• 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.

dual Xeon, HT question....

M

MustangWorld

Weaksauce
Joined
Oct 4, 2004
Messages
97
So when i setup FAH on the machine, what machine ID's should i use? 1 & 3? I'm currently running 4 instances and want to cut it down to two so I can actually work on this machine too
 
any other applications will be given as much cpu time as they need from f@h, its priority should be set to only take idle cpu cycles

 
HAve you got all four set to run big packets? Cos if its feeling sluggish the reason will be ram, not cpu time, if its caused by FAH and not crapware anyway. Try turning two of the clients to normal wus, bear in mind that a 600 point gromacs is ~100mb, so you need at leas 512mb just for folding +os, without taking your programs into account, if youre on one of the qmd cores its even worse.

As far as what ids to use i dont belive it matters, so 1+3 shoudl be fine.
 
Yeah, I had all 4 running big packets. Grabbed 4 QMD cores and each of those took up about 350mb RAM each. :eek:

Also, Is it normal to only see a process take 25% usage, but the graph say its running full speed?
 
With a dual HT-enabled machine, you can run 4 instances... each would take "25%" of the whole.... But I would recommend against running all 4 with big packets... 1 or 2, yes... but not all 4...... Unless you have a few GBs of RAM :D.... F@H gets trumped by every other processor request, so it shouldn't be the cause of sluggishness (excluding the 4 instances on big packets).... RAM was likely what was slowly you down...

I would run 4 instances... 2 with big packets and 2 with regular units... You could even set the 2 regular unit instances up to run at 50% usage.... and that could allow you a little headroom with the processors (if you're concerned about using the PC regularly).... But getting rid of the big packets on a couple of the clients should solve your sluggishness....


Keep on Folding!!

 
If you do want to cut down to two instances, though, it doesn't matter what machine IDs you use as long as they don't match.
 
OSUguy98 said:
With a dual HT-enabled machine, you can run 4 instances... each would take "25%" of the whole.... But I would recommend against running all 4 with big packets... 1 or 2, yes... but not all 4...... Unless you have a few GBs of RAM :D.... F@H gets trumped by every other processor request, so it shouldn't be the cause of sluggishness (excluding the 4 instances on big packets).... RAM was likely what was slowly you down...

I would run 4 instances... 2 with big packets and 2 with regular units... You could even set the 2 regular unit instances up to run at 50% usage.... and that could allow you a little headroom with the processors (if you're concerned about using the PC regularly).... But getting rid of the big packets on a couple of the clients should solve your sluggishness....


Keep on Folding!!

Click to expand...


Thanks for the info, I'll give that a shot :)
 
i run a dual Xeon w/HT at home. I run four instances, with each FAH setup having its own Processor ID. I am impressed with how quickly the HT threads get stuff done. The PC doesnot normally do much when im not home, so it gets alot of work done.

It has 2GB of RAM, which it never comes close to using.
 
Steel Chicken said:
i run a dual Xeon w/HT at home. I run four instances, with each FAH setup having its own Processor ID. I am impressed with how quickly the HT threads get stuff done. The PC doesnot normally do much when im not home, so it gets alot of work done.

It has 2GB of RAM, which it never comes close to using.
Click to expand...


How did you configure yours? big packets? advmethods? etc....
 
MustangWorld said:
How did you configure yours? big packets? advmethods? etc....
Click to expand...

sorry was short on the details. I configured the HT instances just like the regular ones, big packets ok, not timeless, etc. They seem to be working fine.

if you want more details, I can post a copy of the config file when I get home.
 
Steel Chicken said:
sorry was short on the details. I configured the HT instances just like the regular ones, big packets ok, not timeless, etc. They seem to be working fine.

if you want more details, I can post a copy of the config file when I get home.
Click to expand...


That would be cool. Either post it or PM, dont matter to me. Thanks a ton
 
MustangWorld said:
That would be cool. Either post it or PM, dont matter to me. Thanks a ton
Click to expand...

simple settings. all i do is change the machineid, from 1-4 for all the processors.
you would think the logical HT cores wouldnot do much, but they seem to do alright

[settings]
username=[H]Steel_Chicken
team=33
asknet=no
bigpackets=yes
machineid=1
local=13

[http]
active=no
host=localhost
port=8080
usereg=no

[core]
disableassembly=no

[power]
battery=no
 
weird. I have the same config, running as a service, with the -local -advmethods -forceasm flags in the service parameters.

Guess the reason its going so slow is no O/C on the xeons yet, hmmm. Time to find some thin speaker wire, lol
 
You can also cut apart floppy drive cables. They've got some really thin stranded wire. ;)
 
You must log in or register to reply here.
Back
Top