Client 5.02 Released

D

dr_dirtnap

Limp Gawd
Joined
Aug 4, 2004
Messages
339
Folding@Home 5.02
August 20, 2004

----------------------------------------------------------------------

----------------------------------------------------------------------
RECENT CHANGES (Client):

5.02:
- Error checking on queue, related to big packet option.
 
If you do get bothered enough to install it, and you use Larry's most excellent EM3 to monitor yer boxen, rename the program to FAH500-Console.exe.

Otherwise, EM3 doesn't recognize it as a valid program.

This public service announcement brought to you by the guy who wasted 45 minutes trying to get the thing working... ;)
 
Strikemaster said:
If you do get bothered enough to install it, and you use Larry's most excellent EM3 to monitor yer boxen, rename the program to FAH500-Console.exe.

Otherwise, EM3 doesn't recognize it as a valid program.

This public service announcement brought to you by the guy who wasted 45 minutes trying to get the thing working... ;)
Click to expand...
ah thx, this would be renaming the winFAH.exe file?

i was wondering why EMIII would not work :p
 
In your client config:

[settings]
username=dirtnap
team=33
asknet=no
bigpackets=yes
machineid=1

Just add the bigpackets=yes
 
Don't know the points for the big work units, I think one of the ones I got was worth 242.

Carnival,

For the time being, the p130X series will only be available to advmethod clients
Use the advmethod flag to get them.

Quoted from the F@H forums.
 
EM is an uber monitoring program that will monitor your F@H clients. It can be run across a network and is by far the best monitoring program out there.

For more information and to download go to http://www.em-dc.com
 
So since I'm just adding a few more boxes and I haven't been around for a while, could someone tell me which switches to use? They seem to change with every new release and the last one I had was 4.0
 
dr_dirtnap said:
Don't know the points for the big work units, I think one of the ones I got was worth 242.

Carnival,

For the time being, the p130X series will only be available to advmethod clients
Use the advmethod flag to get them.

Quoted from the F@H forums.
Click to expand...
p130x? whassat?

lucky for me, i'm using the -advmethod flag :)

EMIII says my current protein is: Protein: 1252 p724_Abeta21-43-amberGS 63.0

what does that mean?
 
Number of atoms ?
p1301_1RYP_AAAA_UM, Type Gromac, Atoms 119,496, Preferred 24 Days, Deadline 35 Days.
p1302_1ryp_AAAA_free, Type Gromac, Atoms 119,496, Preferred 24 Days, Deadline 35 Days.

The next closes one atom sizewise are p1122 at 84,764 & p1113 at 76,754.
Then it p724 at 47,747.
Eveything else is sub 20,000 atoms.
Tinkers are all sub 1,000 atoms.

I wonder if you put bigpackets=no you will get more Tinkers ?

Luck.......... :D
u=Tigerbiten.gif
<- I've got my fourth [H].
 
Tigerbiten said:
Number of atoms ?
p1301_1RYP_AAAA_UM, Type Gromac, Atoms 119,496, Preferred 24 Days, Deadline 35 Days.
p1302_1ryp_AAAA_free, Type Gromac, Atoms 119,496, Preferred 24 Days, Deadline 35 Days.

The next closes one atom sizewise are p1122 at 84,764 & p1113 at 76,754.
Then it p724 at 47,747.
Eveything else is sub 20,000 atoms.
Tinkers are all sub 1,000 atoms.

I wonder if you put bigpackets=no you will get more Tinkers ?

Luck.......... :D
u=Tigerbiten.gif
<- I've got my fourth [H].
Click to expand...

Tigerbiten,

From my experience, if you don't use -advmethods and request no-timelimit wu's, you will get all tinkers.
 
dr_dirtnap: that could be very useful experience. ATM i think all my athlon boxen are crunching gromacs cores (all three:rolleyes: i'm trying... gimme a break) and the p4 i've got going is doing a tinker... that could improve my output a bit.

are there any performance increases with 5.02? is it worth updating if bigpackets is off? (dialup necessitates it)
 
Nah, the cores are still the same, so the performance stays the same.

Yeah, the way stanford has it setup currently, if you don't have -advmethods and request no deadline WU's, you will always get a Tinker. I could be wrong, but this has held true 100% of the time for me. So basically I let the Intel boxes crunch the Gro's, and the AMD's get the Tinkers.

I don't know if Stanford endorses this method of WU selection, but to me it allows me to be a more efficient participant in the program.
 
Just done two timeless Tinkers.
p1104_Kiefhaber_5, Speed 2.4 Ghz, Points 49, Time 07:43:20, PpHpG 2.645.
p1108_Kiefhaber_20,Speed 2.0 Ghz, Points 37, Time 10:41:20, PpHpG 1.731.

First one was slightly quicker than a Gromac on that box.
Second one is definatly slower than any Gromac I've done.
Nothing near the 3.5 PpHpG of a normal Tinker.
So I'm just going to fold normaly and hope to get normal Tinkers.

Luck......... :D
u=Tigerbiten.gif
 
You must log in or register to reply here.
Back
Top