Printable View
Okay, I've tried all of that and still get failures on test 8. However, there seem to be less - which I want to verify later. Should I apply more vdimm? When I get hme from work I'll see what slackening the timings achieves.Quote:
Originally Posted by mdzcpa
Should I raise vcore?
You shouldn't need more than 3v, but the DFI needs extra voltage to the RAM (more than other boards). So give it a try.
In order to clear up the errors, I'd try the voltages higher and lower. Also you can try adjusting the memory timings a little at a time. Perhaps 2.5 4-4-8, etc
I'll give it a whirl but won't I ge getting close to the edge here? I don't want to fry my RAM.Quote:
Originally Posted by mdzcpa
I'm having an issue with high HTT. Everything above 285/290 it's unstable and reboots, either by Bios or ClockGen. Got LTD @ 3x, 1.4v HT, 1.8v Chipset, Ram @ 100/133/150/166 (tried all), tested with multi 7, 7.5 and 8 and it's no CPU limit (3200+ able to do 2.63Ghzs Stable), no go above 285HTT +-. Any ideias? =)
It's an Ultra-D btw, Bios 217 and 218 tested.
Can everybody give 3.5V+ to the Vdimm under winxp? Becous everytime I give 3.5 it stops when loging in.. 3.4V works, but non over that.
Cheers
Lol... Prommy RoX! :slobber:Quote:
Originally Posted by :banana::banana::banana:in
Nice for the VGA with waterblock, but for the other one, the Leadtek dissy is not more performancer than NV5? :confused: ;)
Regards :toast: .
I can give up to 4.0v without problem. ;)Quote:
Originally Posted by Razor_cut
Maybe it is the ram...
I've managed to run a clean set of tests at 275Mhz FSB, 6xCPU Multi, 3.1 vdimm and 2.5-4-4-10 timings and 4xHTT. There is a downside though. If I raise FSB speed the errors in test 8 return.Quote:
Originally Posted by mdzcpa
My memory is already highly geared voltage-wise. Should I be raising LDT and/or chipset votlages? And if I do, what are "safe" levels to raise them too?
I'd throw an exta .1v to both and see if it helps. Another thing to look at is your CPU voltage. The memory controller is on the CPU so a bump might help there too.
Hi
i have only 1 problem with bios 2.17 and other Betas
C&Q dont work if i set vcore to something else than AUTO.
And if the Vcore is set to Auto i cant use VDIMm Voltage over 3.2Volt.
Also Windows tools like Cristalcpuid cant change Vcore
under Windows if CPU Vcore in Bios is set to something else than AUTO.
Bad for my VX Ram, i can go to DDR533 1.5-2-2-2-7 3.6Volt :slobber:
bat with 3.2 only DDR460 2-3-3-7 :mad:
Pleas help :help: :bsod: :idea:
Well its OCZ VX. So I dont have a clue what the problem may be..Quote:
Originally Posted by Zac89
well i tried 270*9....1.65v in bios...not stable ...so i give up...this CBBID sux.
conclusion : threw 500$ for nothing.
CPU Voltage is at 1.45v, RAM at 3.1v, LDT at 1.3v and chipset at 1.6v and I'm still generating 100s of errors at 6.5xMulti. CPU bios core temp. is being reported at 51C.Quote:
Originally Posted by mdzcpa
I'm not sure how high these voltages are in O/C terms but the extraordinarily low clock rate I'm having to run the CPU at in order to even try and clear memtest makes me think something's not right.
The gain being achieved per notch on the voltage scale is too small to get anywheere near decent clock rates for the cpu. I'm also seeing that any changes which take reported memory bandwidth over 2400MB/s generate bucketloads of errors.
Every time I try to overclock, my SATA drives disappear. Has this happened to anyone else? I cant even raise my fsb 10mhz before it disappears!!
- Awperator
My system
W3200+
2x512 VX
OCZ Powerstream 520W
The system was running OK at 260 2-2-2-5 prime and 24/7 stable 1550Vcore + 110% and 3.6vdimm from the 5V line.
No problems until two weeks ago. Then, my DFI began refusing to boot. 3 leds on (memory error). I had to try it for several times, until it booted normally. However, as days has passed, it has become more and more difficult to make it boot. The last times I managed to boot it, I have had to try it for several HOURS, trying with safe mode jumer and clear cmos jumer until I got it. When it booted, the system was 100% stable and I could reboot without problems, but when I shutted down the system, the nightmare started once again.
Currently, it seems to be impossible to boot anymore with my mem but I have no problem with a pair of PQI pc3200.
I've tried with several bios (0201, 0217 and 0218) and none of them solves the problem. I'm completely desesperated and planning to buy a pair of OCZ Platinum rev.2 'cause I have no more ideas about how to get the system boot with my config.
The system doesn't boot with only one module of the VX :stick: pair on any of the banks; the result is the same. I've tried it after clearing cmos, booting the PQI pair, and switching off the psu.
Has anyone experienced something like this?
any idea?
:confused:
I bought the VX mem just for this mobo and now I'm starting to think about selling them :(
Sounds like a capacitor breaking down to me.
