8800GT causing improper boot

I'm not sure if this is the problem, but I've narrowed it down to it. When I press the power button on my computer, no boot screen or video appears, I can hear my 8800GT fan spin at 100%, but it goes longer than usual, 10 seconds normally, then the computer shuts off immediately. 5 Seconds later it attempts to boot again, and does the same thing, over and over. What I did was unplug the power, and fiddle around with the 8800GT, take it out and put it back in, fiddle with the cords a little, and then the computer boots fine. I don't know what the problem could be. sometimes it boots perfectly. Anyone know what the problem could be, and a possible remedy?

Hardware:
Corsair 520HX psu
dual core e6750 with ninja hs
gigabyte GA-P35-DS3L
2gb gskill ram
bfg 8800gt
120gb WD
XP Pro

Also to add, i recently installed Vista on another partition i made for dual boot. if that has anything to do with it

Thanks
mackry

bump, please anyone? this is driving me nuts

Does that board have another PCI-E slot? Try changing slots, and try booting without the card in at all and see if it doesn't power off immediately. Do you have another PCI-E power connector you can use? Basics of troubleshooting my friend .

No it's a single PCI-E board. But I'll try taking it out completely.

Okay, I took out the card completely, and it seemed to boot properly (even though I had no video to tell, however it did take me to the "do you want to boot windows normally or in safe mode" menu so that's why I guess it booted right.) I also switched out the PCIE power cord with the other one that came with the Corsair hx520w psu.

Okay, I tried booting it this morning, and it happened again. It seems to do this after being off for a longer period of time, maybe after it cools down. I tried taking the card out completely again and it started and shut off even faster. I don't know what the problem is, every cord is in tight. It's like it doesn't have enough power, but that's hard to believe with the hardware : psu ratio

Do you have anything overclocked? I have that board and CPU and if I overclock to some settings it doesn't like it will try to boot and fail, then tries to boot with secondary options and fail (because I don't set option 2 in the BIOS ) and then finally will boot at defaults.

Make sure all your CPU and memory settings are correct.

Do you have another powersupply to test with? Corsair makes some good ones, but you may still have a defective one. If not, return the card for a new one, that will eliminate one potential.

I don't know how much BIOS tweaking you've done, and I know it can be a pain in the butt, but try resetting the CMOS.

I had the same issue here. The GT hype got to me and I picked one up at Best Buy. In my perfectly stable computer that's had several video cards in it including a GTS, this card would randomly go blank video output (machine still working) and eventually after one such blanks it reverted to exactly what you're getting; that is, 100% spin and no boot.

Promptly returned to BB and now happily back to my trusty (and quiet) GTS.

The only thing I touched in the BIOS was disabled the Auto CPU fan so it spins at full speed. I have nothing overclocked. I eliminated my video card being the source of this problem because I completely took it out and it still does it. I really hope I didn't get a defective PSU. Everyone i see on this board has this one. Why me!?

Test the video card in another system......there are reports randomly of this problem, although they are isolated.
Put in another video card and see if you still have the same problem with your system.

I had it in my old rig (one in my sig) for about 3 days before I put my new one together. Everything worked fine. I've already decided that it can't be the video card because I took it out completely and it did the same thing.

Sounds like a Gigabyte cold boot reset. PSU might be not be able to supply enough current on all rails on power up. There is a huge current surge from the PSU whenever the system is started. Make sure you have your 12V rails sorted between devices. You can also try resetting the BIOS to optimal defaults and see if that makes any diff.

i8088 said:

Sounds like a Gigabyte cold boot reset. PSU might be not be able to supply enough current on all rails on power up. There is a huge current surge from the PSU whenever the system is started. Make sure you have your 12V rails sorted between devices. You can also try resetting the BIOS to optimal defaults and see if that makes any diff.

Click to expand...

What do you mean by "Make sure you have your 12V rails sorted between devices." ? Also, how would I go about reseting the BIOS? Should I try updating the BIOS if there is one? Thanks

Most PSUs have multiple 12V rails. Your model seems to have 3. Make sure your splitting the load between all the 12V lines. Ie, If you have multiple PCI-E power connectors, use one that is not sharing with the mobo/SATA/etc. Most of the time they are labeled. On my PSU the mobo / sata gets fed from 12V1, with a current limit of 18A. It would not be a good idea to use PCI-E AUX from 12V1 but rather 12V2 rail that is less taxed.

The BIOS reference was to eliminate the possibility of an erroneous value in your BIOS causing the computer to have a cold boot problem , ie insufficient vcore, memory speed / cas errors.

I would test the system with the minimum amount of components installed, ie CPU/ 1xDDR2 /PSU / then add components one at a time to make sure its not the PSU failing or a current / rail problem.

i8088 said:

Most PSUs have multiple 12V rails. Your model seems to have 3. Make sure your splitting the load between all the 12V lines. Ie, If you have multiple PCI-E power connectors, use one that is not sharing with the mobo/SATA/etc. Most of the time they are labeled. On my PSU the mobo / sata gets fed from 12V1, with a current limit of 18A. It would not be a good idea to use PCI-E AUX from 12V1 but rather 12V2 rail that is less taxed.

