Stuttering P5N32

[frosty024]

Having an issue where the system freezes for 5-10 seconds then resumes working, most noticeable in games (of course). Really wrecking havoc when trying to play racing games. Any ideas as to what this could be? I am NOT doing any everyclocking, just stock. Average temp of CPU 100ºF, motherboard 95ºF (dunno what celcius would be).

Rig:
Motherboard: Asus P5N32-SLI Deluxe
Processor: Intel P4HT 3.20GHz
RAM: 2GB - 2 x 1GB (Corsair DDR2-667)
Video: eVGA Geforce 7600GTOC PCIe x16
WD 80GB IDE 7200RPM Hard drive Master
Maxtor 20GB IDE 7200RPM Hard drive Slave
BenQ DW1640 IDE DVD+RW Master
Samsung SW240B IDE CD+RW Slave
Rosewill RP500 Power Supply (500watts)

 

[frosty024]

Here's a little bit better of a description as to what I'm seeing...

I am running an eVGA 7600GT-CO and it does not have any power inputs on it. For the first 10 minutes or so, everything works great. Then, every 2-3 minutes or so, games freeze up for about 30 seconds, then return to normal (sometimes with bad distortion/colors). Now, I have a mystery Molex plug on my Asus P5N32SLI-Deluxe motherboard. It is literally right next to one of the PCIe16 slots. Also, I see no square 4pin plug on my motherboard. I looked in my motherboard's manual and it makes no mention of that Molex connector, nor is it labeled on the board itself. Should I be plugging a lead from the power supply in that Molex plug to power the video card? Could that be why I'm experiencing this?

 

[bobonut2001]

For an pcie video card there is normally an pcie power connection which i am assuming you plug in. Double check that you plug in the pcie power connection to your video card.

This problem seem like the video card is overheating because the fan on the video card is not working maybe due to lack of power to turn on the fan.

The 4 pin plug on the p5n32 sli is used only if you have an SLI setup which mean you have TWO video card. Read the ASUS manual. YOu need to plug this 4 pin supply if you have two video card.

I also have a p5n32sli + evga6800gs CO. It's working fine and i am not plugging in 4pin plug.

Also double check if you are plugging in the 8 pin EPS on the motherboard.

Also are you setting everything on the mobo to AUTO? P5n32sli does undervoltage the CPU. I got my p5n32sli stable when i set my bios to manual and set vcore, vmemory, NB and NB to higher voltage.

 

[TabrisDarkPeace]

http://www.intel.com.au

Download & Install latest chipset drivers for the Intel 900 series chipset.

Once done disable AI NOS (and related features) in the BIOS, if enabled (often by default, user unaware), they cause problems with timing sensitive code, such as racing games and and applications where precision is required.

The mainboard manual should cover the AI NOS and related features.

Asus do BIOS updates to their boards that often disable the troublesome features btw. Updating the BIOS may fix it, assuming default settings and reset and were changed since the factory BIOS. (With Asus they release like a BIOS a day for these reasons).

Check for updates / patches to the game(s) in question.

Download & Install "PageDefrag" from SysInternals (defrags fragmented pagefiles, which defrag can't do). I assume you already defraged the disk before posting.

 

[frosty024]

the card does NOT have a power connection it, simply drawing from the pcie slot. my mobo does not have a 4pin connecter other than the EZPLUG/Molex next to pcie16 slot #1.

fan DOES run. the card is idling around 52 degrees celcius. under load, it comes nowhere close to the "cause for concern" temperatures.

i have all latest drivers for bios, and have tried multiple old current and beta drivers for the card. my mobo has nvidia chipset. i disabled auto-clocking in the bios (which meant setting it to manual and putting in the numbers myself). it didn't help any. games are up to date.

my mobo idles around 88 degrees fahrenheit, 95 under load.
cpu idles around 92 degrees fahrenheit, 110ish under load.

 

[TabrisDarkPeace]

I normally never suggest this, but, consider disabling the pagefile to see if the problem goes away.

It if does it points at either low HDD performance (usually driver) or a fragmented pagefile.

Since most DEFRAG tools won't touch the PageFile then: Download & Install "PageDefrag" from SysInternals (defrags fragmented pagefiles, which defrag can't do). I assume you already defraged the disk before posting.

