Multiplier wackiness on my Duron 600 and MSI K7T Turbo

[Birdie Lou]

Hi Guys,

What follows is a 4 part (till 3:00am each morning) saga regarding 133fsb and my Duron on the MSI K7T Turbo RAID. I have posted to other spots on the web as well.
I successfully got to 136 x 7.5 earlier on the first day by powering off right after saving multiplier changes in the bios, only to lose that ability the next morning on my first bootup.

Here is the latest attempt at a solution....it works OK.
I had been running back at the 100 FSB speed after my failures to get back to 133fsb mentioned above.

I decided to try and get back to a bootable 7.5 multiplier (whether it be at 100 or 133) I just wanted to see if I could ever get machine to hold at 7.5 again.

I kept bus at 100 fsb and set Duron600 for 9x. I hit F10 to save and rebooted OK..no power off and bleed this time. I did NOT hit F10 and power off immediately as describe in my earlier post. I just did F10 and saw if PC got video sync or not. That is same for what is described below too.

Next I went for 8.5 and saved. Got no video sync after save reboot, so powered off and bled PC by depressing power button with Power strip off. This time machine DID come back at 8.5 x 100.

Next I went down to 8x then finally the desired 7.5x multiplier following exact process mentioned above about powering off and bleeding.......woo hoo I got machine to talk to 7.5 again.

Final step was to change jumper to 133fsb and see if machine came up.....YES success, and once again I am writing this from a PC actually running 7.5 x 136 fsb = 1020MHhz.

I only pray that tomorrow when I turn the PC on again it will come up at 1020 and not fail as it did earlier today.

I don't know if the changing back to 100 fsb and then starting to go down from 9x to 7.5 is really necessary.......If I have to do this again I will try leaving the fsb at 133 and coming down from 9x to 7.5 (if my Duron can handle 9 x133 just to get the wierd process going on way down to 7.5

I would hope that something here can be logically explained and fixed in future bios release. I read something about Hi and Low states of the Durons and maybe my strange procedure somehow switches those states and lets me re-attain the 7.5 multiplier....I'm not sure.

Thx,
Jon

UPDATE:
Have had couple of days of consistent boots at fsb 136 x 7.5 except once when Windows had to restart my machine and then I got "no video sync". I did the Power off and bleed thing. Next boot PC came up as 816 (default multiplier again). I then did another power off bleed and shutdown and then came up correctly at 136fsb X 7.5 = 1020.

So it's working, just not as solid as I'd like.
Thx,
Jon

 

[ken257]

Check your L1 bridges, if the pencil job is not just so you will get that kind of erratic behavior. I used a defroster repair kit on mine because the pencil wore off on me once and I said never again. Also I used a 14x glass to check my work cause those things are small.

 

[OpenFriday]

what pencil did you use?

 

[Birdie Lou]

Thanks for the replies.

My concern is that this chip has always worked perfect in an unlocked state at 10 x 101 = 1010 Mhz. Prime 95 for days on end and glitch free boots.

Wouldn't the issue show up there as well as when I'm just trying to go down to a 7.5 multiplier.

 

[ExtremeOC]

i've heard of this...
try getting the newest BIOS....
i'll tell ya if this happens to me on my K7T turbo when i get it soon...
i'll have a duron 700 or 750 w/ FOP32-1...
then we'll see.....

 

[FuxOC]

I'm really disappointed in this board. I was running a K7T-Pro2-A. I also had a conductive pencil unlocked Duron 800mhz. On the old Pro2-A i managed to squeeze 1007mhz out of it, via 106fsb, 9.5x. I got an opportunity to sell the board, so I did, and purchased a brand new K7T-Turbo Raid. I figured I'd be able to squeeze a few more mhz out of the fsb-friendly 133A chipset. Boy was I wrong. The thing started acting up when I changed the multiplier to 8.5x, 100fsb for starters. Anything over the default multiplier refuses to post at plain old 100fsb. I get no video at all, and i have to reset the bios manually with the jumper. I did get 110fsb.. but without the multiplier what good is that? Do you guys think it would be wise to load the overclocker's bios and try to hit 133+fsb with a lower multiplier? This really sucks and I just recently learned of all the multiplier problems with this board & chipset.. but I don't want to have to buy yet another motherboard. Anything else I can try? I would re-pencil but those conductive pens are supposed to be permanent, and it was working perfectly in my old board. I already tried the 2.5bios from MSI, and the b16 overclocker's bios, but I went back to the old bios because it was saying 'unknown flash type' on bootup and I don't know if this is normal. Please help if you have any ideas.

 

[ophidia]

I was having issues like that with mine until I jacked up the vcore. Now it's fine.

There is a BIOS update out there for it now (version 2.5). It does not seem to address these problems, however.
Chris