This is why GTX590s blow up. Infrared thermography: GTX590 vs HD6990

munkle said:
If its a crap psu, a good psu will shut down. I recently plugged in a hard drive that had a short in it. Guess what happened, my psu shut down the rails the hdd was using (I could tell as my other mechanical hdd shut down also) and the computer kept running even though the rail the hdd was on was overloaded by a short.

Sorry but if it was designed good it would shut down before it damaged itself. It's not a new concept even cpu's shut down when they reach there limit (thermally, you can still kill them by overvolting). If you make the protective limits bypassable then whats the point still a flawed design.
Click to expand...

yes but cpus have a huge amount of resistors, caps and regulators keeping it safe. A gpus thermal throttling are within its drivers as well as the protections.
And a psu can't shut down when its protections is removed that was my point I will go poof. Same as the card
But anyway Nvidia THE WAY ITS MEANT TO EXPLODE <- THey should change their slogan now
 
Anyways. Geforce boards have performance levels printed in their bios. When it boots it will read of a table setting it to a certain safe clocks so that the system can always boot with it. The driver reads the info for the clocks of that table and will set it based on a performance mode. When windows is loaded and it detects a auxiliary power connector it will then switch out of safe mode to a 2d mode. During runtime the driver will then track the gpu state and switch it to a higher performance level when gpu power is needed or to a lower one when the gpu is not being used...... Now the drivers use robustchannels allows the gpu to recover.
That alone is telling me the modding of the cards was with its bios and the performance tables.
 
Lorien said:
So it appears that the 6990 is a cooler card than the GTX590 after all even in AUSUM 450W mode. Pictures speak for themselves:
GTX 590 idle vs load:
gtx590img0031534.jpg
gtx590img0031535.jpg


HD 6990 idle vs load@375W vs load@450W:
hd6990img0031264.jpg
hd6990img0031265.jpg
hd6990img0031266.jpg


Source
Click to expand...

Exactly.
And 590 burned on POST, so it must be drivers fault?? :rolleyes:

Apparently those hotter temperatures are from the back of the PCB, but for some reason they call it "Carte graphique en charge" - graphic card at load(right?),

Further more they don't explicitly say exactly where that particular temperature was measured(whole PCB is uniformly heated?), and they never use it in their temp. tables.

Anyhow I really doubt those Frenchies know how to use thermal camera properly.
Of course I may be wrong, but their readings are all over the place, showing higher readings for Crossfire vs same card single, but lower SLI then single etc etc.

OK so we covered why 590 died :rolleyes:
Now the bad news:

Radeon 3450, and 4870 Xfire users -- ACHTUNG!!! ACHTUNG!!! -- Your PCB is melting as we speak!

3450
thermic-imaging-radeon-hd-3450.jpg



4870 crossfire
thermic-imaging-radeon-hd-4870-crossfire.jpg
 
To be fair, a lot of HD4870 cards did die under OC'd Furmark loads, prompting nVidia and AMD to revise their drivers to throttle certain programs ("power viruses"), and drove AMD's team to include temp based VRM throttling on the HD5000 series to prevent furmark from running loose like that ;) :)

and the HD3450 is a passive card... LOL. Not a straight comparison
 
Wolfkin said:
Aren't you guys missing some points here?
The cards that blew up were overclocked, was overvolted, was using a driver that put power limiter out of action.
Click to expand...

Indeed. These images have pretty much nothing to do with why they blew up... Just a cheap shot at nVidia.
 
Rossi~ said:
Indeed. These images have pretty much nothing to do with why they blew up... Just a cheap shot at nVidia.
Click to expand...

WTF?? They have been doing these pics since at least the 8800gtx came out. It's not a "cheap shot at nVidia." It's what they have been doing as a part of their reviews for a long time, now. Would a review showing nVidia being slower at Surround/Eyefinity resolutions be a "cheap shot at nVidia," since it wouldn't show the GTX590 pulling ahead at 1600x1200?
 
