18492 WU's will expire on RPi4B

[ETA_2025]

WU 18490 is exceeding the timeout of 2 days, on my Raspberry Pi hardware. It's taking about 2 days and 20 hours for the Pi's to complete these WUs.

WU 18491 is exceeding the timeout of 2 days, on my Raspberry Pi hardware. It's taking about 3 days and 9.5 hours for the Pi's to complete these WUs.

WU 18492 is exceeding the expiry of 3.5 days, on my Raspberry Pi hardware. It's taking about 4 days and 12.75 hours for the Pi's to complete these WUs.

WU 18493 is exceeding the expiry of 3.5 days, on my Raspberry Pi hardware. It's taking about 4 days and 3 hours for the Pi's to complete these WUs.

I'll be dumping all 18492-18493 WUs.

Should I also dump 18490-18491 WUs too?

[Joe_H]

I have brought this to the researcher's attention, the WS should stop assigning these to ARM systems.

[ETA_2025]

Project: 18492 (Run 36, Clone 61, Gen 111) and Project: 18492 (Run 30, Clone 54, Gen 131) will both take more than 4.5 days to be completed on my over-clocked RPI4B's. They both have 2 day timeouts and 4 day expiries.

I will be dumping them, as it's just wasting power.

[bikeaddict]

Others have stated that RPI4 was too slow to complete Folding tasks long ago. It's time to retire or repurpose this hardware.

viewtopic.php?p=365160&hilit=Raspberry+Pi#p365160
https://www.reddit.com/r/Folding/commen ... is_remote/

[ETA_2025]

18494 WU's have the same issue (timing same as for 18490 WU's). I'll be dumping these.

[ETA_2025]

I got a couple of 18490 WU's that have the same timing as above. They'll be dumped.

[Joe_H]

I have notified the researcher about this. There has been a recurring problem with assignment settings on the servers reverting to some default settings, they have not tracked down the cause.

[ETA_2025]

I have notified the researcher about this...

Thanks Joe_H

[ETA_2025]

Others have stated that RPI4 was too slow to complete Folding tasks long ago. It's time to retire or repurpose this hardware.

That's not true. I have completed more than 4,000 WU's on RPi4B's. The current problem is that WU's that shouldn't be assigned to RPi4B's, are being assigned to RPi4B's. That's not a problem with RPi4B's, it's a server problem.

See this post: viewtopic.php?p=365443#p365443

[bikeaddict]

Some of your WUs are being returned late and end up being sent out to be run by another user.

Three examples just from recent days:

https://apps.foldingathome.org/wu?p=184 ... e=49&gen=1
https://apps.foldingathome.org/wu?p=184 ... e=88&gen=0
https://apps.foldingathome.org/wu?p=184 ... e=53&gen=1

[Joe_H]

Some of your WUs are being returned late and end up being sent out to be run by another user.

Three examples just from recent days:

https://apps.foldingathome.org/wu?p=184 ... e=49&gen=1
https://apps.foldingathome.org/wu?p=184 ... e=88&gen=0
https://apps.foldingathome.org/wu?p=184 ... e=53&gen=1

Your examples are all WUs from a project, named in this topic - viewtopic.php?t=39581, that is not supposed to be assigned to Raspberry Pi. You are not demonstrating what you think you are. This is a recurring problem. The two current topics are also an example of people not checking to see if there already is an existing topic before posting.

[bikeaddict]

It's still undeniable that people who run potato farms with slow hardware are actually slowing down the scientists who need the results to come back quickly before new WUs can be generated. This is why the quick return bonus exists. People are taking 2-4 days to process when someone else with faster hardware could have finished it in well under a day. The cumulative delay of a project over a period of weeks or months must be substantial and frustrating for researchers. It's even worse when WUs expire and the work was done for nothing while the results are further delayed waiting for another client to do the work again.

It's also undeniable that the Folding project has a problem with properly detecting clients who are frequently missing deadlines. This problem continues to occur because the client doesn't give users adequate feedback when a WU expires and they are doing more harm than good. This problem continues because the server doesn't detect and ban clients that are repeatedly too slow. And this problem continues because a lot of people are ignorant and don't understand how inefficient their hardware is and how little actual work they are doing for the cost of their electricity.

I will continue my crusade to educate people about the folly of using a potato farm for distributed computing. Please invest in a small number of modern 8-16 multi-core computer, people, please. I've learned over the last 4.5 years of folding that it's way more efficient and productive.

[Joe_H]

Then do it more civilly. For now the F@h Project supports use of the Raspberry Pi 4, it just is having some problems with larger projects losing settings that would prevent assignment to these inappropriately. There usually, not always are smaller projects available.

P.S. Next time before pulling up failures in an attempt to do whatever you were trying there, try checking their completion rate. The app is also there to do that. Currently it is in excess of 96%.

[calxalot]

I think most people want to fold on what they have.
Buying hardware just for folding is a niche.

The client and servers could use the ability to better estimate the speed of a client machine.
Manually setting constraints on projects does not seem sufficient to me.

[ETA_2025]

Some of your WUs are being returned late and end up being sent out to be run by another user.

Three examples just from recent days:

https://apps.foldingathome.org/wu?p=184 ... e=49&gen=1
https://apps.foldingathome.org/wu?p=184 ... e=88&gen=0
https://apps.foldingathome.org/wu?p=184 ... e=53&gen=1

Those WU's were almost complete (>80%) when I discovered they wouldn't be finished before the Timeout. Dumping them then, would be a waste, as my understanding is that every work unit is completed at least twice, by different users.

However, if that is not the case, I'll just dump any WU as soon as I discover it won't be completed before the Timeout.