NVIDIA Defends GTX 480's Heat & Power Consumption

[Rob Williams]

A little over a week ago, NVIDIA finally took the veil off of its first GF100-based graphics cards, which came in the form of the GeForce GTX 470 and GTX 480. As we quickly found out, whether or not the cards were worth the wait depended on your point-of-view, but as it seemed, the majority of people (based on comments around the Web) were hoping for a little bit more.

​

You can read the rest of our post here.

 

[x.clay]

I think Nvidia had a hand in the high expectations. Combine Nvidia PR with wild rumors and speculation and a 6 month delay and I think most people were at least hoping the long wait would be worth it.

 

[b1lk1]

I still feel the wait was worth it, it just isn't what people were dreaming of. So many people were hoping it would be double the performance of the HD5870 and that was a foolish and unrealistic expectation.

My take is this. Nvidia has a completely different market in mind with these cards than ATI does with their HD series cards. Nvidia is aiming at professionals and super computing while ATI is more focused on gaming. There is no middle ground here. ATI has the market cornered in gaming if you ask me right now, especially with Eyefinity and how easy it is to setup. They have hard product out there filling all our needs and Nvidia's paper launch did them more harm than good. People were already resentful enough and for Nvidia to foolishly launch these things without hard product is just laughable. They did themselves no favor in this and they can only blame themselves.

Bottom line, GTX480 is the fastest single GPU card made. Sure it can cook toast, but does that really matter in the end to the people buying them? I highly doubt it. This is a successful launch and there is no doubt in my mind they will not be able to meed demand for atleast the next 6 months.

The cards are super hot, there is no doubt there, but isn't that why we have these new super-duper high airflow gaming chassis anyways???

PS: Hey Rob, there are quite a few Ferrari models that are track only cars, but again, the people buying them could care less.

 

[Kougar]

I was expecting a minimum of a 30% performance advantage against the HD 5870, that would have been sufficient to justify the heat, power, and cost negatives. 15% in my opinion doesn't justify squat. Take a game that runs 100FPS on a 5870... is it worth it to buy a GTX 480 just to run 115FPS? I think not. No game I know of requires a GTX 480 to achieve playable framerates.

It certainly does matter, I chose to not buy one after waiting eagerly for this card. I'm an avid gamer and I do plenty of Folding@home crunching on the GPU, in addition to whatever else will run on it that is neat to play with. Which means NVIDIA appeals to me for two different markets, meaning I have above-average interest in their products. But I'm not going to pay $100 more for a GPU that is only 15% faster in games, uses MORE power than a 5970 (!), and runs so incredibly hot that I would worry about running it in 24/7 loads for Folding@home (or any other super-computing applications that you mention) for extended periods.

They lost my business with this one, that's just fact.If ATI cards weren't worthless at folding (for the time being, this will change soon), I would already own a 5870 instead. Once Stanford shifts away from CUDA, it will be anyone's game as to who gets my business again.

 

[Optix]

"General Manager of GeForce"

Bwahahaha!