HohnyF said:
Exactly.
And 590 burned on POST, so it must be drivers fault?? :rolleyes:

Apparently those hotter temperatures are from the back of the PCB, but for some reason they call it "Carte graphique en charge" - graphic card at load(right?),

Further more they don't explicitly say exactly where that particular temperature was measured(whole PCB is uniformly heated?), and they never use it in their temp. tables.

Anyhow I really doubt those Frenchies know how to use thermal camera properly.
Of course I may be wrong, but their readings are all over the place, showing higher readings for Crossfire vs same card single, but lower SLI then single etc etc.

OK so we covered why 590 died :rolleyes:
Now the bad news:

Radeon 3450, and 4870 Xfire users -- ACHTUNG!!! ACHTUNG!!! -- Your PCB is melting as we speak!

3450
thermic-imaging-radeon-hd-3450.jpg



4870 crossfire
thermic-imaging-radeon-hd-4870-crossfire.jpg
Click to expand...
well if you look at the cards the GTX is fairly evenly heated so if you would look... I have a Asus 3450 in my htpc, that thing only gets to about 70c if your getting that kind of temps there must not be any airflow.
 
jeremyshaw said:
To be fair, a lot of HD4870 cards did die under OC'd Furmark loads, prompting nVidia and AMD to revise their drivers to throttle certain programs ("power viruses"), and drove AMD's team to include temp based VRM throttling on the HD5000 series to prevent furmark from running loose like that ;) :)

and the HD3450 is a passive card... LOL. Not a straight comparison
Click to expand...

No cards died, these were the first cards to use GDDR5 and the early GDDR5 (Qimonda IIRC) degraded while heavily overclocked while running a lot of furmark. That is a different issue.

The fact is that 4870/90/x2 did have overcurrent protection at a hardware level as seen here. You can try it yourself. Those volterra vrms, the very same that were on the GTX280 where quality vrms.
 
BababooeyHTJ said:
No cards died, these were the first cards to use GDDR5 and the early GDDR5 (Qimonda IIRC) degraded while heavily overclocked while running a lot of furmark. That is a different issue.

The fact is that 4870/90/x2 did have overcurrent protection at a hardware level as seen here. You can try it yourself. Those volterra vrms, the very same that were on the GTX280 where quality vrms.
Click to expand...

ah, reviews of the 9.8 (??) or so drivers lead me to believe otherwise. That's when AMD started inducing driver based throttling, and people started renaming furmark to get around that, before AMD eventually used load types to determine driver throttling.
 
HohnyF said:
Exactly.
And 590 burned on POST, so it must be drivers fault?? :rolleyes:

Apparently those hotter temperatures are from the back of the PCB, but for some reason they call it "Carte graphique en charge" - graphic card at load(right?),

Further more they don't explicitly say exactly where that particular temperature was measured(whole PCB is uniformly heated?), and they never use it in their temp. tables.

Anyhow I really doubt those Frenchies know how to use thermal camera properly.
Of course I may be wrong, but their readings are all over the place, showing higher readings for Crossfire vs same card single, but lower SLI then single etc etc.

OK so we covered why 590 died :rolleyes:
Now the bad news:

Radeon 3450, and 4870 Xfire users -- ACHTUNG!!! ACHTUNG!!! -- Your PCB is melting as we speak!

3450
thermic-imaging-radeon-hd-3450.jpg



4870 crossfire
thermic-imaging-radeon-hd-4870-crossfire.jpg
Click to expand...

its impossible for it to burn at post. The card falls back on its low clocks and in a safe mode which it gets from a table in its bios. it uses the table till the drivers kick in.
Now to disable that touch needed a modded bios that hides the performance table. To do that you either set the performance table size to zero (which Asus always use to bypass thermal throttle) or to zap the pointer showing to the performance table. No driver can burn a card unless you mod its bios and its performance and voltage tables.
 
Spooony said:
its impossible for it to burn at post. The card falls back on its low clocks and in a safe mode which it gets from a table in its bios. it uses the table till the drivers kick in.
Now to disable that touch needed a modded bios that hides the performance table. To do that you either set the performance table size to zero (which Asus always use to bypass thermal throttle) or to zap the pointer showing to the performance table. No driver can burn a card unless you mod its bios and its performance and voltage tables.
Click to expand...

or the driver shuts down the fan control system while ignoring thermal throttling (well, at least fo the G92 cards :p).
 
