i5 2500k, BSOD under load!

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

Puppies04

Diamond Member
Apr 25, 2011
5,909
17
76
Let me clear a few things up that have been posted in this thread. First up, Impulse69, there is ALWAYS a reason for BSOD's ;) and generally the error code created by them will tell you what it is.

As far as EIST aka speed step, you are 100% safe leaving it ON. There is NO reason to turn it off unless you are pushing to the ABSOLUTE max *think 5.8ghz + on SB chips with tons of voltages* aka things that aren't done much on these forums <3.

With that being said the best way to stabilize SB chips is through vcore. If you run LinX with AVX pack your temps and stability will REALLY be tested. P95 is a decent app, but it doesn't even come remotely close to stressing a system out like LinX or IBT with AVX.

If you aren't sure what AVX is, it increases the GFLOPS and stress through the system. LinX/IBT are also mathmatically A LOT more stressful then P95 and you WILL see much much higher temps, roughly 10 degrees more then p95 temps.

If you run a stability program like LinX/IBT and you fail it, simply up the voltage until you become stable. Now for 24/7 use I wouldn't run more then 1.4 to 1.425 voltage if you want the chip to last forever. If you are going to upgrade to IB and you don't really care about the chip, by all means push the voltages as much as you want as long as temps are fine.

We really don't know the degradation point / limits of SB yet because they haven't been out long enough. I know plenty of real benchers who have ran tons of volts *think close to 1.7* through SB chips and they are still up and running just fine today. In SHORT bursts you can run a lot of voltage through SB chips and they will be completely fine *as long as temps are within reason*.

Anyway I am getting a bit off topic heh. Back to the OP Lorddac, a lot of SB chips need a ton more voltage to go from the sub 4.5 ghz range to over it. For example I know several of my chips / bench chips need around 1.35 for 4.5ghz, but for 4.7+ they need 1.4+ vcore. It really solely depends 100% on the chip.

Tweakboy - Running 1.5volts through a SB chip for 24/7 use is NOT a good idea no matter what kind of cooling you are running. For a quick benchmark sure, but for every day use, even with EIST that is a horrible plan ;).

Lorrdac - as long as your temps are within reason up the voltage to 1.4vcore or 1.425 and see what the highest stable overclock you can achieve is. You for sure will be able to hit 4.5ghz with ease at 1.4vcore LOAD *make sure you factor in vdrop, for example if you set 1.4vcore in bios and at load you are getting 1.37, then run 1.43 vcore in bios so at load its 1.4vcore*. You can probably hit anywhere from 4.7 to 4.8ghz with 1.4 to 1.425 vcore load from my experience with SB chips *which is pretty vast ;)*

Wish you the best of luck overclocking man and hopefully you have a great chip on your hands!

I agree, pick a voltage you are happy running at then find out how fast you can get your chip to run at that speed. Also OCing with offset mode instead of manual voltage (if available and with speedstep on) could prolong the life of the chip because it will enable you to put a lot less voltage through it at idle and it will clock itself down to 1.6ghz
 

ImpulsE69

Lifer
Jan 8, 2010
14,946
1,077
126
Yes, sorry bad choice of words. My reason appeared to be Virtu. J/S. :) Once I stopped using Virtu, I have had no issues. It was confusing because it worked for a few weeks no issues, and survived all the "tests" as well.

Lots of good advice there btw.
 
Last edited:

Puppies04

Diamond Member
Apr 25, 2011
5,909
17
76
Was there an actual issue with virtu or was it just the power overhead associated with the iGPU meaning that your system needed more voltage to stay stable?
 

ImpulsE69

Lifer
Jan 8, 2010
14,946
1,077
126
Well I only am oc'ing to 4.2Ghz. I tried upping the voltage up to 1.45v with same issues so I don't think it was due to voltage. I was stable prior to using Virtu and after I stopped using Virtu. I do not know if it's software related or hardware related. I do know that (on my board) OC'ing the chip also OC's the igpu. I couldn't figure out how to prevent this.
 

fastamdman

Golden Member
Nov 18, 2011
1,335
70
91
ImpulsE69 - what is your board, what problem are you having with virtu and are you overclocking via mutliplier or via BCLK?

Either way, 1.45vcore is WAY to much for 4.2ghz lol.

Generally the issues related with Virtu are software related and it can be a pain to use at times but once everything is figured out all is good. Let me know your issues, your os and your hardware and I will fix it for you.
 

ImpulsE69

Lifer
Jan 8, 2010
14,946
1,077
126
I'm not having issues, I'm happy where I'm at. I'm not at 1.4v I only ran it up there to troubleshoot. I was simply pointing out that something with Virtu seemed to be a issue I was having that he could look at if he's using it.

I had posted about my problems way back when I was having those issues. The board I'm using is the ASRock Professional Gen 3 (Fatality board)