Is this for me? if so, what should I do? 'cause if it's a capacitor, the situation could become even worse and it could stop working with other memories too, couldn't it? Btw I don't understand why it boots everytime with the pqi pack and it doesn't boot with the vx ones even at stock speeds and voltage. With the VX ones at 2 3 3 8 at 2.6V it should boot :confused:Quote:
Originally Posted by stu_allen
Well, not really a problem since i believe im one of the 2.5 people that booted and everything went just peachy from the get go but i didnt think this deserved its own thread. Im using bios 2/17 with some good ole twinmos BH5 and a crappy winny 3000+ thatll brick wall at 275x9.
But anywho, i have an ultra-d and of course, modded it (pencil #2 works just fine, by the way).
Shows up as SLI in the bios, nvidia drivers says i can enable SLI, i do so, asks me to reboot, that i do. Then, when i try to run a game, any game, it does try to run in alternate frame rendering, but only 1 graphic card is at the job, so i get a flicker between image/gray screen.
Has anyone modded an ultra-d and actually used it to do SLI without a bridge?
Toms hardware did it http://graphics.tomshardware.com/gra...rmance-30.html (bottom of the page) but i cannot for the life of me get it to work.
Sounds like a driver issue but i tried from 66.74 to 71.84 71.90 and still no luck.
My dfi just took up smoking. I was playing some doom3 thinking how much i was liking the new x800xl. Then there was a small pop noise and the screen froze, i figgured it was just a sound effect when the game crashed. I was wrong. I powered the pc off, then back on and got a good whiff of electrical smoke and flipped the power switch on the psu as fast as possible. Some portion of the cpu voltage circuit decided to take up smoking. I just hope it didn't damage my chip... If it did, there will be another DFI in the fs section after i get the replacement board. :mad: :upset: :flame:
Sounds like your RAM is getting hot. Set a fan blowinng on the modules AND the MOSFET with the heastsink near the memory and try.Quote:
Originally Posted by Razor_cut
fyi guys, I tried the 2-18 bios and everytime I went high fsb's my raptor got corrupted on the install, using bios 2-17 seems to have fixed that issue, no matter how high I oc my fsb now, it doesn't corrupt. 2-17 seems more robust for high fsb oc's, 250-300+ range I am talking about now.
uh?Quote:
Originally Posted by laserbeam
at stock settings and volts my cpu only got 39°C max after MAAAANY hours of prime...
so if u ask me, ur cpu heatsink is on the wrong chip or something ;)
I have got what is described in my signature for about a 1.5 months.
Last weekend a moved the rig for a small lan and i was moving it very careful.
When i arrive my RAID0 Stripe was faulty Posting and flashing read.
Turned off the system for a while and then it was working again.
Then things like this happens:
*lock ups in windows.
*hal.dll Renames to -> HAL.DLL (Wont boot)
*ITE Smart Guardian gets error 48 going into windows (Seems SP2 Solved this)
*Games freezing
*Reboots at logging into windows
*Blue screen at desktop.
*Rundll32.dll failure when clicking "automatic updates" tab in Properties for "my computer" (Seems SP2 Solved this
*NVraid service US crasheh upon login (Seems SP2 Solved this)
*Nvidia RAID0 Flashing red, error. It is always the same drive that is missing. (WD360GD-00FNA0)
Tested ALL 5 drives with Ultimate bootcd and datalife guard "full media scan"
No errors.
This is Stable at P95 for 41 Hours, SuperPI stable 1M, Memtest 86+ v1.51 tes #5 for 25 loops.
I have reinstalled windows XP Pro a couple of times but wont help.
Got service pack 2 and any other critical update.
Norton Internet security 2005 Full update.
Nforce 6.39
Forceware 71.90
Could it be the SATA port for that specific drive?
I switched the shielded +12v molex connected to the on board molex. I placed this connector to have it together with the other shielded +12v that is going for my second 6800 ultra.
The PSU got 2 +12v rails, so is it possible that im now somehow drawing most of the power from one of the rails/weakest one?
Board rev A01
BIOS: 218
I have it 250*11, 1:1
LDT 4
Vcore 1.58
vDIMM 3.3
2-2-2-5-7
ITE Smartguardian showing at IDLE:
Temps
CPU : 43-45
PWM IC: 45
Chipset: 42
Voltage
CPU: 1.58
LDT Bus: 1.16
ATX +5.0v: 5.02
DRAM: 3.26
ATX+12v: 11.96
NB +1.5v: 1.47
ATX +3.3v: 3.2
VBAT+3.0v: 3.05
This is what the setup looks like.
http://web.telia.com/~u31526178/Projects/DSC02502.png
Malmedahl, I had this happen to me with two 36G Raptors in RAID 0. They seem to have loose conectors to my PSU Molex connectors. They are more loose in the cool room in the winter.
Interesting, since when it first happened the computer wasnt FROZEN cold but cold and i have constructed this molex line myself so perhaps it could be something going on there?Quote:
Originally Posted by sauria
Metal expands during warmup so that very little might have changed things.
Nice tips
http://web.telia.com/~u31526178/Projects/DSC02468.jpg