New Bios for evga 680i just released!

[Babrbarossa]

I just wrote this in another thread- but I am hoping this is good enough news to warrant it's own-- Can't wait to see how this works (very curious to see if it deals with the Quad-core bottleneck).

Edit: I'd flash it now if I was using a quad-core, but as it is, I'll wait for the release notes and some feedback before I make the leap from p23.

Link to EVGA forum announcement

[Babrbarossa]

Just to update and disappoint- so far it doesn't seem like this bios fixes the quad bottleneck-- the changelist is still not posted, but it appears that it fixes Vdroop.

I've given up on this board, and I'm back on my insulated P5W DH. The audio frustrations I've been having over the last two months have been ridiculous. That and it just hasn't been a very stable platform for me. Good luck to everyone who's continuing to use it. I have decided SLI simply isn't worth all the hassle it's caused me, and I'm putting together a secondary little media rig to house my other 7950GT.

[Babrbarossa]

New bios released but it's not very useful and some people are claiming that it has a destabilizing effect. It claims to fix wireless PCI card problems, and it removes the linkboost option from the bios (you can still adjust PCIe and Hypertransport yourself though).

http://www.evga.com/community/messag...TOPIC_ID=29940

EDIT: Jacob at evga revealed that the bios was also supposed to increase fsb overclocking by 5-20 on the fsb, but it's not guaranteed so they didn't post it in the description.

[Babrbarossa]

There's another new bios- these guys are amazing- I have to say- I flashed p26 from p24 and gained .5GHz on my oc- this bios is geared for quad core improvements- I might wait a little bit to see whta comes out of it as I'm dual core.

EVGA | Message Boards

[tomb18]

The P27 bios made no difference to my fsb. The highest I can get is 1440 but this is not fully stable at any voltage (up to 1.6) with a 10x multiplier giving 3.6 GHz.
On the other hand, at 1400 with 10X I'm running 3.5 GHz 24/7 with 1.45V. It is extremely stable running quad prime 95 small FFT's for 24+ hours at 65C max.

The interesting thing is if I lower my fsb and use an 11x multi, I can get stable at 3.7GHz with 1.55V. This gives a boost in all benchmarks for example 3dmark06 goes from 17700 to 18560 from 3.5 to 3.7 gHz. All superpi benches are faster as well.
Clock spped is more important than fsb at least at this level.
I finally settled on 3.5 GHz.

[Babrbarossa]

It boosted my e6400 from 3.45 to 3.84, but for some reason, mine isn't happy at around the 3.6 mark (1800QDR)- more stable at 3.7 with the same voltage. Have you done any messing around with the lower multipliers like 8 and 9?

[Scratch]

wow I flashed my bios and what a story lol

I also posted fantastic results at first going to 3.8ghz running dual SPi 1mil crappy score but still nice OC then I thought...lets go higher lol
So I tried and had a major crash at 3.86ghz after trying to up my ram speed to 1100mhz (do not always believe memtest!)
I was unable to get my pc stable at any OC today and could not figure it out. I was crashing all over the place instability everywhere...I thought o well it only cost me 250 but it was a nice chip lol
Then I received a windows update notification saying my system was going to shutdown if I didn't activate wtf
I call Vista support 2 1/2 hours later we figure it out...my pc reset itself to Jan 1 2003
Needles to say I'm back stable at 3.6ghz synced with 900mhz ram @1.55vcore

BTW mine hates 3.7

[Babrbarossa]

Sorry! that's right- I was having a brain fart- ot was 3.7 that it didn't like for me too!

Wow sounds like a trepidacious experience!

[Beermaster]

Wow thats sounds like quite the horendus experience, glad it all worked out for you and you didnt have to go buy a new chip lol