 |
12-29-2004, 08:49 AM
|
#1
|
|
Xtreme Member
Join Date: Nov 2004
Location: Koh Samui, Thailand
Posts: 198
|
I think I've killed my X800XT PE!....
I let AtiTool go to work finding max clock speed over night... When I checked the machine this morning it had rebooted but locked up (no screen) - oh dear me thinks....
Anyway, now in every game and benchmark at stock GPU and CPU speeds the machine will reboot itself after maybe 30 seconds... I've rebuilt the machine twice and it's still the same, even with just the os, chipset and gfx card drivers it will still reboot.
I think I've damaged the GPU! Maybe I was putting too much voltage through?... I thought 1.75v would be fine on water for this card...
__________________
Desktop:
Intel Q9550 SLB8V CPU un-opened in Box
Gigabyte EG45M-UD2H un-opened in Box
8Gb DDR2 Crucial Memory un-opened in Box
2 x 1Tb WD Caviar Green (RAID-1 Storage)
80Gb Intel X25-M SSD (System)
No Graphics Card (yet)
No PSU (yet)
No Case (yet)
Laptop:
Lenovo Thinkpad X200s
80Gb Intel X25-M SSD
4Gb DDR3 Ram

|
|
|
12-29-2004, 08:57 AM
|
#2
|
|
Registered User
Join Date: Nov 2004
Location: Portugal
Posts: 33
|
1.6v it's enough. You don´t need 1.75v to do 630/610.
If you think you killed your graphic, it´s better you test it in another motherboard. Or in another system. The problem could not be the graphic card.You change the AGP frequency (default 66mhz)??
__________________
|
|
|
12-29-2004, 08:59 AM
|
#3
|
|
.
Join Date: Jun 2003
Posts: 3,371
|
Sorry to hear about your card.....these pe cards are very flaky IMO.
I have had two die on me overclocking and 2 arrive doa 
I feel your pain...if your gpu
BTW--1.75v seems like alot for water cooling and everyday volts.
I bet thats what did it.
|
|
|
12-29-2004, 09:00 AM
|
#4
|
|
Xtreme Member
Join Date: Jul 2004
Location: n^n!
Posts: 488
|
I learned it the hard way, that when OCing at the edge, always be around in case the system freezes/locks. I think, when it freezes a couple of seconds are alright before restarting, but minutes/hours in this limbo can cause some components to fail/die. I lost a DFI UT NF3 mobo this way, lucky for me none of the critical/expensive components(not that a DFI UT is cheap, but better that than really expensive RAM, VPU, or CPU).
I think you're right and the card or other component has been damaged
Checked with another vid card and can you RMA the X800XT PE, if that is the cause of the problem? Hope you get it sorted
|
|
|
12-29-2004, 09:08 AM
|
#5
|
|
Xtreme Member
Join Date: May 2004
Location: Sweden
Posts: 138
|
Check so it's really the gfx that have died and nothing else.
1.75vGpu @ idle ~ on the x800 should not kill it if u use good watercooling.
(keeping temps below.. 50c @ load)
__________________
|
|
|
12-29-2004, 09:30 AM
|
#6
|
|
Banned
Join Date: Dec 2004
Posts: 232
|
There are alot of other parts on the card that when stressed to much can die. The vcore regulation isn't designed for the kind of load you placed on it for all hours of the night. Did you have a fan over the front of the card near the "Fet" area buy the vga port?
The core itself Is really the last place you would get a serious problem if the contact was great.. These cores can handle temp's in the 80-90c range and with water and 1.8v @say 700mhz core your still not going up over 70c so Its then other things you need to watch out for. Like Kingpin said you have to be very careful with these cards we also at the shop here have had 1 DOA and 2 die from the current being draw on the card burn up the card badly.
I don't think I have seen anyone Burn out the card's core on a x800 series card Its alwasy a trace, bridge or ddr that goes from what I have observed.
|
|
|
12-29-2004, 10:47 AM
|
#7
|
|
Xtreme Member
Join Date: Sep 2002
Posts: 162
|
Lo, if you ain't going to RMA it, do a OVP mod, classic sign of OVP cutting in.
look it up.
Cheers
|
|
|
12-29-2004, 02:59 PM
|
#8
|
|
Xtreme Member
Join Date: Nov 2004
Location: Koh Samui, Thailand
Posts: 198
|
I'm going to clean the graphite of the resistor shortly and see if that changes anything. If not I'll have to RMA it and get myself another board.
I should have brought the voltage up slightly at each point the overclock failed to get the max o/c like I did with my cpu. This time I just rushed in, got the voltage to 1.75v and then o/c'd from there. Ooopsss. Then again, like Bulldog say's it could be OVP so it might work if I reduce the voltage (touch wood)
I really don't want to pull this card from my rig as it's a lot of hassle draining the system, etc.
__________________
Desktop:
Intel Q9550 SLB8V CPU un-opened in Box
Gigabyte EG45M-UD2H un-opened in Box
8Gb DDR2 Crucial Memory un-opened in Box
2 x 1Tb WD Caviar Green (RAID-1 Storage)
80Gb Intel X25-M SSD (System)
No Graphics Card (yet)
No PSU (yet)
No Case (yet)
Laptop:
Lenovo Thinkpad X200s
80Gb Intel X25-M SSD
4Gb DDR3 Ram

