Boot problems--no display from 4890

[tortnotes]

So, I've had everything but the power supply for the system in my sig for quite a while now. The original Cooler Master PSU was a dud and I was getting the runaround from them on RMA.

I just bought a Rosewill PSU to get the rig running for now.

Everything was already assembled, so I connected the new PSU and away we go... not.

When I power it on, the POST code screen runs through its checklist, and then repeats the following sequence: 68, 69, flicker, F3. Then it flickers and repeats. The flicker between 69 and F3 looks like A8 (I used a camera to make a video, then found the single frame where the code was displayed)

Specs are as follows:
i7 920 D0
EVGA x58 Micro mobo
6GB (3x2GB) OCZ Platinum DDR3
Asus 4890
WD HDD
Generic DVD/RW
Rosewill PSU
CM Storm Scout case

There's no display output and no beep.

I haven't totally exhausted the standard boot issue checklist, but I wanted to post now to see if anyone had a similar issue and direct my efforts in a particular direction...

First on my list is RAM. I'll update this thread after I've done some testing.

Thanks!

 

[tortnotes]

It's not the memory. Tested each stick in multiple slots to no effect.

After it had been running for a while, the CPU cooler (Stock Intel) was slightly warm to the touch, and the NB heatsink was toasty.

With no memory, the screen flashes 00, E8, 01, 62, (pause), repeats.

 

[tortnotes]

Now we're getting somewhere again... hopefully.

I have removed the videocard and replaced all of the memory.

It beeped (the long, angry type of beep) as the motherboard was trying to detect and initialize the graphics card. It finally settles on error code 7F, which according to my manual means that there was an error and the mobo is waiting for user input before proceeding.

 

[Alvin Smith]

nice GPU !

You do know that it DOES require TWO 6pin PCIe pwr connectors ?

Are they both plugged in?

You did not link the RoseWill PSU ... model ? (That GPU sucks power like no other !)

Reseat the GPU and RAM ... You might (must) have bumped some connection loose or jarred a module, while you were in there working on it.

... Assume you bumped something loose ... even a drive cable ... I do see a lot of these !

In the end ?? Some unrelated connection got jostled.

= Al =

do recheck you psu ratings vs. requirements. Use newegg power calculator ...
... Go to the PSU section ... it is down on the left.

= Al =

 

[Alvin Smith]

Now we're getting somewhere again... hopefully.

I have removed the videocard and replaced all of the memory.

It beeped (the long, angry type of beep) as the motherboard was trying to detect and initialize the graphics card. It finally settles on error code 7F, which according to my manual means that there was an error and the mobo is waiting for user input before proceeding.

Reseat all your ram, yet again ... pull the gpu again (OUT ! ) ... swap the 2 PCIe 6pin GPU pwr cnctrs ...
... Line that GPU up and lean straight down on it ... put some weight behind it (carefully but seriously till it feels like the mobo is about to "give"). ... well ... A Ultra Durable board can be pretty stiff, so ...

... yer call ...

 

[tortnotes]

I reinserted the video card and it seems to beep at the same time, but with a short beep.

It still settles on 7F, and there's no display output on either DVI port.

 

[Alvin Smith]

other x16 slot ?

 

[FALC0N]

nice GPU !

You do know that it DOES require TWO 6pin PCIe pwr connectors ?

Are they both plugged in?

You did not link the RoseWill PSU ... model ? (That GPU sucks power like no other !)

sucks power like no other? How bout some of those Nvidia cards?

 

[tortnotes]

There is a link to the PSU.

It has 41 amps on the single 12V rail, so that should be enough to at least power everything up. I made sure to check at the power supply forum here before I purchased it. It's also 80 Plus certified.

The card has two power connections, one six-pin and one 8 pin. The power supply only has one 6-pin and one 6+2 type pin connection. I'd switch them or try another line if I could.

I have reseated the GPU twice. There's no space between the rear panel bracket and the case where it screws together, so it's not going in any more. It also looks and feels to be securely mounted. Same with the RAM, but I'll try it again.

I'll also try the other PCI-E slot.

 

[Alvin Smith]

Casualty swapping ... if u can ...

try other 6pin type GPU ... if u have access to.

try THIS 4890 in other puter ... sure! ... right ?

= Al =

 

[Alvin Smith]

DO NOT overlook the possibilty of "marginal" power issues (DEFECTIVE PSU!)

... RoseWill is a good brand and they vet their stuff very well but even Corsair puts out a lemon, every once in awhile ... you may simply have a defective PSU (new or not).

Someone coulda fried it and returned it, too.

= Al =

 

[tortnotes]

I wish I had another desktop I could use to isolate the GPU as the problem. :/

I know the motherboard is OK. I know the CPU is OK. The RAM was working too. (these have been tested with another PSU and GPU and I was able to enter the BIOS)

I'm going to pull the motherboard out of the case and breadboard it.

You're right about the PSU. I think it's pretty unlikely that I'd get two duds in a row, but it could happen. If my testing can't determine what's wrong, I'll RMA the PSU.

Also: When I tried using my RAM in the second set of slots, the system rebooted spontaneously every few seconds.

 

[tortnotes]

Well, this is rather depressing.

Now, the system is rebooting every few seconds no matter what. It seems like something is shorting it out, but I don't know what. I have removed everything from the case and have just the motherboard, PSU, CPU, and heatsink set up.
Upon hitting the motherboard power button, the power comes on for a couple of seconds, and then goes off--the fans on the CPU and PSU spin down. Then before they've stopped, they get power again.

The scariest thing? Sometimes AFTER I've turned off the PSU power switch, it starts to spin up the fans again for just a moment. Power stored in capacitors, I suppose.

 

[tortnotes]

Okay... So, I tested the power supply (by using a paperclip to short the blue and gray wires) and it wasn't exhibiting any strange behavior when it wasn't connected to the mobo.

I put the RAM back in and cursed at the thing as it did it's constant reboot sequence. Then I wiggled the integrated power button around to make sure it wasn't causing any problems... and the system is getting up to 7F again. Yay! I think!

I'm calling it a night... I'll probably bump this thread up again tomorrow after work.

Thanks everybody for the replies so far, I really appreciate it!

 

[tortnotes]

So, here's where I'm at now. I'm continuing this thread in case there's someone else with a similar issue looking for reference. I think I've sufficiently narrowed down the problem.

I called EVGA's phone support. They didn't have any suggestions, but told me that my testing had effectively ruled out a bad CPU, motherboard, and RAM.

There are three remaining possibilities as I see it:

1) The power supply is malfunctioning and not delivering enough power to the video card
2) the video card is a dud and needs to be replaced
3) both parts are fine, but the video card is drawing too much power for the PSU to handle.

Fingers crossed for the power supply, and I can RMA it to Newegg easily. I haven't heard a ton of good things about ASUS support and I'd rather not deal with them if I can avoid it.

Then again, if the GPU is bad and they don't have any 4890s laying around, maybe they'd send me a 5850 or something.

 

[Alvin Smith]

I am tracking and will be interested to hear what the fix was ... way to hang.

.. please report back to us ..

= Al =