Printable View
Just flashed my DQ6 from F3G to F4D (from the link provided above) and it seems to be slightly better at overclocking - perhaps another 25-50MHz - I'm still priming to see if 500Mhz FSB is stable.
Oddly, though, F9 in the BIOS shows:
F2
06/29/2006
6A79LGOG
Is it possible it's just renamed?
No, you have to "clear DMI data" when flashing.
i dont even see an F5g on that webpage...
And F4e for the DQ6 is not listed on that page....are we missing something?
Ah, I see. I just flashed using the flash895 utility. Oddly my memory is stable at 1000MHz if I use a 2.5 multiplier and run the FSB lower; it's cheap Crucial 667MHz stuff, it shouldn't be my motherboard limiting me!
looks like 1000mhz 5-5-5-15 is working. using the 2.50 divider
but 4-4-4-15 wont work....
so the chipset doesnt like the tighter ram timings.
so.. keep working on it gigabyte.. you have a ways to go yet but your getting there.
New BIOS doesn't seem to solve my "unable to boot after a while of inactivity" problem after all. Bummer... Neither does setting the vcore back to normal which gives me a cpuz reading of over 1.4v.
btw, I experienced the same problem with DQ6. I didn't need to wait at all though. Most reboots from Windows would require re-enabling OC settings, yet, reboot from BIOS were usually fine (unless I was playing/testing settings). Statement valid to F4a. RMA'd due to MB failure after that. (Thinking about it now, there is probably a connection between the reset cycle that I couldn't overcome and the resetting of OC settings on controlled reboot. :rolleyes: )Quote:
Originally Posted by WackyWookie
From memory, this occurred mostly with OC's of 400fsb and above.
I am also fairly sure but not positive that I could successfully boot to Windows @450+, but not stable.
Has anyone experienced CPU-Z telling them that their memory is running at CAS 3 at a certain speed when it's set to CAS 6 only when the memory is running 1:1, yet (with the same manual settings) CPU-Z says it's correctly at CAS 6 with a 2:3 divider? It's very frustrating; if CPU-Z is correct, it may be why I'm having difficulty.
might wanna try actually using the latest version of cpu-z and the latest bios.
you'll see that issue go away
So you're saying my mobo might be busted and i should RMA?Quote:
Originally Posted by GAM
I was using CPU-Z 1.35, and DQ6 BIOS F4D. Updating CPU-Z to 1.36 hasn't changed the reading.
thats not the latest bios now is it :P
infact thats a pre final F4 bios
No. I wasn't saying that.Quote:
Originally Posted by WackyWookie
I am suggesting that there may be an issue with the (DQ6) board that causes this undue disabling of the OC settings when we know (and have tested) it to boot fine and that there 'may' be a connection between this issue and my (and others') stuffed board.
It didn't occur to me that it is isolated to selected boards, I was thinking more along the lines of potential problem in all DQ6 boards which may result in a stuffed board, given the right circumstances.
...the more I think (gotta stop doing that) about it, maybe your thought is right. Why don't you start a thread for DQ6 owners about this issue of 'reset' oc settings after reboot/cold boot/etc. I assumed our experience wasn't uncommon.
I can't find a later BIOS than F4D for the DQ6. If you can, I'd be very grateful for a link! :)Quote:
Originally Posted by Lestat
Lestat's mixing up his mobo's, I think ;)
If I go into the memory settings in BIOS and set just the CAS delay to 6, leaving everything else at auto, at 500MHz CPU-Z shows CAS 6. However, it doesn't make it any more stable :( bleh. I'm trying to figure out if there's some voltage setting for the MCH/FSB that I can use, but I've tried up to +0.3v on both (and together) without much success.
I won't because i own a DS3 not a DQ6 :) But you're right... maybe a new topic dedicated solely to booting issues involving GB mobo's would be an idea...Quote:
Originally Posted by GAM
Sorry, I for some reason I thought that but too lazy to verify.Quote:
Originally Posted by WackyWookie
This is even more interesting though, because DS3 doesn't have Quad BIOS, right? Which would mean no connection to this (an earlier suspicion).
well the entire line of gigabyte boards share the booting issues.
so your not alone.
i am not liking this F5g bios...... seems like its the smae b.s. that i reported with the F5c total garbage.
