• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

How many instances

I would suggest 1. That way that one will get 100% of the proccesor. If you have 2, each will get 50%, and really you are doing the same amount of work; it will just take twice as long to turn the work units in.
 
Intel proccesor that have hyper-threading enabled are the systems that benefit when two instances are run.
 
Two instances or more slow down the completion time of the work units which isn't always a bad thing. You don't want to fly through a wu only to find out a servers down and you have to wait to download another wu while your box sits idle. The more instances the fewer connections to Stanford which in the long run equals better productivitiy. Don't worry about pilling up stats fast, you'll get the same amount of pts. one way or the other, but potentially more pts. with more than one instance.
 
Originally posted by pageian
Two instances or more slow down the completion time of the work units which isn't always a bad thing. You don't want to fly through a wu only to find out a servers down and you have to wait to download another wu while your box sits idle. The more instances the fewer connections to Stanford which in the long run equals better productivitiy. Don't worry about pilling up stats fast, you'll get the same amount of pts. one way or the other, but potentially more pts. with more than one instance.
Click to expand...

Well, actually piling up points means nothing to me, how many points does it take to buy a cup of coffee... As for where I stand on the list what difference does that make either. There are hundreds ahead of me on the list running 5 or 10 computers that I will never be able to pass so why worry about it.

My main point in running more than one instance was to get more work done to help the project itself and the team totals (Darn Aussies), not my standing in the team which will probably not get any better than it is now.

At the moment I have three instances running and so far I really can't tell a big slow down in production. I am sure there probably is, but not that bad. I started all three instances right about the same time with new work units and right now they show:

1. Doing 124000 steps on #25
2. Doing frames on frame #33
3. Doing 250000 steps on #54

They have been running for about 6 hours, does this sound like about right? It shows 96% cpu usage and a cpu temp of 40C, but the room temp is kind of high tonight by about 8F over normal temp.
 
Originally posted by James McGee
Well, actually piling up points means nothing to me, how many points does it take to buy a cup of coffee... As for where I stand on the list what difference does that make either. There are hundreds ahead of me on the list running 5 or 10 computers that I will never be able to pass so why worry about it.

My main point in running more than one instance was to get more work done to help the project itself and the team totals (Darn Aussies), not my standing in the team which will probably not get any better than it is now.

At the moment I have three instances running and so far I really can't tell a big slow down in production. I am sure there probably is, but not that bad. I started all three instances right about the same time with new work units and right now they show:

1. Doing 124000 steps on #25
2. Doing frames on frame #33
3. Doing 250000 steps on #54

They have been running for about 6 hours, does this sound like about right? It shows 96% cpu usage and a cpu temp of 40C, but the room temp is kind of high tonight by about 8F over normal temp.
Click to expand...

Three instances should be okay. As long as you see them making progress you shouldn't have to worry. As for the points I was just pointing out that a lot of new folders don't like the idea of more than one instance because it slows down their output and they don't get that instant gratification of seeing numbers posted. More than one instance really is more efficient as long a the cpu can handle it, and yours certainly can.

p
 
Running multiple instances on a non-HT or non-MP machine does have some issues.

Should there be no communication issues with Stanford, your production might suffer slightly. Reason, a little thing called process or context switching. The OS will have to take some cycles to switch from one instance to another.

Given that occasionally there are communications issues with Stanford, then having 2 instances might help production as the instances would not normally be trying to send/receive from stanford at the same time. Therefore, if one instance is waiting to communicate with stanford the other will merrily continue using 100% of the CPU.

Also you need to consider the amount of RAM that you have. Under no circumstances do you want your machine paging memory and running a DC project at the same time. So make sure you have enough RAM to do the job.

You need to make sure that your machine will finish the workunits within the given timeframe allowed. I believe that no points are awarded for late workunits.

YMMV, but I would only suggest running at most 2 instances on a non-HT, non-MP machine.
 
I would strongly suggest you run 2 instances - 1 of normal folding with priority set to low (still should not interfere with other apps) and 1 of a timeless Tinker (gah preference selected) at idle priority.
I've found that on systems running the "double Gromacs" that 3% to 6% of my cycles are still wasted.
 
You must log in or register to reply here.
Back
Top