NVIDIA GeForce GTX 480 Fermi

[HillBeast]

He used 9.12. And in all fairness its only a 15% with 10.3a difference vs Nvidias immature drivers. Seems about fair.

The thing is, NVIDIAs drivers have always been on the ball. They always come out pretty decent and drivers do improve performance but not as much as ATIs drivers do.

The other thing is, these chips have been in production since like January and they have had all that time to improve the drivers, so you can't really say they drivers are immature. Especially when the card was 6 months late.

 

[iamverysmart]

ATI: Catalyst 9.12

 

[human_error]

I feel i may have been a little harsh in my initial read of the review - so much so i forgot to say thanks w1zz for doing this. Give it a few days after availability and we'll have more current driver numbers than you can shake a stick at for comparisons. I think we've all forgotten w1zz doesn't get paid for these reviews and as others have said they take a long time to bench.

That doesn't mean i don't think this card fails hard - it does - i just feel we should be greatful to have a set of benchmarks out the minute the NDA was lifted and a SLI scaling review too.

 

[csendesmark]

Fermi? you mean FAIL-me?
320W? OMG

 

[SUPERREDDEVIL]

when NVIDIA GTX495 comes out.. ATI already have the HD6970 for the win... Admit it everyone! ATI has the crown right now!

 

[Melvis]

Seems to out perform the 5870 in just about everthing, well done Nvidia for that, but if you have a 4870X2 or a GTX295 then this wouldnt be an upgrade (unless you want DX11) otherwsie any other card...then yes that would be an upgrade.

Shame its one hot power hungry card for a single GPU wow

 

[mlee49]

A dual GPU card with this GPU would probably burn down the entirety of Lawrence.

History could repeat itself(Lawrence KS was burned down in the late 1800's by a bunch of Missouri A-holes- hence Bleeding Kansas).


Dx 11 needs to be developed before we can really utilize these new cards, yet another point on why not to buy.

 

[sneekypeet]

as a cooler guy, this card has make or break writtten all over it for aftermarket cooler manufacturers. If you can cool a Fermi with it, you got yourself one helluva product

 

[Cold Storm]

There is more to look at then the factor of things being "Up to Date".. in order for a video card to work right, you get drivers that "mature" it. Drivers work to "fix", "help" the video card go along.

Yes, Nvidia has had 6m to work on things.. But, how many months have ATI had a card out, to make a driver "mature" due to the factor of having, Millions, and I mean MILLIONS, of people to work out, and find, all the problems that the drivers have.

Now, if you really want to go and play hard ball.. driver wise on 9.12.. here's some ss to help you along.

9.12 drivers

169k reports

10.1 drivers

389k reports

10.2 drivers

360k reports



As you can see from the SS, the "safer" drivers to have is the 9.12 drivers. Yes, they are not the most "mature" drivers, but they've made people feel Safe in the factor of using.

Also, you gotta look at the factor of this.. 10.3 became "official" as of March 24th.. A little over 2 days go.. W1z uses 25 benches to do his reviews. 5 tests for each.. 25x5=125. So, 125 benches he would have to redo in 2 days.. Did, some do it.. Yes.. Why? How? Because they probably don't have anything else to do but get that going.

W1zzard has more going on then just a "RELEASE" of a card. GPU-z gotta get a work out now. You got services that have to get going in order to get this new "series" going.. He's not here just to do a review. Not like some, probably most, sites have it going..




Just my 2 cents...

 

[HillBeast]

I feel i may have been a little harsh in my initial read of the review - so much so i forgot to say thanks w1zz for doing this

Same here. W1zzard worked hard to get the review out so thanks to him for it. I don't hold it against him for using 9.12 drivers because it shows how useless GF100 is compared to a Radeon 5870 which isn't at it's potential.

 

[HillBeast]

9.12 drivers
169k reports

10.1 drivers
389k reports

10.2 drivers
360k reports

As you can see from the SS, the "safer" drivers to have is the 9.12 drivers. Yes, they are not the most "mature" drivers, but they've made people feel Safe in the factor of using

Not really a fair test seeing it's Google. 160k of that will most likely be porn or unrelated.