|
|
|
12-29-2004, 08:03 PM
|
#9
|
|
Xtreme Member
Join Date: Jul 2004
Location: Switzerland DownTown
Posts: 229
|
try to reinstall drivers ...
__________________
3000+ Winchester
DFI SLI-D
2x512 OCZ EL Rev.2
OCZ 520Watt
2x6800GT Inno3D SLI
|
|
|
12-29-2004, 08:16 PM
|
#10
|
|
Registered User
Join Date: Apr 2004
Posts: 85
|
might be the atitool max overclocking overnight part, i had atitool go 650mhz+ on the core on a stock x800pro before it crashed and i had to hard reboot it
__________________
A64 3200+ 2.65ghz | 2x512 OCZ VX 265mhz | Connect3D x850xt PE 680/650 | DFI NF4-Ultra
|
|
|
12-29-2004, 08:51 PM
|
#11
|
|
Xtreme Guru
Join Date: May 2004
Location: the cold place Temp: 5C
Posts: 3,813
|
R420 cores don't artifact, they just crash. ATItool can't detect the max stable clocks (yet) because of this. There might be some special code that will make them artifact, but apparantly the pipelines are now very strong on ATI boards and there is something else limiting the cores.
__________________
For those of you about to post:
|
|
|
12-30-2004, 02:22 AM
|
#12
|
|
Xtreme Member
Join Date: Nov 2004
Location: Koh Samui, Thailand
Posts: 198
|
Well I took all the graphite off and core voltage was back to 1.4v, now it works OK!...
The screen jumps about in windows slightly but unticking the alternate dvi method fixes this problem. I had this when I first got the card.
I'm going to start overclocking and raise the core voltage a small bit at a time.
I have a strange problem that I get artifacts within games/benchmarks when I reach my max overclock.... When I run 240 x 12 it's fine but when I go to 250 x 12 I get loads of artifacting on the screen!?... It's strange. I've tried different bios's as I thought the AGP lock wasn't working (It's set to 67) but it's still the same...
__________________
Desktop:
Intel Q9550 SLB8V CPU un-opened in Box
Gigabyte EG45M-UD2H un-opened in Box
8Gb DDR2 Crucial Memory un-opened in Box
2 x 1Tb WD Caviar Green (RAID-1 Storage)
80Gb Intel X25-M SSD (System)
No Graphics Card (yet)
No PSU (yet)
No Case (yet)
Laptop:
Lenovo Thinkpad X200s
80Gb Intel X25-M SSD
4Gb DDR3 Ram

