Fermi is out. Thumbs up or down you decide with poll!

[akugami]

I know there's no standard for measuring TDP, but that ought to qualify as false advertisement. TDP is supposed to be a maximum value that the product should almost never reach in normal usage, but it looks like the 480 is a good deal past that under normal and way past that in furmark.

Lol, I think I remember Toyota or someone else saying what Charlie reported as Fermi's TDP wasn't bad since real numbers would be lower, but this seems to be the exact opposite scenario.

The problem is as you stated that there is no standard for measuring TDP. Just like how there are LCD monitors that make crazy response time claims but will actually lose in a head to head with a supposedly "slower" LCD. nVidia can make all sorts of crazy claims as long as it's true, probably with qualifiers in the fineprint.

Here's a case in point with AMD and Intel as examples. While the link is to CPU's it can serve just as well for GPU's in that each company can use a different metric by which they measure TDP. In this case with the GTX 480, nVidia is clearly listing something that is lower than the max power draw. And as shown in several reviews, when playing a demanding game such as Crysis, TDP goes well over the 250W listed. Now, technically they are not lying since there is no standard way to report TDP. They can use different metrics for each card in their lineup if they wanted to. Deceptive? For sure. But it's legal. At least for now.

 

[SolMiester]

Nvidia = Master of unethical corporate practices.

I'm not trolling. Think what you want, it's a free world. My comment simply reflect what Nvidia have said and done in the last couple of years, and more recently with the whole Fermi project fiasco. ATI is far from being all white, but it's not even remotely close to how twisted Nvidia has become.

That being said, the 480 is a gigantic, abysmal, epic >FAIL< of cosmological proportion. 95c and 312W of power consumption under load for *ONLY* 10-15% improvement on average over the 5870? All that despite *TWICE* the transistors, *50%* more RAM and 100.00$ more expensive? ARE YOU FREAKIN KIDDING ME??? How embarrassing for Nvidia to have made us all wait over 6 months for *THIS*. Especially when the 5970 is still beating it badly, runs cooler and quieter.

The 480 fails on all front: Price, power, heat, noise VS performance ratio. No value add either. Let me rephrase that: It has no value, period. Physx is great on paper but a joke in reality; only supported in about 20 games, out of which 17 are very bad. Both Physx and CUDA are closed, proprietary standard and will soon be replaced by OpenCL -open- platform anyway. No Dolby TrueHD Master audio over HDMI, no triple display support.

The only thing going for it is its great folding / GPGPU capabilities. But do you picture running this thing full load 24/7 at folding and sucking all the electricity in your house, over heating your case and components in the process? How about 2 of them in SLI requiring a monster 1.2KW, 350.00$+ power supply?? Your honor, I rest my case...

A little 5XXX series refresh / price adjustment will put this 480 abomination out of its misery and the HD 6000 series coming in Q3 2010 will make it just a bad dream all will want to forget.

I didn't like Nvidia very much before. But with all the lies, trickery, manipulation, re-badging, puppy wood screw hacked sawed board and everything else in between, I've learned to hate them. Profoundly.

Feel free to buy yourself I nice and shinny overpriced, under performing, power hog and simply uncompetitive 480 or 470 if you like. It's your choice. But never forget that you also vote with your wallet. Please, just for once, put your fanatical green ego on the side and don't support immoral corporate behavior. You'll feel good about it and won't get screwed buying something that don't deserve your hard earned cash.

LOL....yawn!

 

[dguy6789]

I don't think anyone can really argue against the fact that Nvidia is a highly unethical company. Their history is not very pretty at all in that regard.

 

[Ramon Zarat]

@RamonZarat

As far as I'm aware nvidia hasn't done anything as morally questionable as what Citibank did in the Enron affair or Union Carbide's behaviour over Bhopal, so can't we keep a sense of proportion? So they are a bit pushy in their tactics in the video card market, it's hardly selling chemical weapons to Saddam, is it?

And I think its too early to say the 480 is a complete failure. It's not a very good card in itself, true, but given the constraints I think perhaps nVidia's engineers did a good job of getting it to work at all. They succeeded in removing just enough so it won't burn a hole through the bottom of your case and go on to contaminate the water-table on its way to China, while still keeping it just ahead of the 5870, so from their point of view its "job done". Pity about the price tag, but I bet that's the minimum they can get away with also.

