**Official ABIT IX38 QuadGT Review/Overclock/Guide Thread**

[jcool]

I really like what I've seen so far.
Major bugs/problems yet anyone?

[pinto]

after a crash the bios always reset oc settings.
if you completly power off the system for a ram exchange or something else the bios will reset too.
not really important but it's a little boring...

[jcool]

Yeah well, you just have to make extensive use of the F6 - F7 keys then I guess
But the old habit of forgetting everything when being completely disconnected from power really IS annoying.
Can you verify whether this happens if you just disconnect from power? Or only when doing sth else like swapping rams etc

[dinos22]

scary to think phase reach those temps on watercooling

imagine having 1.9V+ on the quad on this board on Dice or LN2 eek

it will explode to oblivion lol

[jcool]

I thought he only had the cpu watercooled?
For stock cooling w/o additional fan blowing onto PWMs 70C is pretty impressive actually, if you consider his setup. These PWMs are stable at around ~110C anyway, so I wouldn't worry bout it. On the P965 QuadGT I get phase temps of 95-100C (thanks to my B3 Xeon) running 24/7... and this is on water - so they really did improve the digital PWM

[JPxM0Dz]

Living Review: http://www.clunk.org.uk/forums/revie...ng-review.html

[aster]

On AB9 QuadGT, phases on water gives 60° when priming Q6600@3800@1.53V

[nordien]

Results on AIR with a Tuniq Tower and a E6600:



OCZ Ti Alpha @ DDR 1000 @ 4-4-4-15 @ 2.35v
CPU VTT @ 1.6v

3dMark06 gives a 13153 @ these settings.........



With these settings the digital pwm's are about 60 C

I like this board so far !! Can't wait to put a Penryn in.

[Ademcg]

Any trouble fitting the Tuniq with this board I just orderd it and wanted to make sure

[nordien]

Any trouble fitting the Tuniq with this board I just orderd it and wanted to make sure

Fits perfectly, no mods needed. This mobo has tons of clearance around the cpu socket

[Ademcg]

Thanks for that.....now ordered

Have you had to update to latest bios yet [Ver 11]??
I am looking to put the latest on a USB pen drive in readiness trying to use the HP USB Disk Format Storage tool,......but it states "there are no valid DOS system files at the specified location" but all the bios files (there are files : RUNME; AbitFAE ; awardflash; M704B_11) are present ?? Will it work in any case ?

[jcool]

On AB9 QuadGT, phases on water gives 60° when priming Q6600@3800@1.53V

Not if you're using unmodified mips waterblock... and yes I did tighten the screws properly
It's made that way to prevent the phase chips from breaking but I guess they have overdone it a little...

[Clunk]

Anyone with a quad managed to get stable over 400/410FSB without random BIOS freezes? Does your board need massive amounts of voltage to get stable at 400FSB and below?

This one of mine is really bad. At stock, I enter the BIOS and it freezes and needs a CMOS clear to fix.

I can then go to windows, and do whatever I was doing, restart and the BIOS has frozen again.

A friend of mine has the same problems, no doubt he will be posting soon.

[jcool]

Uh-oh. That doesn't sound too good. I guess I'll wait for a new bios then..
I mean, I can do 400 with my B3 and the old QuadGT w/o high Voltage (1,35V mch) perfectly well...

[Ace-a-Rue]

ut-oh!...doesn't sound good with the freezes...now...i am seriously considering the Gigabyte X38T board...hmmmmm...darn...it must be the same team that developed the AB9 QuadGT board...it wasn't very good in the beginnning with bios help.

[Ace-a-Rue]

Anyone with a quad managed to get stable over 400/410FSB without random BIOS freezes? Does your board need massive amounts of voltage to get stable at 400FSB and below?

This one of mine is really bad. At stock, I enter the BIOS and it freezes and needs a CMOS clear to fix.

I can then go to windows, and do whatever I was doing, restart and the BIOS has frozen again.

A friend of mine has the same problems, no doubt he will be posting soon.

have you done any heatpipe mod work?

i am wondering if this digital PWM is a POS...never like what was on the AB9 QuadGT board..i know it says that it is all improved, which is a lot of bullsh*t in my book.

[jcool]

ut-oh!...doesn't sound good with the freezes...now...i am seriously considering the Gigabyte X38T board...hmmmmm...darn...it must be the same team that developed the AB9 QuadGT board...it wasn't very good in the beginnning with bios help.

