(Sorry if this has been covered before. A search for 'folding' on the forums returned no hits at all).

Can WUProp detect F@H WU's and track the time?
If not, how much effort would that take?

Folding@Home work has grown significantly since the pandemic research and some of my team is pushing to migrate their computers there. I dedicated 2 cores last spring to add to the WU collection (we all understand that urge here, don't we?).

FAH is not a BOINC Project. So no, it can't be added here as WUProp collects info for BOINC projects.

FAH is not a BOINC Project. So no, it can't be added here as WUProp collects info for BOINC projects.

This was already apparent to anyone that installed F@H and so why I asked "If not, how much effort would that take?"

Will we track BOINC competitor WU's as they arise (and they are).

FAH is not a BOINC Project. So no, it can't be added here as WUProp collects info for BOINC projects.

This was already apparent to anyone that installed F@H and so why I asked "If not, how much effort would that take?"

Will we track BOINC competitor WU's as they arise (and they are).

Personally I think that's a very slippery slope, if they do that do they then track the next thing and the one after that etc etc? Tracking involves getting the data from the Projects, why would the non Boinc Projects even want to share their data with me for example when I don't crunch for them. In this case "with me" means me being able to see your data, and if they mandate it be private what's the point as you already have that information. I also think if it's private but included that could cause someone to add up the numbers and say ie 'WTF is going on with marmot's numbers because they are waaaay out of whack'.

I'm learning about how the exports worked (will work again?) and have not formulated a complete opinion about this yet, but since our team's vote on accepting work from F@H is trending 70%+ YES vote; I better look into it.
It seems that F@H exported to Free-DC, where you could see a user contribution points, and it was tied to the BOINC CPID using F@H passkeys to an identical user name on Free-DC. They claim their servers became overwhelmed from the sheer number of new users during the COVID19 research phase so they only export team numbers now.

if they do that do they then track the next thing and the one after that etc etc? Tracking involves getting the data from the Projects,

WUProps already tracks "the next thing and the one after that"; they just all happen to be BOINC "next things". (I wish WUProps had been tracking my hours when I was crunching SETI@Home in 1999->)

Viewing other scientific computing systems' WUs in RAM would be simple. Any task manager can see the WU's; it seems to be just a matter of WUProp DB additions but maybe some extra code would be needed. F@H might willingly share but, if not, the F@H calling process and the child WU name is visible right there in your client's RAM.

I see WUProp as having a philosophy of tracking user hours of contribution to scientific projects (per WU) and so I don't understand an objection to tracking non-BOINC. I'm loyal to research; it doesn't have to be BOINC (whose devs, still, stubbornly refuse to offer easy, specific core assignment mechanisms.)

Tracking involves getting the data from the Projects...

That's not correct.

WUProp does not get statistics from the projects. It gets its data by directly monitoring what's happening on your computer. The data the projects export concern credits. WUProp isn't interested in credits, but it does track information such as hours, memory usage, type of computer and operating system. None of that is exported by the projects.

However, I do agree completely with this not being something WUProp should do, for many reasons. But it's not up to us.

---

Want to find one of the largest known primes? Try PrimeGrid. Or help cure disease at WCG.

Tracking involves getting the data from the Projects...

That's not correct.

WUProp does not get statistics from the projects. It gets its data by directly monitoring what's happening on your computer. The data the projects export concern credits. WUProp isn't interested in credits, but it does track information such as hours, memory usage, type of computer and operating system. None of that is exported by the projects.

You are correct I wrongly conflated the too!!

Stanford provides flat stat files (text files) that other sites can pull in if they want. This is what EOC pulls in and is updated every 3 hours. Free-DC could do the same thing.

WUProp probably uses boinccmd functions to gather info on the tasks that are running in that BOINC session. Host info (OS, CPU type, # of threads, etc) BOINC task memory usage, BOINC task disk usage, etc. Its the info available at the Results link at the bottom of the WUProp site. WUProp can only see BOINC tasks running with the same rpc port. If you run multiple clients (I do to keep CPU and GPU queues separate), then WUProp needs to also look at the separate rpc port that the boinc client is using.

Since FAH is not run through the BOINC client, using boinc cmd functions will never see FAH as a running project. FAH does not have the same type of functions available as boinc cmd. If WUProp is opened up to query other things about your PC, 1- I would guess that goes against the spirit or some rule of BOINC, 2- then some project could basically become Google/Facebook/Amazon and gather everything you do via BOINC. Aka spyware.

I don't exactly see a wealth of WUProp admins to even make said changes.

another way is possible, if you find an enthusiast to run a hybrid BOINC project, something like
- Moo! wrapper (to run jobs one by one from under BOINC)
- DHEP (stub job, interface for someone else's process)
- LHC VM (launching someone else's agent via VBox/docker wrapper)