Barton that does not run 200 FSB

[stevejst]

I know many are faced with this problem that some Bartons cannot run 200 FSB and blame memory for that. I am experience the same and while the CPU easily runs 2.2 GHz itself on 13x170 for example whenever I go over 185 FSB I get memtest86 errors though it is not the memory since the same board, NF7-S v 1.1, runs fine with XP 2100 on 200 FSB. I have the newest BIOS update, whole system is tested stable at 200 FSB.
There is some HardOCP review of Barton 2500 that states this.
Any similar experiences? Is this related to a stepping or just a luck of core.

By the way mine is AQXDA 0321MPM.

 

[blade]

You try increasing the cpu voltage? Mine wouldn't go past 190 till I upped it to 1.7.

 

[pelikan]

Many people have the same problem with the Barton and NF7 rev. 1.x. You need to do the L12 mod to make your board see the cpu as a 266MHz proc instead of a 333.

 

[stevejst]

Many people have the same problem with the Barton and NF7 rev. 1.x. You need to do the L12 mod to make your board see the cpu as a 266MHz proc instead of a 333.

Thanks man, appreciate your input. I am just switching it to another board, I'll see if it works.

You try increasing the cpu voltage? Mine wouldn't go past 190 till I upped it to 1.7

That makes no difference, even upping to 1.75 gives me not 1 MHz more on FSB.
The kick is definitely something about chipset/cpu thing because XP2100 runs 200 FSB just fine, I can actually get 200 FSB on Barton too only I have memory errors, lockups, freezes and similar great things no matter what multiplier I set it to. So I was guessing it is CPU but of course I can switch mobo and see if it works. I'll try tomorrow with my Chaintech that supports 400 MHz CPUs.
This is just a note to all of you to avoid Abit NF7-S v1.0, v1.1, and v1.2. Abit should actually recall those boards, bastards.

 

[MatthewF01]

my friend, cut the L12.




about 1 week ago i was pulling my hair out not being able to exceed 185fsb without windows not loading, and i added voltage only to increase my temps but still not boot. messed with mem timings, went from 6-3-3-cas2 to 11-3-3-cas2 and windows actually LOADED at 200, then rebooted, then i kept bumping down 5mhz til i was back at 185 and windows would again load and be stable.

so i said what the hell and took an exacto knife to the chip.

http://www.sudhian.com/showdocs.cfm?aid=307

now im at 200fsb, 2230mhz and gonna try for more! (my 1800+ TbredA did 217*8.5 easy)

go for it my friend, it really works.

 

[pspada]

I had a chip that would not do 200 either. Then I changed to a different motherboard, and the chip did do 200 just fine.

 

[pelikan]

There are two issues with the L12 bridges on the NF7 and they only apply when the MB is not the cause of inability to reach a high FSB (usually over 200MHz on a rev. 1.x board and over 220MHz on a rev. 2 board):

1. A 333MHz processor on a NF7 rev. 1.x will often not reach high FSB. If the 3rd L12 bridge is cut, the mb will see it as a 266MHz processor and higher FSB is attainable.

2. A 266MHz processor on a NF7 rev. 2 will often not reach high FSB. If the 3rd L12 bridge is connected the mb will see it as a 333MHz processor and higher FSB is attainable.

 

[MatthewF01]

Originally posted by: pelikan
There are two issues with the L12 bridges on the NF7 and they only apply when the MB is not the cause of inability to reach a high FSB (usually over 200MHz on a rev. 1.x board and over 220MHz on a rev. 2 board

thats not neccessarily true, but most of the times that IS the case. I have seen many users such as myself who are having trouble just passing walls SUB-200fsb. I KNEW my system would go beyond 200fsb with my old chip. as soon as i swapped to the 2500+ i couldnt pass 185.

So my guess is that the bridge mod will work in any application where you KNOW the proc. is the wall (ie: memory is rated for 3200, 2700 and has done it before; motherboard can safely handle the FSB speed; PSU is willing; cooling is decent).


he might as well go for it.