Gigabyte GA-X48-DS4 wont OC Q6600

C

carma

n00b
Joined
Jun 16, 2008
Messages
7
First time poster, long time reader. Decided to register finally, since I have this brand new setup that has been a challenge to figure out. I upgraded to a new Gigabyte GA-X48-DS4 recently, and decided to set it all up over the weekend. I have an existing MB with a Q6600 OC to 3.5-3.6, which I just swapped over to the new Gigabyte MB, and bought some new OCZ Reaper PC-8500 DDR2 RAM (2x2GB) 4GB. It seems to work flowless, if it's not overclocked. For whatever reason I cannot get the MB to overclock the CPU. Every time I change the FSB to anything but 266 the system will power-reset three times then finally POST, and revert back to the original 266. Even if I change the FSB to 267 it wont like it. I'm stumped. No matter what vcore I change it to (without going past 1.53) it will not overclock. I'll throw in my old E4300 tonight and see if that overclocks at all, which will tell me if it's a MB or CPU issue. I also updated the BIOS even though there is only one version on the website, F1 I believe it's called. Thought maybe the BIOS got corrupt. Any help is much appreciated!
 
What the ram voltage? My Gigabyte EX38-DS4 had mine set to 1.8 instead of 2.1.
 
I have a x48 -dq6 and it overclocks the q6600 just fine, make sure you have the bios set up right, all the energy saving stuff turned off before oc'ing it. shoudl work without problems.
 
Well, I swapped the RAM out with my old PC6400 stuff (2x1GB) and it acted the same. It has to be the CPU. I'll have to check the energy saving stuff. Why would energy saving being enabled hinder overclocking though?
 
This just sounds like it could just be Gigabyte BIOS "stupid" mode as well. It does not matter what you do, it will run the CPU at stock settings right now.

Shut off the power supply, unplug the power cable, wait a couple minutes, then reset the CMOS, wait a few minutes longer, then power it all back up, let it boot into Windows once, then reset the PC, adjust the FSB up to 333 (just as a simple test) and see if that puts you at 3GHz.
 
Thanks ocellaris, I am going to try that tonight. It really does sound like a MB issue. First thing is to swap out the Q6600 for my 4300 and try increasing the FSB. Then I'll try clearing the CMOS as you suggested. One thing I didn't mention, which I do not believe has any bearing, though is worth mentioning. I lapped the Q6600 before I put it in the new MB, and hadn't tested it on the old MB after lapping. Will update tonight. Thank you for the recommendations!
 
carma said:
Thanks ocellaris, I am going to try that tonight. It really does sound like a MB issue. First thing is to swap out the Q6600 for my 4300 and try increasing the FSB. Then I'll try clearing the CMOS as you suggested. One thing I didn't mention, which I do not believe has any bearing, though is worth mentioning. I lapped the Q6600 before I put it in the new MB, and hadn't tested it on the old MB after lapping. Will update tonight. Thank you for the recommendations!
Click to expand...

Do not bother trying the 4300 first, go with the CMOS reset, make sure you wait and leave the system off power for a while when you do it (people will argue this, I say give it at least 10 minutes with no power just from my own experience). No sense to keep swapping CPUs when you do not have to :D

Anyway I am sure the 6600 is fine, you can not really make it go bad or anything from lapping it.
 
OK, I left the computer off for a good hour all unplugged, then reset the CMOS, still the same story. I also swapped the CPU for the 4300 and that was the same story as well, would not allow FSB overclock, even 1MHz. I thought maybe there was an onboard jumper that controls FSB, allowing or disallowing overclocking but couldn't find anything in the manual. Fed up at this point, it has been about 3 days since I send an email to Gigabyte support still no reply. I sent a request to MWave to get it exchanged, see what comes of that. This is my first and last Gigabyte MB.
 
carma said:
OK, I left the computer off for a good hour all unplugged, then reset the CMOS, still the same story. I also swapped the CPU for the 4300 and that was the same story as well, would not allow FSB overclock, even 1MHz. I thought maybe there was an onboard jumper that controls FSB, allowing or disallowing overclocking but couldn't find anything in the manual. Fed up at this point, it has been about 3 days since I send an email to Gigabyte support still no reply. I sent a request to MWave to get it exchanged, see what comes of that. This is my first and last Gigabyte MB.
Click to expand...

