• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

Whats the highest stable fsb you've reach?

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.
It is a CPU test that gets your processor to run full blastl untill you tell it to stop.
In SANDRA it reruns whatever tests you want it to as many times as you want
so if you do the 2 CPU tests, thats a CPU burn in.

WCPU was on that screenshot though and its an informational tool
to display CPU info.
 
Last edited:
Dirty_Punk said:
250mhz DC completelly stable but with pc4000, timigns 2.5-3-4-11 cpu interface enable
waiting for my KHXPC3500 to come back, them can hit 250mhz easily

ah, 1.7v vdd, no vmod, stock cooling or watercooling NB, no difference... :(

ops, write wrong... 1.5v vdd
ah, tested another nf7, this can do around 260mhz cpu int enable stable, but only with mulitplier form 7x to 8.5x :D :p
so nothing special..
 
yes, but with 1.6v and 1.7v the board seems a bit more unstable at very hight fsb...
for example at 1.5v i can hit 264mhz, with 1.7v i can't, and it's not cooling problem because i have tryed with watercooling and 3°C water, but no gain...
with 1.4v the board won't boot

i have noticed the same think in the other my 2 NF7 and NF7-S both V2.0, so don't know if are the boards or an other common component that i use for testing (like cpu or psu..)
 
hmm interesting, I will try it next time I get the chance to overclock my system. I would like to know why though, how tight is your timing when you are running at 264 fsb?
 
Mr.Rotory said:
hmm interesting, I will try it next time I get the chance to overclock my system. I would like to know why though, how tight is your timing when you are running at 264 fsb?

was a test with A-Data PC4000, because i have sent my KHXPC3500 to a group of friend for a test under CO2, and now i'm waiting them back.. when they return i'll vmod the mobo, i must do 19k with my ti4200

this was with 2*PC4000, cpu int enable, 2500+ cpu

pc4000.JPG

clockgen used because i'm with bios 10 that have at 250mhz the limit of FSB
 
so you are saying at 1.5V you can get higher fsb than 1.7? I havent even tried to use 1.5 or 1.6 D:
 
Cow2kie said:
so you are saying at 1.5V you can get higher fsb than 1.7? I havent even tried to use 1.5 or 1.6 D:

yes, with all the 3 NF7 that i have tryed...
strange, but tested vdd up to 2.2v and the best is always 1.5v....
 
thats an awesome overclock there Dirty_Punk, I am sure these numbers are only possible under extreme cooling, what conditions do you have there dude:eek:
 
mata2974 said:
thats an awesome overclock there Dirty_Punk, I am sure these numbers are only possible under extreme cooling, what conditions do you have there dude:eek:

yes, cpu is good clockers and under t-rex cooling (a termoelectric waterblock with 226w tec), it's perfectly stable at 2750 2v, so for 2800mhz need about 2.06v (values selected by bios, real voltages are around 1.95v for 2750 and 2v for 2800)

howether no vmod

i think that it's impossible without any vmod to do 4gb/s at sandramem with any type of superlucky BH5 memory on NF7... only pc4000 can do this, not so bad right? ;)
 
Mr.Rotory said:
what does cpu interface do?

give a good boost to the performance, probably like 10mhz of fsb, and it doens't stress the ram more, so if you are limited by the ram and not by the mobo you can get a gratis boost of performance, in the opposite case your system won't boot and you must clear cmos

with bios different from 10 it's near impossible to enable cpu interface with fsb>220 with multi 10x, only bios 10 can, this is why so much overclockers prefer bios 10
 
I see, i gotta say I love this forum. Even though I might be asking dumb questions, I'm getting a lot of info. And the stickies are really nice also.
 
Using a locked barton now that will finally do 2.5ghz and 227fsb. Haven't fully tested it yet so i don't know if the 2.8vdimm, 1.7vdd, cpu interface disabled, 7-3-2-2 timings and high vcore are necessary. Burning in now. I am looking at extreme cooling however, and with a locked cpu, my only hope for 2.7-2.8ghz would be with a hearty helping of even more FSB. I'd have to hit that 250+ barrier. Oh well, a challenge ain't a challenge if its easy.

/can't wait to smoke some INTEL boys and their inefficient cpu's
 
I know how you feel, but I've been reading around and found this site http://www.lowyat.net/guides/athlonxp/ They say that in the case of the t - breds, when AMD's palamino production went out, and the demand for 1700+ 's still remained, AMD retaged the 2600+ and sent them off as 1700+, hence so easy to overclock a 1700+ t-bred B. Now on the site he also mentions all the good batches and good weeks along with identity of your chip, check that out for your convinience, but also he says that some cpus did not run stable at 2600+ speeds and where still retaged to 1700+, which to me, doesnt sound like very nice quality control from AMD.

Now I wonder why are some people are able to oc their cpu much easier then others, this would explain it (ofcourse also putting into accountability the possibility of some chips that were on the wafer made better then others, as far as higher oc'ing goes) and since AMD did this with the t-breds, I'm guessing this is why most new locked bartons are able to oc (using the 11 multi) better then some unlocked ones, giving me 50/50 reason to go with a locked/unlocked barton. I'm guessing here but the locked bartons have been tested possibly well beyond 200 fsb and have passed, it might even be argued to go with a locked on over an unlocked one.
 
Back