First of all, lets get this out of the way -the hd7970 will undeniably be the fastest single GPU on the planet when it's officially available. And yes, at only $50 more than the MSRP of the gtx580 1.5gb, it is a better value. But therein lies the problems. The gtx580 was generally a horrible value to begin with AND can be had quite regularly for $450-470. At $450, the gtx580 provides the same performance per dollar as the hd7970 at $550. So what AMD is doing is essentially pricing the hd7970 based on it's performance against 40nm GPU's, instead of pricing it more aggressively at around $450 (which is still a significant premium over the last several generation of AMD/ATI gpu's) and completely obsoleting the gtx580.

I know super high end GPU's are generally regarded as halo products, but I think AMD is going to end up with less revenue and overall profit by pricing the card along the lines of current cards that are out in much the same manner of the hd5870/5850 launch (only in a reversal of price situation). Nvidia did a very similar thing with the gtx280/260 launch, but their mistake was extremely short lived and they quickly adjusted it's pricing tiers.

AMD is making a pricing mistake. Sure there will be a handful of internet braggers that will grab that card, but beyond that the expected price will be a hard sell.

Everything you say is right. However, the HD 7970 will only be a pricing success IF they can get the follow-up to HD 7970 out the door in a timely fashion and it is a real blow-for-blow competitor to Kepler
(why i keep saying that some parts might be disabled or else they will get regular clockspeeds over 1100MHz)

THIS is what might make AMD's strategy work. Their confidence that Nvidia will be unable to deliver before late Spring.

i believe it also could backfire on AMD - IF Nvidia can blow OverClocked Tahiti into the weeds with Kepler, then AMD will have failed in their pricing strategy and they will alienate their enthusiast fan base like they did with Bulldozer. Right now, the HD 7970 is something for them to be proud of.

I wonder how much truth there is to the last few sentences on this page: http://www.legitreviews.com/article/1805/16/

"We talked to NVIDIA today and they said they plan on releasing their next generation processor in the first half of 2012 and our sources off the record said it would be at the end of that time frame. That means AMD might have the lead for the next six month!"

Well, my sources off the record say (Nvidia's) Q1


Off the record is completely meaningless unless your source is JHH

Let's look at this logically.

Nvidia was expecting the HD 7970 in January - Q1 no doubt and probably they also expected faster than what we see
--if they were expecting a January launch, we should expect that they planned for AMD's launch better than they did with Fermi GF100.

Do you see any real pressure on Nvidia with this AMD Launch? Do they need to be in a hurry, marketshare-wise?

Imho,

What they're doing is using the 40nm status quo pricing and not really adding any performance value or redefining price-points even though it is a new arch with a significant new node. Hey, I think I've typed this a few times, hehe!:)

They're simply raising or attaching another premium and somehow the GTX-580 has become an objective pricing barometer but already has a significant premium that is over a year old.

Execution has its advantages and pricing flexibility is one of them. AMD may enjoy the advantages and fruits of this and hopefully they can garner some mind share for the brand-names, revenue, profit and impressive margins. I don't have a problem with this.

The problem I have is the ironic stance of some posters -- really enjoy the irony from the sweet spot and how the GTX 580 was insane premiums to now premiums and higher prices are good. Only in Forums and video cards, hehe!:)

The key is if the HD 7970 is around 40 percent faster than a HD 6970 and the starting point is 549, what kind of value/performance will there be at the sweet spot this round? Curious here.

There could be an opportunity here from nVidia to expose if they get more performance out of this process for their Kepler generation but the key is when? And how much performance? Will nVidia counter by adding or attaching more premiums or try to redefine the GTX-580 and GTX 560 TI performance for their original price-points?

Another point: Since AMD has priced their sku's higher.......nvidia's Fermi product line may be still very competitive, with some differentiation and strong brand-name -- nVidia may still keep market share, revenue and margins 'till Kepler. There in much better shape than what happened with Fermi and competing against a lower cost HD 5870 and derivatives with a big die DirectX 10 product family and derivatives.

All this is really talking out of my ass but it's fun to speculate without getting so emotional.

Short answer. No. I think they know exactly what they're doing.

AMD's PR, a guys who actually WRITES the PR and does the slides; one whose JOB depends on the HD 7970 launch is a ringleader there and he coordinates replies with a few of the supporting choir who also gets their kit from AMD.

Their purpose there is to SELL video cards and to smear the competitors. They have a stake in it. They have a free license to attack and your character and motives while they themselves can get away with anything.

Now back to the topic. THIS is one of the most amazing chess matches between two companies - in the short history of tech; beginning with ATi vs nVidia and now with AMD. i love it!!


AMD launched the HD 2900XT with the beginning of a change from the way ATi traditionally countered Nvidia. THEY were the ones who decided the strategy was to go after the midrange and they came up with the "value" and "performance sweet spot". Well, they just admitted that Nvidia was right all along and they have done an amazing about face

Nvidia's response will be interesting to see. If they drop prices on the GTX 580, it will tell us something



Of course, they know what they are doing. What we need to see is Nvidia response, Kepler's arrival and THEN how AMD responds to see if they are successful at their new (old) strategy.

Do they?

They have changed directions a few times since they picked up ATI. This is another direction -- another chapter.

Good luck with them trying to attack my character -- I have over a decade of posting and consistent -- will enjoy the challenge.