Help to Overclock i5-2500k on MSI P67A-G45

IdiotCoder said:
Yeh... they are there (when Turbo is enabled), but it wouldn't let me change the ratios. No idea why because they weren't greyed out to suggest they weren't changeable.
Click to expand...

Did you try double clicking each individual value? Rather than the text on the left hand side? Never heard of anyone having a problem like this on this range of boards before! Though always a first I suppose.

IE. You need to set each value separately.
 
mikeo said:
Did you try double clicking each individual value? Rather than the text on the left hand side? Never heard of anyone having a problem like this on this range of boards before! Though always a first I suppose.

IE. You need to set each value separately.
Click to expand...

Yep, I double-clicked the numbers and nothing happened =/
I'll try again later, atm I just would like to know whether my current overclock would be sufficiently stable. I thought with voltage being on Auto it wouldn't be likely to fail because I assumed it would adjust voltage appropriately.

The only thing I can think of that could be changed is one of the other settings such as PLL.
 
I've found in these "new" BIOS's that sometimes a double click doesn't work. Try pointing at the setting you want to change and pressing ENTER.

I can't remember which setting it was (maybe these) but there was something that double clicking on did not seem to work.
 
Yeh, I noticed the Click BIOS to be useless sometimes.

By the way, what do you think of the 64minutes of prime stable until BSOD. Would you regard this as not being stable enough? I hear a lot of people saying they run Prime95 for just one hour to check stability, I ran mine for 4 minutes longer before it BSOD'd. I mean, the people who run it for an hour, what's to say there's wouldn't do the same if they'd have carried on for a little while longer?

Arrgghhh I don't know what to do ]=
I thought Auto Voltage would automagically make it fine. ]=
 
IdiotCoder said:
By the way, what do you think of the 64minutes of prime stable until BSOD. Would you regard this as not being stable enough? I hear a lot of people saying they run Prime95 for just one hour to check stability, I ran mine for 4 minutes longer before it BSOD'd. I mean, the people who run it for an hour, what's to say there's wouldn't do the same if they'd have carried on for a little while longer?
Click to expand...

Final stability checks are best run for 8 hours to be certain really, an hour still leaves time for error, as you found. Mine crashed after an hour when my volts were 0.015 too low, so you shouldn't have to push it up too far to make it stable, hopefully.
 
Tend to agree. Ultimate test (only my opinion from what I've read over the years).

- 10 runs of Intel burn test with memory set at very high or max (though max can take ages as it genuinely appears to use all free memory).
- Followed by 8 hours of Prime95 blend.

If it passes that. Then I think you would have to be VERY unlucky to have stability problems.

As to changing the turbo boost (as per my suggestion). Just checked in the BIOS for you. And it is a little different. Just click each value once, then enter the figure you want (EG. 42). IE. you don't use the + / - to adjust the value. You simply enter it direct. Remembering to save your settings of course.
 
@mikeo
When I put everything back to original settings, then changed just DRAM timings. THen changed the ratios you told me about it didn't seem to work.

When I ran prime95 and opened CPU-Z, it only went up to 3400MHz =/

Anyway, at the moment I've gone back to the settings I was on when I BSOD'd out of Prime95 after just over an hour, but this time round I set Voltage myself to 1.32V.

2hr 15mins into Prime95 so far temps @ 62-65 max, I opened my window to let some fresh air in and now all cores are running at 57-60C.

Reasoning for setting voltage to 1.32V is when on Auto it was at 1.312V at times, and that failed so I think 1.32V is the minimum which seems a bit high just for 4.4GHz?

If all goes well with these current settings, I could try setting it at 1.315V (next lowest) but even if this ends up being stable, is there much point in trying, its only 0.005V less.
 
IdiotCoder said:
imag0047xk.jpg
Click to expand...
C1E disabled - that's probably why your multiplier wasnt dropping on idle
 
ScottiB said:
C1E disabled - that's probably why your multiplier wasnt dropping on idle
Click to expand...

Nope... you would think so wouldn't you.

From what I understand. There is a certain amount of cross-over between C1E and EIST (which is on). Also I think the C-state has an effect as well. And as commented earlier in this thread, it's this that the BIOS has a habit of switching off if you OC the core ratio. Hence why the OC appears permanent (IE. multiplier doesn't go up/down with load). Though I'm the first to admit that I'm no expert.

