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

[Supershanks]

No need to translate, I already gave you the main points

Sorry jcool, must have been nodding off after lunch, missed it:

Cheers Ace & you m8.

[jcool]

No problem - and a merry Christmas from me too

[Supershanks]

No problem - and a merry Christmas from me too

All the very best

[Ace-a-Rue]

lets have a "GROUP" hug.

[jcool]

[Amey]

Wanted to contribute my initial test results,didnt know where to post, i guess this would be the right place,

Here we go, nothing great in the results

3DMark 2001


3DMark 2005


3DMark 2006


A quick Super Pi 1M test


This was a quick and hasty run on the IX38 Quad GT, will do some more testing later on, maybe with some more DICE

Cheers !
Darky

[jcool]

Quick and hasty, wow.. what phase unit?

Edit: Sorry misread the DICE part, thought you were gonna try DICE later

[Supershanks]

lets have a "GROUP" hug.

[Ace-a-Rue]

i moved on to the DFI LT X38 board...it is about the same in performance as the Pro...probably needs bios maturation to reach or even the Pro and i presume abit's IX38.

[jcool]

Well let us know which one is better
Sadly I have a very annoying problem with my IX38... it's making horrible crackling noises, just like the old QuadGT but much louder and... in idle
The old one only did this under full load and only when highly OCed - funny thing is, when I put the IX38 under load it shuts up
I think I'm gonna have to RMA cause its really really loud (thanks to me having an ultra-silent system...)

[Ace-a-Rue]

crackling sound that comes out of the speakers?

[Supershanks]

i moved on to the DFI LT X38 board...it is about the same in performance as the Pro.

have you got some stuff up on that?? have been watching Georges thread looks quiet tasty.

[Ace-a-Rue]

i have one video bench, 3d2001.

what's nice about the DFI board is low vdroop...some say there is no vdroop but it has it...about 0.015v...DFI employs the vdroop switch in the bios...what that does is actually raise the vcore above what you set in the bios when the switch is enabled...tricky...hehehe...actually, it will droop under load...now understand this!...when droop switch is disabled, you get actual vcore equaling bios set vcore and the vcore will not droop at ALL during a wPrime 1024M bench...scratching head here...LOL

the other nice thing, you have numerous bios adjustments for DRAM...there is a huge learning curve here but it does pay off...you can get DDR-whatever to run about 200 points or more above its stated rating...that said, it is lagging behind in at least one video bench...i scored a 1000 points higher on the IP35 Pro than on this DFI board..obviously bios maturation is lagging at the moment.

the problem could be this latest bios that is no longer posted on DFI's website...apparently it was issued to allow PCIe 2.0 standards but one user posted that the bios set his VC to 1x...mine is working at 16x as you will see after posting 3d2001.

[jcool]

crackling sound that comes out of the speakers?

Naw... I'm using an X-Fi anyways. No, the crackling/buzzing/hissing noise comes from the board itself, namely from the PWM Aaea... dammit. And everybody was hoping they fixed this issue with the rev. 2 of these digital PWMs. I guess they didn't since it's a lot worse than it was with the old QuadGT.
Dunno why they started only yesterday, but it's very annoying

[Ace-a-Rue]

running silent can be a curse!

[bicycle_wreck]

I just got this board for Christmas and have been setting everything up. So far I've had a couple of headaches, but one is really irritating me.

I have my E2180 running stock (2GHZ) at 1.325v, and my memory is strapped 1:2.00 to run at DDR2-800 (stock) speeds. I am continually failing Prime95--one thread--after about an hour. I flashed 12.B06, and it hasn't seemed to have helped.

Since I am having problems with Prime95, I decided to go ahead and do some memtest86+ testing. I ran both sticks of ram with memtest86+ for four passes and no errors were found. Later, I ran the test again and found 1 error on a third pass, so I ran each stick individually in the same slot (slot 1), and they each passed four passes. All of this was done on the shipped BIOS, and I have not run memtest86+ yet with 12.B06.

