rick6 wrote:

Oh mine is already on the way, but i've payed more than that. You don't get good deals from ebay over to Portugal without high shipping costs. Or even worse, the seller won't send to any other country that not his own.

Note: Never mind!! You were talking about the CPU right? Fail on me. I plan to put my 6800 Ultra on a Athlon XP for a ultimate Windows 98 retro gaming machine.

Oh sorry, yes the CPU, sorry on the confusion. I paid about $35 with shipping for my 6800 ultra.

mr_bigmouth_502 wrote:

I'll have to remember that. I used to just use rubbing alcohol.

Arctic Silver actually dis-suggests using automotive cleaners/solvents to clean CPUs and heatsinks, their argument is that if it has any sort of oil base (which many do) it can adversely affect adhesion of TIM<-> devices and cause poor performance (or other issues), and that some chemicals may damage the package or seep into places they ought not seep. I can't speak specifically to carb'n'choke cleaner, and I'm guessing there isn't a unified "carb'n'choke cleaner formula" - different makers probably have different blends (and I'd add further that the specific brand/blend/whatever that kithlyn is using is probably OK for computer parts, especially if he/she has been using it for years - kithlyn can you share specifically what brand/ingredients/whatever make up what you're using?). The primary things to avoid are oil-based chemicals or things that will damage plastic/fiberglass/etc (as it will also potentially damage computer parts).

Having said all of that, acetone is a decent chemical to remove TIM from heatsinks/blocks/etc (I wouldn't put it on a CPU itself, as it may damage the packaging - it can damage some plastics after all). You can buy more or less "straight" acetone as nail polish remover at most grocery stores (and it shouldn't cost more than a dollar or two). Just make sure to read the label - some contain scents, coloring, etc (which you don't want), and may also not even contain acetone.

Rubbing alcohol is an equally effective cleaner, and has the advantage of not leaving any residue and drying very quickly. I wouldn't think you should need to change anything if that's what you're already using. 😀

If you aren't keen on using chemicals, AS' suggested method of cleaning heatsinks and CPUs is to use a somewhat coarse (but lint free) rag, like what you might find as a shop rag or similar. And then go over the part with a rubber eraser, and then again with the rag. I've found this to work extremely well on chip dies and polished heatsink bases, and there's no concern about chemicals possibly damaging things, or wait-time for things to dry out after being washed.

obobskivich wrote:

Arctic Silver actually dis-suggests using automotive cleaners/solvents to clean CPUs and heatsinks, their argument is that if it […]
Show full quote

Arctic Silver actually dis-suggests using automotive cleaners/solvents to clean CPUs and heatsinks, their argument is that if it has any sort of oil base (which many do) it can adversely affect adhesion of TIM<-> devices and cause poor performance (or other issues), and that some chemicals may damage the package or seep into places they ought not seep. I can't speak specifically to carb'n'choke cleaner, and I'm guessing there isn't a unified "carb'n'choke cleaner formula" - different makers probably have different blends (and I'd add further that the specific brand/blend/whatever that kithlyn is using is probably OK for computer parts, especially if he/she has been using it for years - kithlyn can you share specifically what brand/ingredients/whatever make up what you're using?). The primary things to avoid are oil-based chemicals or things that will damage plastic/fiberglass/etc (as it will also potentially damage computer parts).

Having said all of that, acetone is a decent chemical to remove TIM from heatsinks/blocks/etc (I wouldn't put it on a CPU itself, as it may damage the packaging - it can damage some plastics after all). You can buy more or less "straight" acetone as nail polish remover at most grocery stores (and it shouldn't cost more than a dollar or two). Just make sure to read the label - some contain scents, coloring, etc (which you don't want), and may also not even contain acetone.

Rubbing alcohol is an equally effective cleaner, and has the advantage of not leaving any residue and drying very quickly. I wouldn't think you should need to change anything if that's what you're already using. 😀

If you aren't keen on using chemicals, AS' suggested method of cleaning heatsinks and CPUs is to use a somewhat coarse (but lint free) rag, like what you might find as a shop rag or similar. And then go over the part with a rubber eraser, and then again with the rag. I've found this to work extremely well on chip dies and polished heatsink bases, and there's no concern about chemicals possibly damaging things, or wait-time for things to dry out after being washed.

The reason I don't like using Rubbing alcohol is it doesn't dissolve very quickly and usually requires multiple applications (5+) to ever get a heat-sink "clean". Where as other solvents just work the first time in a few seconds and you're done. I'll photo what I've been using but it's just the generic Walmart brand. I also used an electrical contact cleaner for years, but I stopped using it after reading it had hexane in it and how bad that stuff is to breathe, so I tossed it out and switched to Walmart choke cleaner years ago and haven't ever had a problem. In fact I have to use so very little of the stuff to actually clean heat-sinks that one can of it lasts me like nearly 5 years.