1000mhz is it.. once again another wall.
something gigabyte is doing with the bios is rejecting the micron D9 from the Teamgroup and gskill ram. and gigabyte needs to fix that crap.
there is no reason why the 2 best ram kits on the market shouldnt be working.
just the opposite. these two kits should be doing what they are doing in all other boards. hauling a$$ and taking names.
Quote:
Originally Posted by Lestat
100% Agree!
DS3 F5 final is out , go to gigabyte website or get it here http://www.cranox.com/Gigabyte/DS3/
ggrrrNOOOOO its not final yet damn it!!! they havent fixed it
nope they didnt fix it.
still cant run 1000mhz stable. or rather i should say i cant run it anymore.
they did something though my 1m Pi is faster.
i dont have time before work for a 32m so ill try it on lunch
F4 out for DS4 users
get it here: http://www.jzelectronic.de/jz2/index.php
why are these new bio's not on the gigabyte websites if they are final?
I see DQ6 owners are getting screwed once again...no bios for us!
Hey! This F5 Final really helped me alot. On previous bios versions I got memory errors with memtest (in 10 minutes) and bluescreens with right stock ram settings (667mhz, 1.8V), but now with this new bios I haven't got any BSOD and also memtest run totally without any errors, so I'd say this helped my alot. So like they say on F5 Final description: "Fix some DDR667 compatibility issues", they really fixed that.
Quote:
Originally Posted by a5h
I have tried to flash the F4 on my DS4 from flash895 but it says "bios file size error" .....before trying to force from the qflash and to make damages, someone has tried this bios? :stick:
What's wrong with F3? I can do 490 FSB, and mem works fine (including D9).
Eldonko what board and what D9 ram ?
and are you able to go over 950mhz on the ram ? ie: 1000+ mhz 4-4-4
DQ6
BIOS F3G
Mush Redline 8k (D9GMH)
Tested to 502 3-4-3 @ 2.2v so far
http://members.shaw.ca/Eldonko1/dq6-8k-3.JPG
well thats the DQ6 lol..
ds3 has issues with the gskill 6400 HZ i can never remember which D9 that is.
and CAS 3 ?? ehhh you sure about that.... ?
and i did some testing.
my 32m times are slower at the same clock speeds than from the F4 wtf gigabyte wtf!!
i dont mean 1 sec slow either. im talking 10 seconds
infact also I arrive to 495 of fsb and the memories works fine with F3 and F4C bios for DS4...doesn't change nothing, but the coupled DS4+6600 Boxed don't get on well together at least to me.....and therefore I would want to try new bios :D
http://img159.imageshack.us/img159/576/495hd3.th.jpg
P.S. - sorry for my English :slap: :D
EDIT: Eldonko you have the DQ6 I spoke of the DS4...however nobody knows nothing for the error of the flash?
I set cas 3 myself with memset so unless it doesnt work properly then cas 3 is correct. :confused:Quote:
Originally Posted by Lestat
1 issue I have though is the memory timings appear in the BIOS on occasion but 99% of the time they are not there and I dont know what I did to make em appear. Same w/ EIST, it just disappears sometimes... :o
Not sure shark2, never got that error when flashing before.
bios file size error means its a bad download
redownload the bios and try again.
if it does it again
try downloading from a different place
Hmmmm................now if only the micron D9MGH sticks work well things will be looking up. Hopefully by the time I get ready to upgrade to a 2 GB kit that will happen.Quote:
Originally Posted by ClockCatcher
Quote:
Originally Posted by Phosphate
i got teamgroup PC5300 D9MGH and no problems (446 FSB 1:1 ):
http://img528.imageshack.us/img528/3351/3570144vlg6.jpg
Strange things happening tonight. I updated the bios on my ds3 some hopurs ago, tested it and was unable to acheive any extra OC, stuck at 450FSB. Just been playing with it again (the bios:cool: ) and can now hit a FSB of 477 and thats with out adding any extra volts to PCI-e, FSB or GMCH. Not sure what happened to suddenly allow it to go but it has. On a side note any one have any ideas on the re cord for air cooled e6400, just got mine into windows at 3.8Ghz but was not quite stable enough to run SPI
How's the F5g for the DS3?
Quote:
Originally Posted by Delphi00
oh boy you really need to get with the program kiddo.
