8800GTX SQUEAL!

[Spazilton]

I hate to beat the dead horse here, but stop, take a deep breath, and think about this a moment.

You can have a top-line, name-brand, high-quality PSU.

That PSU can be rated at enough power to light up an African village.

Every meter, oscilliscope, and Star Trek tricorder can show that the output is so perfect, so smooth, so tightly within spec that the very angels in heaven sing at the sight of it.

But until you try a different make and model of PSU, you haven't eliminated the possibility of some offbeat mismatch between a totally new and different GPU and your PSU.

You can shake your fist at the gods and declare that it's impossible, but that's not troubleshooting--it's only denial.

At least one user has done the sensible thing and actually TESTED the hypothesis by replacing his PSU, which fixed the problem. If that had not happened, all the reasons why it shouldn't fix the problem would carry more weight. But because it did happen, it shows us that there's something going on here that isn't in our data banks yet when we try to just think through the problem.

There's no substitute for hands-on troubleshooting because there are just too many unknowns in any complex interactive system. This alternative may not fix your problem, but there's only one way to find out: try it. All other indirect tests and measurements fail to get to the root of the issue.

And if it doesn't fix the problem, don't complain that I screwed you over with bad advice. Just cross that possibility off your list and move on to the next one. Really, friends, this is Troubleshooting 101 we're talking about here.

No what I was talking about is troubleshooting 101.

In case you missed my posts. The only common component to my issue IS the video card.

With the tests I have preformed I can completly eliminate my power supplies... all 3 of them.

Voltage is spot on. Meaning that the power supply is not having issues delivering a constant voltage at a specified load.

Power is clean, Meaning the rectifiers and filtering circuts are doing thier job. Oscillscope proves that, I am getting less than 5mv of ripple on the 12v line going directly into both inputs of the video card. On both the 510 ATX and 1KW. The 750 Silencer due to its lower end design does have a little bit more ripple at full load but still well under ATX guide lines.

Power draw is no where near the rated capacity of any of my power supplies.

This has been duplicated on both a 680i and 975x Motherboard.


This off beat mismatch you describe is frankly a load of bullshit. Solid electronics theory will beat your conjectures any day of the week.

Please do not proceed to lecture me on the steps in troubleshooting.

I have done electronics repair for the greater part of my adult life, and I am 100% confident in my abilities to identify a problem like this.

Blindly swapping components without disproving a component is bad, is troubleshooting for the uninformed.

 

[Spazilton]

ROFL! @ Commander Suzdal.

The round component, is it a piezo speaker? If it was then you would think it would be mentioned in the user manual for the card (I've never heard of an alarm speaker on a video card, but I suppose it's possible).

So, has anyone RTFM??

Could someone take a close-up picture of the suspect component?

Actually I do have to agree with him here it is a Pizeo speaker.

 

[Jomanscool2]

According to mwforbes on the EVGA board, this is what part was making the noise on his card [link].

It may be this, or it may be something else, but thats what the buzzing was on his card.

Re - GoodBoy, I have buzzers like that, this is not the same. It is an electronic hiss more than a buzzer. Unless this buzzer is nothing like normal buzzers. Edit: Unless it is different than what similar buzzers i've seen sound like.

Regarding swaping PSU's, none of my friends have over a 400 watt, or name brand PSU's. I have already RMAed the GPU, and it is shipping off to them today. I could test the board with an aircooled 6800GT, but I doubt that would do anything to the PSU.

Regarding the PCI-E slot not providing enough power, I HOPE NOT! I would HATE to have to RMA my mobo, ripping out my ENTIRE system.. ='(

Well, it would give me time to dremel some holes for wires... and rewire my entire system...

Btw, reason I am resistant about replacing the PSU, is I would litterally have to hold the PSU in my hand, because I have the case wired so well (imo) that I do not want to rip out the PSU... But if replacing the GPU itself doesn't work, I will run to fry's, and pick up a new PSU so I can test it, then return it XD.

I would test the lines of the PSU with a volt meter, but how the hell do you do that when the plugs are plugged into the card?!?!?

P.s., replying to 'This needs a poll now', how would I do that?

 

[Commander Suzdal]

With the tests I have preformed I can completly eliminate my power supplies... all 3 of them.

If you tested with a different PSU powering your system, my comments were obviously not relevant to you.

Voltage is spot on. Meaning that the power supply is not having issues delivering a constant voltage at a specified load.

Power is clean, Meaning the rectifiers and filtering circuts are doing thier job. Oscillscope proves that, I am getting less than 5mv of ripple on the 12v line going directly into both inputs of the video card. On both the 510 ATX and 1KW. The 750 Silencer due to its lower end design does have a little bit more ripple at full load but still well under ATX guide lines.

Power draw is no where near the rated capacity of any of my power supplies.

None of which means dick if the problem isn't being caused by the quality of the power, but by some other issue.

This has been duplicated on both a 680i and 975x Motherboard.