Sorry that did not work out for you. I am very curious as to what the cause of this problem really is so keep us updated, maybe someone else will have ideas as well. Personally I had a Gigabyte P35 DS3L and that will be my last Gigabyte board as well.
 
OK.....

I have the exact same MB, exact same CPU, AND I had the exact same problem initially.
I had my Q6600 easily to 3.2 in my old computer. This board would reset off and on while I was trying to get a stable OC as well.

Here's what I did.....may work, may not.....but currently I'm running my Q6600 at 3.25 GHz and rock stable.

disable or do not load the energy saving software.
disable all the thermal Intel stuff (C1E, EIST, what have you)
enable manual cpu host clock
disable that CIA thing
run the enhansement on "turbo" (you can't disable it, I tried all three settings and turbo seemed the most stable, it's the default)
disable auto memory
set voltages to manual
I chose to set my memory clock to "B" 333 FSB and 2.5:1 ratio (I'm using Corsair XMS2 PC 6400 C4 running at @840 MHz and 1.875V)
MCH +.1V
VCore 1.3675V
I thought I was the biggest idiot that ever lived, I know how you feel.....I spent a whole night trying to get a stable OC.....I thought I'd gotten a bad board......then I hit on the stuff I have listed above and things just settled down and have been fine ever since.
One thing I did notice.....after I adjusted my OC and went from the BIOS to windows.....the board would shut off for about 2 or 3 seconds and then restart and off we'd go.........I think that is a given with this board.........that endless cycling off-on is not and I cannot for the life of me figure it out.
 
I believe I have tried all of that, sans disabling the auto memory. I've NEVER owned a MB that wouldn't overclock out of the box without serious tweaking that perportedly fixes this issue. I finially did get a reply from Gigabyte, but have since setup an RMA already through Mwave. Ordered a DFI x48 MB, shouldbe here tomorrow. Here's the reply from gigabyte:

there has no any newer version bios then F1 could be one of the memory was not stable cause overclock reject try remove all memory from board take onboard cell batterry out for 5 sec clear CMOS then test with single stick memory on slot close to cpu , note : mother board bios memory voltage setting is base on 1.8v ( default voltage ), check current model memory voltage spec if it's higher then 1.8v go in bios under M/B intelligent tweaker memory voltage control item adjust memory voltage to match memory voltage needs .
for example : current memory is 2.1v go in bios M/B intelligent tweaker on memory voltage control item change to +0.3v ( 1.8v + 0.3v = 2.1v ) this will match the current model memory voltage spec .
if problem still persist with correct memory voltage setting and all memory tested was stable that could be some thing wrong with the bios on mother board please refer to our warranty :
All motherboards carries an 3 year warranty from the date it was manufacture ( according to serial number ), products which was purchased within 30 days from invoice will be handled directly from your vendor or dealer they will replace mother board . Over 30 days please log on our online RMA service http://rma.gigabyte-usa.com go to end-user click on add new request fill out the RMA request form click on submit.

But this doesn't really apply because I've tried different RAM, PC6400, and I verified it had the correct voltage and it still wouldn't OC.
 
really too bad.
my board works like silk now, but I admit it was a bit picky at first.
however if you thought the GB was stubborn, those DFI boards take real TLC.
good luck.
btw, mwave has very slick rma service. I order alot from them.:D
 
Hello, My name is Mike.


So I also have the x48 ds4 motherboard. This is my first build and I must say all the worriers were for nothing. It was really really easy.

So I am having the same problems as the others with overclocking this. I have read alot and tried the settings. It will load the giga mobo background then go to the black screen and load others... sometimes it says succeccful other times it doesnt say anything the n go to another black screen saying windows was shut down wrong... start in safe, start normally, ect... Then when I go to start normally it comes back as a black screen and doesnt do anything(the HDD is silent).