 

[senninex]

GTX480?
I'm surprise that not surprise.

 

[pentastar111]

as a cooler guy, this card has make or break writtten all over it for aftermarket cooler manufacturers. If you can cool a Fermi with it, you got yourself one helluva product

That is very true! lol

 

[Cold Storm]

Not really a fair test seeing it's Google. 160k of that will most likely be porn or unrelated.

hey, all good.. want to play even more. go deeper. go site to site, and see what the "SEARCH" button might find you..

Or, look at what people here have said..

Nvidia had drivers out not to long ago.. I think it was the first 191 drivers.. People will tell others to use them. Why? because their a "stronger" driver suit, then any of the newer ones. Yeah, after a few "newer" drivers, you got one that was better, but you still had people say that they tried, and they did, the newer ones, just to go back to the certain one.

 

[HillBeast]

Nvidia had drivers out not to long ago.. I think it was the first 191 drivers.. People will tell others to use them. Why? because their a "stronger" driver suit, then any of the newer ones. Yeah, after a few "newer" drivers, you got one that was better, but you still had people say that they tried, and they did, the newer ones, just to go back to the certain one.

I use the latest ATI 10.3 drivers on my computer and have no bugs I can report, but what I can report is a huge boost in performance over older drivers which is what matters in a review. If the game crashed in the benchmark then they should mention it. They used to do that with the original SLI cards they would say 'We gave the 7950GX2 0 FPS because it crashed in benchmarks'.

By using old drivers it is making the GTX480 look faster. Which it isn't. Not really.

 

[Cold Storm]

I use the latest ATI 10.3 drivers on my computer and have no bugs I can report, but what I can report is a huge boost in performance over older drivers which is what matters in a review. If the game crashed in the benchmark then they should mention it. They used to do that with the original SLI cards they would say 'We gave the 7950GX2 0 FPS because it crashed in benchmarks'.

By using old drivers it is making the GTX480 look faster. Which it isn't. Not really.

Nice that your using the latest drivers. Cookie should go your way for it.. But, you took one part of my post and now going away from the whole thing..

Drivers came out 2 days ago. 25 benchmarks are used. 125 redone bench marks have to be done. Can it be done in 2 days? yeah.. Maybe..

His reviews are done a little different then others.. Yeah, most do "up to date" games. Just Cause 2, all the latest games.. Not him... He, probably, does the games in which most people will still have in their computer. Have the factor of "replaying" then when a new driver came out.
^^^ this is what I "think he does" and do not respond to what he "really" does^^^

 

[yuriylsh]

ATI 9.12 driver?

unfortunately had to skip the review after seeing that 9.12 direvers were used for ATI cards

 

[HillBeast]

Nice that your using the latest drivers. Cookie should go your way for it.. But, you took one part of my post and now going away from the whole thing..

You go off at me for going away when you were talking about how they are using old driver because they are more reliable and I rebute it by saying the newer drivers are just as good. Just because NVIDIA can't make drivers for s**t anymore doesn't mean ATI are following suit. Your Google search is a completely inaccurate way for finding how many bugs there are with the new drivers.

What I was trying to say is that if bugs arrised in the benchmark with using new drivers he should have mentioned them after the graphs like they used to do in reviews.

If you think the review is fair then get a GTX 480. I am just trying to say the review isn't accurate enough to make a fair point on it. TechSpot used 10.2 drivers in their review and the 480 barely managed to get past and in most cases was beaten by the 5870 let alone the 5970 which is what this card is more going to compete against seeing as how they have roughly equal TDPs and prices...

 

[Roph]

I took my browser elsewhere also, when I noticed that ancient 9.12 drivers were used, and thus all results would be skewed in Nvidia's favour. I recommend firingsqaud's review, it's quite thorough and also uses drivers actually released this year.

I think I'll call this card the Geforce FX 480. It sounds more fitting . They might as well have just renamed an extra one hundred into the model number, so we could have and easily recognise another "5 series" geforce failure.

Overall I'm very disappointed.