But the BIOS on these boards sets C1E "off" even at default safe settings and the multiplier / voltage changes depending on load.

Mines running fine with 4.2 GHz turbo boost. Just into hour number 7 in prime95 blend. and left all settings as per default safe settings, except for upping the turbo boost to 4.2 GHz
 
Ok so, I now have my 2500k at 4.4GHz, I ran Prime95 for 3hrs 20mins. I know it's not the 8 hours you guys recommended but I'm happy enough with over 3 hours table in Prime95.

My vCore stays at 1.32V pretty much constantly, on Idle it only goes down to 1.312V. Is this safe to have it constantly on 1.32V? =/
 
IdiotCoder - Am having a similar problem getting the cpu to lower its multiplier; at what page of the link you provided did you find your answer?
 
IdiotCoder said:
Ok so, I now have my 2500k at 4.4GHz, I ran Prime95 for 3hrs 20mins. I know it's not the 8 hours you guys recommended but I'm happy enough with over 3 hours table in Prime95.

My vCore stays at 1.32V pretty much constantly, on Idle it only goes down to 1.312V. Is this safe to have it constantly on 1.32V? =/
Click to expand...

I'd also recommend a few hours of LinX (All Memory), I tend to use short bursts of Prime95 and LinX to see if it is stable. LinX first, 30 minutes then Prime95 blend 1 hour. If it passes both go for longer stretches. What I've found is certain overclocks fail on LinX quickly or Prime95 quickly, but they seem to pick up different issues. Hence why short bursts of one after another is a good indicator that the clock is stable. No point doing 8 hours of Prime95 if LinX makes it fail after 30 minutes.

I'd suggest being quite careful leaving things like vCore on Auto, you need to keep an eye on it at higher clocks as it tends to push things up to silly levels.

There is also no real need to change much more than Muliplier, Vcore and CPU I/O (to bring the memory speed up). Turbo, EIST, C1E can all be left to their default settings for regular overclocks (sub 4.8GHz) in my experience.
 
Sorry if I'm being a bit thick but I read Post 30 upon the lines that the fact that C1E was disabled/enabled did not make any difference viz a viz the multiplier dropping. - so enabling it would not make any difference.

Also
But the BIOS on these boards sets C1E "off" even at default safe settings and the multiplier / voltage changes depending on load.
Click to expand...
I read to mean that this was its standard position and as I'm running at stock it should be dropping when 'idle' etc. which it is not.

Is it better to have this enabled?

Am running the GD65 Board myself - I tried the fixing each core at 42 you suggested and it recorded the same in the bios but no effect at all when running still stuck on 33.

It's early days but am beginning to regret leaving Asus.- I found overclocking the Q9650 an easier proposition.
 
AgedIdiot said:
IdiotCoder - Am having a similar problem getting the cpu to lower its multiplier; at what page of the link you provided did you find your answer?
Click to expand...

Seems to be Intel C-State under CPU Features, at first this kept somehow being disabled automagically in the BIOS.
Enable it and you should notice the multiplier lowering down to 16 when idle.
I also have EIST enabled.

I don't know how I managed to stop it from disabling, but thinking back it may be due to the OC being unstable at that moment in time, because I BSOD'd a lot.
 
IdiotCoder said:
Seems to be Intel C-State under CPU Features, at first this kept somehow being disabled automagically in the BIOS.
Enable it and you should notice the multiplier lowering down to 16 when idle.
I also have EIST enabled.

I don't know how I managed to stop it from disabling, but thinking back it may be due to the OC being unstable at that moment in time, because I BSOD'd a lot.
Click to expand...

Found the same thing, ignore the turbo bit, its the C-State and EIST options that turn themselves off. Enable them again and it should fix it. Seems to be a bug in the BIOS or it happens when you first overclock.
 
Will spend a bit of time tomorrow in the Bios to see if I can bottom the problem and get this cpu running at 4.2 at less than 1.3 v.:)

Will have a look to see what alterations , if any , have been made in the Bios - I did keep a note of all the different entries so that if the board ever had 'an asus' moment (sudden, often, inexplicable reversion to default - which happened from time to time with my last asus board) I'd know what I started with.

Will def. have a look at C-state.

Not sure if it was a typo or not but do like the 'autoMagically' reference:D
 
You must log in or register to reply here.
Back
Top Bottom