nf7s v2.0: what's the better overclock?

Archived from groups: alt.comp.periphs.mainboard.abit (More info?)

"Doug" <pigdos@nospam.com> wrote in message
news:92YMe.2830$Z%6.760@newssvr17.news.prodigy.com...
>I can get my mobile athlon to 2550 @ 221 Mhz FSB (synched, 2.5-3-3-11 CPC
>disabled) but I can get 230 Mhz FSB (synched, 2.5-3-3-11 CPC disabled) if I
>lower my CPU overclock to 2530. Since I've got an XP (even a mobile)
>shouldn't I be aiming for the highest CPU clockspeed I can get?
>
There is practically nothing between the to set-ups. The only way you could
tell would be through a bench test like PCMark, etc. The 11x230 might be a
few seconds faster on a DVD to AVI conversion. Use the one which runs the
longest in Prime95 torture test, etc.

Dave

Archived from groups: alt.comp.periphs.mainboard.abit (More info?)

Thanks Dave. I'll give that a try. What do you think of the PI test (the one
which calculates the value of PI)?

--
there is no .sig
"Dave" <dave2972@hotmail.com> wrote in message
news:gh1Ne.3853$FA3.1636@news-server.bigpond.net.au...
>
> "Doug" <pigdos@nospam.com> wrote in message
> news:92YMe.2830$Z%6.760@newssvr17.news.prodigy.com...
>>I can get my mobile athlon to 2550 @ 221 Mhz FSB (synched, 2.5-3-3-11 CPC
>>disabled) but I can get 230 Mhz FSB (synched, 2.5-3-3-11 CPC disabled) if
>>I lower my CPU overclock to 2530. Since I've got an XP (even a mobile)
>>shouldn't I be aiming for the highest CPU clockspeed I can get?
>>
> There is practically nothing between the to set-ups. The only way you
> could tell would be through a bench test like PCMark, etc. The 11x230
> might be a few seconds faster on a DVD to AVI conversion. Use the one
> which runs the longest in Prime95 torture test, etc.
>
> Dave
>
>
>

Archived from groups: alt.comp.periphs.mainboard.abit (More info?)

Doug,

As Dave said there is little difference between these two settings.
20Mhz <1% difference.

However, in general the difference between using the the multiplier or
the FSB to get your overclock reflects the capabilities of the bus,
CPU and memory. For instance, if you had a main application that was
very FPU intensive, you would want to emphasize the CPU clock even if
you had to run at a slower memory (FSB). i.e. you want want to use a
high multiplier. The other anlogous situation would be if your memory
speed was limiting the net clock speed, you would also want to raise
the multiplier to get a faster CPU clock. and vice versa if yur
processor limits you.

Hope that helps a bit.
Acc7

Archived from groups: alt.comp.periphs.mainboard.abit (More info?)

Superpi is a great prog to test your pc stability.
"Doug" <pigdos@nospam.com> wrote in message
news:1u4Ne.887$3F6.257@newssvr13.news.prodigy.com...
> Thanks Dave. I'll give that a try. What do you think of the PI test (the
> one which calculates the value of PI)?
>
> --
> there is no .sig
> "Dave" <dave2972@hotmail.com> wrote in message
> news:gh1Ne.3853$FA3.1636@news-server.bigpond.net.au...
>>
>> "Doug" <pigdos@nospam.com> wrote in message
>> news:92YMe.2830$Z%6.760@newssvr17.news.prodigy.com...
>>>I can get my mobile athlon to 2550 @ 221 Mhz FSB (synched, 2.5-3-3-11 CPC
>>>disabled) but I can get 230 Mhz FSB (synched, 2.5-3-3-11 CPC disabled) if
>>>I lower my CPU overclock to 2530. Since I've got an XP (even a mobile)
>>>shouldn't I be aiming for the highest CPU clockspeed I can get?
>>>
>> There is practically nothing between the to set-ups. The only way you
>> could tell would be through a bench test like PCMark, etc. The 11x230
>> might be a few seconds faster on a DVD to AVI conversion. Use the one
>> which runs the longest in Prime95 torture test, etc.
>>
>> Dave
>>
>>
>>
>
>

Archived from groups: alt.comp.periphs.mainboard.abit (More info?)

Well, it turns out than even though memtest can run all day at a 230Mhz
FSB/memory bus Half Life 2 didn't like it. So now I'm at 221 FSB/memory bus.

--
there is no .sig
"Lester Piglet" <lesross@nullandvoid.com> wrote in message
news:430737dd$0$12920$cc9e4d1f@news.dial.pipex.com...
> Superpi is a great prog to test your pc stability.
> "Doug" <pigdos@nospam.com> wrote in message
> news:1u4Ne.887$3F6.257@newssvr13.news.prodigy.com...
>> Thanks Dave. I'll give that a try. What do you think of the PI test (the
>> one which calculates the value of PI)?
>>
>> --
>> there is no .sig
>> "Dave" <dave2972@hotmail.com> wrote in message
>> news:gh1Ne.3853$FA3.1636@news-server.bigpond.net.au...
>>>
>>> "Doug" <pigdos@nospam.com> wrote in message
>>> news:92YMe.2830$Z%6.760@newssvr17.news.prodigy.com...
>>>>I can get my mobile athlon to 2550 @ 221 Mhz FSB (synched, 2.5-3-3-11
>>>>CPC disabled) but I can get 230 Mhz FSB (synched, 2.5-3-3-11 CPC
>>>>disabled) if I lower my CPU overclock to 2530. Since I've got an XP
>>>>(even a mobile) shouldn't I be aiming for the highest CPU clockspeed I
>>>>can get?
>>>>
>>> There is practically nothing between the to set-ups. The only way you
>>> could tell would be through a bench test like PCMark, etc. The 11x230
>>> might be a few seconds faster on a DVD to AVI conversion. Use the one
>>> which runs the longest in Prime95 torture test, etc.
>>>
>>> Dave
>>>
>>>
>>>
>>
>>
>
>

Archived from groups: alt.comp.periphs.mainboard.abit (More info?)

Doug,

In general, for your OC, you find the max you can get it to boot and
run at all at. And then back the OC down to get good stability for
the programs you want to run.

Make sure you monitor the temps during this.

A process for doing this is to start at or just below the max tou can
set and get the system to boot and not crash. Then run Super_PI to
look for quickly at stability. If It passes the 1M digits calc, next
move on to Prime95. Run the Prime95 benchmark. Again if it passes
that (runs it without errors), move on to the Prime95 "torture test".
Try it first at 15 minutes nd work up to where Prime95 torture test
can run for 24 hours.

Most often there will be one level at which you can "set records" but
not have the longer term behavior oyu need. Then ther is a lower
"speed" level at which you can run with stability for your favorite
programs.

Hopethis helps some.

Acc7

>Well, it turns out than even though memtest can run all day at a 230Mhz
>FSB/memory bus Half Life 2 didn't like it. So now I'm at 221 FSB/memory bus.

Archived from groups: alt.comp.periphs.mainboard.abit (More info?)

btw, Memtest is not a good "torture test" Use Prime95 as a better
measure of long term stability.

And watch the temps.

Acc7

>Well, it turns out than even though memtest can run all day at a 230Mhz
>FSB/memory bus Half Life 2 didn't like it. So now I'm at 221 FSB/memory bus.