't require much compute any time soon, and a bunch of services in containers/VMs that are spiky which I've bound to the two free cores for duration of the event.

I have yet to figure out how to get any of my CPU’s to fold without dragging down the GPU’s PPD significantly… 

[leadeater]

@Windows7ge Your client configs still aren't right, still at ~600k ppd

 

 

Edit:

Maybe could be ok, will need to see next 2x 3 hourly updates