Problems OCing a 2500+ Barton to 3200+

N

nylint

Gawd
Joined
Oct 8, 2004
Messages
551
Well, I see the majority of people get 3200+ speeds out of their Barton 2500+...and I'm a little sad because I can only manage about 2.10ghz before my 2500+/A7N8X Deluxe 2.0 combo craps out on me. I've tried vcore bumping to 1.7v, and I know I'm not memory limited because I can run at 10.5x200 CL2 6-3-3 with my two sticks of GEiL ULL PC3200 all day. I have good cooling (CNPS3100+ with a 120mm antec fan, hsf base needed no lapping and it's got a nice thin coat of AS5 on it)

Any other things I should try? Or have I just not seen the posts from the thousands of people whose 2500+'s aren't OC demons? :(
 
ah, you probably have a locked multi don't you?
try lowering the ram multi down (just for testing purposes) and seeing what your max cpu o/c is. it seems strange that you have to raise the voltage at that speed..
 
No locked multi. The default multi of the 2500+ is 11.
Being as a lot of people are getting 11x200 at default vcore, isn't 1.7 a little much?
What's the most I should up it to?
 
2.0v ?? it really depends.. i went up to 1.85 on my cpu.. and decided that i felt more comfortable with 1.75ish.. yours starts out stock at 1.65 though.. so you should have to raise it more to oc more..

anyways... what are your temps? don't go over 60 degres celcius.. try to keep it under 40.. the higher the voltage is, the more heat the cpu puts out.. so people running water cooling and phase change and stuff can use way over 2.0v because they can cool the extra heat..

anyways.. try 1.8 and 11 x 200.... it should boot for sure.. then drop it to 1.775, then 1.75, 1.725.. see how low it can go and be stable...

then after that, see how high your ram can go... i would rather have a computer running 210 x 10 than 200 x 11.. the extra memory bandwidth makes a difference... basically just play with everything and find its limits.. test the cpu max ghz with the memory doing something it can easily handle.. then drop the cpu speed down to 1.6ghz or something and see how high the ram goes.. so you know one isn't affecting the other...
 
Wow thats really odd because out of the box I was able to go to FSB 200 and overclock to 3200+ with stock voltage (1.65) the Multi is set at 11. I have really crappy ram so posting my ram timings wouldn't help at all. I can overclock to a 3400+ with a voltage of 1.75 and multi set at 11.5. So it might seem that the problem is your mobo? :confused:
 
I'm thinking the problem MIGHT be my mobo. I forget the particular stepping code this core has, but I remember looking it up on a guide somewhere and finding that people generally had good results.

On my Raidmax PSU, the +5v rail is a little weak, but 3.3v and 12v look good under load. CPU stability is dependent on the 12v rail, isn't it?
 
nylint said:
I'm thinking the problem MIGHT be my mobo. I forget the particular stepping code this core has, but I remember looking it up on a guide somewhere and finding that people generally had good results.

On my Raidmax PSU, the +5v rail is a little weak, but 3.3v and 12v look good under load. CPU stability is dependent on the 12v rail, isn't it?
Click to expand...
Socket A platform chews on only the finest of 5V rails. :eek:
This makes for a decent overclocking limit, since not many PSUs have a beefy enough 5V rail. As soon as I put mine to 2405MHz and 1.85V, the 35A rating on my 5V line is overused. The PSU gets damn hawt when that happens. If I have enough spare cash kicking around, I'll grab a Zalman 400B PSU (Fortron 530 in disguise). Something tells me that my oc'd 9800Pro and HL2 will justify going over 3200+, cause Doom3 made no difference.

My week 30 Barton went to 3200+ at stock voltages while only gaining a few degrees C. Going to 2405MHz requires max voltage on water, and my cheapo RAM is limiting me on that.
 
come to think of it, i've wanted a new PSU. dunno what i want but i know i want to move it into a PC-V1000 someday when I scrape up the cash.

so a PSU with a strong 5v rating it is.
 
I replaced my 430w Enermax PSU with a 475w Enermax. Big difference was the dual 12v rails instead of just one. With one for the mainboard/cards, and one for the drives, my stability has increased dramatically.
 
i had the same setup with a locked multi and was able to get 200x11 no problem. i know that on the asus board has a FSB jumper that you can set to 200fsb or 333fsb. make sure that you have it set to 333fsb.
 
jackbauer said:
i had the same setup with a locked multi and was able to get 200x11 no problem. i know that on the asus board has a FSB jumper that you can set to 200fsb or 333fsb. make sure that you have it set to 333fsb.
Click to expand...
...huh?

Trust me, if it was a jumper issue or something easy, I'd have found it by now. I've been doing this for awhile.

I already said that the chip runs fine at 10.5x200 and 11x166, stock....so why would a FSB jumper have anything to do with the problem I'm having?
 
it was just a idea jackass!!!! dont have to be a dick. that is the last time i try to help, noobie
 
No, I was asking you what a jumper has to do with MY problem.

If it was just a 100/133/166 fsb jumper issue, I wouldn't have been asking the question I asked. You didn't read my first post.

AND you resorted to name calling. Who's the noobie?
 
Jack, heed Arnolds advice.

stfu.jpg
 
nylint:

Ever hear the saying, ymmv? Meaning Your Mileage May Very.... Basically, not all chips will act the same. I had a Barton 2500+ that was not a stable overclock at 3200+ 400fsb 11x200 multipler. It would crash out games even though it was cool enough. I had to set it at 10x2000 for it to be stable and not crash anything.

That's when I went to the mobile chips and haven't had a problem since.
 
nylint:

1.7V ain't gonna cut it for many chips. My 2500+'s all require at least 1.85V to be stable at 2.2GHz. The power supply issue may be the problem, but up the voltage and see what happens.

Jackbauer,

Arnold has it right. Shut it before the mods hear you breaking wind.
 
ShepsCrook said:
nylint:

Ever hear the saying, ymmv? Meaning Your Mileage May Very.... Basically, not all chips will act the same. I had a Barton 2500+ that was not a stable overclock at 3200+ 400fsb 11x200 multipler. It would crash out games even though it was cool enough. I had to set it at 10x2000 for it to be stable and not crash anything.

That's when I went to the mobile chips and haven't had a problem since.
Click to expand...
Yeah, I'm aware of YMMV, but I was hoping it wasn't one of those things :)
Thanks for the input, though. Like I said, I never seem to see posts where people say 'OMGWTFBBQ, MY BARTON 2500+ ISN'T STABLE AT 2 BILLION MEGAHURTS'.

mwarps said:
nylint:

1.7V ain't gonna cut it for many chips. My 2500+'s all require at least 1.85V to be stable at 2.2GHz. The power supply issue may be the problem, but up the voltage and see what happens.
Click to expand...
1.85v, eh? I heard these chips are okay at 2v even. I'll up the vcore and replace the PSU.

Thanks for everyone's input. :D
 
nylint said:
1.85v, eh? I heard these chips are okay at 2v even. I'll up the vcore and replace the PSU.
Click to expand...
getting a new 2600+ mobile would probably be cheaper and more productive than getting a new psu :D
 
that's damn true, except this raidmax 350w has been through 3 cases so far, has a suck-ass 5v rail and is getting a little long in the transformer. (tooth?)

that's not to say a 35w mobile isn't in my future. i'm waiting for nforce4/soundstorm2/939+PCI-E like everyone else.
 
You must log in or register to reply here.
Back
Top