• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

FAH new WU? and it's a monster

jfb9301

[H]ard DCOTM x4
2FA
Joined
Jan 11, 2005
Messages
1,751
I got a new WU last evening, and it was a proper monster unit. Like Big-ADV monster. If you have a higher end Nvidia GPU, it could really boost your PPD.

Has anyone else received one of these, or one like it?

My GTX5080 normally gets 22-24,000 PPD, this was 58KPPD.

Screenshot From 2026-09-23 14-27-54.png
 
26M Points, for that beast. and took my card 12 hours.

Working on Alzheimer's. Not cancer, but I totally support the cause.

Screenshot From 2026-09-23 17-07-26.png
 
Well that makes sense. Omega2 is the only machine I have that has 64GB of RAM, so it's the only one to catch one of those WUs. It would be nice to catch some more, the PPD boost is huge, and I don't mind the mentioned penalties that come with running the unit.
 
Looks like im going to be picking up another 32GB kit tomorrow...
 
These units still seem a bit rare, I have only caught one so far.

I did shutdown the CPU core to ensure all RAM is available.

On reflection, I would not recommend running out to buy RAM (at today's prices 🤮) just to catch a rare WU. I would wait for these to become more frequent. I do suppose as folding becomes more complex, they will become more common.
 
Last edited:
Lucky you....

I've only caught the one unit 5 days ago. What is your setup? Maybe there is something I can change to get more.
 
High end, but relatively standard otherwise. I did check the beta WU box in my client settings on the box. Four in a row now.

5090 FE
9800X3D
64GB
 
AMD Ryzen 7 9950X3D, 5080, and 64GB.

I didn't have the beta flag, so I added that to the GPU settings. We'll see what happens
 
I'm not sure if the flag made a difference, but I only got the WU after checking it.
 
Beta flag did the trick, I just got one.

Weird, I didn't need the flag a few days ago, but now I do. Oh well, I have one now, and we know the "real requirements" to get these units.
 
now if only I can find a solution to this nonsense:
1790615799220.png


It is super annoying to wait over an hour to get a WU, when the machine can finish a WU in under 3 hours..... it's a lot of downtime.
 
I'll see how it goes for the next week then might upgrade my other rig lol
 
It seems there are several iterations of the WU. TPF varies 5:03-7:20. I will update to the latest nVidia driver to see if that makes a difference.


1790732376926.png
 
I guess your assessment is correct, a 32GB kit is still cheaper than a RTX50 series card, and now that we know how to get the units regularly, it's mildly tempting to upgrade my RTX507ti rig.
 
It seems there are several iterations of the WU. TPF varies 5:03-7:20. I will update to the latest nVidia driver to see if that makes a difference.


View attachment 828712
Yes. Looking at my logs, though much less dramatic, I have 1 that is 2 sec/frame faster, but is 1.5MPPPD LESS. But in general they are all around 7:10 to 7:20 TPF, seems like the baseline for an RTX5080 is right around 60MPPD (about double normal).
 
I wonder if there is an artificial limit on the 5090. I'll check the GPU utilization.

Edit: updated to the latest driver (617.14) and times are back down to ~5:0x

1790797899630.png


Watching it further.
 
Last edited:
Driver update holding steady. There is one that came in very high.

1790967730373.png
 
Last edited:
  • Like
Reactions: Nobu
like this
I bit the bullet and installed that new driver.

In the day and age of 32 cores, 64GB Ram, and solid state drives, IT SHOULD NOT TAKE 5 MINUTES to reboot a machine. We've gone back to the '90's. I also installed WinBlows updates and finally got around to fixing the conflicts between 3 pieces of software. 3 reboots and 40 minutes later, we shall see what results I get. So far it would seem there is a large (25 sec) variance between frame times.
 
On the 5090, I will see variations of 10-12 seconds per frame at times. So that doesn't seem to be too much of an oddity.
 
having finished 23 frames on the 617.14 driver , I then plugged my log into Excel.....

7:03 minutes a frame, which looks like a slight improvement, but the units vary from 6:58 to 7:19 minutes.

I will take the improvement, but the variance is troublesome. I am used to units finishing frames like clockwork.
 
And you replied while I was typing.... LOL. OK, so maybe this is just how it is going to be with these units.

I will see how the next unit goes.
 
Modern GPUs are very sensitive to heat load (some maybe more than others), so if it's operations which generate varying amounts of heat, the time to complete them will also vary, because the frequency will also vary with temperature.
 
Modern GPUs are very sensitive to heat load (some maybe more than others), so if it's operations which generate varying amounts of heat, the time to complete them will also vary, because the frequency will also vary with temperature.

I always run my GPUs at 70-75% PL for that reason. That and to prolong the life of the card.

Edit: nVidia efficiency curves suck past ~75%, so there's really no point.
 
  • Like
Reactions: Nobu
like this
Back
Top