 

[frosty024]

Disabling the pagefile didn't do anything. Yes, I have defragged.

 

[TabrisDarkPeace]

Try disconnecting your 'Maxtor 20GB IDE 7200RPM Hard drive Slave' HDD, and rejumper the Western Digital 80 GB as Master only (if required).

Then see if it goes away.

Is it really 20 GB @ 7200 rpm ?

 

[frosty024]

Also didn't do anything. Yeah lol 20GB, it's just for mp3's.

I'm starting to think it's the vid card, just because this is only happening in 3d games (Oblivion, NFS Most Wanted). Initially I thought maybe my power suppy was screwy, so I disconnected everything that would draw power, aside from the USB keyboard/mouse, fans, and the master HD and DVD. Still had the same problem.

 

[TabrisDarkPeace]

Here's a little bit better of a description as to what I'm seeing...

I am running an eVGA 7600GT-CO and it does not have any power inputs on it. For the first 10 minutes or so, everything works great. Then, every 2-3 minutes or so, games freeze up for about 30 seconds, then return to normal (sometimes with bad distortion/colors). Now, I have a mystery Molex plug on my Asus P5N32SLI-Deluxe motherboard. It is literally right next to one of the PCIe16 slots. Also, I see no square 4pin plug on my motherboard. I looked in my motherboard's manual and it makes no mention of that Molex connector, nor is it labeled on the board itself. Should I be plugging a lead from the power supply in that Molex plug to power the video card? Could that be why I'm experiencing this?

Contact ASUS Technical support regarding that connector.

Are you are 100% certain that AI NOS and PEG Link Mode are Disabled / Set to slowest setting ?

Could also be video card overheating, tried underclocking it ?

You don't have a joystick or steering wheel + pedals installed do you ? (Can cause stuttering, but if it is affecting the video card )

 

[frosty024]

Yeppers, I am 100% sure. I don't think it's overheating... It's only running around 60 degrees celcius (according to NVTray anyway) under load.. I tried to underclock it, but no matter what I try to change speeds to, it tells me I can't do it because it "failed test". That happened in both Coolbits and Rivatuner. I don't have and joysticks/wheels.

 

[frosty024]

For those who are monitoring this thread:

After tearing my hair out all week on this, I decided to just RMA the video card and see if a new one would make a difference. Upon packing the card into it's original box, I found a diode in the box! It must broken off during shipping, and I'm guessing it's the culprit! I'll update once the replacement arrives.

 

[frosty024]

Just a quick update:

After a nail-biting week-long wait for my RMA w/Newegg, my replacement video card arrived. I shoved the sucker in and *poof!* everything works now! wooot!

New question though:

This is my first build, and I'm looking for temperature guidelines.. Any ideas as to what my components should be at when idle and when under-load? With my fans at 80% of full speed, I'm seeing the CPU idling around 95 degrees when my apartment is in the upper 70's, and the mobo idling around 92 degrees. Under load, they are appoximately 115 and 100 degrees, respectively.

Asus P5N32SLI-Deluxe
Intel P4HT 3.20GHz (640 Prescott)

 

[Psimon]

Do NOT trust NvTray's temperature readings.

I did, and it led me on a month-long wild goose chase trying to troubleshoot what turned out to be an overheating 6800GT AGP (and heat was the first thing I and the card vendor expected).

I don't know if its because the chips rapidly heat up & cool down, NvTray is just retarded, or what . . . . my problem involved a system-wide crash, so I did immediate rebootages to get temp readings, anyway (and that gave the chip at least 15-30 seconds to cool down). Still, NvTray's readings were wildly off of what later diagnostics revealed.

I finally found the problem (yes, it was heat) by using Rivatuner. It not only has a nice feature where you can monitor FPS, Core, Memory, and Ambient temps while playing games, it can also create a diagnostic log you can review later (very helpful since my board was crashin' the entire box).

Oh, and some use it for OC'ing . . I actually first got it to underclock my GPU to see if it'd fix my crashes (and no, it didn't).

Anyway, just a lil' rah-rah here for Rivatuner. Its got some good features and you can't beat the price (free) . The only way I found that was better to get temps was to use a laser temp reader.