Is my new comp a lemon?

Hey there all,

I've just recently got a computer, pre-assembled and shipped to me. Everything seems to be running just fine, except for high-end graphics stuff like gaming.

Specs: AMD Athlon 64 X2 Dual core Proc 4200+ S939
2 gigs OCE gold 500 ddr Ram
ASUS A8N5X Mobo (Nforce 4 chipset)
VGA E-GEFORCE 7600 GT CO Superclocked card (Via PCI-E)
Samung SyncMaster 740N Monitor
Seagate Barracuda 7200.8 300GB 8MB IDE HD

More precise specs can be found in this thread: http://forums.majorgeeks.com/showthread.php?t=91894

What happens, when playing a higher end game like Oblivion or Black and White 2, or even benchmarking software like 3dMark05, is that the program starts fine, and then after a period of time (usually just a few minutes), I get sporatic freezes or flashes of multicolored graphical artifacts, increasing in frequency and usually ending in a hard freeze or, rarely, a bluescreen stop error.

The bluescreen I saw was a Stop 0x000000EA in NV4_disp. I haven't seen that sceen in the last couple of attempts, though, after trying some things with the display drivers.

What I have done:
- Had someone who knows hardware reseat all hardware and check connections. Also check to make sure the heatsink is properly on the video card. Everything came up green.
- Updated to most recent chipset and video drivers
- Nuked said drivers with driver cleanup pro and reinstalled them
- Tryed playing games with lowered settings and/or with stuff like antialiasing or antisotropic filtering disabled. Little or no noticable change.
- Tryed changing my BIOS overclock settings from "Automatic" to "standard"
- Tryed a checkdisk /r from recovery console (Hey, why not! )

What I plan to do but haven't got a chance to yet:
- Run memtest overnight to find out if the ram is bad.
- Clean install with XP.

Now, I know the obvious thing to do would be to ship the computer back and have it tested/have the card replaced, but that will take time, effort, and I am a very, very stubborn man who wants to fix the damn thing himself if he can.

So, any ideas? feel free to ask for more info, and thank you for reading this book of a post.

 

It's the Nividia Card. sometimes they get weird problems no matter how fast they are. the blue stop error is being cause by the card. note the NV at the end. your best course of action is to get another brand like ATI.

 

Your decribing the classic symptoms of video card overheating,freezing,rainbowing and finally a BSOD related to the nvidia driver

First I'd check what temperature your video card is running at in vid card properties see if its in spec,try removing the side of the case and run a deskfan blowing cool air into the card if theres any increase in the amount of time before the card starts to atifact then you can be sure its the card

You may be able to work something out with the retailer see if you can exchange the card alone,I would also want an upgraded card or threaten to return the whole computer,I disagree with baboomin,stick with nvidia make sure the card you get is SLI capable

Hang in there mate

 

definetly call up the retailer, if you haven't talked to them already.. That is something that should have been checked before you even got the system... I personally run every computer i build for 72 hours trying to push the hardware abit so i know it is going to last. since you just got the system there should be no reason why you can't get a replacement board. I can usually replace parts for upto a year without a problem. good luck.

 

Thanks Ricky, it does sound like the Card. It is running at 47c when idle, at about 57c when playing a game, which seems pretty hot to me.

Probably happened in shipping, since the builder apparently ran the benchmark utility before sending it, but I paid extra for shipping insurance and will hopefully be able to ship the card back for a replacement.

Thanks for the help guys, I really appreciate it! Please let me know if I am off with the heating numbers; my understanding is that over 50 degrees definatley means trouble!

 

That's not hot, that temperature is fine.

In fact, it's pretty good.

Doesn't mean it's not the card though, it could very well be faulty.

 

ATI has crossfire which is the same as SLI.

if you want to have the two cards and not deal with the Nivida problems. no amount of cooling will fix the driver errors they get.

 

It may not be a driver error. There are problems that occur before the crash.

 

ya which lead to the conclusion the card is faulty or overheating or both. but the blue stop error is a driver error. I got the same thing as well as a friend when playing games with Nividia. I switch to using ATI and I don't get problems anymore.

 

I have an Ati and Nvidia system here.

The Ati system throws VPU recover errors all the time, then after a few minutes locks up the machine. Doesn't matter which drivers I use.

My Nvidia machine is golden.

What does that tell you?

The exact opposite you said. Which means jumping ship to ATi is not the answer. Stick with what you got, and make it work. If it doesn't, get it fixed.

 

looks like certain brands of cards don't like us.