F5 final is released F5g is a beta.
listen we have discussed this soooo many times before and this is why i harp on people to READ THE GOD BLASTED THREAD!!
Gigabyte names the BETA bios's with a letter, such as F5a, b, c, d, e,f , g
their RETAIL bios does not have a bios. ie,,, F5
the next beta bios to be released should be F6a or b or c or something like that.
also let me restate the whole Micron D9 issue.
the micron D9 ram runs just FINE up to around 950mhz..... after that.. its down the toilet.. not even 5-5-5-15 is stable and being forced by gigabyte to use 5-5-5-15 on ram that is VERY capable of 4-4-4-12 is absolutely retarded.
Gigabyte has programed the bios in such a way as to totally fuxxor anyone with the D9 ram that is in the Teamgroup and Gskill 800 and 667 kits.
again this is the DS3 not the DQ4 or 6... i have seen people say that DQ6 doesnt have any issues so god knows wtf is wrong with the DS3.. but gigabyte can fix it they just wont...
so saying 445mhz is no problem doesnt show anything.
go into the bios and set the ram so it runs 1000mhz 4-4-4-15 and se if you can even reboot. i bet it wont boot if it does itsdefinately not stable.
now try 2.3 or 2.4v which again is taking the voltage of the ram out of spec and not good so be sure you have active cooling.
i bet it still wotn work.
now set 5-5-5-15 it might work it might not... now try running some tests see what happens... in windows please DOS driven memtest has shown itself to be buggy with this setup..
try it see what happens.
anything over 950mhz is flakey at best. and isnt stable at all.
Sorry, didn't see it listed in the main gigabyte website but I found it anyway.Quote:
Originally Posted by Lestat
Do I NEED to load fail-safe and then reset my cmos after I flash?
Yes, load optimised defaults.
yup just as richard said.Quote:
Originally Posted by Richard Dower
go into the bios hit F8 then Yes.,,, then update from floppy...
then it will show you the bios name it sees on the floppy...
ok wait.. why am i explaining this. this is the most simple and easiest bios flashing ever lol.
when its done reload and laod setup defaults.. save and exit.. then go back into the bios and start overclocking.
Where you all get the F3G bios for th eDQ6?
DQ6 (current) beta: F4dQuote:
Originally Posted by N15M0
Above BIOS, older BIOS and new BIOS (when available) for various GB boards can be downloaded fom this page: http://www.jzelectronic.de/jz2/index.php
Go to end of page and select applicable BIOS from popup menu.
My SATA2 seagate harddisk running in ACHI mode (Gigabyte RAID) seems to be unable to boot up at times. It's only in rare occassions, but it happens. The Gigabyte RAID bios detected the HDD, but it doesn't try to boot from the HDD at all. All I get is "Disk boot failure" after seeking for boot sector in my floppy (didn't have a startup disk in there).
Amazingly though, once I press CTRL+ALT+DEL, and it restarted, the drive will boot just fine. Again, this happens once in a blue moon.
Don't know if this is BIOS related, but I started seeing this problem only after F4 beta. The F5s didn't fix it. I can't test each bios to be certain since it's intermittent. Anyone else facing this problem?
Resetting itself to backup bios, maybe? I did see someone else post a similar if not same problem. Don't recall where though, sorry.
Im currentely running my Corsair 6400C4 at 890 , 4-4-4-12 2.1V , gonna try to run then higher than that
Wow it sucks seeing people having these problems with the DS3 and D9...
Guess I just got lucky
http://img304.imageshack.us/img304/8...21880mr.th.jpg
500MHz 4-4-4-12 Dual 32M on DS3 w/F4 Bios
http://img136.imageshack.us/img136/2...d32mpo7.th.png
What CPU are you using?Quote:
Originally Posted by TheMeatFrog
Is it a e6300?
Indeed it is
Shouldn't be the backup bios running since my bios settings are still intact. I'm running a Pentium XE 955 ES, so it'll default to a multi of 12x when it's running backup bios. Multi is supposed to be 13 for stock frequency at 266MHz FSB. I'm waiting for my Conroe, so this system is not overclocked yet.Quote:
Originally Posted by GAM
nevermind F5 did not help at all
And how exactly are we DQ6 users getting screwed again??Quote:
Originally Posted by Richard Dower
Ahh, that's right, because YOU can't get YOUR system running properly...:rolleyes:
There're plenty of DQ6 users who haven't had a single problem with their board.