Have you applied your test equipment to the power pin(s) of the 16X PCI-E slot on these boards? For that matter, have you tested the 12V supply on the ATX connectors of your PSUs, not just the aux PCI-E connectors?

This off beat mismatch you describe is frankly a load of bullshit. Solid electronics theory will beat your conjectures any day of the week.

Except for Teknokid who stopped his buzzing by swapping PSUs. Empirical evidence beats theoretical ass-gassing any day of the week, too.

Please do not proceed to lecture me on the steps in troubleshooting.

I have done electronics repair for the greater part of my adult life, and I am 100% confident in my abilities to identify a problem like this.

And yet, your card is still buzzing, while Teknokid's is not.

Blindly swapping components without disproving a component is bad, is troubleshooting for the uninformed.

Reasonable, systematic empirical tests, i.e. swapping components (but not blindly) is a lot faster and easier than breaking out the oscilliscope, especially for people who don't have access to one.

As I said at the beginning of my last post, take a deep breath. Lay off the Red Bull. Whatever.

 

[Spazilton]

If you tested with a different PSU powering your system, my comments were obviously not relevant to you.



None of which means dick if the problem isn't being caused by the quality of the power, but by some other issue.

Lets see. Power supplies while magical to some are nothing more than voltage regulators, converters and filters. If they are delivering the proper voltage, with good filtering and can hold that quality of power at load, than the power supply is fine.

Considering all 3 powersupplies are single rails, aside from the difference in voltage drop from the resistance of the different lengths of cable, it should not matter where in the specific voltage circut you test. And for your information, I tested right at the output of the power supply before the harness and directly on the input of the video card. And yes I did check the 5v and 3.3 volt rails.

Have you applied your test equipment to the power pin(s) of the 16X PCI-E slot on these boards? For that matter, have you tested the 12V supply on the ATX connectors of your PSUs, not just the aux PCI-E connectors?

See above.

Except for Teknokid who stopped his buzzing by swapping PSUs. Empirical evidence beats theoretical ass-gassing any day of the week, too.

Hate to break it to you, a Law breaks your Emperical Evidence. Ever heard of Ohm's LAW. I have provided Emperical evidence, with tests to back up those claims.

Going off my testing, it is most definatly a problem with the video card itself period. It is not the power supply or anything related to it.

And yet, your card is still buzzing, while Teknokid's is not.

Hence its a problem with the video card itself, and not my other equipment.

Reasonable, systematic empirical tests, i.e. swapping components (but not blindly) is a lot faster and easier than breaking out the oscilliscope, especially for people who don't have access to one.

While it is a lot faster and easier, it is not the *Correct* way to do proper troubleshooting. Blame it on a tendancy to do things correctly and not half assed.

As I said at the beginning of my last post, take a deep breath. Lay off the Red Bull. Whatever.

Yea... Whatever.

 

[Jomanscool2]

Sorry to say, last two posters, your posts were completely not beneficiary to the topic, and were complete flames.

Its one thing to say 'just because you tested your psu with oscillator/whatever, it does not disprove the PSU could cause the problem in my mind', its another thing to bring in personal attacks.

We all have different opinions on how troubleshooting should be done. Only way to know which one works better, is to see which one finds the absolute problem the fastest, and cheapest, and proves it by getting it replaced.

 

[Spazilton]

We all have different opinions on how troubleshooting should be done. Only way to know which one works better, is to see which one finds the absolute problem the fastest, and cheapest, and proves it by getting it replaced.

I do have to say you are correct on this. My last comment was a tad harsh.

However when swapping out components fails, and all other avenues have been exausted as in my case, extended and more indepth tests are warranted.

 

[Commander Suzdal]

Sorry to say, last two posters, your posts were completely not beneficiary to the topic, and were complete flames.

Its one thing to say 'just because you tested your psu with oscillator/whatever, it does not disprove the PSU could cause the problem in my mind', its another thing to bring in personal attacks.

We all have different opinions on how troubleshooting should be done. Only way to know which one works better, is to see which one finds the absolute problem the fastest, and cheapest, and proves it by getting it replaced.

At no time in my original post or in my reply to Spazilton's post did I engage in personal attacks. I did, however, exercise my option to defend myself from one.

I do have to say you are correct on this. My last comment was a tad harsh.

Thank you for recognizing that.

However when swapping out components fails, and all other avenues have been exausted as in my case, extended and more indepth tests are warranted.

The point of my original post was that Jomanscool2 was skipping swapping out the PSU because he was convinced his PSU could not be the problem, and that you were strongly advocating that he ignore Teknokid's advice because the PSU turned out not to be the problem for you when you did try swapping it out.

Jomanscool2 did not complete your first step above, and you were saying that was fine. Now do you both understand what I was trying to point out?

 

[GoodBoy]

Lol.....