My question is: since each stick of RAM passed individually, is it some sort of memory controller issue or what? Also, I know some people run memtest86+ all night or whatever, but I figure in four passes you have what you have, is this wrong?

My memory: Patriot 2 x 2GB PDC24G6400ELK. I have set all of the timings manually (5-5-5-12 [stock]), and have the voltage set to 2.0v--at which they are rated.

I am running a fresh install of XP Pro 32-bit, my CPU temps are around 30-32*C, and I have C1E and EIST disabled. I am testing with Super PI right now, so far so good.

The rest of my rig in sig. Any help/advice would be appreciated.

[AMDfreak]

Hi all! Hope everyone's had a Merry Christmas.

I've had little time to test the past few days between a nasty cold, my 6-month old, visiting family, etc. For me, lower multis are still a no go with C1E disabled. No amount of voltage will get me anything better than a BSOD at Vista startup @450 FSB. 1:1.25 also is not stable at all, so I've settled for 9x412 1:1.20.

Started to OC my 1950xtx and was running 700MHz core stable for a day when I started to get that electronic buzzing and sudden instability in video apps. Oh well, what better excuse to have for getting an HD 3870? After catching up on the thread though I'm really hoping the noise isn't actually from the IX38 PWM...

[jcool]

Yeah, it just started suddenly, there's still no reason to me as to why these noises didn't appear earlier... haven't even changed my OC.
I hope your card is all right, I had to bury my X1900XTX earlier this year... just died on me while surfing (40 C Core max so no overheating..)
Made me go for Nvidia - so far I didn't regret it. Just played Crysis on my new 24" in 1920x1200 with everything set to high (DX9 though) - runs unbelievably smooth

[aster]

Hey Guys, I wanted to share my cooling method for IX38.

For Northbridge, I use an Apogee GT with AM2-fix cut-off and drilled
For Mosfets, watercool #7 block is perfect for 30€
For Southbridge, MCW30 is fine.



This pic was taken during filling, don't worry for bubles

I've crunched the mosfets cause I've mounted my waterblock on AS instead of thermal pad. Some CMS have made a shortcut, dead mobo
So, since i'm waiting for a replacement one, i'm gonna make a new fix for ApogeeGT with 5mm plexy.

[jcool]

Always stick to the thermal pad when it comes to mosfets (even if it hurts) you learned that now
You should've read the manual, I think it's even printed in bold
Good look getting a replacement!
That tubing of yours looks sick... I don't see anything but tubes Apogee's looking nice though on the NB.

[R*r]

today i just change my memory to G.Skill F2-8000CL5D-4GBPQ

previously my d9 cannot run with 1:1.25 regardless th voltage that i give.

but now i can run it with 8x444 / 555@5-5-5-15

but if i use 9x444 it give me c1,which is so strange.i put 1:1 can post n stable.

so i think th divider problem got smt to do with memory ic

[jcool]

today i just change my memory to G.Skill F2-8000CL5D-4GBPQ

previously my d9 cannot run with 1:1.25 regardless th voltage that i give.

but now i can run it with 8x444 / 555@5-5-5-15

but if i use 9x444 it give me c1,which is so strange.i put 1:1 can post n stable.

so i think th divider problem got smt to do with memory ic

Then why can I run 1:1.25 with 4 different types of D9 ?
Crucial Value 667, Cellshock PC6400, Crucial Ballistix 800 (all D9GMH)
OCZ SLI 1066 (D9GKX)
They all go up to ~1050-1100 stable with 4:5

I think we need a new final bios...

[Humble]

Running my Q6600 on this board now at 3.0 Ghz speed and DDR2 1000 Mhz. The voltages it needs is unbelieveable low.
Vcore is default (high possibility lower voltage will works), vMCH is 1.3v and vFSB is 1.25v and vDIMM is 2.0v.