Exactly, I sometimes suspect that some are expecting far too much, if it can't overclock to 500fsb then it's immediately assumed that something's wrong with the board.Quote:
Originally Posted by BiOvOrE
I'd agree, except there ARE issues with the DQ6 that are NOT limited to FSB's of 500MHz. I was/would be quite content at the stable 425 or so I experienced before it DIED!
The board/bios needs some attention. Gigabyte?
This is exactly my point.Quote:
Originally Posted by GAM
Firstly, you try and run the board way over specification and then complain when/if it fails.
Secondly, I seem to recall that you also physically modified your DQ6 and lo and behold soon after it is rendered inoperable, or to put it in your own words, you "killed" it.
And thirdly, you'll be the one whinging when/if your new CPU that you lapped to such a nice shiny finish also stops working , but of course it wouldn't be your fault.
Now now. This IS xtreme systems, right?
The board provides overclocking facilities, is designed to do so and promoted as such.
I don't think I've complained about the DQ6. I like and want it back. The only physical modification I made to the board is to fit a heatsink and place some cardboard on it--I 'could' have damaged it in this process, and if so it IS my fault. Whether I stated I 'killed' it or not, the statement is out of context. And, No, I would not complain if my CPU died. I am, obviously, not concerned with it's warranty.
Please vent on someone else.
You've got almost identical hardware to me, so how is it your board can reach 500FSB and mine can only go to 338FSB????Quote:
Originally Posted by BiOvOrE
Good question, mate.Quote:
Originally Posted by Richard Dower
But if I'm using almost identical hardware to you, I'd say that very fact pretty much rules out the cause being a faulty Gigabyte BIOS.
I've used most of the BIOS betas (currently on F4d) and been able to boot (but not necessarily bench) at more than 445FSB, without much tweaking at all (basically, more and more vcore.)
I'm guessing you're just plain unlucky and have a mobo/CPU and/or RAM that is a dud overclocker for whatever reason, OR you're doing something seriously wrong with your setup.
The 500+ FSB's may be a different story, but there's something seriously wrong somewhere if you can't hit 338, and I strongly doubt that ANY BIOS that Gigabyte releases will fix that problem.
Just installed my F5 bios last night on the DS3. I then spent 3 hours trying to find a good overclock and jut could not. NOthing I did was dual prime stable and I was getting weird choppiness and studdering in 3dmark2006.
HOWEVER. I finally said screw it and set everything to "normal" for voltages including VCORE,GMCH,VFSB,and VDIMM. I set the FSB speed to 400 and PCIE freq to 105 since that is the only setting that will allow my 7950gx2 to functioning. Memory timing is set to 2x and I set timings to automatic.
Low and behold my ds3 booted up at 2.1 vdimm, 1.425 vcore(which sounds about right), and memory set to 5,5,5,15 timings. The system is completely ROCK STABLE dual prime stable 3dmark2006 stable. In fact this is the most stable I have seen my system since it's birth. I ran a 1M pi time of 18.5s
I changed the memory settings and manually set it to 4,4,4,12 and left vdimm to auto. Everything is still completely stable and 1M pi time is 18.4
So I changed the memory divider to 2.5 which puts me at 1000mhz memory speed at 5,5,5,15. I ran a 1M pi time of 18.2
I then changed my memory to 4,4,4,12 at 2.5 divider for 1000mhz memory speed. I ran a 1M pi time of 18.1.
So it looks like the F5 bios has better memory compatiability(I can for the first time easily run my Gskill HZ memory at 1000mhz even though I really don't see much of a difference at all. What the heck does the memory speed increase anyways? I was expecting better gains upping it to 1000mhz. Even my 1M pi times didn't improve much.
I know its a sin to leave everything at auto but it looks like the F5 bios handles automatic voltages VERY WELL. SO WELL in fact that my system is more stable than ever before.
F4 Final for the DQ6 is out.
http://www.jzelectronic.de/jz2/index.php
Thank christ!...though i doubt i'll have any luck.
Quote:
Originally Posted by BiOvOrE
I've changed out the memory to an 0627 G.SKILL HZ kit, it's been Orthos dual prime tested for 12 hours @ 900MHz. 2.2V with 4-4-4-12 timings, 2GB kit.
