The FAQ reads
"Select the option that best meets the needs of your system. The project would prefer you select times from 8 hours and up"

Is this 8 hours or more? Or 8 hours or less?

Just want to be clear.

---

Is this 8 hours or more? Or 8 hours or less?

Just want to be clear.

That would be anywhere from 8-24hrs. This is also discussed in a QA item on minimizing connect time for dialup users. Note the caution stated there.

---

Add this signature to your EMail:
Running Microsoft's "System Idle Process" will never help cure cancer, AIDS nor Alzheimer's. But running Rosetta@home just might!
https://boinc.bakerlab.org/rosetta/

Thanks for you quick reply. I checked out the cautions in the Q & A.
Have a great Day.

God Bless

---

some WUs seem to be much faster than the configured time though.

resultid=25245120 ran 4 hours even though it would have been allowed 6 hours.

(I've increased to 10 now)

some WUs seem to be much faster than the configured time though.

resultid=25245120 ran 4 hours even though it would have been allowed 6 hours.

(I've increased to 10 now)

In the case of that WU, it only completed two models in 4.19 hours, and so it figured out that a third model would probably take you past the 6hr preference and so it ended rather than starting a third model. You should expect your preference to be met to the nearest complete model. And that's what happened in the case you describe. Many of the CASP proteins are larger and so each model takes longer to process, so this tends to make the rounding to the nearest complete model a bit more noticible. No preference change is required.

Smaller proteins that can process a model every 10 or 15 minutes would never be more than 10 or 15 minutes away from the target preference. But these larger ones, in your case taking more than 2 hours per model, it can miss the target by up to those two hours.

As you increase your WU runtime preference, those 2 hours become a smaller percentage of the total time and so become less noticible and the actual runtimes become more consistent (even though still off by up to 2 hours). Just use caution when changing the preference as described in this QA item, to avoid getting too much work downloaded.

---

Add this signature to your EMail:
Running Microsoft's "System Idle Process" will never help cure cancer, AIDS nor Alzheimer's. But running Rosetta@home just might!
https://boinc.bakerlab.org/rosetta/

I'm still a bit confused. The FAQ says "If you do not change this setting, all of your Rosetta@home Work Units will run in approximately 8 hours." But the page where you modify the settings says "Target CPU run time (not selected defaults to 3 hours)" So, is the default 3 or 8? Also, which would help the science better (or does it make any difference)? Finally, does it impact credits either way?

Thanks,
Erik

---

Reno, NV
Team: SETI.USA

I think running for longer is better for science, although shorter runs may also contain important results, if you see what I mean... Particularly, longer runs require the users to go change the settings, so there's fewer doing those, I would think...

I don't know what the default value is, but if it says 3 hours on the settings site, then it probably is 3 hours, rather than 8 in the documentation. It's not unusual for a value to either be changed or miswritten in the docs (FAQ in this case).

Credit SHOULD not be affected, other than in the sense that if you run for (say) 24 hours, you won't get credit until after 24 hours, while if you run 8 3 hour jobs, you'll get credit once every three hours or so... But the credit is based ona calculation of how fast your machine is, and how long it took to run the job, so a longer job will give you more credit for the job, but not more credit per our of work, if you see what I mean...

--
Mats

---

I'm still a bit confused. The FAQ says "If you do not change this setting, all of your Rosetta@home Work Units will run in approximately 8 hours." But the page where you modify the settings says "Target CPU run time (not selected defaults to 3 hours)" So, is the default 3 or 8? Also, which would help the science better (or does it make any difference)? Finally, does it impact credits either way?

The default value is 3, the "recommended value", for "the typical user" is 8hrs. It really just depends upon your personal preference and how you use your computer. If you have high bandwidth, it's not really going to matter. If you have a dial up modem, then you can reduce your number (and total time) of downloads by going with a longer runtime.

The science is the same either way. In the long run, the credits are the same either way. Any mention of it affecting credits is only for the highly competitive... if you crunch 3 hrs and report in, you get credit for 3 hours... and if you do that every three hours, then after 21 hours, you'll have racked up 21hrs of credit... meanwhile I will still be crunching my 24hr WU and have not yet completed it. So my credit reflects none of today's work... but then 3hrs later I rack up 24hrs of credit and now we're equal for a brief moment each day. The rest of the time, your numbers are a fraction of a day ahead of mine (assuming you took the 3hr default). We're both contributing equally to the science (all other things being equal, CPU speed, etc.)

Be sure to heed the caution noted in the QA item on the subject when making any changes to the runtime value.

---

Add this signature to your EMail:
Running Microsoft's "System Idle Process" will never help cure cancer, AIDS nor Alzheimer's. But running Rosetta@home just might!
https://boinc.bakerlab.org/rosetta/

I'm still a bit confused. The FAQ says "If you do not change this setting, all of your Rosetta@home Work Units will run in approximately 8 hours." But the page where you modify the settings says "Target CPU run time (not selected defaults to 3 hours)" So, is the default 3 or 8? Also, which would help the science better (or does it make any difference)? Finally, does it impact credits either way?

The default value is 3, the "recommended value", for "the typical user" is 8hrs...

Thanks for all the info. Looks like the FAQ needs to be updated.

Erik

---

Reno, NV
Team: SETI.USA

I'm sure they've already seen our posts here pointing out the discrepancy... but you might want to explicitly call the moderators attention to this by posting in the moderators contact thread.

---

Add this signature to your EMail:
Running Microsoft's "System Idle Process" will never help cure cancer, AIDS nor Alzheimer's. But running Rosetta@home just might!
https://boinc.bakerlab.org/rosetta/