New build, finally! E4300/Gigabyte 965P-DS3 @ 3.0GHz

So my week 44 cheapie processor finally came in and I was able to just start
messing around with it today. I used Easytune to ramp up the FSB to get an
idea of where my proc was at. I started off at 200 and got up to only 9x283
(2550MHz) before Orthos started immediately erroring. Hmm... disappointing.

So I thought about all those posts I saw about FSB walling and decided to
drop the multi to 8 and up the FSB. I must have ended up in a higher strap
because the FSB kept going up and up until XP BSOD'ed. It crashed at 8x386
(3088MHz )and I flipped over the multi back to 9. Crashed again at 3070MHz.
Looking good... it seems I've found my proc's limit. 3G's was my target and
it's nice to have found it so quickly. I notice too that the power
consumption at the wall is quite relatively low compared to the old AthXP
150W load.

Currently I'm pushing the PC6400 memory (Corsair XMS2 5-5-5-12 1GBx2) to see
if it can handle a 12% OC @ 900MHz while the CPU is hovering at 1350MHz.
Seems to be doing okay at this point. Will report results soon. Afterwards
I'm going to be cutting down power to see what kind of power consumption
advantages I can get.

Have any of you used an E4300 with a pinmod to get the higher strap? Any
positive results?

Phil

"Phil, Non-Squid" <phil_leeDELETEME@hotmail.com> wrote in message
news:4Ga%h.436505$Pi4.122462@newsfe14.lga...
> So my week 44 cheapie processor finally came in and I was able to just
> start messing around with it today. I used Easytune to ramp up the FSB to
> get an idea of where my proc was at. I started off at 200 and got up to
> only 9x283 (2550MHz) before Orthos started immediately erroring. Hmm...
> disappointing.
>
> So I thought about all those posts I saw about FSB walling and decided to
> drop the multi to 8 and up the FSB. I must have ended up in a higher
> strap because the FSB kept going up and up until XP BSOD'ed. It crashed
> at 8x386 (3088MHz )and I flipped over the multi back to 9. Crashed again
> at 3070MHz. Looking good... it seems I've found my proc's limit. 3G's was
> my target and it's nice to have found it so quickly. I notice too that
> the power consumption at the wall is quite relatively low compared to the
> old AthXP 150W load.
>
> Currently I'm pushing the PC6400 memory (Corsair XMS2 5-5-5-12 1GBx2) to
> see if it can handle a 12% OC @ 900MHz while the CPU is hovering at
> 1350MHz. Seems to be doing okay at this point. Will report results soon.
> Afterwards I'm going to be cutting down power to see what kind of power
> consumption advantages I can get.
>
> Have any of you used an E4300 with a pinmod to get the higher strap? Any
> positive results?
>
> Phil

So bumping up voltage 7% to 1.45V gives a 7% increase to 3.174GHz. Not
worth it. Gonna clock it back down to ~3G.

Phil

"Phil, Non-Squid" <phil_leeDELETEME@hotmail.com> wrote in message
news:05b%h.20419$ZD3.17133@newsfe01.lga...
>
> "Phil, Non-Squid" <phil_leeDELETEME@hotmail.com> wrote in message
> news:4Ga%h.436505$Pi4.122462@newsfe14.lga...
>> So my week 44 cheapie processor finally came in and I was able to just
>> start messing around with it today. I used Easytune to ramp up the FSB
>> to get an idea of where my proc was at. I started off at 200 and got up
>> to only 9x283 (2550MHz) before Orthos started immediately erroring.
>> Hmm... disappointing.
>>
>> So I thought about all those posts I saw about FSB walling and decided to
>> drop the multi to 8 and up the FSB. I must have ended up in a higher
>> strap because the FSB kept going up and up until XP BSOD'ed. It crashed
>> at 8x386 (3088MHz )and I flipped over the multi back to 9. Crashed again
>> at 3070MHz. Looking good... it seems I've found my proc's limit. 3G's
>> was my target and it's nice to have found it so quickly. I notice too
>> that the power consumption at the wall is quite relatively low compared
>> to the old AthXP 150W load.
>>
>> Currently I'm pushing the PC6400 memory (Corsair XMS2 5-5-5-12 1GBx2) to
>> see if it can handle a 12% OC @ 900MHz while the CPU is hovering at
>> 1350MHz. Seems to be doing okay at this point. Will report results soon.
>> Afterwards I'm going to be cutting down power to see what kind of power
>> consumption advantages I can get.
>>
>> Have any of you used an E4300 with a pinmod to get the higher strap? Any
>> positive results?
>>
>> Phil
>
> So bumping up voltage 7% to 1.45V gives a 7% increase to 3.174GHz. Not
> worth it. Gonna clock it back down to ~3G.
>
> Phil
>

Stable and oh-so-silent at 2.9GHz... Able to play multiple HDTV 1080p
streams at once where before I was unable to play one 720p stream. Probably
quicktime's intel-optimized algorithms...

Phil

Phil, Non-Squid wrote:
> Stable and oh-so-silent at 2.9GHz... Able to play multiple HDTV 1080p
> streams at once where before I was unable to play one 720p stream.
> Probably quicktime's intel-optimized algorithms...

Phil, what FSB settings are you using? Still using 8x multiplier? I cannot
seem to get my E4300 to work below 1333 :-( At 1333, it needs too much
voltage (1.47) to be stable in an Orthos stress test (BSOD).

--
Met vriendelijke groeten, Thomas van der Horst

"Thomas" <ThomasH@lycosmail.nl> wrote in message
news:463e0343$0$16940$9a622dc7@news.kpnplanet.nl.. .
> Phil, Non-Squid wrote:
>> Stable and oh-so-silent at 2.9GHz... Able to play multiple HDTV 1080p
>> streams at once where before I was unable to play one 720p stream.
>> Probably quicktime's intel-optimized algorithms...
>
> Phil, what FSB settings are you using? Still using 8x multiplier? I cannot
> seem to get my E4300 to work below 1333 :-( At 1333, it needs too much
> voltage (1.47) to be stable in an Orthos stress test (BSOD).

I bumped everything down to optimize the heat/noise and went with the 9x
multi because it let me OC my memory by ~8%. I figured the NB was hot
enough and higher FSBs might begin melting it down .

CPU
9x324MHz=2916MHz

Memory
324x2.66=862MHz (OCed from 800)
5-5-5-12 default timings (might tune this later)

Voltages
Default CPU (1.325V)
Default Memory (1.9V)
Default NB
Default PCIE

So you're saying that you need a higher voltage with a higher FSB? I never
knew how that related really. I did read that the PCIE bus being OCed a
little bit helps somewhat.

The highest I got on 1.45V was ~3.2G but the voltage increase was beginning
to have diminishing returns (1% increase in Vcore yielded less than 1% of an
OC).

I feel like I'm done... anyone have other suggestions? I'm moving out of my
apartment today so I will tweak my timings later.

Phil

Phil, Non-Squid wrote:
> So you're saying that you need a higher voltage with a higher FSB? I
> never knew how that related really. I did read that the PCIE bus
> being OCed a little bit helps somewhat.

Hmmmm maybe I'll try that then. Well, to get higher FSB doesnt need extra
voltage, but to let my CPU run 66% over spec DOES need some extra juice :-)

It sounds like you're indeed done, and I'm still searching for answers with
thuis build. For now, it's only rock steady at 1066 MHz FSB (for 2.4 GHz
CPU).

Good luck on the moving. We just moved ourselves, can be a real pain.

--
Met vriendelijke groeten, Thomas van der Horst