Strategy? Why, the only PIMP strategy is to straight up PIMP SLAP a fool who needs it.  Fill your fine and fly Folding Machine with the Pimpest of Pimp-Juice and stand back as that engine goes from a fine purrrrrrr to the loudest roar on the block!

In short, my fellow Pimps and Pimpettes, UnLEASH that folding fury and let those WUs FALL before us!
 
d

My strategy for future contests: don’t trust myself to estimate my own PPD when F@H is throwing a few errors lol.

My 87 year old mother gave me an odd look when I told her she was a PIMP.  LOL she said she'd pimp for a good cause.

I am getting Stalled work units on both of my 3090's.  No overclocking.  This is with Beta units and Advanced units.  If I switch to normal units, I don't get units assigned sometimes.
 
is anyone else seeing these issues?
 
Copy and paste of the log below.  Again, both 3090's, no overclocks, with beta and advanced units. 
 
******************************* Date: 2021-02-19 *******************************
11:32:31:WARNING:WU00:FS02:FahCore returned an unknown error code which probably indicates that it crashed
11:32:31:WARNING:WU00:FS02:FahCore returned: WU_STALLED (127 = 0x7f)
******************************* Date: 2021-02-19 *******************************
14:17:39:WARNING:WU02:FS01:FahCore returned an unknown error code which probably indicates that it crashed
14:17:39:WARNING:WU02:FS01:FahCore returned: WU_STALLED (127 = 0x7f)

If it were me I would get rid of the beta flag
I have the advanced flag and having no issues

Or turn them off for twenty four hours. And try again

But seriously. Drop the beta flag and see

Switched back to advanced, got a stalled unit right away, so that wasn’t a fix for sure.

I have removed all flags and will see what happens next.

Xavier Zepherious
1300 watt PSU -1kw for GPU - yes they can draw that high mining i saw over 900W
 
that leaves 300 watts - - 150 for CPU(and im being modest her an overclocked CPU can draw 400W) - 30 watts for ram... hd will draw 12
you have fans, coolers, lights wifi/nic -usb,keyboard mouse
 
you lose 5-10% every year on the power avail because the parts wear out and fail - less efficient
 
 
yes if you want to run two 3090's i would suggest a 1500W-1600W PSU - 1300 is cutting it close
yes that can be TDR
 
christ i have a Watercooled rig with two 780 WC classified that are maxing out a 1300watt PSU
i had to remove hardware to get it to boot- no the hardware is fine i checked it
 
any crash to Black screen/shutdown is usually power

 
Maybe I mistyped above.  I will try again.
 
GPU's are on a dedicated 1300G2 power supply separate from the rest of the system.   Even at 500w each, they would still be in a good power range.
 
 
Dedicated, as in completely separate from the rest of the system.  That is 1300w to the GPU's.  Not the motherboard, not the CPU, not the fans, not the SSD's, not the hard drives, not the pumps... Just the GPU's.  Dedicated to them. 
 
The rest of the system runs on a completely separate PSU.
 
I don't have any black screens or shutdowns.  I simply have stalled work units, which comes down to the work unit.  I can remove the stalled work units by removing the beta AND the advanced flags, but the whole reason I was running beta and advanced is because I wasn't getting work units from the servers. Adding the Advanced and Beta Flags got me work units, but occasionally they produced the error above. The only fix is, you may have guessed, remove the flags. The 3080 system isn't having this issue as has had the advanced flag for about 2 weeks, so I am curious if the issue is with the 3090's, because Advanced and Beta units are not running the system at 100%, it stays around 85% GPU usage normally.
 
Now, anyone with a 3090, are you seeing this with advanced or beta flags?  I would like to hear from those running the same system, because I would like to be able to report them over at the folding forums, and we already know how that is going to go.  I feel like it will be exactly like above, where the repeating gets old, especially repeating what was already repeated. Bruce is going to do exactly what is happening here, and it is going to push me away from folding and I don't want that.
 
