ASRock 775Dual-VSTA OC Help

[Dread]

I'm using SuperPI since it echos what Window users would see.
Will post some results in a little bit.

I actually found (through your links, and some of my own searches), hundreds of linux benchmark apps. Unfortunately most of them target specific sub-systems other than CPU / memory and would have no worthwhile results for an over clock.

But am keeping them in my bookmarks for when I decide to get a little loose and wild with linux. Thanks :)

EDIT - Did 3 separate runs with SuperPi C2D E4300@2.026 GHz 2x256 MB DDR-266 RAM Ubuntu 6.10-64 with F@H running :) Before I add new RAM will run again without F@H loaded.

Results:

------ Started super_pi run : Wed Feb 14 19:06:36 CST 2007
------ Ended super_pi run : Wed Feb 14 19:07:14 CST 2007
Version 2.0 of the super_pi for Linux OS
Fortran source program was translated into C program with version 19981204 of
f2c, then generated C source program was optimized manually.
pgcc 3.2-3 with compile option of "-fast -tp px -Mbuiltin -Minline=size:1000 -Mnoframe -Mnobounds -Mcache_align -Mdalign -Mnoreentrant" was used for the
compilation.
Start of PI calculation up to 1048576 decimal digits
End of initialization. Time= 0.468 Sec.
I= 1 L= 0 Time= 1.644 Sec.
I= 2 L= 0 Time= 1.928 Sec.
I= 3 L= 1 Time= 1.912 Sec.
I= 4 L= 2 Time= 1.928 Sec.
I= 5 L= 5 Time= 1.912 Sec.
I= 6 L= 10 Time= 1.916 Sec.
I= 7 L= 21 Time= 1.928 Sec.
I= 8 L= 43 Time= 1.912 Sec.
I= 9 L= 87 Time= 1.920 Sec.
I=10 L= 174 Time= 1.928 Sec.
I=11 L= 349 Time= 1.904 Sec.
I=12 L= 698 Time= 1.936 Sec.
I=13 L= 1396 Time= 1.912 Sec.
I=14 L= 2794 Time= 1.912 Sec.
I=15 L= 5588 Time= 1.916 Sec.
I=16 L= 11176 Time= 1.908 Sec.
I=17 L= 22353 Time= 1.900 Sec.
I=18 L= 44707 Time= 1.876 Sec.
I=19 L= 89415 Time= 1.784 Sec.
End of main loop
End of calculation. Time= 38.002 Sec.
End of data output. Time= 0.124 Sec.
Total calculation(I/O) time= 38.126( 1.408) Sec.

[gvblake22]

Looks like a good start. If you want to really make sure it is stable, do a 32M run of SuperPI when you have a half-hour or so to spare.

And you don't have to list the time for every iteration of the PI calculation, just record the total time it took to complete and which size calculation it was (1M, 2M, 16M, 32M, etc).

[Quakindude]

All I have to say is:



And

[Dread]

Well, got it to 2.43 GHz on stock air stable. SuperPI 1M 19.82 seconds (50% faster than before). Going to let it run at this speed for a few days. Tomorrow will more than likely do
what gvblake22 suggests, and run SuperPI at 32M to get a better feel for what it can do.
Then I'll start increasing the clock, maybe to around 2.7-2.8 GHz. I don't want to go much higher on stock HSF. I probably shouldn't go any higher at all until I get another Zalman
9700 or something.

Also noticed a significant increase in folding speed :)

Quakindude, you're 1 Ghz faster clock, and only 5 seconds quicker on the SuperPI :) Whats up with that?

[Quakindude]

Originally Posted by Dread

Quakindude, you're 1 Ghz faster clock, and only 5 seconds quicker on the SuperPI :) Whats up with that?

Wait until you try to get Super Pi under 16 seconds. For some reason, that can be tough to do. I never could do it on my AMD X2 4800 rig.

[Dread]

System stable @2.43 GHz still, no changes on the OC yet.
Re-ran Super-Pi, this time without the Gnome Desktop Manager loaded
and shaved off almost couple seconds from it. I think without the GDM, and adding a couple hundred MHz to the CPU, I may be able to break the 16 second barrier. Proof of concept won't happen until I get around to ordering that Zalman 9700 though.

1M = Time = 17.96 Sec

For whatever reason, when closing GDM on this box, the system
will hang (not crash) for about 5 minutes before loading the terminal.
So for all future runs will leave the GDM loaded and just assume it's marginally quicker without it.