 

[HillBeast]

Quoting Guru3Ds review (http://www.guru3d.com/article/geforce-gtx-470-480-review/35):

ATI reaps mucho benefits from the release of their Catalyst 10.3 driver (used in this review). The new driver brings significant performance boosts throughout the Radeon HD 5800 and 5900 series. Performance was enhanced in a lot of game titles. Would ATI not have released Catalyst 10.3 on time, then this review would have looked different. You'll probably notice a review or two out on the web using older drivers.

I rest my case.

 

[Cold Storm]

You go off at me for going away when you were talking about how they are using old driver because they are more reliable and I rebute it by saying the newer drivers are just as good. Just because NVIDIA can't make drivers for s**t anymore doesn't mean ATI are following suit. Your Google search is a completely inaccurate way for finding how many bugs there are with the new drivers.

What I was trying to say is that if bugs arrised in the benchmark with using new drivers he should have mentioned them after the graphs like they used to do in reviews.

If you think the review is fair then get a GTX 480. I am just trying to say the review isn't accurate enough to make a fair point on it. TechSpot used 10.2 drivers in their review and the 480 barely managed to get past and in most cases was beaten by the 5870 let alone the 5970 which is what this card is more going to compete against seeing as how they have roughly equal TDPs and prices...

If you want me to go off at you. I can. Trust me, that wouldn't be a hard thing to do. But, yet I haven't.

I get the factor of other sites using the newer drivers. But, you have to look at how reviewers do their review. Every review is different. A lot of times, a review will even have to use a "certain" thing in order to do their review. Or, the factor that they are "sponsored" by someone, so they have to watch their reviews with a Microscope.

I never said the review was "fair". I was just trying to back, and give site, into why a person would, may, use those "certain" drivers.

Yeah, Google wouldn't be the "idea" way to search the problem. but it's a Start. Yes, a lot of "strange" things may come across with the "numbering" system. But, that was just a factor on what "might be"..

I remember a review I read a few days ago.. Great review on a product.. But, when someone asked why he didn't use a certain product in his review. Reviewers answer.. Give me one and I'll show you the numbers..

Now, I'm glad I got to get in your head HB. Yeah, these nvidia cards are showing to be some crazyness. Making us wonder what the hell has been going on..

To me, a person that has went back and forth to both camps, I'm glad I have tried out the 5870's.. But.. I do wish I still had my GTX 295.

 

[Atom_Anti]

So the Ati cards why did not use the newest 10.3 driver?? Unfortunately this test do not give real picture.

 

[HillBeast]

I get the factor of other sites using the newer drivers. But, you have to look at how reviewers do their review. Every review is different. A lot of times, a review will even have to use a "certain" thing in order to do their review. Or, the factor that they are "sponsored" by someone, so they have to watch their reviews with a Microscope.

I know argueing in forums is immature, but I just want you to see my point. Guru3D used the newer 10.3 drivers and said they did because they are so much better, and because of that the GTX480 was slaughtered by the 5870. Now I'm not trying to say the 480 is fail because ATIs drivers are better (well I am but that's not my point). What I'm trying to say is that the 480 isn't what it was hyped up to be. Even if you consider NVIDIAs drivers being immature and ATIs drivers being superior so they made it even by using release drivers you get with your card, which makes no sense because reviews are for the end user to look at and see at this point in time this card is best, but you simply can get that conclusion from a review that uses horribly outdated drivers which are 6 months old.

There is no excuse for it in my opinion, sponsers or no. If they wanted NVIDIA to be happy then I suppose using old drivers for the ATI is understandable but if they wanted ATI to be happy then they would use the new stuff.

 

[HillBeast]

Also I should mention (and I'm not accusing TPU of anything) but I have always found TPU to be more AMD biased so saying sponsors doesn't make sense because then in that case TPU have just lost AMD as a sponsor.

 

[Cold Storm]

Hey, just like everyone has. A person has their opinion to things. Yeah, might of went "off" wrong in wording. It's internet. I was just throwing out factors on which a person "may" use that sort of thing. Calm the review bashing down.