I feel like it is important to say this again, There is no overclock on the system or the GPU's, the GPU's are on a dedicated power supply, I am NOT having crashes, and advice about crashes, temps, and overclocks probably will not help.  I just need to know if anyone is seeing the same thing with their 3090's so that I can report it to the folding forums if it is an issue others are experiencing.
 
 
 

I just came to the basement to find a stalled work unit. The culprit is 14561 (0, 1033, 68) running on Linux Mint 20.1 on the 3060ti. The screen was black and I was unable to "wake it up". Reboot and according to the log about 4 hours down time. It stalled near 80% and picked back up from there.
It's finishing now. This my second one in this contest.
I'm returning to Win 10 after it finishes.

yodap
I just came to the basement to find a stalled work unit. The culprit is 14561 (0, 1033, 68) running on Linux Mint 20.1 on the 3060ti. The screen was black and I was unable to "wake it up". Reboot and according to the log about 4 hours down time. It stalled near 80% and picked back up from there.
It's finishing now. This my second one in this contest.
I'm returning to Win 10 after it finishes.

What does it say in the full note list for you?
 
I typically check the "Warnings & Errors" box, so it shows that a unit stalled but not why.
 
If I open the full log, and look at the work unit just before stalling, it is showing a watchdog shutdown failed at 59%:
 
16:25:30:WU01:FS01:0x22:Project: 13440 (Run 3842, Clone 29, Gen 1)
 
20:28:13:WU03:FS02:0x22:Project: 17800 (Run 20, Clone 12, Gen 108)
21:01:51:WU03:FS02:0x22:Watchdog triggered, requesting soft shutdown down
21:11:52:WU03:FS02:0x22:Watchdog shutdown failed, hard shutdown triggered
21:11:52:WARNING:WU03:FS02:FahCore returned an unknown error code which probably indicates that it crashed
21:11:52:WARNING:WU03:FS02:FahCore returned: WU_STALLED (127 = 0x7f)


 
Watchdog errors seem to be a windows/CPU error, so I am going to reset my BIOS and see if it helps at all.   I don't know what else it could be. 

Because of the reboot (I guess) I can only show the end of the first part and the second part. Also a little over 2 hours missing not 4. That red line was actually red in the log. IMHO not much to see here.
 
19:02:11:WU01:FS01:0x22:Completed 390000 out of 500000 steps (78%)
19:03:02:WU01:FS01:0x22:Completed 395000 out of 500000 steps (79%)
19:03:53:WU01:FS01:0x22:Completed 400000 out of 500000 steps (80%)
19:03:54:WU01:FS01:0x22:Checkpoint completed at step 400000
19:04:45:WU01:FS01:0x22:Completed 405000 out of 500000 steps (81%)
19:05:35:WU01:FS01:0x22:Completed 410000 out of 500000 steps (82%)
19:06:26:WU01:FS01:0x22:Completed 415000 out of 500000 steps (83%)
\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\
 