High proof alcohol should dissolve/evaporate fairly quickly ("quickly" is certainly relative though - I guess if you want instant evaporation that's another story 🤣). That said, good TIM (like AS5) shouldn't need chemicals to remove (I just use a shop rag and an eraser and can get good results). Honestly the worst I've seen in a long time was the factory ceramic gunk on my VP880, but a little elbow grease and it was gone. 😀

+1 on not using stuff with hexane or other nasty chemicals like that. I have some of that DeOXIT stuff and honestly feel like it should only be used outdoors or in a big open space (like a garage); it's just rancid to spray out. 😵

kithylin wrote:

Standard Def Steve wrote:

That's an excellent score for an S754/GF6 setup. Interested in adding it to the Mega Thread?

I think I'll go do that, after dinner, thanks for reminding me. 😀

Thanks man! 😀

Alright. I just got my 6800 Ultra agp, and this is what it displays:

Now i just want to kill someone!

rick6 wrote:

Alright. I just got my 6800 Ultra agp, and this is what it displays: […]
Show full quote

Alright. I just got my 6800 Ultra agp, and this is what it displays:

<snip>

<snip>

Now i just want to kill someone!

It's a pretty sure bet that it's a dead GPU Jim. Are you sure you have both the power plugs connected securely, and a decent power supply? these cards need at least a PSU with I would say at least 25 amps on +12v line(s). You can get decent 500 watt ones new cheap these days with around 35-40 amps on +12v, almost no excuse for not having one.

kithylin wrote:

It's a pretty sure bet that it's a dead GPU Jim.

Really looks to be the case. Sorry to see that rick6! 😢

Out of curiosity (before we entirely write this card off for recycling): does it do this on both video outputs? (or all three, if it has a TV output as well) And is this connected via DVI or VGA? Does switching from DVI to VGA change anything? If you're using a DVI-VGA adapter, can you try another adapter? (IME a bad adapter has never caused this; it will usually result in a dropped color channel, but you never know).

Are you sure you have both the power plugs connected securely, and a decent power supply? these cards need at least a PSU with I would say at least 25 amps on +12v line(s). You can get decent 500 watt ones new cheap these days with around 35-40 amps on +12v, almost no excuse for not having one.

If I remember right my 6800GT had a little buzzer on it that would go off if you forgot to connect the extra leads. I'm not sure if that's a standard feature for 6800s or not. I'd expect to receive some sort of error message if the card didn't have sufficient power on start-up as opposed to this madness. 😵 (I know my Radeon 9700 would just pop up a message box before the system POSTed and demand a shut down and re-connection of its aux power lead; if I remember right the GeForce FX cards will actually boot sans the aux power lead, but you'll get error messages in Windows.

I'd entirely agree on "almost no excuse for not having one" on a decent 500W+ PSU as well. Even a basic Thermaltake or Antec isn't terribly expensive these days. 😀

Yes i have a decent power supply, dual rail, but still those 25 amps are really only required when the card is being stressed out with benchmarks or games. For testing the video card running my computer bios i shouldn't get artifacts, even if the power supply isn't strong enough. The card would beep or crash if not enough juice was being supplied while gaming or normal usage.

Let's see how the seller reacts, if for any reason i end up keeping the card i'll try to reflow the chip and if it works, then i should reball the chip for a long term solution :s

obobskivich wrote:

Really looks to be the case. Sorry to see that rick6! 😢

Out of curiosity (before we entirely write this card off for recycling): does it do this on both video outputs? (or all three, if it has a TV output as well) And is this connected via DVI or VGA? Does switching from DVI to VGA change anything? If you're using a DVI-VGA adapter, can you try another adapter? (IME a bad adapter has never caused this; it will usually result in a dropped color channel, but you never know).

The video card has 2 DVI's, and only one displays this blockomania, changing the adapter shows the same results. It sure has bad solder joints. Either that or the card was damaged by electrostatic damage caused by the bubble wrap used by the seller.
If i endup keeping the card i'm going to try to fix it first, try as hell! If not then it will become a pretty paperweight.

rick6 wrote:

obobskivich wrote:

Really looks to be the case. Sorry to see that rick6! 😢

Out of curiosity (before we entirely write this card off for recycling): does it do this on both video outputs? (or all three, if it has a TV output as well) And is this connected via DVI or VGA? Does switching from DVI to VGA change anything? If you're using a DVI-VGA adapter, can you try another adapter? (IME a bad adapter has never caused this; it will usually result in a dropped color channel, but you never know).

The video card has 2 DVI's, and only one displays this blockomania, changing the adapter shows the same results. It sure has bad solder joints. Either that or the card was damaged by electrostatic damage caused by the bubble wrap used by the seller.
If i endup keeping the card i'm going to try to fix it first, try as hell! If not then it will become a pretty paperweight.

So just to clarify: you're connecting DVI to VGA on one output and getting problems? The other DVI is okay? How does the "bad" output work when sending DVI?

Sounds like the connector is bad more than anything else, especially if the machine isn't locking up or displaying any other quirkiness/problems with the card installed. Tried wiggling things?