 

Yeah, after I posted here I did a bit more research and realized 60 degrees is well within specs. I just haven't had a high end computer in a while, I guess. :s

I still think it is the card itself, though. I ran Memtest overnight, which passed with flying colors, and I get the same thing after clean installing and applying most recent Mobo and video card drivers, so I think I will email the manufacturer later today and find out what they can do for me!

 

where you bought it from will make all the difference with that avenue.

 

I have been having the exact same problem you have been having. My specs are:

TruePowerII TPII-550 ATX12V 550W power supply
AMD Athlon 64 X2 3800+ Manchester 2000MHz Dual Core Processor
GeForce 7600GT 250MB x16 video card
ASUS A8N-E nForce4 Ultra ATX AMD motherboard

I have the crazy color polygons when I play Morrowind (even after using imagecfg to assign core 0 to the game exe), I have random pausing/freezing when I play Civilization IV (again, even with assigning core 0), and Diablo II freezes 100% of the time (messing with core affinities corrupts the game file so battle.net can't recognize the version number). I've run memtest for 24 hours on each of my two RAM sticks, all passed 100% of the time. I also rarely get the blue screens with the following error:

STOP 0x000000EA THREAD_STUCK_IN_DEVICE_DRIVER

Also, when running eventvwr.msc, I have a TON of logged errors. I've listed the ones just from today (no programs were running):

Error code 100000ea, parameter1 88b18840, parameter2 89bcef60, parameter3 bacebcbc, parameter4 00000001. (Source = system error)

Error code 1000008e, parameter1 c0000005, parameter2 bfaac308, parameter3 adc20aec, parameter4 00000000. (Source = system error)

The driver nv4_disp for the display device \Device\Video0 got stuck in an infinite loop. This usually indicates a problem with the device itself or with the device driver programming the hardware incorrectly. Please check with your hardware device vendor for any driver updates. (Source = nv)

The driver detected a controller error on \Device\CdRom0. (Source = cdrom)

Note that all these errors were after all updated drivers were installed. This machine was custom ordered and built, and I've had two different network managers dismantle and reassemble my entire computer from scratch, so none of the connections are incorrect. I'm getting from the rest of the posts that this is most likely a problem with the video card itself? My video card runs at 48C even after letting a game run for 14 hours overnight, so I'm sure that's not the problem. Is my DVD-RW drive also faulty according to the error messages above?Thanks in advance for any help.

 

Gecks, you updated the chipset. What about the BIOS?

What about you dumediat?

 

BIOS is up-to-date, as well as the chipset.

 

Unless it doesn't interact well with your board.....
PERIOD.
All brand layalties aside, use something that works. Obviously Nvidia isn't working very well for this guy. Does it mean taht he should jump ship to ATI? No.
however, does it mean ATI will not work either? No. My advice is to root out the problem first. Such as, is there a friend's computer where you ca try out your video card/RAM to make sure all are working optimally?

At first reading I almsot said out loud, "Heat problem" but your card seems to be running at pretty normal temps. With that said, what is the temp of your CPU at idle/load? Please let us know?
I am almost doubting that it is an heat issue at this point, but let's find out before you get that Vcard replaced (which sounds like a good idea)

Sorry I can't be more helpful,
Randy O.

 

Make sure it's not the manufacturer of the card that's the problem. I've got a Sapphire manufactured ATI card, and never had any problems. Perhaps that manufacturer does something different with your cards that others wouldn't. They shouldn't, but shouldn't doesn't mean doesn't.

 

You could DL SiSoftware Sandra and firstly start the "Environment Monitor" setting the increments to say 10-20 sec, this will monitor ALL system sensors (CPU, HDD, mobo and if avail I think GPU etc) with this running now you can try the "CPU Multi-Media Benchmark" to push the graphics compnents of your system and you could also try the "Burn In Wizard". This will allow you to reflect back on the systems sensors over time. There are a multitude of benchmarks with this app for testing any and all components of your system and best of all ITS FREE:

http://majorgeeks.com/download.php?det=4664
Latest

http://majorgeeks.com/download.php?det=92

The options avail within this app are amazing and you almost need a degree to get the most out of it.

 

Idle temp = 30C, temp with programs running briefly = roughly 35-38C (huuuge case with 3 fans ), temp with programs running overnight about 45-48C.

 

Well, it's definately a signature error with the Nvida cards, the only question is if it is a physical issue with the card, the environment, or a driver issue.

On my system, card is currently in transit for an RMA, I will let you all know how that works out.

To answer Hipster's question, no I haven't updated the BIOS of the card or the MoBo, but both are new enough that I would be very surprised if it is a compatibility issue. I have tried killing overclocking and "PEG" (some sort of PCI-E acceleration, I think) in the BIOS, nogo.

I would definately imagine that a video card could be busted but still not overheat. If anything, the ultra-low temps might be a sign of trouble...