 
And
 
21:17:21:FS01:Finishing
21:17:27:WU01:FS01:0x22:Attempting to create CUDA context:
21:17:27:WU01:FS01:0x22:  Configuring platform CUDA
21:17:37:WU01:FS01:0x22:  Using CUDA and gpu 0
21:17:38:WU01:FS01:0x22:Completed 400000 out of 500000 steps (80%)
21:18:29:WU01:FS01:0x22:Completed 405000 out of 500000 steps (81%)
21:19:21:WU01:FS01:0x22:Completed 410000 out of 500000 steps (82%)
21:20:13:WU01:FS01:0x22:Completed 415000 out of 500000 steps (83%)
21:21:05:WU01:FS01:0x22:Completed 420000 out of 500000 steps (84%)
21:21:56:WU01:FS01:0x22:Completed 425000 out of 500000 steps (85%)
21:21:57:WU01:FS01:0x22:Checkpoint completed at step 425000
21:22:48:WU01:FS01:0x22:Completed 430000 out of 500000 steps (86%)
21:23:39:WU01:FS01:0x22:Completed 435000 out of 500000 steps (87%)
21:24:30:WU01:FS01:0x22:Completed 440000 out of 500000 steps (88%)
21:25:21:WU01:FS01:0x22:Completed 445000 out of 500000 steps (89%)
21:26:12:WU01:FS01:0x22:Completed 450000 out of 500000 steps (90%)
21:26:13:WU01:FS01:0x22:Checkpoint completed at step 450000
21:27:04:WU01:FS01:0x22:Completed 455000 out of 500000 steps (91%)
21:27:55:WU01:FS01:0x22:Completed 460000 out of 500000 steps (92%)
21:28:46:WU01:FS01:0x22:Completed 465000 out of 500000 steps (93%)
21:29:37:WU01:FS01:0x22:Completed 470000 out of 500000 steps (94%)
21:30:28:WU01:FS01:0x22:Completed 475000 out of 500000 steps (95%)
21:30:30:WU01:FS01:0x22:Checkpoint completed at step 475000
21:31:21:WU01:FS01:0x22:Completed 480000 out of 500000 steps (96%)
21:32:12:WU01:FS01:0x22:Completed 485000 out of 500000 steps (97%)
21:33:03:WU01:FS01:0x22:Completed 490000 out of 500000 steps (98%)
21:33:54:WU01:FS01:0x22:Completed 495000 out of 500000 steps (99%)
21:34:45:WU01:FS01:0x22:Completed 500000 out of 500000 steps (100%)
21:34:45:WU01:FS01:0x22:Average performance: 13.5529 ns/day
21:34:46:WU01:FS01:0x22:Checkpoint completed at step 500000
21:34:53:WU01:FS01:0x22:Saving result file ../logfile_01.txt
21:34:53:WU01:FS01:0x22:Saving result file checkpointIntegrator.xml
21:34:53:WU01:FS01:0x22:Saving result file checkpointState.xml
21:34:58:WU01:FS01:0x22:Saving result file positions.xtc
21:34:58:WU01:FS01:0x22:Saving result file science.log
21:34:58:WU01:FS01:0x22:Folding@home Core Shutdown: FINISHED_UNIT
21:34:58:WU01:FS01:FahCore returned: FINISHED_UNIT (100 = 0x64)
21:34:58:WU01:FS01:Sending unit results: id:01 state:SEND error:NO_ERROR project:14561 run:0 clone:1033 gen:68 core:0x22 unit:0x0000005480fccb025f3305ddd21cf629
21:34:58:WU01:FS01:Uploading 42.81MiB to 128.252.203.2
21:34:58:WU01:FS01:Connecting to 128.252.203.2:8080
21:35:04:WU01:FS01:Upload 18.10%
21:35:10:WU01:FS01:Upload 37.67%
21:35:16:WU01:FS01:Upload 57.38%
21:35:22:WU01:FS01:Upload 76.94%
21:35:28:WU01:FS01:Upload 96.22%
21:35:31:WU01:FS01:Upload complete
21:35:31:WU01:FS01:Server responded WORK_ACK (400)
21:35:31:WU01:FS01:Final credit estimate, 167750.00 points
21:35:31:WU01:FS01:Cleaning up

I’ll try but the rig was black screened and frozen so reboot was only option I could find. No keyboard combos worked and Linux. Learning curve? You bet!
 

Finally stopped getting stalled work units, and started getting "no work unit assigned." 
 
I am out of town for, so it is up to the folks at Stanford from here on out.  

Xavier Zepherious
Better pull your bling out - your falling behind 
as the prophets would say "god is on their side?

I’m sad. I actually tried to sneak to a microcenter and try picking up some reinforcements... there were no reinforcements in stock.