In Spaz' case, since he's tried 3 different PSU's, 2 mobo's, his problem very likely is the card itself. (Ocillosope not needed )

The component circled in the linked picture is NOT a piezo buzzer, its either a voltage regulator or rectifier. Since so much of the board is covered up by the heatsink, that may not be the actual source of the noise. If it is, the part indicated is likely defective. RMA the card.


These are times when the EVGA cross-ship RMA thing (new deal they got with 8800 and newer cards) is handy. The overnight replacement option was only like $12 or something.

EDIT: Found a picture in this thread http://www.hardforum.com/showthread.php?t=1171151 where at the bottom right you can in fact see a piezo buzzer. This suggests that the majority of the squealing issues will be (insufficient) power related.

 

[LOCO LAPTOP]

well im sorry but it solved my whine

didnt solve mine.

 

[Spoudazo]

Mine squeals when I turn off the PC or restart it, but only for a second or less.

This is with a 750w Toughpower Thermaltake PSU (modular).

 

[Jomanscool2]

Uggh, got the new card setup and running. with orthos and RTHDRIBL (full quality settings, 1920x1124 res).

Still makes a squeal, definately NOT comming from the buzzer, definately IS comming from the card. But this time the noise is much less loud, but that tmay be because all the components are kept cooler with a YL 12CM fan blowing directly.

BTW, the card, after 10 minutes of full load, is running at 50C, when it idles at 40C, while both cores are idlig at 30C (all according to nTune).

Thankfully, this time (at least from what I can remember) the squeal is much lower in volume.

Oh, and whats the best way to find the clocks on things?

If nTune is right, my CPU is running at 1869Mhz (e6600) while at 100% load on both cores, and my GPU core is running at 600 MHZ (above stock, and below OCed edition, weird), while my GPU Memory is running at 900Mhz (possibly *2 = 1800Mhz(stock) for dual channel?)

 

[Grimlaking]

Dumb question I know. It probably isn't this... but you did connect both of the 12 volt feeds for the card right?

 

[Teknokid]

its the psu!! said it all along...
the mems double data rate so its 900mhz but ddr makes it 1800mhz.
temps are great so its defo not the card!

 

[Jomanscool2]

Yes, both 12V feeds, or more specifically, dedicated PCIe Power connectors are connected.

If it is indeed the PSU, its the PSU. Seems its the only thing left to try in this thing (don't see how a MB could f-up the cards power, unless it has like transistors built into the PCIe Connector).

Maybe this spring break I will grab another PSU. I'm still trying to convince my overly rich friend to let me build him a QX system with SLI OCed 8800GTXs and a P&C1kw PSU.

Or maybe I'll win one at the lan party im going to this weekend =D, but I doubt it, cuz im not that great at games....

But, I got school in 7 hours, so I'll post back 2morrow morning.

 

[GoodBoy]

Hope you get it working. If you lived in okc I'd let you borrow a 480w psu I have for testing.

You might try going into a pc club, and ask the same question. The guys are usually geeky, and might help you troubleshoot. Don't know if they would charge anything.

However, in you case, 2 cards both have the squeal I'm inclined to believe that the card isn't the issue. The PSU, and next the mobo are the items I'd point to. With the PSU the more likely culprit.

 

[Teknokid]

was for me-what i said ages ago.

 

[Jomanscool2]

Yeah, we'll see. I'll be gaming a bunch this weekend. I don't think it will be making the squeal while gaming.

If it does, f@$k, I'll try to find someone with a PSU I can borrow...

 

[Teknokid]

the squeel gets worse in game, or it did for me. id get a new psu if i were you, maybe even a different model.

 

[Jomanscool2]

the squeel gets worse in game, or it did for me. id get a new psu if i were you, maybe even a different model.

Hope not...

 

[Eldest_One]

Wow this is funny finding this thread I had the same thing on a BFG7800 GS AGP card. The first card I had was RMA'd back to BFG and the card I received back from them did the same thing. It only did it when under a high graphics load and the sound did come from a little speaker on the card. I was always suspicious of the PSU but never got a new one. Kind of a mute point now as the M/B i was using the AGP slot died on it.

 

[newls1]

well im sorry but it solved my whine

Hey bud, how is your PSU

 

[Teknokid]

great.. just mobo less. mobo blew.

 

[sgt1111]

I just bought a new system with an evga 8800GTX and i have been having a similar problem. The card squeals loudly when the system has been at rest for an hour or so. I mean really loud. It sounds just like a smoke detector in pitch and loudness. These other problems seem to happen only during high load but mine works fine, no noise at all from the card after hours of gaming (BF2142) with maxed out settings. I have an e6600 cpu, 2 gigs corsair ram, an asus p5be motherboard, creative X-fi gamer sound card and a corsair 620 wat power supply running vista home premium. I don't think this issue is the power supply as like i said it only happens when the system is doing nothing. It locks up and starts the squeal. First time it did it windows had a display init error when i rebooted the machine.

I tried the latest drivers from evga, and the nvidia ones as well. I think i will return this card and try another one.