• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Folding@Home: The Now Outdated FAQ

Hi jliechty!
I am very new and I don't know much or what I am doing, so these are some ideas that might or might not be helpful.

I would disagree with this:
Quote
A dialup connection that must be initiated manually will be much harder to deal with; you might be able to work around this by using a monitoring program (discussed below) and carefully watching the work unit completion times to ensure that you can provide the dialup connection just before F@H gets done with its work unit, but this would be a very big pain in a certain area of the body, and I don't recommend it if you want to retain your sanity and not get burnt out on F@H in a very short time.
Unquote

If you have set the client to "Ask before using Network", it IS NOT "much harder to deal with" and you DO NOT have to "carefully watch" anything! When a work unit is finished, a small dialog opens that says ready to connect. When a person dials up to check email, they just click OK on that dialog and it sends results and gets another WU. There might be a few hours that the computer is not folding, but is that a reason to discourage a potential folder from joining the TeAm?

Also: I noticed on another thread that some of the other DC clients use lots of memory (20MB on one and 50MB for another). With the console and core combined F@H only uses 2.5MB on my system it looks like. If people knew this it might influence them to do F@H instead of something else. Or someone running a different client that has hyperthreading computer could run their client and F@H because it uses less memory. Maybe? I don't know.

Anyway, thanks for the FAQ, I added it to my sig.
 
Originally posted by: mikecel79
Originally posted by: lnguyen
i like this service install:
http://www.bluetentacle.co.uk/dc/fahservice.htm
Click to expand...
Linky no worky for me....
Click to expand...
It works ok for me right now. 😕

Try the Google cache of the page, if there's a disturbance of the for... er... internet between your provider and theirs. 😱

Gleem, please don't feel neglected; I'll address your concerns later, as I don't have much time right now...

Edit for speling. 😉
 
Originally posted by: GLeeM
Hi jliechty!
I am very new and I don't know much or what I am doing, so these are some ideas that might or might not be helpful.

I would disagree with this:
Quote
A dialup connection that must be initiated manually will be much harder to deal with; you might be able to work around this by using a monitoring program (discussed below) and carefully watching the work unit completion times to ensure that you can provide the dialup connection just before F@H gets done with its work unit, but this would be a very big pain in a certain area of the body, and I don't recommend it if you want to retain your sanity and not get burnt out on F@H in a very short time.
Unquote

If you have set the client to "Ask before using Network", it IS NOT "much harder to deal with" and you DO NOT have to "carefully watch" anything! When a work unit is finished, a small dialog opens that says ready to connect. When a person dials up to check email, they just click OK on that dialog and it sends results and gets another WU. There might be a few hours that the computer is not folding, but is that a reason to discourage a potential folder from joining the TeAm?
Click to expand...
Well, true, but admittedly it is quite a bit more of a hassle than using any of the clients that can cache work (SETI with SETIQ, RC5 [bleh], DPAD, and some of the other medical projects). I don't want to discourage any potential TeAmmates, but I don't want to get people's hopes up, either, only to have them disappointed.
Also: I noticed on another thread that some of the other DC clients use lots of memory (20MB on one and 50MB for another). With the console and core combined F@H only uses 2.5MB on my system it looks like. If people knew this it might influence them to do F@H instead of something else. Or someone running a different client that has hyperthreading computer could run their client and F@H because it uses less memory. Maybe? I don't know.

Anyway, thanks for the FAQ, I added it to my sig.
Click to expand...
That number is not very accurate. Do this experiment: close your F@H client, open the task manager, observe the "commit charge" (the first number in the right panel of the status bar). Now, open the F@H client, and see how much that number jumps. It will vary depending on the WU, but with a p520_BBA5_pf, I'm showing a 25MB increase in memory usage when F@H loads. I have seen it go as high as 50MB in the past with particularly complex WUs.
 
Yeah I didn't know how to figure ram usage. Thanks for the info.

My rig sat several hours today only working on one WU while I was at work. Being able to cache one WU would be nice.
But the WU was waiting for me to connect.
 
The WUs(?) are getting bigger for FaH right? I almost remember someone mentioning that it is taking longer to dump an FaH WU than it used to or something.

I only bring this up because I started running FaH this morning/afternoon sometime, and haven't seen anything on the stats yet. Dual p3 866, running two clients. Since it's the only Linux machine I have, it will probably be the only one running FaH, unless someone wants to donate a hyper threading p4... 😉
 
There are many different size work units. They can take anywhere from 5-7 hours to 6-7 days depending on the WU and the speed of the system you are using.

Linux users will get more Tinker core WUs than Windows users will get. It'll probably take a few days to crunch through the WUs on your dual 866.
 
And speaking of the other kind of WU sizes, a subject of interest to those of us who have dialup connections, the amount of data downloaded for Tinker vs. Gromacs is not a noticeable difference to me on a 56K line. However, completed Tinkers upload rather quickly, while I'd suppose on the other hand that a finished Gromacs WU is probably the greater part of a megabyte.
 
Originally posted by: Hyperfocal
There are many different size work units. They can take anywhere from 5-7 hours to 6-7 days depending on the WU and the speed of the system you are using.

Linux users will get more Tinker core WUs than Windows users will get. It'll probably take a few days to crunch through the WUs on your dual 866.
Click to expand...

Thanks for the info. I wanted to make sure I wasn't wasting cycles or anything. 😉
 
Originally posted by: Hyperfocal
Bump and a question.

How do you set up F@H to load automatically on boot in Linux? Preferably in the background.
Click to expand...

You should be able to write a quick startup script to start it at boot.

Look at /etc/init.d
 
Originally posted by: Insidious
My Q:

what are the prefered switches for maximum pointage?

Your A:

tia! :beer:

-Sid
Click to expand...


With the new points values, I would say just use -forceasm (for Intel) and -forceSSE (for AMD)

All the point values went up, but the Tinker and Gromacs units are now worth pretty much the same.

-advmethods now just indicates you are willing to fold beta units.
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top