And surely it can't be judged entirely on its own merits, there's the question of where it leads, what comes next? Vista wasn't exactly a great success (though software is more post-release fixable than hardware) but it did seem to get certain fundamental changes out of the way, leaving room for the problems to be cleaned up later. Time will tell whether the 480 is a Vista or an Edsel. Obviously I wouldn't actually buy one myself (I can't even afford a 5870), but at least they got this one out the door and out of the way.

I feel a bit sorry for them. Designing GPUs appears to be quite difficult - who'd have thought it?

You logic is flawed. You are using a straw man fallacy argument by trying to associate Citybank, Enron and chemical weapons (!!!!) with Nvidia. Don't mix things up like they are connected in some way when in fact there is no relation whatsoever beyond the boundary of your own argument. One bad thing don't make another bad thing good, as you try to imply. Each case must be judged according to its own reality. Nvidia burned all their bridges in the business and lost a lot of credibility lately; It's not CityBank, Enron or any chemical weapon's fault!

You surely have a impressive sense of theatrical exaggeration by saying such things as "it won't burn a hole through the bottom of your case and go on to contaminate the water-table on its way to China". That made laugh. Fermi is way over acceptable limit for power draw and heat dissipation, can you live with that fact? Yes, Nvidia engineers tried hard, A+ for the effort. Unfortunately for them, we live in a world driven by results. For that, they get a C-, period.

Your next statement "And surely it can't be judged entirely on its own merits" is very questionable, to say the least. Since when *anything* is not judged on its own merits?? I judge Fermi for what it is, not what I want it to be or what it could hypothetically becomes. When and IF the 512sp, true 250W, running at below 80c version comes out at 450$ or less, THEN we will talk. Until then, 480 = FAIL.

Yes, designing a GPU is VERY hard. But Fermi is not a GPU. It's a GPGPU and that's even harder to do. Cudos for trying, but maybe Nvidia chewed at something far too big for their own good. That, unfortunately, was a lack of judgment on their part. In this business, that's the kind of fatal mistake that doesn't forgive. Too bad for them, I have no pity whatsoever.

That's called evolution by natural selection or survival of the fittest. If it has to go that way, something stronger and more competent will replace them and that's all. Nearly happened to ATI before AMD bought them. Can happen to anyone in the business if you are not careful enough or become too complaisant.

Nvidia came out of nowhere to eventually buy 3DFX and became the dominant player. They could be gone as rapidly as they came if nothing changes within the organization...

 

[Genx87]

You logic is flawed. You are using a straw man fallacy argument by trying to associate Citybank, Enron and chemical weapons (!!!!) with Nvidia. Don't mix things up like they are connected in some way when in fact there is no relation whatsoever beyond the boundary of your own argument. One bad thing don't make another bad thing good, as you try to imply. Each case must be judged according to its own reality. Nvidia burned all their bridges in the business and lost a lot of credibility lately; It's not CityBank, Enron or any chemical weapon's fault!

You surely have a impressive sense of theatrical exaggeration by saying such things as "it won't burn a hole through the bottom of your case and go on to contaminate the water-table on its way to China". That made laugh. Fermi is way over acceptable limit for power draw and heat dissipation, can you live with that fact? Yes, Nvidia engineers tried hard, A+ for the effort. Unfortunately for them, we live in a world driven by results. For that, they get a C-, period.

Your next statement "And surely it can't be judged entirely on its own merits" is very questionable, to say the least. Since when *anything* is not judged on its own merits?? I judge Fermi for what it is, not what I want it to be or what it could hypothetically becomes. When and IF the 512sp, true 250W, running at below 80c version comes out at 450$ or less, THEN we will talk. Until then, 480 = FAIL.

Yes, designing a GPU is VERY hard. But Fermi is not a GPU. It's a GPGPU and that's even harder to do. Cudos for trying, but maybe Nvidia chewed at something far too big for their own good. That, unfortunately, was a lack of judgment on their part. In this business, that's the kind of fatal mistake that doesn't forgive. Too bad for them, I have no pity whatsoever.

