2500+ OCing Results

(edit: adding updated results to the first post)

I've started working with this rig now that it's been running a few days breaking in a little. Here are the specs of what I'm running before I start with the early results and initial problems I'm having:

NF7-S rev 2.0
Barton Athlon XP 2500+ AXDA2500KV4D AQZFA0402VPMW
SLK-900U w/ 92mm Evercool @ 2200 RPM
512meg Corsair XMS PC3200 LL DDR CL2 w/ Platinum Heat Spreader
Enermax 460W Whisper Dual Fan
Ati Radeon 9200 128meg
2X 80mm Panaflo Ultra Queit Med Flow Intake
(not sure on the rpm on these, my enermax fan control unit isn't giving an rpm read though fans are turned to max)
2X 80mm Antec (generic?) Linked Exhaust
40GB Maxtor Diamond Max +8 7200rpm
(note: chose this hard drive for its very low profile being a single platter and the fact that under load it only puts out 12C temp, nice since I have it mounted low behind the top intake fan)

Memory timming I left at default 7-4-3-3.0

Now for the early results:

Defualt:
11X166mhz = 1826mhz @ 1.65v mem @ 364mhz idle=22/32 load=24/37

11X173mhz = 1903mhz @ 1.65v mem @ 414mhz idle=22/33 load=24/38

11X182mhz = 2002mhz @ 1.65v mem @ ???mhz (no post)

11X182mhz = 2002 mhz @ 1.675v mem @ ???mhz (no post)

11X182mhz = 2002 mhz @ 1.75v mem @ ???mhz (no post)

11X182mhz = 2002 mhz @ 1.65v mem @ 364mhz idle=22/34 load=25/39

11X191mhz = 2101 mhz @ 1.65v mem @ 382mhz (error)

11X191mhz = 2101 mhz @ 1.675v mem @ 382mhz idle=22/35 load=25/40

11X200mhz = 2200 mhz @ 1.675v mem @ 400mhz idle=23/35 load=25/40 (prime95 failure before 30minutes)

11X200mhz = 2200 mhz @ 1.7v mem @ 400mhz idle=23/35 load=26/42 (prime95 failure before 1hour)

11X200mhz = 2200 mhz @ 1.725v mem @ 400mhz idle=23/35 load=27/43

11X209mhz = 2299 mhz @ 1.75v mem @ 409mhz idle=25/38 load=(crashes before windows finishes loading)

There seems to be a problem with the memory so far. Basically if I set the memory to auto there isn't a problem though it does run it at 364 instead of 333 at default. Setting the memory to run at speed it does do 207 with out a voltage change. However once I pass that it fails to post. I'm going to keep working on it and see how far it goes. So far the temps I'm getting have me hoping that this chip will do very well.

Corsair loves voltage

How much Vdimm have you put to it ?

Thunda

I've left the ram voltage at the defualt 2.6v I'm not sure if I'll actually need to push it over 200 yet though. This chip might end up becomming a furnace before I hit the 2200mhz mark. Still my aim is the 2500mhz mark keeping it under 50C load.

Looks like I'm stopped for a bit while I figure out what's holding me back. I've checked the rails and the PSU is holding up fine however I think that my ram may want some more voltage before it's willing to take the leap to 209 though it did work well at 207.

Only thing holding me back on that is the idea of running the ram at the next voltage step available in the bios which is 2.9v since it was at 2.6v default. I would much rather prefer being able to change it to 2.7v first and be able to move it up in 0.1v increments. I'm certain that this is not the limit of the CPU as it's still not really close to being hot and I haven't really even boosted the cpu fan yet (it's only up to 2500rpm).

Any thoughts on this would be welcome. I'm stopping at 2200mhz until I find out a bit more about my ram limits.

Interesting stuff. I just went through this with my 2500+ (locked)... I'm running an MSI K7N2 Delta-L mobo and 1GB GeIL Golden Dragon Value PC3200 in dual channel.

I had problems with posting and such until I set the tRAS to 11. I read another thread on here in the memory forum stating that there's an anomaly with nForce2 chipsets that they perform better with tRAS at 11. After mucho tweaking and playing around, I was able to run at 400 fsb with 2-3-3-11 timings, but Prime kept failing. (Prime would fail at 333 fsb with those timings as well). I finally ended up at 2.5-4-4-11, and Prime has been running for nearly 24 hours now without an error. Vdimm is at 2.7.

Here's my same Sandra screen print:

dag yo... you are runnin 2.2 at 1.85vcore? ouch

i figured most if not all 2500+ should run 2.2 at stock voltage....

i am runnin mine at 2.32ghz at 1.675vcore with the ilsr mobo

Yeah, it didn't seem to want to stabilize at anything less. I'm still playing with it, but it's stable, and my temp at full load hovers around 47C. (I'm on a Prime95 team, so my machine runs it constantly. I'm never at idle.)

Well I did some more testing with it today and it's not looking overly promising now. To get into windows and prevent an un-expected reboot @ 2299mhz I had to do several things.

First I had to up the memory voltage a hair to 2.7v this got me a bit further into loading at first but still caused intermitent resets.

Next I raised the CPU core voltage to 1.775v and this stopped the resets. When I had made the step to 1.75v from 1.725v windows would load and I could start programs but shortly after I would end up with a reset.

After I had a reliable windows boot I checked temps at 22/35 idle (the house was about 4C cooler than last attempts). SiSoft Sandra offered a nice benchmark stomping a 3200+ mark into the ground which I thought was cool. However Prime95 fails after about 1-2 minutes. MBM5 showed the temp only reaching 24/40 before prime failed however I'd expect it could climb much higher given time. Possibly to the 26/44 range though that's only a guess. I looked at the average overclock for the 2500+ and looks like about half of them needed 1.8v to reach a stable 2.3ghz though I didn't see any listed there with this newer stepping yet.

Anyways if anyone has thoughts on this so far and especially how much farther I should try pushing this your comments would be most welcome.

I'm telling ya your memory timings will affect Prime stability. My Prime got stable when I slowed my memory down and changed nothing else. I would have upped the voltage more on the memory, but my mobo maxes out at 2.7 Vdimm.