I'm connecting DVI to VGA yes, but i've tried everything else regarding that and believe me, it's not a connector problem. This is a classic BGA solder problem most likely. It's perfectly normal for the card to work while displaying this stuff and not lock up. I was already given several cards with these symptoms over the years (ati 9700\9800 pros, geforce 6600GT and 7900GT) and in almost all i had to do for a short term fix was a reflow with a bit a flux and a hot air gun. This card is no different (so i hope).

I just thought I would put a little update to this build. I'm not entirely sure what happened.. as I wasn't really running the thing that far overclocked or anything.. just from the 2.6 ghz stock up to 3085 Mhz, and used it in a well-ventalated case (Antec three hundred). Well, it was running perfectly when I last used it and then I shut it off and left it sitting in the other room for a few months then 3 weeks ago I went to try and use it again and turned it on and it wouldn't come on no matter what I did, wouldn't POST. Machine "ran", as in everything turned on and was active (fans, hard drives, etc) but no video output. So I shut it off and looked inside, and all the capacitors around the cpu socket had bulged and leaked brown stuff. So... that board's dead in the water for now. I have a line on a local friend that will help me recap it some day and so I haven't given up on getting it to live again, but still.. can't use it now, sadly. Maybe using a mobile chip in it caused it to run out of spec some how, I'm not sure yet. Some of the capacitors were already slightly bulged when I put it together last time, I guess I just finally pushed em over too far.

I recently bought a KT-600 core AMD Socket 462 motherboard and plan to (in the short term anyway) replace the purpose of this 754 machine with the new-to-me (I think almost new in general!) Socket 462 machine so might move most kit over there and make a thread on it soon, and then store this 754 board and mobile chip in a box in a closet somewhere to be fixed some day.

I'm not sure what your obsession is with overclocking but you don't overclock a board with bulging caps and expect it to live much longer...

Half-Saint wrote:

I'm not sure what your obsession is with overclocking but you don't overclock a board with bulging caps and expect it to live much longer...

It's more of the "Trying to get the most out of a system" I've overclocked lots of stuff before with bulging caps and had it last several years, this one I just got a little unlucky. The whole kit with the 754 I got incredibly cheap so I don't really care either way, and it's fixable some day, so doesn't matter at all.

I had an A8n32-SLI Deluxe motherboard with 2 bulging caps that ran seriously overclocked (way more than this system was) and it worked fine for almost 3 years.

The only stuff I -don't- clock over though is the really old stuff like 486's and pentiums, but Super Socket 7 and anything newer than that where we can still buy replacement parts and they're not expensive, all game and no foul.

EDIT: This thing had bad caps when I put it together again recently. I already knew it was going to die some time soon anyway, it was just a matter of time. It doesn't matter if I killed it now or if it died later, it was going to die from caps anyway, so I really don't care.

Board is fixable and you had me wanting to build a system like that myself. Got a Pentium Mobile desktop going, pretty epic with a nice overclock and a 8800 Ultra.

nforce4max wrote:

Board is fixable and you had me wanting to build a system like that myself. Got a Pentium Mobile desktop going, pretty epic with a nice overclock and a 8800 Ultra.

I'm hoping this KT600 board I bought will accept mobile chips, I'd love to get my paws on one of the mobile AthlonXP's with an unlocked multiplier and try clocking that thing up. We'll see what happens.

kithylin wrote:

I'm hoping this KT600 board I bought will accept mobile chips, I'd love to get my paws on one of the mobile AthlonXP's with an unlocked multiplier and try clocking that thing up. We'll see what happens.

I've never seen a 462 board not accept mobile chips, HOWEVER if you can't adjust multiplier, FSB, and vCore they will generally only run at something like 500-600MHz (even on boards you can adjust they generally default to those settings - basically minimum multi, FSB, etc because they can't correctly read off the clock settings from the chip). Basically, if it allows overclocking, you should be good to go. 😀

No idea how common or rare mobile XPs are these days though. 😊

obobskivich wrote:

I've never seen a 462 board not accept mobile chips, HOWEVER if you can't adjust multiplier, FSB, and vCore they will generally only run at something like 500-600MHz (even on boards you can adjust they generally default to those settings - basically minimum multi, FSB, etc because they can't correctly read off the clock settings from the chip). Basically, if it allows overclocking, you should be good to go. 😀

No idea how common or rare mobile XPs are these days though. 😊

Almost all AthlonXP chips are dirt cheap these days. We can get the 3200+, 400 Mhz FSB Barton chips for $20 even. I saw someone selling a Mobile AthlonXP 2500+ (claiming it has unlocked multiplier?) for $12 with free shipping, including a copper desktop heatsink with it.

All Mobile Athlon XP chips have unlocked multi. Usually the problem is with the BIOS not having multiplier above 12.5X. This can be solved via some software in Windows - but it usually still won't go higher then the default multi of the chip itself - remember that Socket A mobiles are FSB 133 and have multipliers over 13X. So if you use some old motherboard with support for only 133MHz FSB then the high multiplier is a must. But if you use newer MB with FSB support of 200MHz or more, then even max BIOS multi of 12.5X is practically enough (12.5x200=2500 MHz).