That's called evolution by natural selection or survival of the fittest. If it has to go that way, something stronger and more competent will replace them and that's all. Nearly happened to ATI before AMD bought them. Can happen to anyone in the business if you are not careful enough or become too complaisant.

Nvidia came out of nowhere to eventually buy 3DFX and became the dominant player. They could be gone as rapidly as they came if nothing changes within the organization...

Did Craig234 make a new account? lol

 

[Dark4ng3l]

The guy has a point. Nvidia has been falling behind big time in the performance per die space and per watt because they were migrating more and more towards the ideal tesla board than the idead Geforce board. The reality is they probably should have built 2 products. The reality is that for some reason G80 probably made them feel invincible and that even though they were focusing less and less on the graphics card aspect that in the end their infallible superior design would always compensate for the required lost efficiency.

Maybe they thought that throwing more money into the design could overcome those problems. Fermi aint that bad but it's performance per die space and per watt is not impressive at all. But in the end its biggest problem is that its just too big and too ambitious a chip.

 

[Ramon Zarat]

LOL....yawn!

Please, elaborate because I found illogical to laught and yawn at the same thing! Don't you?

What exactly in what I've said you think is funny? If it made you laugh, I assume you read my entire post, right? Then, why the yawn comment if it was so entertaining that you've read it all?

Ah ha! I got you there!

 

[Ramon Zarat]

Did Craig234 make a new account? lol

My name is Ramon Zarat.

I will start by asking you, please, refrain from quoting me and writing at the third person, like I'm not present. That's very impolite and insulting. I will add that If me and my position are not welcomed here, just say it straight in my face or report me to the moderator if you think I was out of line.

In the mean time, please comment on the facts I put forward instead of simply laughing. You don't agree with me? Fine, that's what forum are for: civilized debate.

 

[Ramon Zarat]

Snip pointless troll.



I like the 470 because it has a nice price\performance ratio for me. Nvidia also handles TVs better than ATI in my experience. My new rig if I build it will be used on a 67" DLP. I went ATI last generation with a 4850. I want to try the other side of the world now. Time frame isnt a big deal for me either. I have to gather the parts for a new rig anyways.

Fair enough, thanks for taking the time to answer my interrogations.

 

[pmv]

@Ramon Z

I entirely agree that Fermi isn't a good product. If I had the money I'd go ATI/AMD, no question. But I just think that, not for the first time on this forum, someone seems to be treating things a bit more seriously or dramatically than they merit. "Master of unethical coporate practices" just seems like hyperbole in a world where corporations are generally amoral by definition and where the worst of them actually end up killing people. Probably nvidia have been a bit more ruthless about marketing than ATI have so far managed - I suspect its more likely that ATI/AMD will learn from them and descend to their level than the reverse though. I don't think its 'morals' that have kept them from doing so, corporations don't have morals, its probably just a slower-moving and more conservative corporate culture with respect to marketing.

I also still think that its too early to say what the GTX480 means for the longer term. Would it have been better if nvidia had given up completely and not released anything? I guess I just have a huge respect for engineers in general (though not for marketing departments) and feel some sympathy for those nVidia guys trying to salvage something from what was clearly a difficult situation. It may be that for the bigger, long-term picture they've done a good-enough job, as opposed to simply surrendering entirely and going off to do an arts degree. Time will tell whether this was a total dead-end or a temporary mis-match between design and resources. (I'd certainly love to read an honest insider's account of the development of Fermi.)

And, though the nvidia loyalists can write annoying posts, in a way I'm glad some of them are so committed that they'll buy Fermi whether its really objectively a good deal or not, because its in everyone's interest that a competitor to ATI/AMD continues to exist and gets the funds to develop something better.

 

[Kalessian]

I entirely agree that Fermi isn't a good product. If I had the money I'd go ATI/AMD, no question. But I just think that, not for the first time on this forum, someone seems to be treating things a bit more seriously or dramatically than they merit.

Meh, suit yourself, I think nV should die for this. And I hope all their employees starve to death. I mean, how dare they release a subpar toy and act like they haven't done anything wrong?

