my 60GXP doesnt like me overclocking

[monkeyspank]

I seem to be limited to around 160 FSB and i am almost certain it is my hard drive.
I have:
IBM 60GXP 40 gig
epox 8KHA+
1.33 athlon AYHJA
512MB crucial 2100
asus v8200 t2
350W enermax

all other components have been removed and i am still limited. The temperatures aren;t the problem (43 under load, 39 idle) at 160x9.5 (1525). the processor speed isnt the problem cos this also happens with an 8X multiplier.

When i try + boot up at over 160 fsb, i get 'boot volume unmountable'

What i am getting at is - are there any good performing HDD out there that have a reputation as being good overclockers?

If i find a good drive i might be able to use this one as storage space + see if windows boots ok then. (winXP, btw)

thanks for any input fellas.

-- There are no answers to find in the bottom of a glass. But you can have fun looking for them.

 

[labdog]

try scsiu3w.


if you know you don't know, the way could be more easy ...

 

[monkeyspank]

i'd love to dude but i dont have the cash.

Maybe i could whittle one out of cheese...

-- There are no answers to find in the bottom of a glass. But you can have fun looking for them.

 

[FatBurger]

It's not your hard drive, it's the IDE bus. Try changing the multiplier for the PCI, that should also change your IDE.

<font color=blue>If you don't buy Windows, then the terrorists have already won!</font color=blue> - Microsoft

 

[labdog]

lol.


if you know you don't know, the way could be more easy ...

 

[unoc]

If you want speed up your system try to change the CPU.
I wonder that your system is able to sustain such FSB, 160 MHz ??? Everything is running out of the specification, particularly the IDE channel and the PCI bus. Probably the disk is the last thing that can suffer for such FSB.
Also the processor is out from the specification. It has a FSB of 133 MHz and even you take the frequency around the nominal (at 8x) the input stage has to trust 166 MHz instead of 133. I believe that your system cannot work stable over 153÷155 MHz. Also RAM has its problems at that frequency. Did you tried some PC 2400 ? and what about CAS. It should be the first thing to check before an overclocking.
Rememeber also that you are losing so much time for a risible gain of time.

 

[monkeyspank]

i have changed my cpu speed. a 1.33 athlon running at 1520?

+ yeah i'm sitting here typing and my system is running at 160 FSB (x9.5 = 1520 Mhz). I'm fairly proud as I only spent £20 on a thermoengine with a delta fan. But i'll never be satisfied. It's stable too. I haven't tried any faster ram. maybe i'll do that next. cas is currently at 2 but even if i set it to 2.5 i cant get a higher FSB. I've got it set at turbo settings too cos it doesnt make any difference to the FSB I can reach.

Thanks for the input mate.

i'm not wasting time - i'm enjoying myself



-- There are no answers to find in the bottom of a glass. But you can have fun looking for them.

 

[unoc]

In my last post I mentioned 166 MHz instead of 160. Sorry for the mistake but I was thinking about the new chipset from VIA tech which is able to sustain only 6 MHz more then yours but with alot of difficulties to solve and some not yet solved. I believe that the big bottleneck for your job is the chipset also before the PCI bus. If your RAM memory is able to trust with a CAS2, you have good chip installed. The increase of FSB does not influence more then a bit the possibility to maintain a CAS latency 2 instead of 2.5

 

[monkeyspank]

do you think if i tried to cool my chipset better it might go a little further, or do you think that it has simply reached its max?

btw, I edited my last post. my maths was bad, i meant to put 160*9.5=1520.

i think i'm gonna give 166 motherboards and memory a miss and wait til the next spec is available. i heard some rumours about ddr2 - but i dont know how far off it is, or if it will be any good.

-- There are no answers to find in the bottom of a glass. But you can have fun looking for them.

 

[monkeyspank]

Fatburger, are you talking about the divider?

I can change the pci latency. What kind of effect would that have?

-- There are no answers to find in the bottom of a glass. But you can have fun looking for them.

 

[unoc]

You should increase also the Voltage supplyed to the chipset to reduce the electric noise, but it is not possible.
Because the electric noise is originated by the temperature you can try to cool the chipset but you should reach a very low temperature. Also the RAM should be cooled as every component which is involved in the general frequency increase.
You also should try to isolate the failing component by limiting the overclocking to one component a time. Does your motherboard allow that ?
PCI should works at a frequency not far from 33.33 MHz but probably you are using for it a frequency of about 40 MHz that is really too high for the PCI controller (32 bit 33 MHz).
THG made some tests overclocking an Athlon. In such test is clear that if you want to reach the limit for the CPU you should be able to change only the multiplier and not the FSB. At least not too much.
The next step go in two different direction: a dual mobo or 400 MHz KT??? motherboard.
DDR2 should be able of a transfer rate of 3.2÷4.8 GBytes per second (DDR 133 is "only" 2.1 GBytes/s) depending on the frequency (up to 600 MHz or 300MHz DDR)

 

[FatBurger]

The PCI divider, not the latency.

<font color=blue>If you don't buy Windows, then the terrorists have already won!</font color=blue> - Microsoft