Anyway, like I said, I will let you guys know if a new card makes a difference.

 

I'm gonna call EVGA tomorrow and see what they can tell me. I'll pass what they tell me onto you, gecks, because it sounds like our problems are very similar. I'm actually hoping its a physical problem with the card itself, rather than a compatibility issue. That way, it is an easy fix. I've had this thing for almost a month now, it would be nice if it would work for once...

 

I hear that! These rigs aren't cheap, so it is pretty dang frustrating when they don't work right.

Let me know what EVGA says for sure, dude, there may be some simple thing we've both been over-looking.

BTW, is your card a factory OCed or "superclocked" version? I've been kind of wondering if going superclocked was a mistake...

 

Mines the OCed version, but I've turned it off the second I started having problems with it (5 minutes after turning it on). I had the AI booster on when I first set up my BIOS, and when I tried to intall windows Xp pro, it froze 7 times during installation. When I finally figured out what was wrong, I reset the BIOS, turned AI off, and installed windows with no problem.

Edit: I just got an e-mail back from EVGA, and they said to check the +12V rail to make sure it doesn't drop below 11.75V. Mine's running at 11.968V, so I don't think that is the problem. Check your's to make sure, gecks.

 

Sorry, to double-post, but I thought I'd update you with what I've been doing. I called eVGA, and after about 5 minutes on the phone they gave me an RMA and had me send in the video card, which I will do in the next couple of days. Hopefully, this will solve the problem. I'll let you know what happens after eVGA receives the card.

 

New card arrived today via RMA. Popped the card in, same problem.

Checked the 12 volt rail, did not drop below 11.95V.

I'll admit, at this point I am pretty much at a loss, probably going to end up returning the whole computer...

I hope dumediat has better luck than I do, and if anyone has any other ideas, please chime in, willing to try just about anything at this point.

 

i had a problem like that it was a driver issue
i downloaded nvidias beta drivers and it worked maybe you should try that

http://www.nzone.com/object/nzone_downloads_rel70betadriver.html

 

Beta drivers eh? Why not, everything is worth a try at this point!

I was in contact with the Manufacturer, they recommended a BIOS update (which was already recommended here, I know). I had been holding off due to worries that I could pooch the Motherboard and void the warrenty (just general wussiness, I guess!), but I guess if the manufacturer tells me to do it it is OK.

Failing that, the manufacturer suggested I RMA the Mobo AND the video card (AGAIN! ), but that may be more trouble than is worthwhile for this...

 

Sorry to reply my own post, but to update, the BIOS flash went fine but did not help, and the beta drivers (actually release to the public as full drivers on the 23rd) also did not help.

At this point, it looks like I am going to either look into returning the whole computer or just break down and take it to a tech shop.

Thanks again for all your help, guys, I really appreciate all of the suggestions. Will let you know if I ever figure out what the problem is!

 

Well, I finally got my RMA from EVGA, and I've been having new problems with it. It doesn't happen when I'm playing games, though, just when I'm on the internet. I've gotten the following BSOD twice within 10 minutes, but not since then:

MACHINE_CHECK_EXCEPTION STOP 0x0000009C (0x00000004 0x8054D5F0 0xB2000000 0x00070F0F)

Also, I've had my computer freeze twice with no other errors, and my monitor completely go into power save mode once. I've check Event Viewer, and there are no error messages in there.

I called EVGA about this, and they said it might have something to do with DirectX v9.0c, as there have been several problems with it. The tech mentioned that I should re-download the newest version of DirectX. Once I did that, I stopped getting the BSOD, but my computer now crashes now and then as described above. He also mentioned running dxdiag and disabling DirectDraw and Direct3D, which I haven't tried yet.

Does anyone have any advice for me?

 

Yes what this looks like to me like "hexidecimal refrence address" errors, basically address's on your RAM modules and that data is becoming corrupted. You may have a failing RAM module that is causing you grief.

What you can do to test for this is to download either memtest86, or 86+ or memtest from 1 of the 3 links below ar even better still if you have speed internet DL UBCD (THE BEST OPTION POSSIBLE) from this site and will have all of these apps plus many more.

http://majorgeeks.com/download.php?det=1247

http://majorgeeks.com/download.php?det=4226

http://majorgeeks.com/download.php?det=350

If you have a floppy copy only 1 of the files onto it, check in BIOS to make sure floppy is selected first and away you go. If no floppy goto memtest86+ (via google) and read how to burn the .iso to CD, once again going to BIOS and making sure CD/DVD is before HDD in boot-up seqeunce. Post back if any ques. Run at least 3-4 passes!