It's not the memory, the CPU is a week 24 retail B2 stepping E6600. My POV is even IF the CPU was holding me back i SHOULD be able to set a 6x multi and acheive 400FSB
400x6=2400MHz....thats stock speed, i should be able to do this, but i can't. No amount of MCH or FSB voltage will do anything for me.
So are you saying then that each board can be very different from another?...surely all mobo's are made the same at the factory. Why should one perform better then the other?
Anyways i'm off to flash and test F4.
You might want to gradually reduce vcore. I started at 1.4 and slowly dropped to 1.35. Each time tested Orthos > 4 hours. That's my sweet spot. GMCH & VFSB are set to default. I get 57-59 coreTemp on Load with Orthos.
Quote:
Originally Posted by HeadShotWI
HIGH TEMPS
I have a 965P-DS3 / Corsair 6400C4 / e6400 setup and my temps are abnormally high. Its not my fan as Ive used this fan on my previous setup so I know it can cool, its the Zalman CNPS9500. Ive remounted the fan to check for improper seating, and Ive applied a5 correctly. I bought the e6400 second hand, and he was using a DS3 also and didnt have this high of temps (unless he was lying but ive seen screen shots and i dont think he was). At stock the chip is idling at 50c / 70c peak @ 1.30V. If I bump it up to 3.2Ghz @ 1.35 it goes up to 65c / 83c peak. Ive tried every bios revision thats been released. Should I RMA the board? The chip? Different fan? What else could it be?
Thanks
OSA
That's the way MBs work, they all can be totally different in terms of overclocking even though it's the same board. It sucks but that's oc life.Quote:
Originally Posted by Richard Dower
Have you tried testing fsb limit using various straps? I get a different max with all dividers..
OneStepAhead - What are you measuring temps with?
Measuring temps with CoreTemp, Everest Ultimate, SpeedFan, the BIOS, and my hand :)
All measurements report the same temps
OSA
If you want a comparo, at stock vcore my E6400 is at 35C load (easytune), 48C load (coretemp). Not sure which is correct but that is using stock cooler and 28C ambient. This is DQ6 though, hard to say if temp monitoring is different from DS3.
I don't think that mobo makes CPU ran hotter. That is physically impossible. My temps are with stock cooler (that runs @ 1700rpm) and stock fsb about 43c at idle and about 62c at load. These temps are from CoreTemp, but with other temp apps. temps are about 10c lesser. So I think also that CoreTemp is showing too high.
Here is a cool Tool I ran across a minute ago. Intel Thermal Analysis Tool. Apparently its an OEM tool and isnt on intel.com for download. It measures both core's temps and you can even throttle the CPU to 100%. I downloaded it from here: http://shintai.ambition.cz/files/tat.exe
OSA
That program shows my temp at 55c , bios, and easy tune shows 44 also speedfan shows 44 , wth?Quote:
Originally Posted by OneStepAhead
I have my DQ6 bios f3e, with a E6600 at 400fsb for 24/7, the only problem that i have is that i cant restart. I need to turn off the computer and wait a few seconds before she goes again. Not a big deal.
Does anyone know how to unhide the ram settings in BIOS? Normally it looks like:
http://members.shaw.ca/Eldonko1/biosdq6-1.JPG
But twice when I went into BIOS the ram settings were there, but I have no idea how they got there. Gone again now of course.. :rolleyes:
http://members.shaw.ca/Eldonko1/biosdq6-2.JPG
Did I press something to unhide or is this just some weird bug?
ctrl + F1 at the BIOS main menu
HTH
Press Ctrl+F1 @ BIOS to enable advanced settings.
I cant find any difference between F4d and F4 so far, which is understandable. Ill keep playing and watching.
Edit: Running an e6600 on a DQ6 at 3.33ghz (rock stable at 1.4v vcore, maybe lower).
Ah thanks!
They fixed the spelling of RAID from RIAD. That's all I can tell that they fixed.Quote:
Originally Posted by Scarlet Infidel
Mel
OneStep, have you verified that the heatsink is making proper contact with your CPU? I had the same issue the 2nd time I mounted my heatsink - my E6400 is very slightly concave and I used half as much Arctic Silver 5 the 2nd time, and I discovered that the AS5 made zero contact in the middle, only the outer edges of the heat spreader.