|
|
|
12-30-2004, 05:59 AM
|
#13
|
|
Xtreme Mentor
Join Date: Jul 2004
Location: Ontario
Posts: 2,768
|
Quote:
|
Originally Posted by craig588
R420 cores don't artifact, they just crash. ATItool can't detect the max stable clocks (yet) because of this. There might be some special code that will make them artifact, but apparantly the pipelines are now very strong on ATI boards and there is something else limiting the cores.
|
No clue why people keep saying this. I most certainly get artifacts if doing too high. If going too high on the memory I get black lightning and on the core flashing textures are very apparent and often in more than a few games.
|
|
|
12-30-2004, 07:33 AM
|
#14
|
|
Overclocking X.I.P. Otaku
Join Date: Nov 2004
Location: Kiminomori, Chiba, Japan
Posts: 839
|
cinders, glad to see your card has "recovered"... I'm going to do the pencil mod tonight... it was already a good overclocking card right out of the box (580/600 stable) so I'm going to take it up to 1.5-1.6v (on water) and see what I can get. My goal is a 650 core.... I'm targeting 37000 points in 3DMark2001 so that should do it if my card survives
|
|
|
12-30-2004, 07:42 AM
|
#15
|
|
Xtreme Member
Join Date: Nov 2004
Location: Koh Samui, Thailand
Posts: 198
|
Quote:
|
Originally Posted by gocchin
cinders, glad to see your card has "recovered"... I'm going to do the pencil mod tonight... it was already a good overclocking card right out of the box (580/600 stable) so I'm going to take it up to 1.5-1.6v (on water) and see what I can get. My goal is a 650 core.... I'm targeting 37000 points in 3DMark2001 so that should do it if my card survives 
|
Hey Man, I'm just at work at the moment waiting to go home so I can play about with my card when I get in.
I've already set the memory back to 610 and thats cool, so next I'm going to mod the core to 1.5v when I get in and slowly push the core speed up and see where it max's out. Then I'll push the voltage up say 0.02v at a time - I think I'll stop about 1.6v
I've just ordered myself a prommy mach 2 so i'll be taking the cpu out of my cooling loop meaning my thermochill 120.3 setup will only be cooling my gfx card, so hopefully it might allow me a better o/c...
Good luck with your card!
__________________
Desktop:
Intel Q9550 SLB8V CPU un-opened in Box
Gigabyte EG45M-UD2H un-opened in Box
8Gb DDR2 Crucial Memory un-opened in Box
2 x 1Tb WD Caviar Green (RAID-1 Storage)
80Gb Intel X25-M SSD (System)
No Graphics Card (yet)
No PSU (yet)
No Case (yet)
Laptop:
Lenovo Thinkpad X200s
80Gb Intel X25-M SSD
4Gb DDR3 Ram

|
|
|
12-30-2004, 07:56 AM
|
#16
|
|
Overclocking X.I.P. Otaku
Join Date: Nov 2004
Location: Kiminomori, Chiba, Japan
Posts: 839
|
cinders how many pencil strokes did it take for you to increase the voltage each 0.1v? My multimeter was showing a stock voltage or around 1.38 and as a test I penciled 1597 around 7-8 times and it went up to 1.4!
|
|
|
12-30-2004, 08:14 AM
|
#17
|
|
Xtreme Member
Join Date: Nov 2004
Location: Koh Samui, Thailand
Posts: 198
|
I had to literally pencil mine black to get to 1.75v!... I'm just going to pencil it until I get the desired voltage, or maybe use some isopropyl to reduce it to the desired voltage.
It's just a case of playing around, but I'd aim to go up 1.50 -> 1.52 -> 1.54 -> 1.56 etc..
I found with my 2nd FX-55 that when I increased voltage gradually and slowly I found my best O/C to be at 1.63v and any higher was no good. So more voltage isn't always better.
__________________
Desktop:
Intel Q9550 SLB8V CPU un-opened in Box
Gigabyte EG45M-UD2H un-opened in Box
8Gb DDR2 Crucial Memory un-opened in Box
2 x 1Tb WD Caviar Green (RAID-1 Storage)
80Gb Intel X25-M SSD (System)
No Graphics Card (yet)
No PSU (yet)
No Case (yet)
Laptop:
Lenovo Thinkpad X200s
80Gb Intel X25-M SSD
4Gb DDR3 Ram

|
|
|
 |
| Thread Tools |
|
|
| Display Modes |
Rate This Thread |
Linear Mode
|
|
Posting Rules
|
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts
HTML code is Off
|
|
|
All times are GMT -8. The time now is 01:07 PM.
|