Naieve said:
I wonder what the settings were when the pics were taken. If that's a stock 590 at load, with stock fan profile, yeah, thats Nvidias fault.

Kinda hard to know what those pics are saying without test methodology.
Click to expand...

Google translate says at reat & in charge. The third picture shows a better breakdown of where the heat is generated. Vreg & nf200. Getting 1 of these before winter is over to save on my oil bill. :eek:
 
bunch of sissies. 8800GT got that hot too.

either the card is hot, or it is noisy. pick one.

cards breaking because reviewers OVERVOLTED THEIR FREE GTX 590 without knowing what it would do... is a separate issue.
 
bAMtan2 said:
bunch of sissies. 8800GT got that hot too.

either the card is hot, or it is noisy. pick one.

cards breaking because reviewers OVERVOLTED THEIR FREE GTX 590 without knowing what it would do... is a separate issue.
Click to expand...

What does free have to do with anything? At the end of the day a $700 card should be using a top notch power delivery system like we see on Cayman and Antilles. There is a reason that we aren't seeing these types of issues pop up on AMD's cards.
 
Yeah, like in W1zzard's case, reasons being -

not touching the voltage on 6990
(TPU rarely if ever uses voltage bumps when showcasing an overclock.)

AND using old beta drivers, not even bothering to check if WHQL is out.

AND being trigger happy with 30% overvoltage on 590 because he "couldn't wait to try bumping the voltage".

AND not increasing fan profile

AND not disabling OCP/OVP as per guide the card was supplied with
 
Last edited:
HohnyF said:
Yeah, like in W1zzard's case, reasons being -

not touching the voltage on 6990
(TPU rarely if ever uses voltage bumps when showcasing an overclock.)

AND using old beta drivers, not even bothering to check if WHQL is out.

AND being trigger happy with 30% overvoltage on 590 because he "couldn't wait to try bumping the voltage".

AND not increasing fan profile

AND not disabling OCP/OVP as per guide the card was supplied with
Click to expand...

I guess it was appropriate for W1zard to add E.P.I.C.F.A.I.L subliminally that is exactly what his actions were in this case. Not that the freakin driver should have been available but definitely a RTFM issue by none other than one of the most respected reviewers in the industry. His worst work to date I believe.
 
Lord_Exodia said:
I guess it was appropriate for W1zard to add E.P.I.C.F.A.I.L subliminally that is exactly what his actions were in this case. Not that the freakin driver should have been available but definitely a RTFM issue by none other than one of the most respected reviewers in the industry. His worst work to date I believe.
Click to expand...

Only a blind Nvidia fanboy (with crossfire 6950 in his sig) could write off the issue as just that. Do you even know when he wrote the review? I think that it's safe to say that he was using the latest driver that was available to him. Secondly TPU was not the only reviewer to fry a card. On top of all of that there are all of the pictures of burnt up vrms on GTX570. Yeah, Nvidia is using inferior components on their high end, very expensive cards.
 
BababooeyHTJ said:
Only a blind Nvidia fanboy (with crossfire 6950 in his sig) could write off the issue as just that. Do you even know when he wrote the review? I think that it's safe to say that he was using the latest driver that was available to him. Secondly TPU was not the only reviewer to fry a card. On top of all of that there are all of the pictures of burnt up vrms on GTX570. Yeah, Nvidia is using inferior components on their high end, very expensive cards.
Click to expand...