Adding more AS5 in the middle and spreading it out a little bit with a razor brought the temperatures to a more reasonable level - 53C Idle, 64C Load, as measured by CoreTemp.
Please post details of this, here: http://www.xtremesystems.org/forums/...d.php?t=114726Quote:
Originally Posted by Silves
Quote:
Originally Posted by Eldonko
and you dont need to raise the pci-e fsb or (g)mch for what little bit your overclocking.
turn them back to auto.
you dont even need to touch those until you get well over 400mhz FSB
I have changed those voltages and other settings 100 times since that pic lol, .3 vmch = 490 fsb :)
which mobo do you get that for? and which BIOS?
Dq6, F3g.
Good result with F3g. Have you tried F4 yet?Quote:
Originally Posted by Eldonko
Yeah been using F4D since last night. Seems the same, although I didn't get any reboot loops or fsb resets yet..
Was referring to F4 (Final). Not beta. Good to know about the reset issue though ;)
Using F4 for the DQ6...remember that thread about hidden bios options on the DS3, well i had an OC go bad and the PC shut down and restarted...went into and endless cycle.
Anywhoo...it disabled my OC settings and when i entered bios i saw one of those screens, it was the one with Advanced Chipset Features...something about EHCI.
I looked also in PC Health to see if i could also see the additional display for MCH temps and whatever else was supposed to turn up, i didn't see anything new.
But that Advanced Chipset Features thingy definately was there, i haven't been able to see those screens since.
http://www.xtremesystems.org/forums/...idden+features
FYI: the Intel AHCI bios has been updated to version 05082006. The previous version was 10042006.
The Intel RAID bios has noot ben updated though, still 6.0.0.1022, wheras Intel now have 6.1.0.1002.
Also it appears Asus have an updated JMicron RAID bios, 1.06.56a, whereas Gigabyte have 1.06.53.
Is there any way of making the DQ6 to run Crossfire? :D
That would be :woot:
U sooooo arrogant dude, thats alll u do being negative and arrogantQuote:
Originally Posted by Lestat
How exactly is that arrogance??Quote:
Originally Posted by Ajfina
I, for one, can do 3.7Ghz without touching my PCI-E, FSB, vDimm or MCH voltages.
Well Darkoz, it seems I may have to concede at least half a point here.Quote:
Originally Posted by Darkoz
I've heard back from Gigabyte (indirectly) and apparently a CPU socket pin is busted. I am not sure whether it is broken or bent. So, the 'physically damaged' board will be repaired at my expense.
Although I am not convinced that I did this, and though I don't believe my (documented) actions could be realistically considered as 'physically modifying the board' as Darkoz put it, I do recall some difficulting replacing the plastic protective insert back into the socket for shipping, of all things. (Stupid, I know, but probably due to the variety of socket inserts I've got floating around and the stress I was feeling at the time :rolleyes:.) Otherwise, I can't really imagine how I could have broken a socket pin :confused: I am quite careful with these things, truly... (No smart-arse comments, please ;)).
Nevertheless, I will be interested to know whether the board exhibits the original problem after they repair the socket.
Quote:
Originally Posted by Ajfina
uhm ok...
so which one of you sneaky little turds has the F6a DS3 beta ?
Well Im liking the DS3 F5e BETA (I can get F5 but I cba to change it now lol)
dropped my multi down to 8x, and im running a 430 FSB priming fine currently, having to put 1.55v through it though (may be less but im not finding out untill I get the max)
and Ajfina, he is right, I have had to increase my vMCH up .2v to get 430mhz (might be only 0.1v needed but i want to get as much as I can)
I appear to be running stable at 2.8Ghz on the DQ6 with F4. I have left the (G) MCH and FSB voltages to normal, VDIMM is set to 2.2V and the VCore is set to 1.36250. FSB is 311x9=2799.
FSB : DRAM = 2:3 giving me 466.5MHz on the memory, DDR2 933Mhz, 2.2V with timings set to manual 4-4-4-12.
Orthos dual prime stable for 3 hours now, but will leave run for 12 hours.
F5C beta for DQ6 is up now @ computertechnik. I'll let one of you try it first.. :D