Hello. I'm in a bit of a pinch here since i've noticed that my clocks go down to CnQ mode while running prime95 or any of the two OCCT cpu tests. It goes to these speeds every minute or so for a few seconds, then goes back up to regular speeds.
Thing is, I've disabled C1E and CnQ in bios, same with Microcode Updation and Secure Virtual Machine Mode. I also uninstalled the AMD CPU driver. I'm using Windows XP. I get 4414 CPU Score with the cpu OC'd to 3000. Please tell me what you think.
EDIT: This "throttling" of sorts doesn't seem to occur at stock speeds (2500Mhz). Perhaps it's temperature-related, though i have nothing like that in my BIOS. But at 3000Mhz, if i turn the fan up to maximum (using the Zalman FanMate2) the clocks NEVER go down, regardless of temp.
EDIT2: With the CPU @ 3000Mhz, Fan @ 100%, the CPU goes to CnQ speeds (1250 Mhz) when it hits 57C. This puzzles me to no end. I also added voltage readings from the short OCCT test I did @3000Mhz + [email protected]% .
ok, after seeing that chart dude, your vcore looks nuts. what voltage are you trying to run it at?
also, what does the 12v line look like during that test?
and what options do you have when it comes to fan and temperature detection in your bios?
you know what, I should just go on ensue and yell at you haha
(though, you have 41 amps on the 12v rail, which is MASSIVE, so the 12v should look pretty dead flat)
weird, according to your PSU rating, it's a 550 watt amp, but add up all the wattage maximums and it's 650 watts. maybe it's under rated so they can get that 80+ certification
I'm running Vcore at 1.32(standard) +100mV. That's 1.42V.
The 12v line during the test looks like in that pic, 12.48. But a quick look at HWMonitor now and it reports min as 12.42 and max as 12.67. The lower +12v voltage may be because I plugged in another HDD yesterday.
I only have fan and temperature detection in the hardware monitoring section. CPU temp, MB temp, CPU fan, PWR fan, CASE fan, and voltage readings. And then there's Q-Fan somewhere, which i disabled, because I control the fan through FanMate2. I tried disabling CPU temp, MB temp and CPU fan detection, but that doesn't help much.
As for that 57C reading, it usually goes no higher than 53-54 before the CnQ voltage and multiplier (6.25x) kick in. It only got that high because I turned up the CPU fan to max and this apparently allowed the CPU to burn more before dropping to those weird speeds.
With the CPU at 3100 MHz i need to turn up the voltage another +50mV up to a total of 1.47 (BIOS has no option to increase it higher than +150mV total) and i get CPU Score: 4523 in 3DM06. I can't go to 3200 either because 3DM06 crashes. So the extra 100Mhz is not worth it for the extra juice i'm injecting into the Vcore, and as such keep it at 3000MHz now.
No, it usually vDroops from 1.42 to 1.41-1.39 and holds like that for a few minutes, then vCore drops to 1.17 and the multiplier to 6.25x for a few seconds, after which it goes back up, and the process is repeated ad infinitum. Check the second attachment of the first post and you'll see.
not including those droops to 1.17, it shouldn't ever vary that much. you might have some droop when you first load up the CPU (mine does it) but after that, it should be a totally dead line the entire time. having your voltage jump like that is VERY bad.
I think it jumps like that mainly because of the CnQ speeds. I'm blaming my cheap mobo but it may very well be something else. I'm going to install W7 on an old 120GB HDD I have and see if it performs differently.
EDIT: Results. CPU @ 2800 MHz but the problem still persists, and the behaviour @ 2800 is just like in XP.
EDIT 2: Ok, so I removed my GTX260 216 to see if the PSU was to blame (switching to the mGPU), and it isn't. Weird voltages persisted. That being said, I will buy a new mobo in the near future. Thanks for your help.
Sorry for the double post but this is news to me! I found this someplace else, but I think everybody here could also benefit from this info.
"Congratulations, you guys have been ripped off by Asus the same as I was. While I was running the M3A78 EMH HDMI board, the same thing applies to your board. It's thermal throttling!!! Asus used cheap VRM's in these boards and the board is designed to protect the circuits at temperatures, usually around 42C, which is absurdly low. Check it out for yourself. Open up the Asus Cool N Quiet utility, but keep CNQ turned off in bios and also open the Asus PC Probe utility. Run Prime 95 and as soon as the CPU temp gets near 42C or 43C, you'll see the voltage shown on the Cool N Quiet utility drop in half. Uh, the CPU can run up to the high 50's without harm but you'll never get close to that on these boards. Try all you want, but there are only two solutions. One is open the side of your case and use a regular house fan and have it blow onto the motherboard or junk the board and go with another manufacturer. I chose the latter and went with an MSI board that doesn't do this because it was designed to run Phenom processors while these particular Asus boards weren't despite their advertising. These boards should have been recalled over this, but that will never happen so I'll never buy anything Asus again. Check the Asus forums on my old board and yours. The complaints about this are in there. Asus even has a mention of this in their FAQ's. Look for thermal throttling."
This is an older thread, you may not receive a response, and could
be reviving an old thread. Please consider creating a new thread.
Overclock.net
A forum community dedicated to overclocking enthusiasts and testing the limits of computing. Come join the discussion about computing, builds, collections, displays, models, styles, scales, specifications, reviews, accessories, classifieds, and more!