Hoody

same here

I guess I missed the question. :cool: so if you are running 260x10 and have a RAM divider of 5/6, then your RAM speed is 215MHz...right? what divider are you currently using?

so I read the beginning of that thread, and what I got out of it was that the divider doesn't matter when it comes to performace measurements..what I still don't understand is how come my benchmarks got worse? Angry's test proved that divider/no divider the tests were similar, mine were not. I'm obviously missing something.

The marketting people have you brainwashed!

So maybe some can explain this to me. As stated above running my RAM with no divider 250x10 I get lower benchmarks then stock. I add a divider to the memory and keep CPU at 250, the benchmarks get better...Here are my results with 3DMark03: stock - 14,824 250x10 (no divider - 3-3-3-8) - 13,397 250x10 - 5/6 divider (208MHz 2.5-3-2-5) - 15,461 250x10 - 9/10 divider (227MHz 2.5-3-2-5) - 15,518 Do the timings make that much of a difference? it seems tighter timings are more important than bus speed to the memory; or at least it appears that way when running 3DMark03 as a benchmark. question #2: Is the 5% gain I am seeing (in this benchmark at least) going from 200MHz to 250Mhz normal? [caveat: I have yet to run other benches...]

Thanks!!! I have A02 ;-)

So mine is reporting 18-23 idle and 30 max load as well. my question: The BIOS is set to only turn on my HFS if > 25c if the sensors are in fact incorrect then my CPU is running hotter with no fan. Should I force HSF on 100% of the time since I can't trust the temp sensors?

[n00b question alert!] Where do I get the Rev for my Expert? Is it the one under the 'mainboard' tab in CPUZ? If so, then mine is A3

I rebooted and re-ran the tests. still low. All the settings are the same as they were BEFORE the OC. I want to try and keep the test prior and post OC the same so I have accurate tests...If I tweak the settings in the test than I won't get an accurate reading of my perfomance gain (or loss ;-) ).

I ran Prime95 for 10 hours last night...I was re-running briefly due to a small timing change I made to the memory (benches were the same before/after that change I just mentioned). no, I did not boot piror to running the bench, in fact it has been up 24 hours since I ran Prime95 last night. no divider on the mem...but I have my 2-3-2-5 memory running at 3-3-3-8

I am running stable at 250x10. I ran the benchmarks and I am comming in a lot lower than I should...In fact BEFORE OC I was getting 14,824 in 3DMark03. Now I get 12,375. My aquamark and memory results also seem lower compared to some other's results in teh OC database with similar hardware and overclocks. OC results 3DMark03 Any ideas?

That was exactly the problem. I moved the memory to the orange slots (from the yellow) and now I can hit 250fsb with 1T...what a difference that made!!!!

I could not get my RAM to successfuly pass Memtest past 220 with 1T in yellows. I read this and moved my RAM to orange this morning and I noticed an instant difference...the same settings that weren't working in yellow are now working in orange!!!! game on

I have my 3800+ at 250x10 with 1.4vcore. I have tested higher, but I want to keep the temps down as I really don't want to listen to my HSF humming ALL the time Yesterday I started trying to OC the RAM (see above, and my other thread) and it's actually been 100x harder than the CPU. I hoping this information about the orange slots helps me out a little.

fargin' A...that prolly explains my OC instability!!! I couldn't get crap outa my RAM unless I went to 2T. As soon as I hit 220 I got Memtest errors..cranked up the VDIMM and went 3-3-3-8 1T and nothing...changed to 2T and BAM, easily hit 250. I'm going to switch it over to orange and re-test! thank you so much!