e4300, evga 680i revision A1 overclocking

Babrbarossa

Obliviot
Hi- I've been overclocking an e4300 on both the AR and A1 revision of evga's 680i, and the experience has been very similar on both, but with some peculiarities.

First of all here's my setup:
Arctic Cooling Freezer Pro 7
680i
e4300 at 3.2 GHz, 1.45V (also an e6400 that will clock to 3.45 on this same setup at that voltage)
evga 8800gts
OCZ gold pc6400 (2GB)
ocz GameXStream 700w

All of the dumb bios things like speedstep, CIE, etc are disabled- my other system voltages are set generously (but not too much) and my ram is at 2.1V.

There are a few mysteries I am trying to solve but they are beyond me:

1) For some reason I can't get the CPU heatsink fan to go above 1000rpm anymore.
I've tried: Setting fans to manual and 100% in bios, setting to 100% in speedfan, plugging fan into different jumpers, changing bioses, CHANGING MOBO from AR to A1 revision- no lcuk-- Maybr it's something wrong with the heatsink fan- although it's fairly new.

2) Can't get beyond 3.2 GHz with the e4300 - rather disappointing considering the 3.45 I got with the e6400 and same setup. I tries volting up to 1.50, but no luck- tried linked and synched, unlinked, and even changed mulitpliers- which brings me to the next mystery:

3) 8X multiplier is the only multi that gives decent overclocking results. I don't get much with 9X, and for some reason I don't get anything when I lower the multi to 7X or 6X- I was hoping to try 6X to find out what the max FSB was on the board. The funny thing is that I found the same problem with both board revisions- tried linked and synched, and unlinked.

4) The computer behaves differently when I clock the e4300 too far, compared to when I clock the e6400 too far. When I go too far with the e6400, it will either reset to the prvious values on reboot, or automatically restart in windows, or show an error on Orthos.
With the e4300, it either freezes up windows, or crashes on post- reverting to deafault values, or doesn't post at all- It's very unforgiving- I might be running orthos and instead of an error appearing, windows just freezes. If I even try to go to 3.3Ghz it won't even post. I don't know why it would be perfectly stable at 3.2, but quickly freeze up windows at 3.25..

If anyone recognizes any of these wierdnesses and wouldn't mind sharing their wisdom, it'd be appreciated muchly!

Thanks!
 
Last edited:

Rob Williams

Editor-in-Chief
Staff member
Moderator
I don't know, but to me it might just be the fact that the E4300 is just not willing to go much higher. I didn't have -that- much luck with my E6300.. I believe it topped out at 3.35GHz or around there.

Have you checked temps? Maybe the E4300 for whatever reason, is heating up faster than it should. I have found the eVGA to be a decent overclocking board, but it is finicky as you mentioned. I'm unsure what revision I have here though, but it's been pretty good to me. It doesn't hit a super high FSB, that's the only gripe I have against it.

Oh, and the fact that you need a floppy drive to update the BIOS.

Are you sure the CPU fan is only going 100RPM? That is painfully slow... You could spin it by hand faster ;-)
 

Babrbarossa

Obliviot
Thanks Rob- I was hoping you'd weigh in- I lost a zero on the fan speed- I meant 1000rpm- Also-i've been flashing bioses using a .bin file and a cd and it's worked fine. I'm thinking that I might just have hit the limit on my cpu- the temps aren't too bad- I'm maxing out at 63C on Orthos- I'm just surprised about the 8X multi being the best if FSB is any kind of limitation.
 

Rob Williams

Editor-in-Chief
Staff member
Moderator
I'm curious, where did you get those BIN files? All I could find on their site were .EXEs that were used to create a floppy.

I do agree on the CPU limit... chances are you topped it out. All things considered, it's still an amazing overclock and performance gain over stock.

Also... did you increase the other voltages as well, such as the ICH?
 

Babrbarossa

Obliviot
here are the .bin files- i wasn't having any luck with the iso's which are also posted in another sticky there- (I have all the volatges increased, but none over 1.5)

I really shouldn't complain about an 80% overclock, but all things equal, I should be able to get closer to the e6400- I'm doing a review so I was hoping to get closer to the max speeds that others were finding. I have a great water setup coming this week, si I'll see what I can do at 1.6V when I get it set up- I won't have to worry about fan spped then, But I am doing a bit of a comparison bwtween, AR vs. A1 and e6400 vs. e4300, and air vs. water
 

werty316

Partition Master
If you are having slow fan issues try running the fan on the 12V line of your PSU by using a 3-to-4 pin adapter.

As for your E4300 overclocking issues, most of them hit 3.4GHz with around 1.6V and you should be able to run it rig 1:1 without any problems.

Also remember not all chips are the same so you are not guarrenteed the same overclock when compared to a different chip.
 

Rob Williams

Editor-in-Chief
Staff member
Moderator
I began overclocking with this board the other day, and haven't been able to hit a high FSB like I have seen many others have. I believe my max at this point is 450FSB, but I am using an AR board.

Bab, what has your highest FSB been with the A1 board?
 

Babrbarossa

Obliviot
Mine has been 450 as well, but I flashed to the new p27 bios yesterday just for the heck of it (evga said that it was just for some quad-core oc improvements), but I have the e4300 in there right now, and I increased my overclock from 3.3 to 3.5- pretty amazing improvement- That's still at 9X so the fsb isn't higher than 450, but I'm interested in seeing what my e6400 can do now at 8X.

On p26, I have tried lowering the multiplier on the e4300 but I don't have any luck at all getting it too boot into windows with lower multis for some reason - even at low FSB settings- I'll try it again with the new bios and get back to you.

I have seen people on evga's message boards getting 500FSB with the latest bios.
 

Babrbarossa

Obliviot
I think I realized why I'm having no luck with the e4300 on a 7X multi- is it because of the 800MHx fsb rating of the cpu? Would that would prevent it from going over 400MHz? If so, I'll have to try the e6400 on 8X to find my max again with the new BIOS
 

Rob Williams

Editor-in-Chief
Staff member
Moderator
I flashed to P27 the day it was made available... and it didn't help me at all. I did read on another forum where someone was in a similar situation... 450FSB was his max. Then with P27 he was up over 500FSB.

The 800FSB on the CPU won't mean much when overclocking... it's not a "hard limit". It probably -will- hit an overall limit though, similar to CPU frequency. That comes with the territory of lower binned CPUs.
 

Babrbarossa

Obliviot
An interesting discovery-- For the A1 revisions, they have been replacing the 590 southbridges with 570s- that's the real reason that linkboost is disabled- it's not available on the 570- admittedly a useless feature, and according to EVGA it's the only real difference between the two but it somehow seems underhanded.
 

Babrbarossa

Obliviot
From all I've gathered, the southbridge on that board and all pure 680i boards (such as striker extreme) is a 590- the 680i refers to the northbridge. EVGA has switched the 590 back to a 570 on the revisions- it's not clear whether they are all like this, but you can tell by looking at the chip- it's written on top.
 

Rob Williams

Editor-in-Chief
Staff member
Moderator
Interesting. I guess I didn't look at closely. CPU-z still calls it a 680 SB though. Either way, if that's the only difference between the revisions, I am not sure why A1 "tends" to have better overclocks. I noticed there is a T1 revision on NewEgg as well, although I am not sure whats different about it.
 

Babrbarossa

Obliviot
I compared the two boards (AR, A1) using a couple of C2Ds and didn't find any overclocking or general performance difference between the two versions for the dual-core chips (there definately is for the quad cores, of course). The overclocking improvements I've been finding have just been from bios updates.
 
Top