• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.
  • The forums have been upgraded with support for dark mode. By default it will follow the setting on your system/browser. You may override it by scrolling to the end of the page and clicking the gears icon.

Nvidia GT9600 Crashes often!

Thanks for helping!

I'll look into the retun, if I ran into any problem I'll let you know.

Also if they don't end up accepting RMA's from Canada for some reason.. i can do the RMA from here for you. But any questions just PM me here and I will do everything I can to help you man. :toast:
 
Brand new update! Post has been silent for a good long while.

I managed to get my hands on an older version of the Nvidia graphic card: GT8600. Needless to say that since I'm back, I got the same problems again. All the crashes, all the freezes etc. occurs at the same spot with MORE frequency in the same programs.

Specificly, in Empire Total War, while the crash no longer happens during the grand map scene, it happens for all the battle loading scene. In Voyage Century online, it can't even load the game. Crashes straight to desktop period. I haven't had the time to try any other game at this moment cause... I don't know if it'll be helpful at all finding more programs that crashes with this problem.

As usual... any advice or help is much appreciated. Thanks a whole bunch. Will be checking the RAM chips tomorrow.

A side note:
During the graphic card change over... I noticed that the computer had great difficulty booting up in the first 2 start up. After restarting couple times by force shut down, and insert + reinsert the older graphic card, this problem was elimated... which led me to suspect some problem with the graphic card's socket... I donno. Thoughts?
 
you probably had the old 9600GT which had a lot of problems
 
you probably had the old 9600GT which had a lot of problems

He had the Palit 9600gt sonic. I say run memtest to check the ram. It could also be the motherboard.

Running memtest first though just to make sure.
 
4 memtests are running back at home at 320 MB each (since I have 4 CPU). I'll tell you the results when I get back home. What should I do if there's a problem with the mobo? From the fact that switching video cards still didn't solve the problem, I'm getting a bit worried bout my mobo.
 
Done, 0 error on all 4 tests. Is it also possible that my BIOS setting is messed up?

I haven't dared touch anything there other then the settings suggested by others on this forum.
 

Attachments

  • untitled.JPG
    untitled.JPG
    61 KB · Views: 319
The last crash my computer had was quite bad... couldn't start up the computer untill after 3 restarts... the windows loading screen would freeze giving the "White screen" at start up. Looks like I'll be reformatting the computer soon. I'm still open to all suggestions. I'll also be running the Seagate hddr testing tool to see if there are corruptions.
 
I have sent you a large PM(sorry for the long length) outlining the process. If there is anything else feel free to let me know. Sorry I have been answering by PM instead of the thread. :)
 
I think the psu decided to stop working
 
if strange things happen, and everything is on stock and not tested faulty, then it seems to be the psu... if it starts crashing randomly in windows, then youre probably an inch away from its death
 
Hey, I just got back the mobo from Asus factory yesterday. Honestly, I'm not very happy about the whole procedure. They've never told me how long it'll be until they're finished, and they've never told me anything at all about what they've done with the mobo. From what I can see, they've done nothing other than letting it sit for 7 days in the factory then send it back exactly the same as it was.

I've updated my driver ver. for the Nvidia 9600 GT card and the pc still crashes exactly like how it used to crash.

I've noticed something different though. When I put in only 1 RAM chip (instead of both at the same time), all the mem tests run fine, including the one which failed previously: "prime95". This works for either of the chip, so I dare say that the RAM chips are NOT broken.

One other thing I noticed is that when the computer crashes. Nvidia tried to show an error dialog or a pop-up of some sort, but I can't ever get to see it cause the computer would immediately close and restart. From what I can see in the split second it's something like this.

hxxp://www.experts-exchange.com/Hardware/Components/Video_Cards/Q_23530718.html



Don't know what to do at this point. I'll try out later today with the older graphic card that I had and see if the same error box pops up.

Here's another forum that discussed the similar problem:

hxxp://forums.nvidia.com/index.php?showtopic=95843
 
k nvm I found out why it said insufficient power. I forgot to plug in the 6 prong plug. So that's solved but the crash isn't. Suggestions please.
 
I would definately RMA the graphics card as its the only thing left besides the PSU you have verified to be working by your testing.
 
sure thing, I'll get it done asap. Will post back when I get the RMA back. Any idea why the mem test would pass when it's only one RAM stick but NOT when there are two loaded? Still sounds a bit fishy to me... I mean I've even tried it out with a different graphic card and the problem still persisted... well granted that they were both Nvidia cards, but these are just my wondering thoughts.
 
link to the ram you have.

almost sounds like you're running 2.1v ram at 1.8v
 
link to the ram you have.

almost sounds like you're running 2.1v ram at 1.8v

When it was dropped to 800mhz and 5-5-5-18 timings, it shouldn't have mattered if it was at 1.8 or 2.1.