More thing to praise is that this board don't even do any double boot like most of the other Intel boards do, even when I unplugged the power cord after using it (cold booting). Very nice board indeed.
Gonna climb for higher speed soon.

Naw... I'm using an X-Fi anyways. No, the crackling/buzzing/hissing noise comes from the board itself, namely from the PWM Aaea... dammit. And everybody was hoping they fixed this issue with the rev. 2 of these digital PWMs. I guess they didn't since it's a lot worse than it was with the old QuadGT.
Dunno why they started only yesterday, but it's very annoying

Buzzing sound when the system under load, bro? My previous Commando was, but this board is just running silent without any sound like that.

I just got this board for Christmas and have been setting everything up. So far I've had a couple of headaches, but one is really irritating me.

I have my E2180 running stock (2GHZ) at 1.325v, and my memory is strapped 1:2.00 to run at DDR2-800 (stock) speeds. I am continually failing Prime95--one thread--after about an hour. I flashed 12.B06, and it hasn't seemed to have helped.

Since I am having problems with Prime95, I decided to go ahead and do some memtest86+ testing. I ran both sticks of ram with memtest86+ for four passes and no errors were found. Later, I ran the test again and found 1 error on a third pass, so I ran each stick individually in the same slot (slot 1), and they each passed four passes. All of this was done on the shipped BIOS, and I have not run memtest86+ yet with 12.B06.

My question is: since each stick of RAM passed individually, is it some sort of memory controller issue or what? Also, I know some people run memtest86+ all night or whatever, but I figure in four passes you have what you have, is this wrong?

My memory: Patriot 2 x 2GB PDC24G6400ELK. I have set all of the timings manually (5-5-5-12 [stock]), and have the voltage set to 2.0v--at which they are rated.

I am running a fresh install of XP Pro 32-bit, my CPU temps are around 30-32*C, and I have C1E and EIST disabled. I am testing with Super PI right now, so far so good.

The rest of my rig in sig. Any help/advice would be appreciated.

Why don't you try another divider bro? Try increasing your FSB a bit, lets say to 266 Mhz and try divider 1:1.5, and see how it behaves.

I myself, also have problems with divider 1.6+ and 2.0+ in BIOS 12B06.
Divider 2.0 even has two version, one is 2.0 only and the other is 2.0+ (with plus sign). Anybody know details about the difference between the plus and non plus version?
When I use divider 1.6+ and 2.0+, my system is locked up and can't even POST, so I have to clear CMOS. Meanwhile the other dividers are working just fine.

[R*r]

Then why can I run 1:1.25 with 4 different types of D9 ?
Crucial Value 667, Cellshock PC6400, Crucial Ballistix 800 (all D9GMH)
OCZ SLI 1066 (D9GKX)
They all go up to ~1050-1100 stable with 4:5

I think we need a new final bios...

bro my gskill is using superchip ic my d9 can run 600 with 1:1.5 3d stable,other divider no good but must use slightly higher vdimm than usual

[ust]

Hiya
Hate to break the ongoing discussion, but I have a quick question that needs answering.

I followed Supuershanks 400*9 BIOS settings ( http://www.clunk.org.uk/forums/3914-post30.html) on my Q6600 g0/Ballistix 8500 ( with BETA BIOS b06) but Prime failed immediately so I bumbed down the bus to 389*9 and raised Vcore to 1.4250 and let Prime test while I was outta town.
When I returned core3 had errored after 1hour and 83 minutes while the other three cores were stil running - and they had been running for 53 hours without errors ( ! ) when I stopped the test.

I can't really understand why only core three would fail and the others still running. The Vcore should be high enough and the other settings seem to be failsafe/overkill settings.
Could Prime be the problem? I havn't seen the magnitude of this problem before; I mean the other three cores should flake out at some point, when one of the cores already failed after 1 hour and 80 minutes.

?

Thank you in advance guys

P.S Maybe I should add that the systems is rocksolid (prime tested ofcourse) running 333*9 at default Vcore/VID 1.3125v and BIOS settings.