Right... the AB9 QuadGT's original bios was nothing short of virtual BS.
Bios 11 fixed alot and then 12 and 13... still running 13 w/o any problems.

They did improve the digital PWM I guess, but less heat than on the AB9 still doesn't mean it's cool

[nordien]

Anyone with a quad managed to get stable over 400/410FSB without random BIOS freezes? Does your board need massive amounts of voltage to get stable at 400FSB and below?

This one of mine is really bad. At stock, I enter the BIOS and it freezes and needs a CMOS clear to fix.

I can then go to windows, and do whatever I was doing, restart and the BIOS has frozen again.

A friend of mine has the same problems, no doubt he will be posting soon.

I think there is something wrong with your board. I've playing around for three days now and cannot confirm your problems but i only have a E6600 not a quad. Yesterday i did 8*450 @ ddr 900 @ 4-4-4-10 without problems, the fsb wall on the E6600 is keeping me back, anything over 450 is not stable.

The bios sure isn't finished yet but i only get a freeze when i try oc'ing on the fly (F8) with the wrong settings.

I am getting good pwm temps (60-62) with the E6600

[Clunk]

have you done any heatpipe mod work?

i am wondering if this digital PWM is a POS...never like what was on the AB9 QuadGT board..i know it says that it is all improved, which is a lot of bullsh*t in my book.

No mate, it is absolutely stock. Normally, I would have taken it off, and replaced with something else, but I didnt have anything that would fit - Being realistinc though, the PWM cooler is flawed, in so much as it really needs a fan to cool it, but unless you go out and buy one, the board will overheat relatively quickly.

I have seen 112c on the PWMs when only using 1.5v, and that was with a 92mm fan directly on the heatsink, and the board was built up out of the case.

I'm returning mine on monday, it has no redeeming features.

[Clunk]

I think there is something wrong with youyr board. I've playing around for three days now and cannot confirm your problems but i only have a E6600 not a quad. Yesterday i did 8*450 @ ddr 900 @ 4-4-4-10 without problems, the fsb wall on the E6600 is keeping me back, anything over 450 is not stable.

The bios sure isn't finished yet but i only get a freeze when i try oc'ing on the fly (F8) with the wrong settings.

I am getting good pwm temps (60-62) with the E6600

I would have said there is something wrong with it, but supershanks also has one, and his is the same, he also has a quad, but I will let him tell you of his troubles, if he can ever get back online

If I were to speculate at what is wrong, I would say that the BIOS isnt ready for public consumption, and those PWMs need a better heatsink, or some kind of active cooling included.

[Ace-a-Rue]

No mate, it is absolutely stock. Normally, I would have taken it off, and replaced with something else, but I didnt have anything that would fit - Being realistinc though, the PWM cooler is flawed, in so much as it really needs a fan to cool it, but unless you go out and buy one, the board will overheat relatively quickly.

I have seen 112c on the PWMs when only using 1.5v, and that was with a 92mm fan directly on the heatsink, and the board was built up out of the case.

I'm returning mine on monday, it has no redeeming features.

what i was talking about for the heatsink is yanking it and removing the TIM pads to replace with Ceramique along with using a bolt/nut, where the plastic push-pins with springs are used, to get better clamping action and therefore, better heat dissapation.

[jcool]

Or you could try to fit one of the existing waterblocks onto the phase chips. From what I've seen so far, the PWM area is 100% identical to the AB9 QuadGT's. That's what I would try, anyways...

[Clunk]

what i was talking about for the heatsink is yanking it and removing the TIM pads to replace with Ceramique along with using a bolt/nut, where the plastic push-pins with springs are used, to get better clamping action and therefore, better heat dissapation.

Yep, I had a read of your guide (great stuff by the way ), and I would definitely do that if I was keeping it, but the BIOS problems are just too much.

Or you could try to fit one of the existing waterblocks onto the phase chips. From what I've seen so far, the PWM area is 100% identical to the AB9 QuadGT's. That's what I would try, anyways...

To be honest, I wouldnt want all that heat in my loop either!

[jcool]

Oh really... that's about 20-30W max für the PWMs. Who cares - some people use dual laings, they heat up the water with more than that :O

[Clunk]

Oh really... that's about 20-30W max für the PWMs. Who cares - some people use dual laings, they heat up the water with more than that :O

20-30w? maybe at stock and idle, but if they do indeed hit the 80c+ that I've been getting, and there is 5 of them, certainly seems like more than that.

I have dual loops, but why would I want to add pointless heat into the loop? and also, why bother buying waterblocks for a board that doesnt work properly?