Op's 12v line is running a little high according to software but can't for sure know unless tested via multimeter.
 
sure thing, I'll get it done asap. Will post back when I get the RMA back. Any idea why the mem test would pass when it's only one RAM stick but NOT when there are two loaded? Still sounds a bit fishy to me... I mean I've even tried it out with a different graphic card and the problem still persisted... well granted that they were both Nvidia cards, but these are just my wondering thoughts.

The memtest issue with two sticks versus one would imply an issue with the Northbridge being faulty or the ram's revision not being able to stable run in dual channel with that motherboard. Or simply the ram needs more voltage to be set in Bios.

If the ram has trouble in dual channel, you can set the ram in the slots in the order that the motherboard "signifies" as running in single channel as found in the motherboard manual.

Almost all desktop 4 memory slot LGA775 motherboards I have used you can put the ram in slots 2 and 4 for single channel, rather than 1 and 3 which is almost always dual channel. Then run memtest and see if it comes back with errors. :)
 
Last edited:
When it was dropped to 800mhz and 5-5-5-18 timings, it shouldn't have mattered if it was at 1.8 or 2.1.

unrelated. They're only guaranteed to be stable at stock voltage - and thats whatever the sticker says.

if its 2.1v 4-4-4-12, they wont bother to test it thoroughly at 1.8v 5-5-5-15 - as long as one stick boots so you can get into the BIOS and set the voltage manually, it doesnt matter to the ram companies. You're undervolting it, so they dont have to make it stable.
 
I personally believe that this is a Northbridge error, but ofcourse I don't knoa if this is what the RMA ppl fixed or not since they've told me nothing of what they had done with the mobo. I will test out the 2 RAM sticks in single channel tomorrow, and can someone explain to me how I can go about doing what Mussels has suggested? Thanks a ton for the instant feedback.
 
go into the bios, turn the ram to 2.1V and turn the NB up 1-2 notches. see how that goes.

since its a gigabyte board, 2.1v on the ram will be "+0.3" and it may give you silly warnings about it being 'dangerous'
 
unrelated. They're only guaranteed to be stable at stock voltage - and thats whatever the sticker says.

if its 2.1v 4-4-4-12, they wont bother to test it thoroughly at 1.8v 5-5-5-15 - as long as one stick boots so you can get into the BIOS and set the voltage manually, it doesnt matter to the ram companies. You're undervolting it, so they dont have to make it stable.

Every kit I've had that is either 4-4-4-12 @800 or 5-5-5-15 @ 1066, Has a spd profile for 5-5-5-18 @800 with 1.8v. I understand what you're saying but when there's a profile for it?
 
o...kay so newest update. In the single channel slots, I've placed the two RAM stick and ran the prime95.exe test. Result:

First time run:
Completes the 1024K test (a total of 6 tests) but computer automatically restarts during the middle of the next test (the 8K test)

Second time run:
Completes the 1024K test ( a total of 6 tests) but computer freezes during the middle of the next test (the 8Ktest)

Conclusion:
Overall this is better result than the dual channel version since this one at least finished one series. Previously the computer would "blue-screen death" at the 4th test.

will be trying out the voltage thing in Bios later.

Mussels, btw your "NB" = northbridge? Is this in Bios too? Can you tell me where it is? Thanks.

I see the "System Voltage Optimzed" section has
System Voltage Control
DDR2 OverVoltage Control <--- this the ram one you want me to add 0.3V?
PCI-E OverVoltage Control
FSB OverVOltage Control
(G)MCH OverVoltage Control
CPU Voltage Control

Oh and yeah it still crashes like there's no tomorrow.
 
Last edited:
o...kay so newest update. In the single channel slots, I've placed the two RAM stick and ran the prime95.exe test. Result:

First time run:
Completes the 1024K test (a total of 6 tests) but computer automatically restarts during the middle of the next test (the 8K test)

Second time run:
Completes the 1024K test ( a total of 6 tests) but computer freezes during the middle of the next test (the 8Ktest)

Conclusion:
Overall this is better result than the dual channel version since this one at least finished one series. Previously the computer would "blue-screen death" at the 4th test.

will be trying out the voltage thing in Bios later.

Mussels, btw your "NB" = northbridge? Is this in Bios too? thanks.

Oh and yeah it still crashes like there's no tomorrow.

yup NB=north bridge and the voltage control will be in the bios :)
 
Yeah sorry I still can't find the one that says NB or NBridge or anything like north bridge in the BIOS setting. RAM voltage has been set to +0.3V. If anyone can give me a screen shot or the detailed direction that would be much appreciated.
 
Yeah sorry I still can't find the one that says NB or NBridge or anything like north bridge in the BIOS setting. RAM voltage has been set to +0.3V. If anyone can give me a screen shot or the detailed direction that would be much appreciated.

G(MCH) is going to be your north bridge voltage.
 
Back
Top