My comment was based on the facts. If w1zard is saying that he only had that one driver available, not only did he post a review with a RTFM issue he's full of shit. Brent and many other reviewers wrote their articles days ahead of time and had the correct driver. He didn't and I could be wrong but the other reviewers didn't have asus cards. they may have been burning up other brands as their review kits probably didn't specify what the asus review kit specified. It said which driver to use, what to do when you overclock and over volt it. Basically to not sound like an broken record what I quoted in my last post that infuriated you as AMD fanboys get when someone discredits the bullshit review you are using to go out on your fanboy rants. Oh and FYI there is a seperate review kit issued by Nvidia themselves (and AMD) during the initial launch of a new card and guess which driver that said to use. So keep in mind W1zard didn't read 1 manual he didn't RTFM 2 manuals LOL
 
BababooeyHTJ said:
Only a blind Nvidia fanboy (with crossfire 6950 in his sig) could write off the issue as just that. Do you even know when he wrote the review? I think that it's safe to say that he was using the latest driver that was available to him. Secondly TPU was not the only reviewer to fry a card. On top of all of that there are all of the pictures of burnt up vrms on GTX570. Yeah, Nvidia is using inferior components on their high end, very expensive cards.
Click to expand...

Only and AMD fanboy would say that.

If you ask around, HD 6990s are dying also. A reviewer from OCC just told me the HD 6990s were dying also.
 
Lord_Exodia said:
My comment was based on the facts. If w1zard is saying that he only had that one driver available, not only did he post a review with a RTFM issue he's full of shit. Brent and many other reviewers wrote their articles days ahead of time and had the correct driver. He didn't and I could be wrong but the other reviewers didn't have asus cards. they may have been burning up other brands as their review kits probably didn't specify what the asus review kit specified. It said which driver to use, what to do when you overclock and over volt it. Basically to not sound like an broken record what I quoted in my last post that infuriated you as AMD fanboys get when someone discredits the bullshit review you are using to go out on your fanboy rants. Oh and FYI there is a seperate review kit issued by Nvidia themselves (and AMD) during the initial launch of a new card and guess which driver that said to use. So keep in mind W1zard didn't read 1 manual he didn't RTFM 2 manuals LOL
Click to expand...

There are other sites that have popped a card on more recent drivers, probably the same one on the [H] review. Blaming the death of a video card or any other piece of hardware on a driver alone is a ridiculous statement. Either way Wizz was using the official review driver btw.

Most importantly he wasn't the only one to pop a card.

xIronCrossx said:
Only and AMD fanboy would say that.

If you ask around, HD 6990s are dying also. A reviewer from OCC just told me the HD 6990s were dying also.
Click to expand...

Where?
 
Last edited:
BababooeyHTJ said:
There are other sites that have popped a card on more recent drivers, probably the same one on the [H] review. Blaming the death of a video card or any other piece of hardware on a driver alone is a ridiculous statement. Either way Wizz was using the official review driver btw.

Most importantly he wasn't the only one to pop a card.
Click to expand...

AFAIK noone has yet to pop a card with the right driver. Do you have any linkage to prove what your saying. BTW I do think nvidia sux for using cheap ass VRM's on the reference GTX 570. BTW 6990 FTW in case you or anyone else was wondering what I thought was the better card. I just can't stand all the useless trolling and fanboyism on the forum right now.

I'm gonna have to call shens on your comment that the wiz was using the right driver and other too unless you can provide a few links proving this. Nvidia has contacted everyone and found that it was that one driver that caused the issue.

Also the asus review kit specified when going over the voltage range (like the wiz did) you need to disable the power limiter and up the fan profile, possibly even put a better cooler for long term use. That's why I say RTFM.
 
BababooeyHTJ said:
:rolleyes: Yeah a bunch of confirmed cases. That one guy saying that he knows of a couple of cases. :rolleyes:
Click to expand...

Yeah because overvolting a GTX 590 way beyond spec is a great reason to bash Nvidia :rolleyes:

Honestly, you need to stop being such an AMD fanboy.

Once again, why don't you kids get an HD 6990 and overvolt it way beyond spec? 1.065 is spec for GTX 590, so overvolt your HD 6990 max spec by 0.145V and see how it fares. I am talking about the value that MSI afterburner shows, go 0.145V above that and see how your card fares. But no we don't need to do this because we trust our card and it wasn't done by a reviewer so lets keep bashing the GTX 590 :rolleyes:
 
xIronCrossx said:
Yeah because overvolting a GTX 590 way beyond spec is a great reason to bash Nvidia :rolleyes:

Honestly, you need to stop being such an AMD fanboy.

