Something in my setup ain't right...

All of my mates either have laptops or Intel boards unfortunately. My RAM is two sticks of Kingston 512MB PC3200U. And my PSU is the 460W one shown here.

I guess it's a small comfort that you're running out of ideas as am I - it means I'm probably not losing my marbles. But it is a very small comfort lol. Do you think I should contact MSI?
 
i've done some reading and one article in particular shows similar problems that you're facing... they couldn't even get the system to boot at 225 FSB, this also caused a corruption of their windows installation.
i dont know whether its to do with the design of the board but i've seen the same chipset on other boards do well with overclocking.
 
Could you post a link to the article Hairy Lee? Cheers.

I guess I could always pop round yours and try my CPU :P Just kiddin'
 
Haha didn't twig yours was a 754 board lol.

Cheers for the article. To be honest I think I'll contact MSI and see what they say. Perhaps I'll be able to take it back since it's within warranty and effectively faulty. I bought the board under advisement from Praetor, so I'm really surprised I've been having problems with it.
 
OK, update. I haven't contacted MSI over this, instead I contacted the technical department of scan.co.uk, I'm still waiting for them to get back to me.

BUT, I HAVE narrowed down what the problem is. To cut a long story short, the memory divider is not working as it should. If I set the divider to 133, even keeping a conservative FSB (~250) so as to maintain a RAM speed of under 200, boot faliure occured. But I went down even further to a divider of 100, and lo and behold, even though my memory is crawling along at the speed of a snail, I have successfully upped the FSB to 265 as I write this, and I'm sure it won't stop there.

The question is, any idea why when at stock speeds the memory is fine, at an underclock of FSB250/Div.133 it fails to boot, but at a bigger underclock of FSB265/Div.100 it boots perfectly???
 
OK, fair enough, I didn't know that.

In that case would I have a case if I said it was faulty for that reason, and go for a board who's dividers work?

I suppose I can play around with my CPU multiplier to get my memory back up to stock speeds...
 
On some boards certain dividers dont work, so that definetly could be a problem.
Are problems with memory dividers for individual boards only, or for certain board models?

The reason why I'm asking is that my BIOS keeps crashing (seperate problem) which hasn't helped by a BIOS Flashing, and if I can get a same-board-model replacement I wouldn't have to reinstall anything.
 
But even at 245Hz-250Hz, the system's very unstable, and sometimes fails to boot. I'm sure it's just because I'm missing something, but what is it???
Did you bump vCore?

no, you are just stupid and are trying to push your system too high- idiot
dog.jpg

ok, he expects to get 2.7 out of his system when hes probably not supporting right volatges, etc, how is this an 'immature post'
its the truth
1. 2.7Ghz isnt unattainable, hell, 3Ghz on air isnt absurd
2. If you looked into the product specs before making your post, you'd realize that motherboard kicks ass for overclocking ... which ... in case im being too subtle, means that numerous vCore options are available.
3. Not to be your english teacher or anything, but "he" wouldnt be supporting any voltages ... the CPU and MOBO would be ... which, well, if you read up on the parts, you'd realize, they do, in fact, support those voltages

Edit: I thought the AGP lock was automatic on this board? It's certainly set at 66, so wouldn't that do it?
The board should have AGP locked out. (also, did you update your BIOS? :))

Apparently using the first two SATA ports actually overrides any PCI/AGP lock that I put on. So even though the lock was saying active in the bios, it didn't really matter.
Use the other SATA connects, one pair isnt locked, the rest are :)

I thought setting the AGP to 66 automatically set the PCI to 33.
Correct

I can't find any option to raise the chipset voltage on my mobo
Since you'd lowered your LDT multiplier, you shouldnt have to worry about this (not to mention that board doesnt give you chipset voltage options)

The question is, any idea why when at stock speeds the memory is fine, at an underclock of FSB250/Div.133 it fails to boot, but at a bigger underclock of FSB265/Div.100 it boots perfectly???
When OCing the CPU, the CPU can be "wierd" and since the memory controller is on-die, then its possible that OCing it will produce "wierd results"
 
Thanks for the comprehensive post, Praetor. Rofl with the picture, too.

I successfully overclocked to 2.61Ghz on a 100Mhz divider. However ( a seperate problem by the way), the BIOS has kept crashing and causing POST faliures, even with reset CMOS jumpers, and I'm getting a refund organised from the vendor. People will say that's down to me, but all I can say is I never pushed the hardware that hard.
When OCing the CPU, the CPU can be "wierd" and since the memory controller is on-die, then its possible that OCing it will produce "wierd results"
Does that mean that even in getting a replacement mobo the 133 divider will not function?
 
dog.jpg


HAHAHA owned, I just helped this guy too i gave him my AIM name and he IMed me with the exact same questions your asking so don't let him get you down.
 
HAHAHA owned, I just helped this guy too i gave him my AIM name and he IMed me with the exact same questions your asking so don't let him get you down.
No worries on that score, but I'm still a bit confused as to what extent the CPU on-die memory controller can affect a memory divider when overclocking, as when my replacement motherboard arrives I will be wanting to use one...
 
but I'm still a bit confused as to what extent the CPU on-die memory controller can affect a memory divider when overclocking,
It allows you to somewhat control the memory clock while you increse the CPU clock (call it the poor-man's unlocked multiplier :))
 
Back
Top