 

Actually, I ran memtest on each stick of RAM seperately for 20+ hours a piece, and found nothing wrong. That was actually the first thing I thought of, but unfortunately it didn't fix anything...

Anything else?

 

Jeepers only the obvious things of uninstalling and reinstalling the Vid Card drivers and DL the latest AGP (GART) drivers for the mobo system chips.

 

Well, I did the first thing you said with the video card driver, and I run an ASUS m/b, not a Mobo, so I guess I'm out of luck...

I'm giving my card to one of my more hardware-savvy friends to see if it works in his computer. I'll let you guys know how that goes. I guess if it doesn't work in his computer, I'll have to send the RMA card back to EVGA, and I'll be damned if I pay shipping this time.

I'm actually hoping that the card doesn't work in his computer, because then I know it's the card causing the problems. If it isn't, I'm fearing that it may be my m/b, which means tearing the whole comp down and sending that thing back...

Wish me luck!

 

Just to update, after a fair amount of hair pulling, I found I could stabilize card just by lowering the overclock. I was very leery trusting it because of dumediat's experience (underclocking worked, but only for a while), but it seems to be stable and working fine. I will post a entry in the overclocking section to see weither I am severly underclocking my card, but I'm willing to live with the way things are now.

I definately think that alot of people with GEFORCE 7600 or 7900 series cards are having trouble with ASUS motherboards. I wonder if it is an obscure BIOS setting, a minor incompatibility, or just dumb luck? It's hard to say, because freezing and NV4_disp errors are kind of standard "something is wrong" type errors with NVIDIA video cards; it's sort of like "Page Cannot Be Displayed" in Internet Explorer; almost ANYTHING can cause it...

Anyway, thought I should post what worked for me, fingers crossed that it keeps working!

 

Well your specs are great, including your video card. Maybe your video card isn't good enought for the game (which I think is unlikely) or the card can be broken. I bought a computer that was already assembled and it was also broken. You should have someone to look at the video card.

 

Well, as far as an update for me, part of the problem was solved with re-downloading DirectX 9.0c. After I did that, I stopped getting MACHINE_CHECK_EXCEPTION errors, but my computer does lock up occasionally. I gave my graphics card to my friend, and it worked great, so the problem doesn't lie there. (unless, like gecks said, geforce cards are somehow incompatible with ASUS m/bs)

Another friend was having similar lockup problems with his computer, minus the blue screens, and his problem was fixed by getting a new m/b. It had something to do with the m/b not holding a charge properly. I will call ASUS tomorrow to try and get a replacement. Once that happens, I will test it out and let you know how it goes.

 

Might I recommend getting a PhysX card later down the line, when you fix your problem... I have one with 2 evga 7900gt superclocked in sli and the gaming and reliability on everything is amazing... However, I have them on an nforce board... so I guess I was lucky in that I don't have any clashing with my system... The card takes a great strain off of the cards... it is a bit pricey.. but well worth it... with the physx card and the two cards together I am getting al least 70 gigs of bandwith.... when they are clocked in standard mode... I have not even tried to overclock yet...

 

70 GB's eh? Where and how are you measuring that, and which games and applications can take advantage?



What's the point of getting a PhysX card now or in the near future?

Hardly any games support them, and although support is being promised, it hasn't happened yet.

And the ones that do, don't really have any performance advantage.


Anyway, they are designed to improve graphics performance, not remedy bugs.

 

OK, it seems that I finally got my system to work properly. Here is what I did...

1. I got an RMA card from E-VGA. This got rid of the 0x000000EA BSOD and the errors in the Event Viewer logs. This was apparently a memory error with the card itself, and should be resolved by the manufacturer.

2. Re-download the newest version of DirectX 9.0c from Microsoft. I guess there are some problems with this version, but they will be resolved in further updates in SP2.

3. DELETE THE OLD DRIVER!!! Then, re-install the driver from the CD that came with the card. I am currently using v 6.14.0010.8391 and it has been working fine. DO NOT UPDATE TO THE NEWEST DRIVER ON THE E-VGA WEBSITE!!! This causes a problem with the card, and causes MACHINE_CHECK_EXCEPTION BSOD's. At least, that has been my experience. Also, if you roll back the driver, and you are using an nForce m/b chipset, this may cause problems with your ethernet connection. In this case, you need to uninstall the driver, disable the card in your BIOS, restart the computer, enable the card in BIOS, and then re-install the driver from the CD that came with the m/b.

So far, I haven't tested overclocking with my system under these settings, and I really don't intend to yet until I have about a month without any more related errors/BSOD's. So far, I haven't had any errors in Event Viewer either. I'll let you know if I come across any more problems, though. I hope this helps, Gecks!