Page 13 of 13
Re: Answers to: Reasons for not using F@H.
Posted: Sun Jan 06, 2013 8:00 pm
by ford316
bruce wrote:What kind of computer do you have that cannot meet the deadlines running 20 hours per day? Certainly there are assignments that you can't complete, but there are also ways to get assignments that you CAN complete. In other words, it's a matter of learning how to configure your client -- and there's certainly work that needs to be done in future clients so that that information is readily available to even the most novice Donor who might find out about FAH.
There's really nothing that can be done about no internet, but that's a pretty unusual circumstance, these days.
Bruce it was barely a P4 and I did alot of work on it so fah program didn't have a chance at about 4 hours a day which some days I had to leave a few programs running for the 4 hours it had to play catch up. That computer is long gone now off to the heavens to become something better lol true there may have been ways to still work it but at that time I didn't really have the time to research it. Now I have my gaming rig doing it in which case I am taking about a month or 2 off and letting fah catch up while I also make plans to move and build 2 more roughly of the same specs ya see listed in my signature with the exception I plan on running 2 gpu cards in each of the 3 desktops. It was not but just afew days ago I joined the forums but my team site always reminded me about this and finally I got this rig put together had problems at first 2 days but now working great even testing the new 310.90 driver so far about 13 hours and no problems. Months ago I use to be very busy doing many things and helping run afew websites at first I tried running fah while working but it slowed me down to much and thats when I tried the off time of me 4 hours a day and that didn't work I don't remember the exact numbers but some thing like this days to get it done first 17 final day to get it done 36 my time of getting it done would have and was 44 days
finally it died as I worked it to death and it was time for upgrade. I think it was a 1.6GHz had only 512mb ram it just could not do it in the same time I had for it to do it. Since then I built this rig and my time is not 20 or so hours a day any more so fah can run great now
Re: Answers to: Reasons for not using F@H.
Posted: Sun Jan 06, 2013 8:02 pm
by Kornflake
bruce wrote:ford316 wrote:...the second reason is computer can't handle it I know because sadly the computer setup I had could not run it because I was on it about 20 hours a day and the off time would not have been enough to finish a unit in time "I tried it" No other reason besides no internet I can think of
What kind of computer do you have that cannot meet the deadlines running 20 hours per day? Certainly there are assignments that you can't complete, but there are also ways to get assignments that you CAN complete. In other words, it's a matter of learning how to configure your client -- and there's certainly work that needs to be done in future clients so that that information is readily available to even the most novice Donor who might find out about FAH.
There's really nothing that can be done about no internet, but that's a pretty unusual circumstance, these days.
I was looking for the answer earlier but could not find it, what techniques are currently available to receive smaller WUs (or longer time limits)? I had a poster on a different forum explain they couldn't meet deadlines using V7.
Re: Answers to: Reasons for not using F@H.
Posted: Sun Jan 06, 2013 9:41 pm
by art_l_j_PlanetAMD64
Kornflake wrote:I was looking for the answer earlier but could not find it, what techniques are currently available to receive smaller WUs (or longer time limits)? I had a poster on a different forum explain they couldn't meet deadlines using V7.
To get smaller WUs:
1) in FAHControl, select 'Expert' mode from the dropdown menu at the top right.
2) now click 'Configure' and select the 'Expert' tab.
3) under the 'Extra client options', click on 'Add'.
4) for Name enter
max-packet-size
5) for Value enter
small
6) click on OK, then click Save at the bottom right of the Configure window.
From the
FahControl - FAHClient Guide:
max-packet-size options:
- small = Configures the slot to get small WUs (~5MB)
- normal = Configures the slot to get normal WUs (~10MB)
- big = Configures the slot to get big WUs (~500MB)
Re: Answers to: Reasons for not using F@H.
Posted: Sun Jan 06, 2013 9:50 pm
by Kornflake
Thank you I will pass that on. From reading
http://folding.stanford.edu/English/FAQ-Configure#ntoc3 I got the impression that this was no longer a working configuration option for V7. Additionally, I did not see that flag under:
https://fah-web.stanford.edu/projects/F ... ncesV6ToV7
Re: Answers to: Reasons for not using F@H.
Posted: Sun Jan 06, 2013 10:01 pm
by Joe_H
The small, normal and big tags exist, but have little effect on SMP folding slots. If you were running a system capable of processing bigadv WU's, the big setting could be needed. They do effect slots configured as uniprocessor, which is what is recommended for older and slow systems. Uniprocessor slots are also recommended for some systems that fold intermittently.
Re: Answers to: Reasons for not using F@H.
Posted: Sun Jan 06, 2013 10:16 pm
by Kornflake
Joe_H wrote:The small, normal and big tags exist, but have little effect on SMP folding slots. If you were running a system capable of processing bigadv WU's, the big setting could be needed. They do effect slots configured as uniprocessor, which is what is recommended for older and slow systems. Uniprocessor slots are also recommended for some systems that fold intermittently.
What would be the steps to reconfigure an SMP slot to a Uniprocessor slot? From what you are saying, that sounds like the way to go.
Re: Answers to: Reasons for not using F@H.
Posted: Sun Jan 06, 2013 10:26 pm
by Joe_H
Simplest is to finish the current WU using the Finish button. Then using Configure you can add one or more Uniprocessor slots and delete the existing SMP slot.
Re: Answers to: Reasons for not using F@H.
Posted: Sun Jan 06, 2013 10:39 pm
by art_l_j_PlanetAMD64
Yes, as noted elsewhere, the v7 documentation is a work-in-progress, with even experienced folders finding things to be unclear, confusing, or very difficult to find. Especially when on the
http://folding.stanford.edu/English/FAQ-Configure#ntoc3 web page it states this:
Below we've listed some of the more common switches used for v6 and V7; please see
this page for a complete list.
Where "this page" links to the "ClientDifferencesV6ToV7" web page which
does not contain the "complete list", only the differences.
Only after some further searching did I discover the FAHControl document with the full option list.
I had to go here:
https://fah-web.stanford.edu/projects/F ... umentation
Which led me to go here:
https://fah-web.stanford.edu/projects/F ... FahControl
Which actually (I think) contains the true complete list of options.
To try to find this information by starting at the "Home Page":
http://folding.stanford.edu/English/HomePage
or even the "Installation Guides":
http://folding.stanford.edu/English/Guide
is extremely difficult. I just went back and tried it now, and even knowing in advance where I wanted to end up, I could not find a reasonable path to get there. The "straightforward path" led me to the link that has the false claim of being the "complete list".
Re: Answers to: Reasons for not using F@H.
Posted: Mon Jan 07, 2013 4:45 am
by bruce
Joe_H wrote:The small, normal and big tags exist, but have little effect on SMP folding slots.
In this world of specialization, you can get different answers from different sources. The small/normal/big options are supported by V7. Unfortunately the assignment server has different ideas. Many of the uniprocessor projects are classified as small/normal/big, especially if your computer is limited to 256K or 512K ram. Since all SMP projects need more than that, those Projects have never been classified so the option is now essentially useless.
See also viewtopic.php?f=38&t=23378&p=233149*#p233149
As has already been said, the best way to get what you want is to use one or more uniprocessor slots and/or setting the "extra-core-args -np N" .