Vista home premium 32 (soon to be 64, when the cd comes)
x48 ds4
q6600 G0
2 2gig sticks of OCZ platinum 1066 (4 gigs total) http://www.newegg.com/Product/Product.aspx?Item=N82E16820227298
rosewill cpu http://www.newegg.com/Product/Product.aspx?Item=N82E16835200026



Here are my settings to try and reach 3.0

cpu multi...9x
cpu freq 333
pciE 100
cia2 disable
performance enhance... standard
memory multi... 2.00B----> 667----> so 1:1
DRAM timing manaul 5.5.5.18
DDR2 overvoltage... +.3(reading it as 1.888???)
cpu vcore 1.325


HDD SMART... enable
cpuid disable
no-excute memory protect... disable
cpu halt (c1e) disable
CPU thermal monitor2 (tm2) disable
EIST disable
Virtualization disable


Please help me get it to 3.0.


Thank you,
Michael
 
Got the same motherboard and it OC's like a beast. Love it.

Don't know where I read it, but I think I remember reading that the performance enhance setting should be left on the default which is Turbo otherwise it won't OC.

You might try 2.00D. Your ram might need that 400 strap. Shouldn't but maybe it does?

Try upping the FSB voltage a little at a time. Leave Loadline on auto. Shouldn't need to fool with MCH at 333. Mine runs fine @ 425 on stock.

So, if I was you, first I'd load optimized defaults and start over.

Set your fsb to 333 and leave it at 2.00B and make sure performance enhance is set to Turbo.

If that doesn't work, Try 2.00D.

If that doesn't work, try giving it a little FSB voltage bump.

If none of that works.

Mess around. Reload optimized defaults and start over. See where it stops. 280.......290 ...........300..........maybe it doesn't like 333 but likes 336 or 332. Maybe it likes 400. Try setting multi to 8, fsb to 400, memory to 2.00D and run it @ 3.2.
 
I also have this motherboard. I set the FSB to 400 and haven't looked back since :)

The only thing I noticed was that temperatures were much higher than they should have been. Come to find out the "Auto" setting for the voltages jacked the VCore up to 1.5V :eek: Put it back to stock volts and everything was fine.

So, definitely make sure that the voltage section isn't on Auto and ignore the flashing "YOUR VOLTAGES ARE NOT OPTIMIZED!!!" warning.
 
Does anyone else find the OP switching from Gigabyte to DFI for ease of overclocking funny?
 
so it seems to HATE anything higher then 329 fsb!!!!

when i do 330 with 2.00B and 1.3375 vcore I get BSOD right after i log on(put password in) :mad:

but when I do 329 with 2.00B and 1.3375 vcore It boots


So I think my computer hates me
 
The CPU's themselves are bus limited, Quads are affected more.
It would be unfortunate if you cant get higher though.
X38/X48 boards can get well in excess of 450MHz FSB, my X38 does over 520MHz, others blow mine away!
 
OK
You did. Well almost. 329.

hmmm
Above poster is right that the 65nm quads, B3 especially, were themselves bus limited. I had to crank vtt (different motherboard) on my B3 to the high heavens to get it to 333. You might need to fool around with fsb and gtl ref voltages to procede further. Gonna take a little studying and reading on your part.

But before we do that......

Some, just don't like certain ranges of fsb. Did you try lowering the multiplier and seeing where it will go? Crank the chip down to 6 and start raising fsb. I mean, lets be sure with what we're dealing with here.

Just for fun, set the chip to 6, fsb to 400 mem to 2.00D and see of she boots.
 
I will try that in a little bit...


So When I did the 9x at 329 it booted but as soon as i ras prime95 it dropped core 1 then about 10 seconds later the screen goes black and my cpu cooler goes to max... I waited a couple of seconds nothing... tried buttons nothing... the power and reset woulnt even work, So I had to use the switch on the psu...




EDIT:


2.8ghz using 7x 400 2.00D 1.35 seems stable____ran prime95 for 5 minutes no error


2.9ghz using 7x 428 2.00D 1.35 boots then i ran prime 95 and it restarts... I also added +.2 to ram to get it up to 2.032 instead of 1.888.
 
OK

Looking like your chip is gonna need a little voltage.

I know your saying 1.35. What does cpu-z report your voltage as?
 
ok so i got it to run but it fails prime 95...