Once again, why don't you kids get an HD 6990 and overvolt it way beyond spec? 1.065 is spec for GTX 590, so overvolt your HD 6990 max spec by 0.145V and see how it fares. I am talking about the value that MSI afterburner shows, go 0.145V above that and see how your card fares. But no we don't need to do this because we trust our card and it wasn't done by a reviewer so lets keep bashing the GTX 590 :rolleyes:
Click to expand...

Hmm intesting post, I believe alot of reviewers were treating the GTX 590 like it was a GTX 580 or something. Truth is it's a totally different card and you need to understand the deign well before you do things like volt modding. Asus spent alot of lab hours testing the card and took careful consideration and posted their finding of the tollerances of the VRM's and pcb design.
 
I'm done arguing with you. RTFM, fanboy, lol, troll use whatever buzz words you feel like.

Here is a video from another site and if you have the time read the TPU review since the driver that he used is a addressed there.

http://www.youtube.com/watch?v=sRo-1VFMcbc

I'm done, we'll see how these cards fair in the hands of end users.
 
Lord_Exodia said:
Hmm intesting post, I believe alot of reviewers were treating the GTX 590 like it was a GTX 580 or something. Truth is it's a totally different card and you need to understand the deign well before you do things like volt modding. Asus spent alot of lab hours testing the card and took careful consideration and posted their finding of the tollerances of the VRM's and pcb design.
Click to expand...

I honestly don't understand what the reviewers were thinking. Maybe Nvidia was being an asshole to them, since I know Hardware secrets go butthurt after they didn't receive one. Its odd really.
 
well my GTX 590 right now is clocked at 650/1300 and with gaming and stress testing it never went over 86C in my case and that's with a GTX 465 in there as well.
 
bash bash. HAHA, I own nVidia cards, so that means all AMD GPU can suck my minime.

Now for me to make a fanboi comment:

reviewer much idiot. munchie munchie. 7 other reviewers fried cards? pffsshshhh. Those stupid idiots. They are all incompentent idiots. Only reviewers who didn't blow up GPU matter in my eyes. Luckily, I'm not a TPU forumer, otherwise, I'd slit my fucking wrists. :rolleyes:

HA! You have AMD in your sig! OMG! That means you are an AMD fanboi!! :eek: out, OUT!! ARg, crap, this isn't the nVidiot section! :mad: WTH was the OP thinking!? :rolleyes:
 
BababooeyHTJ said:
I'm done arguing with you. RTFM, fanboy, lol, troll use whatever buzz words you feel like.

Here is a video from another site and if you have the time read the TPU review since the driver that he used is a addressed there.

http://www.youtube.com/watch?v=sRo-1VFMcbc

I'm done, we'll see how these cards fair in the hands of end users.
Click to expand...

Cool. Using old drivers. Got anything else?

Maybe they should get a screenshot of the card instead of just showing it blow up. Its funny how they were filming it just to see it blow up, but no picture of the monitor showing the volts (before they restarted obviously to see what they set it to). I just don't trust them. But honestly, I don't care. You can keep saying what you want, when the HD 6990 gets overvolted and dies you will shut up. But no one will do it because they are scared and just wanna keep bashing Nvidia.
 
BababooeyHTJ said:
I'm done arguing with you. RTFM, fanboy, lol, troll use whatever buzz words you feel like.

Here is a video from another site and if you have the time read the TPU review since the driver that he used is a addressed there.

http://www.youtube.com/watch?v=sRo-1VFMcbc

I'm done, we'll see how these cards fair in the hands of end users.
Click to expand...

Again in the same video you just linked they were using the wrong drivers. It says in the same video do not use 267.52 drivers! LOL thanks for proving my case. W1zard did use the right driver it seems however he took the card well past it's voltage specs without reading the asus review kit and following the appropriate guidelines. If people are blowing their cards it's because they are overdoing the voltage or using the wrong driver.
 
If Nvidia is shitting with us and putting crappy parts on these cards I will be angry too. But I just don't see it, all I see is old drivers and overvolting like a retard killing the cards.
 
You must log in or register to reply here.
Back
Top