Jeez, don't you know? The vid card industry has been completely in balance since the beginning, with both sides perfectly balanced at all times. It's totally not like the CPU industry, where one side will win and then the other wins and it goes back and forth every now and then. Nope, what nV has shown us today is their true colors.

My only regret is that I don't have the time like you guys to write pages and pages of outrage over this disappointing toy.

 

[Zebo]

Thermi's only problem is leakage.

-Makes nV pump more volts to chip to make it run, thus runs very hot and uses lots of power
-Makes them unable to run memory at full speeds but 2/3 instead
-Makes them cut down cores to 480 or less

I will be very interested to see if they can get it undercontrol with either current tech or 28nm. Then this chip will kick ass. At least double 5870's performance if they get 520 core @ 900Mhz & 4800Mhz respectively.

 

[zerocool84]

I don't think anyone can really argue against the fact that Nvidia is a highly unethical company. Their history is not very pretty at all in that regard.

The 2 things that made me go red was the constant renaming and the loooooooooooong wait and only kept themselves alive with hype that went no where for months.

 

[Dark4ng3l]

The 2 things that made me go red was the constant renaming and the loooooooooooong wait and only kept themselves alive with hype that went no where for months.

What about the focus group stealth marketing BS? Sure they are ''open about it'' these days but it really was not like that just a few years ago. Anyways you guys know what I think about the focus group....

 

[BenSkywalker]

Fermi is *currently* the meanest pieces of hardware available for GPGPU application, no one will ever contest that. C++ compiler, ECC memory and great double precision floating point performance is impressive and all, but that's beside the point, isn't it???

Performance per watt, performance per dollar and performance for given space- all categories which they currently dominate for the HPC market, and that is all that matters in that market segment. Nothing in the HPC space is remotely close to Fermi, and they also happen to be the fastest single GPU solution. Given the stated design goals of Fermi, it seems to me that they have done precisely what they intended to do.

Fermi draws a lot more power they said it would and with only 480sp and lower clock, it perform a lot worse they said it would.

At the wall power draw of 320 watts on an 80 certified PSU is 256 watts draw at the board level. In terms of raw performance, the GTX480 is 3.3TFLOPS which is 10&#37; higher then what they claimed at first. So, power comes in at 2% higher then what they claimed with raw performance coming in 10% higher. So to recap, it performs a decent amount better then what they claimed and their performance per watt is also superior to what they had claimed.

You clearly want to bash nV- I would advise you stay far away from the HPC market, their is no question that they utterly dominate that segment and have overdelivered on their promises.

The fact is, Nvidia won't be alone in this market segment for long.

Perhaps, we will have to see how that ends up. Intel already ran away from the market once, ATi has shown no interest in entering it. Maybe they will be displaced next month by some incredible new part from another company, at the moment they soundly dominate the segment with noone else being remotely close using any performance metric you'd like to use.

 

[Lonyo]

Thermi's only problem is leakage.

-Makes nV pump more volts to chip to make it run, thus runs very hot and uses lots of power
-Makes them unable to run memory at full speeds but 2/3 instead
-Makes them cut down cores to 480 or less

I will be very interested to see if they can get it undercontrol with either current tech or 28nm. Then this chip will kick ass. At least double 5870's performance if they get 520 core @ 900Mhz & 4800Mhz respectively.

They do have another "problem" (issue/choice) which makes it inherently less effective than ATI from a size/performance per watt issue in gaming, and that's the focus on HPC, but like you say, it's nothing compared to the problems due to manufacturing which have lead to this giant mess. Arguably by designing a chip which can do more than just video game graphics, you are going to lose some ability to compete on efficiency and cost, but you do gain other benefits which may or may not impact consumers, but will help the company in the long term, but the impact of the HPC elements are nothing compared to the giant mess NV made with their design decisions approaching a new process.

The main things are that:
1) NV is now effectively trying to lie it wasy out of it (with the GTX480 TDP issue)
2) The problems are of their own making
3) They haven't ever helped themselves by admitting there are issues, and either try and blame others (e.g. "TSMC should have zero via defects) or cover it up (see 1).

