Overclock.net › Forums › Intel › Intel CPUs › Overclocking i5 3570K With Only the CPU Multiplier Questions?
New Posts  All Forums:Forum Nav:

Overclocking i5 3570K With Only the CPU Multiplier Questions? - Page 6

post #51 of 72
Quote:
Originally Posted by Ophaq View Post

When IBT finishes a pass, the CPU usage drops to say, around 50% and climbs to 80% for a second before going onto the next pass which loads it at 100%. Is this normal?

Yes, the CPU load drop is normal when it starts a new pass. 96C is starting to get pretty hot. Seems a little too high for ~1.3V and a H80.
post #52 of 72
Thread Starter 
Quote:
Originally Posted by Forceman View Post

Yes, the CPU load drop is normal when it starts a new pass. 96C is starting to get pretty hot. Seems a little too high for ~1.3V and a H80.
It's actually running at 1.284V or 1.296V. And it's not a constant 96, it's more like going from 89 to 94, 88, then 95, and temperatures in between. I'll post a pic of my desktop during the 10th run (I'm on the 8th at the moment).

Also, the air venting out of the fan of the H80 (I think it's the fan of the H80) feels cool and not around 190F. XD

EDIT: Here is the pic (It may be hard to read). http://i551.photobucket.com/albums/ii479/Ip0wnn00bs/IBTResults_zps21b828f5.png


Now I'm going to run Prime95 blend for 12 hours. Does Prime95 automatically detect errors or is that an option that I have to manually set?

EDIT 2:
Prime95 has been running for a couple minutes now. The average temperature is 67-71C but I've seen my cores go up to 78C maximum for a short period, which seems a little high. Should I lower TB's clock speed down to 4.3 or 4.2? I don't really want to mess with the voltages.

EDIT 3:
10-15 minutes in and temperatures are 80-86C which is pretty hot. I'm going to try all the cores at 4.3 and re-run IBT.
Edited by Ophaq - 2/16/13 at 11:42am
post #53 of 72
Thread Starter 
Here are the results of running IBT for 10 passes at all cores at 4.3Ghz:
http://i551.photobucket.com/albums/ii479/Ip0wnn00bs/IBTResults2_zpsf9d6307a.png

I'd say that the temperatures are not bad at the voltage it runs at (it ran at around I think 1.17 or 1.16V with maybe a 1 second drop to 1.07V every now and then during IBT). Now I'm going to test with Prime95 again. I'm really surprised at the difference between 4.4 and 4.3.

Any opinions on the temperatures at the voltages? I circled everything in red so your eyes will notice them faster. XD
Edited by Ophaq - 2/16/13 at 12:11pm
post #54 of 72
The temps are safe enough, given that normal ops will be 20C lower than IBT, but it seems like you should be getting better temps with that cooler. All chips are different, but 1.3V at 44 is not all high so I would expect temps in the 70s with that cooler, not 90s. Have you tried the usual suspects, reseating the cooler, making sure the pump is running full-speed, turning up the radiator fan speed, etc? Might be worth a reseat, if you haven't tried yet.
post #55 of 72
Thread Starter 
Quote:
Originally Posted by Forceman View Post

The temps are safe enough, given that normal ops will be 20C lower than IBT, but it seems like you should be getting better temps with that cooler. All chips are different, but 1.3V at 44 is not all high so I would expect temps in the 70s with that cooler, not 90s. Have you tried the usual suspects, reseating the cooler, making sure the pump is running full-speed, turning up the radiator fan speed, etc? Might be worth a reseat, if you haven't tried yet.
Hm... Well, Ivy does heat up a lot with voltage changes. All I know is that the cooling system is running, and running fine as the idle temps are that of the average temperatures most people get (29-34C, sometime up to 40-43C when running Mozilla with multiple flash videos).

I haven't tried a reseat, nor do I want to because that would involve a lot of work and I don't know how to. I've been running Prime95 for around 18 minutes now and the temperatures are at 53C minimum to 67C maximum. To get a better estimate on the temperature at load during it the summer time, I'd probably add 1 or 2C to that. Also, the Vcore is currently running at 1.092V and the VTT is at 1.068V.

EDIT: Do you think that the temperatures are good at these volts? I'm an hour and 35 minutes in and the lowest i've seen so far is 52C and the highest is 67C on the second core.
Edited by Ophaq - 2/16/13 at 1:45pm
post #56 of 72
Thread Starter 
1/4 of the way done and temperatures are still like said above. Is it really necessary to run Prime95 blend for 12 hours? Not that I'm complaining or anything because I want to make sure my system is stable. XD
post #57 of 72
It's only necessary if you think it is. That's become the standard for whatever reason, but stability testing is up to you. If you are willing to accept some risk then just test it as long as you are comfortable with, and then use it the way you normally would. If nothing happens then you are fine - if it crashes then you should have tested it more. You could check for WHEA errors (there's a thread here somewhere) and if you don't have any of those you are probably fine. They seem to be a way of telling if you are on the edge of stability - not enough to fail Prime but not 100% stable either.

My personal feeling is; I'm not running a nuclear reactor with his thing, so if it passes IBT and a few hours of Prime, I'm willing to roll with it.
post #58 of 72
Thread Starter 
Quote:
Originally Posted by Forceman View Post

It's only necessary if you think it is. That's become the standard for whatever reason, but stability testing is up to you. If you are willing to accept some risk then just test it as long as you are comfortable with, and then use it the way you normally would. If nothing happens then you are fine - if it crashes then you should have tested it more. You could check for WHEA errors (there's a thread here somewhere) and if you don't have any of those you are probably fine. They seem to be a way of telling if you are on the edge of stability - not enough to fail Prime but not 100% stable either.

My personal feeling is; I'm not running a nuclear reactor with his thing, so if it passes IBT and a few hours of Prime, I'm willing to roll with it.
OK, thank you. Your input is really helpful to me! I might only do 6 hours or so of Prime95 then as nothing has happened for the past 4 hours and 15 minutes it has been running.

On the temperature issue, what is your opinion on the temperatures I'm getting (minimum I have seen it 51C and maximum I have seen is 67 or 68C. However, they average at 56C to 65C) at the voltages? smile.gif
Edited by Ophaq - 2/16/13 at 4:30pm
post #59 of 72
Mid 60s is fine. And so you know, you will probably get different opinions on the stability testing. Some people are adamant that only 12 hours is stable, but since I'm not trying to compete for anything I take a more laissez faire attitude.

Are you using the 27.7 version of Prime?
post #60 of 72
Thread Starter 
Quote:
Originally Posted by Forceman View Post

Mid 60s is fine. And so you know, you will probably get different opinions on the stability testing. Some people are adamant that only 12 hours is stable, but since I'm not trying to compete for anything I take a more laissez faire attitude.

Are you using the 27.7 version of Prime?
27.9, actually. Is there something wrong with 27.7 or my current version? And I too, share the same attitude. However, since this computer is new I need to make sure it lives up to itself and not crap out on me.

Also, what if the temperatures were 70C? Would that be a bit on the warm side? I'm just concerned for the longevity of the CPU (although it's not like it's going to be 100% load all the time). XD
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Intel CPUs
Overclock.net › Forums › Intel › Intel CPUs › Overclocking i5 3570K With Only the CPU Multiplier Questions?