Discussion in 'Overclocking & Cooling' started by giorgos th., Aug 17, 2006.
if the database is going to be created it will be ONLY with this version and ONLY if the screenies have the checksum.
Will do, & I will run that later tonite (or because it is Friday, possibly tomorrow morning/nite) so I can get the potential benefit of cooler surrounding ambient temperatures.
heres mine on the 1.5.with cpu at 4.4ghz,mem at 883mhz,fsb=294,vcore=1.45v.
here it is at 2600mhz, on september once i get the PSU and cooler i post the final score:
that`s more like it...
Ok here is my superpi 1M Score nothing fancy
From what I have read online (besides my "reservations" about running the single-threaded model of this on a DualCore machine)?
Some things to note/think about which I stated in this thread here:
A DualCore machine, running the single threaded model of this program?
Technically, that will NOT extoll their full virtues quite possibly on DualCore/SMP rigs!
That is, unless this single-threaded test is run in REALTIME cpu priority!
(Then, a single threaded program will take over an entire core itself IF run in "REALTIME" cpu priority granted it... @ least, in theory!)
The OS Kernel memory & process mgt. subprograms SHOULD then, run ALL ELSE on the least used processor - that is, if already running processes are NOT on the core you intend to run this on...
(Affinity may help here, not only setting processor CPU priority given to a process)...
This also supposes that the test will bear that type of running on it, & NOT all programs will!
* Some "food-for-thought" for the dualcore testers here...
P.S.=> ALSO, from what I understand? Where loosening timings in RAM BIOS stuff can help on ScienceMark 2.0 for 1 example??
Tightening timings SUPPOSEDLY helps on THIS test... but, this supposition/theory??? I am NOT sure of... still, another thing to think about... apk
Ha, X800 that was my really old name.. Anyways here is mine
*I tryed uploading it from here but the file size was to big*
i will post w/ the new mod as soon as the temps drop out here we are getting 100F+ out here so its to hot to firstname.lastname@example.org like i did on my scm2 test
P4 640: 27.000s @ 4.8ghz. The memory was a bit of a bottleneck tho I'll see if i can find the screenshot.
ok these are my score, fastest with a checksum stock and the fastest i could get but couldnt get the pic off befor i bsod so could you pls still accept it even w/o the checksum
My Super-Pi 1.5 mod scores AND CPU-z 1.36 data for CPU & RAM too
CPU-z 1.36 data CPU-mhz-wise:
CPU-z 1.36 data RAM-mhz-wise:
Super-Pi 1.5 mod score:
* There she be, & HIGHEST overclock I have been able to run stable (239fsb x 12 multiplier = 2872mhz CPU) thru/by using 12x multipliers thusfar!
(Stable on THIS Super-Pi test, but NOT ScienceMark 2.0 though)
P.S.=> Using 11x multipliers yielded:
(Earlier tests (photos I put up before on the 3rd page of this thread) did better scoring by using SSE2/SSE3 optimized versions of this program (31 seconds ranges using them &lower than score now) ranges on Super-Pi 1.4 mod)
Using 12x multipliers yielded:
& scored about the same in the 32 seconds ranges (very close using non-SSE2/SSE3 optimized versions builds of this benchmark test, 1.4 build vs. 1.5 build of Super-Pi now):
32.609 current score Super-Pi 1.5 using 12 multipliers
32.578 previous score Super-Pi 1.4 using 11x multipliers
(Tough call here for me: Battle between using 11x mult.'s (worse CPU-mhz, but better RAM-mhz return) vs. 12x mult.'s (better CPU-mhz by far, vs. worse RAM-mhz return) as to really which I am going to stick by in the end here)
I am experimenting with here again using 12x multipliers in lieu of the 11x ones I used on my 3rd page photos posted earlier in this thread... apk
looks like i've got ya on super pi m8.lol
hehehe......suckers my best time so far is 17.00 with my new E6600 B2/4gigs of Teem Xtreem mem @ 1000mhz, 1TB RAID0, ASUS P5WDH, and a 7950GX2 oh yea, and thats only with about 10 min after building it, it got here yesterday. go Intel!!
Alec definetely buy some ram...
Oo_Skyline_oO let me see you break 12" with your E6600..
Yes, I think so, somekind of better "more overclockable/overclocking ready" RAM.
(Don't really know which is best though - Mushkin, OCZ, Patriot, you-name-it!)
P.S.=> All I really need on this machine is 512mb of that, & I'd be set for higher scores, but again - what kind is best from the DDR-400 range you upgrade to? apk
well i i just picked up my new cpu and ram and heres my first score at 1m....35s...so far only oced from 2.4 to 2.52
another AM2 user , but that memory will kill you in games if you don't overclock it, mine at 533 in science mark and cpu at 1800mhz only do 900pts, see if you can lower your timmings to cl3
if your mobo gives 3+V to memory i`d go to 2x256 old school bh-5....kick ass....
well this is what im running at now....what can i do in my bios the only i changed it from 2t to 1t everything else is on auto...
if it's stable at 1t it's pretty good you need to try if you have better performance at lower timmings and lower mhz or more mhz and higher timmings
hey alex, i just passed you (competition is a good thing)
is anyone going to do a score table?
oh thats easy, Mushkin drop me a PM if you want some help picking some godlike RAM
It does... can you give me a linkage to this particular memory you speak of?
* I'll be buying again shortly/most likely, because my RAM is what is constantly holding me "down/back" from better scoring on this benchmark test & ScienceMark 2.0...
(And, doubtless from better overall performance, period!)
Separate names with a comma.