I would think almost everyone on this forum knows that going with a new process and a new architecture is often a pretty terrible idea, as there are typically problems. NV should know from their own experience than going with unproven tech can be a risk (5800Ultra RAM).
Sometimes risks can pay off (e.g. HD48xx series going with GDDR5), but NV don't seem to have planned very much, unlike ATI who started off with the 4770 on TSMC's 40nm and worked through their problems and found solutions, while NV rushed headlong into it with a giant chip and got screwed over.

The root of the problem is production, it should be fixable, Fermi is not inherently bad, but this iteration of it sucks.
It should be easy for NV to come back swinging, but the product in its current state is pretty unarguably poor and due to NVs own choices. Luckily for them, and for competition, it shouldn't be a long lasting problem, and luckily for ATI, it gives them some more breathing room to try and consolidate their market position some more.

 

[AndroidVageta]

Im sticking with what I have! Crossfired 4890's even today have more than enough power in my eyes. Havent come across a single game that I cant max out (maybe not on the AA side of things, but that doesnt matter to me).

I still kinda feel neutral about Fermi. Sure its faster than the 5870, which is something that was expected, but as others have said, its too hot, too loud, and too much power! Now power and heat arent that big a deal to me, but Ive heard the GTX480 full speed and its louder than both my 4890's at 100% fan speed...and thats just completely unacceptable!

Until DX11 hits big or what not, what I have now is something that I am completely comfortable with, and thats saying a lot as Im always itching to upgrade, but Im actually quite happy with the performance that Im getting!

 

[RussianSensation]

Shamino Gets 1165MHz Core Clock on GeForce GTX 480 Video Card Using LN2
http://www.legitreviews.com/news/7697/

Looks like LN2 finally tames teh beast!!

I guess Asus is "safe" to claim up to 50&#37; overclock with voltage tweak (with exotic cooling).

 

[bryanW1995]

Where's the "meh" option. Good card but way too much juice and $ for me at this point. Very glad it's out though and *very* glad it's faster than the 5870 on some things (even if it's not even remotely as good of a value lol). It really seems like this architecture has a lot of promise for the future, maybe moreso than the 5xxx ati architecture. I'm curious what the $250 part in 6-9 months will look like.

it will probably be 480 shaders clocked at 700mhz.

 

[MrK6]

Shamino Gets 1165MHz Core Clock on GeForce GTX 480 Video Card Using LN2
http://www.legitreviews.com/news/7697/

Looks like LN2 finally tames teh beast!!

I guess Asus is "safe" to claim up to 50% overclock with voltage tweak (with exotic cooling).

Great find, thanks for posting! This result shows what this architecture holds. With a revision (and probably a process shrink, although I'm not that savvy with GPU design), this architecture could have a lot of potential. However, this also shows you its problems - heat and leakage. If they can bring the performance down to a realistic TDP/thermal envelope, they'll have a winner on their hands.

 

[Apocalypse23]

Shamino Gets 1165MHz Core Clock on GeForce GTX 480 Video Card Using LN2
http://www.legitreviews.com/news/7697/

Looks like LN2 finally tames teh beast!!

I guess Asus is "safe" to claim up to 50% overclock with voltage tweak (with exotic cooling).

My thoughts exactly

 

[Apocalypse23]

Shamino Gets 1165MHz Core Clock on GeForce GTX 480 Video Card Using LN2
http://www.legitreviews.com/news/7697/

Looks like LN2 finally tames teh beast!!

I guess Asus is "safe" to claim up to 50% overclock with voltage tweak (with exotic cooling).

My thoughts exactly.

 

[mooncancook]

At the wall power draw of 320 watts on an 80 certified PSU is 256 watts draw at the board level.

Another way to look at it relative to the 5870 TDP is that since the 5870 has a 188W TDP, and the 480 drawing about 100W more on load, assuming PSU efficiency of 80-85, I think it comes to be about 270W TDP, not too far off I guess.

 

[kalrith]

The 5850 remains the best bang for the buck, so that's what I'll probably get. Even if the price per performance were the same for Fermi, the excessive heat and power consumption are too big of a turn off for me.

 

[Ayah]

I was sort of hoping that Fermi would be a revolutionary jump and force AMD to drop prices competitively. (Competition = good for us)
Needless to say, I'm not too impressed. We, consumers, will only be shafted.