8x and 380 (3.04)Ghz 2.00D 5.5.5.15 dram +.2 mch +.1 vcore at 1.3875(in bios set)

It failed core 0 instantly saying rounding was .5 expected to be less then .4


then it ran for about 30 seconds and I lost my picture and my cpu fan spooled up, I had to hit the reset button.


I will have to check the vcore with cpuz

CORE TEMP 0.99.1 VID 1.325v

HWmonitor value 1.38____min 1.36____max 1.38

CPU-Z 1.376___

BUT cpu-z also had my core clocks moving... right now it says 2280.2MHZ multi 6.0??? But when I click on say AIM the core speed and mutli jump up to 3040 and mutli at 8

Im stumped... I uninstalled gigabytes power saver... I still have the RAID configurer on. hmm... Any ideas where to go from here
 
Keep an eye on your temps. If their OK, hit it with 1.4 or maybe even a little higher.
 
yes, Unless they got reset...


EDIT


and they did, must have forgot about them after I had to open the case and reset using the jumper.
 
So at 8x__380___2.00D___+.2ram___+.1mch__5.5.5.15

Vcore set at 1.4...... cpu-z said 1.39

It ran and temps at idle were 37,31,32,35


Ran prime 95....core 0 failed then system rebooted.... ran for about 1 minute or so
core0 just got up to about 52. the rest were lower
 
So I might not be on here till monday... I was edging the grass and ummm... I kinda hit the cable cord going into the house :(. I dont know why it wasn't buried more than a 1/4" under ground... Oh wait yeah I do... STUPID COMCAST!!!! So I am just using someone's wireless right now and its very very slow. Took 2 minutes to load this page!!!!
 
lol
Time-Warner does that too. Mines been accidently cut at least 3 times. They finally came over with the trencher.

Well, lots of time to play with it.

Try everything. Watch temps. Stay at or below 1.45. Play with fsb voltage and gtl ref. Sometimes those Q6600 don't need more voltage, they need to "see" the fsb better.... um...well, they need a larger amplitude in the signal that they use to mark the fsb frequency.........or something like that. Right at the edge of my understanding.

http://www.thetechrepository.com/showthread.php?t=87
http://www.bit-tech.net/news/2008/05/12/will-you-overclock-intel-s-p45/1

See what I mean?

I will do some looking around for a gtl ref setting that seems to work with the Gigabyte X48 boards. I'm to the point of squeezzing out the last few megahertz so I need to see for myself anyway.
 
OK
While in the main menu, hit Control F1.

Then go to Motherboard Intelligent Tweaker.

2 new settings there called GTL ref. Pick one and give it the old +3% and see if things change. If not, go back and try the other one. If not, give it to both. If not try the -3% or -6% settings. Sometimes quieting down amplitude can reduce noise allowing better signal. Real voodoo science going on now. Who knows but it doesn't look like they give you enough that you can kill the chip so play around with it.

Gonna look up clock skew control.

holy moly
 
So when i am failing prime95 its because of the round error " .4 expected but got .5" or somthing like that. Or the screen gos blank or the system resets...


So From reading a little bit, This rounding error is caused my the timings of the ram. I have OCZ platinum 1066 and my timings are 5.5.5.15. I really dont know much about ram and the timings. What would you recommend trying?
 
streetmike said:
So when i am failing prime95 its because of the round error " .4 expected but got .5" or somthing like that. Or the screen gos blank or the system resets...


So From reading a little bit, This rounding error is caused my the timings of the ram. I have OCZ platinum 1066 and my timings are 5.5.5.15. I really dont know much about ram and the timings. What would you recommend trying?
Click to expand...

Its caused by something failing to operate correctly.
It could be motherboard, memory or CPU.
 
So it seems my cpu likes the high voltages...:(

3.00Ghz
8x 375
pcie 100
2.00D
5.5.5.19
Dram +.25v
mch +.1
GTL 1 +3%
vcore 1.4bios 1.39CPU-z

Rand prime95 for 10 minutes(yes i know it's not long enough) but I came back with zero error. Longest I have been able to run it.

Dropped the v-core down to 1.35 fatal error (the 5 and 4 one)
1.625v and fatal error was 4.99**** expected 4
 
You must log in or register to reply here.
Back
Top