The BIOS reference was to eliminate the possibility of an erroneous value in your BIOS causing the computer to have a cold boot problem , ie insufficient vcore, memory speed / cas errors.

I would test the system with the minimum amount of components installed, ie CPU/ 1xDDR2 /PSU / then add components one at a time to make sure its not the PSU failing or a current / rail problem.

Click to expand...

Okay, I've tested everything with the minimum amount of components, CPU, 1 stick of ram, the PSU; It failed again. Each component is using its own power cable. 1 for GPU, 1 for dvdrom, 1 for hdd, and 1 for the fans. How can I reset my BIOS?

Swap sticks to make sure its not a memory issue. Gigabyte bios should be F2 at startup, then select "restore optimal settings". Save and reboot. If its doing this with only those basic parts then your mobo might be toast.

Swap sticks to make sure its not a memory issue. Gigabyte bios should be F2 at startup, then select "restore optimal settings". Save and reboot. If its doing this with only those basic parts then your mobo might be toast.

Click to expand...

I tried both memory sticks. I've just updated my BIOS to the newest version and it just booted up fine; doubt it will fix it though. I've been searching this all over Google, even this forum, and apparently this cold boot issue is a plague among many of these P35 DS3 boards. I guess if all comes to an end, I'll RMA it and get another board. Anyone recommend another with my gear? Thanks

Sounds like it oculd be like the old cold boot problems nvidia had in the past with different cards.

RMA the card.

Two weeks and many US dollars later, I have attempted to upgrade my XFX 7600GT with eVGA 512-P3-N802-A1 GeForce 8800GT Superclocked 512MB:

Started with Vista Business 32-bit/XP Pro SP2 32-bit Dual Boot original configuration, components either from Microcenter or newegg, all Retail parts..

Intel E6750 2.66 GHz Dual Core, 8x multiplier w/ 333MHz FSB
* Zalman CNPS7700 AlCu CPU Cooler, applied AS5
MSI P35 Neo-F Motherboard, MS-7360 (Default v1.0 BIOS settings)
* Video Card installed in single PCI-E x16 slot, v1.0a according to this link)
http://www.msicomputer.com/product/p_spec.asp?model=P35_Neo-F&class=mb
2 x 1GB Patriot DDR-2 800MHz PDC22G6400LLK (Default SPD setting)
Antec EathWatts EA500 ATX12V 2.0 500W PS in Antec Sonata III Case
* 12V1: 18A Max -> CPU1
* 12V2: 18A Max -> Video Card, MB Resources, etc.
36GB SATA-I HD Western Digital Raptor
* Vista Business 32-bit, Logical Drive D
500GB SATA-II HD Western Digital
* XP Pro SP-2 32-bit, Logical Drive C
External Samsung USB 2.0 CD/DVD
Samsung 22" LCD Monitor

Swapped out components, one by one...
1. Replaced original MSI Retail MB with identical model from Microcenter
2. Replaced CPU with Retail E6400 2.13 GHz from newegg
3. Replaced Antec with 700W OCZ GameXStream OCZ700GXSSLI from newegg
* 12V1: 18A Max -> CPU1
* 12V2: 18A Max -> PCI-E1, 6-pin connector (Video Card)
* 12V3: 18A Max -> MB Resources, etc.
* 12V4: 18A Max -> PCI-E2/CPU-2, 6+2 connector (Not Used)
** Tried 12V2, then 12V4, then the connector provided with the 8800GT
4. Removed Patriot Memory and installed 1 x 1GB G.Skill DDR-2 800MHz F2-6400CL4D-2GBPK, default SPD timings for 667 MHz.
5. Switched to Viewsonic 19" LCD monitor

Similar symptoms as original User Post. At the beginning of each config, I get the loud 100% fan startup, goes into XP and/or Vista one time with blank screen, then after soft reboot, the system in any config will Post, will go for ~3-5 seconds then reboot, even when I hit the power button on the Case. Forced to manually turn off the PS switch.

Sometimes get into MB BIOS just before next reboot. Tried cold start, disconnected all devices (interal HDs, external CD/DVD, etc, just the USB keyboard), wait overnight, etc. And just to make sure after each failed config, I replace the 8800GT with the 7600GT and everything runs fine.

Is this a Bad card (from newegg, maybe not), MB incompatibility issue, or a Video Card design / 2.0 specification / manufacturing problem? Willing to roll the dice and install everything on my original Asus P5B-VM motherboard in my Antec Sonata III Case, but after reading about others' experiences, PCI-E 1.0a may be the problem. However, RMA will definitely rule out the Video Card as the culprit.