Overclocking the Palomino

blacktankofhopelessness

Senior member
Feb 5, 2003
211
0
71
I'm going to try overclocking my 2000+ a bit later today. I'm getting a Sempron 3100+ soon so I thought I'd push my 2000+ to the max just to see how far it will go. I haven't done any extensive testing earlier, just pushed the FSB to 140 mhz and got it to 1,75 ghz a few months ago without any hitch.

Now, my problem is my mobo. It's an A7V8X and it lacks a PCI/AGP bus master freq lock. I read on another forum about another fellow who got his 2000+ (Palomino core) up to over 2 ghz (!!!) and he claimed that pushing your fsb up to 166 mhz will tell the mobo to lower the Bus Master divisor to 1/5 instead of the standard 1/4 on a board without a bus master lock, thus keeping bus frequencies at around 33 mhz. Is this true on my mobo? I couldn't find any setting in BIOS that lets you specify the bus master divisor manually...

Is it also true that data on my HD will start to corrupt at bus freq around 38mhz (FSB at around 150 mhz with divisor at 1/4)?

 

Yanagi

Golden Member
Jun 8, 2004
1,678
0
0
Hi

Forget overclocking the Palomino. They cant overclock at all. At least most of them cant. I guess you can get a max of 200 MHz out of it.


No PCI/AGP lock will cause HDD corruption at high frequencies yeah. But I doubt you'll get that far.
 

Tiamat

Lifer
Nov 25, 2003
14,068
5
71
I had a Palomino 1700+ and it could only go from 1470 to 1490 without crashing. It didnt overclock at all.

I have overclocked a 2000+ to 142fsb max. Im not sure if it was a Palomino or Tbred-A. It was not multiplier unlocked, nor a good overclocker as the tbred-b's are.
 

Concillian

Diamond Member
May 26, 2004
3,751
8
81
Originally posted by: blacktankofhopelessness
Is it also true that data on my HD will start to corrupt at bus freq around 38mhz (FSB at around 150 mhz with divisor at 1/4)?

Yes, it's true, I had it happen when I was running a Celeron at 75MHz FSB (37.5 PCI speed). In those days there were no speeds between 66MHz and 75MHz.

Since then I keep things locked down or very close to a correct PCI speed. Losing data SUCKS.


I had a 1500+ that went from 1333 to about 1580. I overclocked primarily with multipliers, it's not too hard to unlock a pally if you know what you're doing.

Scotch tape to cover the contacts, then fingernail polish in the burn mark ditches.
Then scotch tape to isolate each bridge individually and paint with conductive ink. Do each bridge individually and let dry 15 minutes or so. then you're good to go with multipliers.

Don't expect to get too far with a 2000+ though, that's pretty close to the top of the Pally line.
 

blacktankofhopelessness

Senior member
Feb 5, 2003
211
0
71
Oh well, I maxed out at 147 mhz FSB and since I couldn't be bothered to unlock the multiplier that's it for my little adventure. I tried raising the Vcore but nothing would get me up to 150 mhz without crashing at Windows startup.

1, 825 Ghz is where I'm at, and seeing as how the 2000+ is at 1,667 Ghz at stock, this translates to a 9,4% overclock... utter crap!
 

ZobarStyl

Senior member
Mar 3, 2004
657
0
0
My 1800+ Palomino was a good ol' chip, but couldn't O/C worth a damn. I got maybe 70 mhz out of it, and the performance gain was slim to none. Today she gets retired (read: given to my girlfriend) and replaced with a shiny new 90nm 3200+ with watercooling...but for all the failure of o/c'ing, I can't say anything bad about that chip...that system just aged in performance terms (see sig) but never crapped out...treated me well.
 

Concillian

Diamond Member
May 26, 2004
3,751
8
81
Originally posted by: blacktankofhopelessness
165 FSB with 2000+ Palomino

How in the hell did he manage that??? 165 mhz FSB would bring the 2000+ to 2075 mhz if he hasn't changed the multiplier (which he never specifies as having done).

In that thread, he's saying that the BIOS wouldn't let him go above 165 MHz. He was trying to get to 166MHz, but couldn't. He didn't actually RUN that speed. He was only able to run up to ~145-ish before Windows had problems loading.

There is really no way to know if your overclock is being limited by your PCI/AGP speeds or your processor if you don't unlock the processor. The fact that voltage doesn't change how far you get suggests that it's probably the PCI/AGP speeds.
 

blacktankofhopelessness

Senior member
Feb 5, 2003
211
0
71
There is really no way to know if your overclock is being limited by your PCI/AGP speeds or your processor if you don't unlock the processor. The fact that voltage doesn't change how far you get suggests that it's probably the PCI/AGP speeds.

Well, actually there is. Windows starts up fine at 146-147 mhz FSB but not higher than that. The PCI bus freq is specified in increments next to the manual FSB setting in the BIOS of my board and an FSB setting between 146-149 equals a 37 mhz PCI Bus speed. Setting the FSB to 150 mhz takes the PCI/AGP divisor to 1/5 thereby making the PCI Bus run at 30mhz between 150-153 mhz. So my CPU is definately maxed out at that multiplier.