post you AMD dual core overclocking results, both the opterons and X2's

lots of info and pics would be nice, such as all of your hardware, voltage, multiplier, htt, etc

to make it easier for us you could post ur computer specs in your signature, but make sure that ur signature is not too big

Edited December 16, 200520 yr by ripken204

my 170 opteron is stable at 260x10 1.475v 1:1 3-4-3-8

ive been trying to get to 270x10 but i just cant, ive tried up to 1.58v and no luck. ill be trying to get it for a little while longer but i might be staying at 2600

Edited December 31, 200520 yr by ripken204

That's pretty bad Some cpus are like that, I had one Winchester CBBID week that could hardly do 2.25Ghz stable, and that's single core.

FX60 looks **** good, I keep an eye on thread posted by "guess" at XtremeSystems about FX60, he got 2 of them

How did he manage that? As far as I know, they're not released...

well alot of companies like to sell "test" products to people at XS so that they can promote the product. they will get ppl all excited about a dual core cpu going 3ghz on air with 1.55v. then when it finally is released, a bunch of ppl will want it right away b/c they know what it can do.

MSI K8N SLI Platinum / Athlon64 X2 4200+ Manchester.

[My full specs]

That page still contains info for the "safe" overclock I've been doing until today. It's now set to:

- 2320Mhz clock speed (290x8.0)

- 870Mhz HT speed (290x3)

The memory is set to 133Mhz in the BIOS now, which results in 193Mhz with the settings above.

The "NF4 voltage" is set to 1.55V. It was set to that before but I forgot to mention that on the page.

[Screenshot] (904x800; 104kb)

Edited December 28, 200520 yr by Tomcat76

run prime for more than 22min, more like 22hrs to see if its rly stable. 22min means nothing. also that many volts might be a little too much, just dont go any higher.

I know. I just don't have the time.

It would probably show errors after 15 hours or so, but that's fine; it doesn't take that long to render a video.

I'll leave it on overnight. That's the best I can do.

The memory voltage can probably go back down now. It was running at 208Mhz before.

I'll test the rest too.

CPU: 1.325V (slighly lowered)

MEM: 2.70V (under spec)

NF4: 1.55V (same)

Updated BIOS.

Ran Prime95 for 8 hours and 3 minutes.

By the way... Is there a program similar to Prime95? Most of the time, it doesn't run properly on systems set to "ACPI Multiprocessor PC". It starts out using 100% of CPU, but releases everything after about 5 to 10 seconds and uses 50% of memory instead. After several hours it's still on test 1 subtest 1. When you stop it, it displays the total time the test has been running and claims there were no errors. This problem does not occur when the system is set to "ACPI Uniprocessor PC" but that's hardly the point of an Athlon64 X2...

ahh, forgot to mention this to you

you need to make a copy of the prime folder that was installed, so that you have to folders. you then need to open up the 1st one and set tha affinity to 0, the open the 2nd and set the affinity to 1. this tells prime to only test 1 of the cores for one program, and the other core for the other program. check in task manager to make sure that they are both 100%.

you need to make a copy of the prime folder that was installed, so that you have to folders. you then need to open up the 1st one and set tha affinity to 0, the open the 2nd and set the affinity to 1. this tells prime to only test 1 of the cores for one program, and the other core for the other program.

I've been doing that ever since I got the X2. Both the folders are on the F: drive (FAT32) and are called "Prime95" and "Prime952" respectively. One checks CPU0 and the other CPU1 (Advanced > Affinity).

check in task manager to make sure that they are both 100%.

The processes of the Prime95 instances or the CPU usage history? The latter is as I described in my previous post. I'll check the processes tab now.

Both instances are at full load for roughly 10 seconds, then everything goes down and only the memory is being tested. If I then halt (for example) instance #2, instance #1 goes back to full load on that CPU core. It looks like one is interfering with the other.

Edited December 29, 200520 yr by Tomcat76

check the infinity in the task manager, set those too.

Thanks for the tip. They don't fully use the CPU simultaneously, but one of them does use 100% (aka "50%") of CPU now. This is how the program used to work for me in the past.

Edited December 29, 200520 yr by Tomcat76

9 hours and 11 minutes. It was a long night...