Your second post says that, not the original one.
Then I believe you're mighty pissed at how poor the GX2 scaling is as of today ..I tend to criticize the leader for it's shortcomings.
Sorry if i didn't meet your expectations.
NVIDIA GT200 Rumours & Speculation Thread
- Thread starter Arun
- Start date
- Status
Then I believe you're mighty pissed at how poor the GX2 scaling is as of today ..
I'm mighty pissed that dual-GPU cards from any vendor even have to exist (to fill in the blank until the "real" next generation chip).
you'll love this quote then
"Advanced Micro Devices' ATI graphics chip unit doesn't want to build "huge" chips like rival Nvidia, an executive says."
""We took two chips and put it on one board (X2). By doing that we have a smaller chip that is much more power efficient," said Matt Skynner, vice president of marketing for the graphics products group at AMD.
"We believe this is a much stronger strategy than going for a huge, monolithic chip that is very expensive and eats a lot of power and really can only be used for a small portion of the market," he said. "Scaling that large chip down into the performance segment doesn't make sense--because of the power and because of the size."
Skynner said that AMD tries to design GPUs (graphics processing units) for the mainstream segment of the market, then ratchet up performance by adding GPUs rather than designing one large, very-high-performance chip.
Nvidia's "strategy is to design for the highest performance at all cost. And we believe designing for the sweet spot and then leveraging for the extreme enthusiast market with multiple GPUs is the preferred"
"Advanced Micro Devices' ATI graphics chip unit doesn't want to build "huge" chips like rival Nvidia, an executive says."
""We took two chips and put it on one board (X2). By doing that we have a smaller chip that is much more power efficient," said Matt Skynner, vice president of marketing for the graphics products group at AMD.
"We believe this is a much stronger strategy than going for a huge, monolithic chip that is very expensive and eats a lot of power and really can only be used for a small portion of the market," he said. "Scaling that large chip down into the performance segment doesn't make sense--because of the power and because of the size."
Skynner said that AMD tries to design GPUs (graphics processing units) for the mainstream segment of the market, then ratchet up performance by adding GPUs rather than designing one large, very-high-performance chip.
Nvidia's "strategy is to design for the highest performance at all cost. And we believe designing for the sweet spot and then leveraging for the extreme enthusiast market with multiple GPUs is the preferred"
But ATI never intended the prices to be that low; they were forced to drop the prices to that level because the performance turned out to be so dismal that it was the only way they could sell any.
Unknown Soldier
Veteran
8800GTX: $599
2900XT: $399
8800GTS 512MB: $299
8800GT: $249
3870: $219
Whats next? Comparing a $169 9600GT to a $100 3450?
Sorry serenity, but if the 2900XT had actually preformed, we would've see a comparable $549 or there abouts price tag on it. It didn't and AMD dropped the price even further.
Same thing with the 3870, it's supposed to be a superior part. It isn't, hence the related price difference.
US
Courtesy: Inq
I agree that AMD had to drop the price on the 2900XT. But that doesnt make the comparison (2900XT v 8800GTX) right.
3870 performed as per their expectations, I might even add it enabled them to win some market share back from Nvidia.
Last edited by a moderator:
The multi-GPU strategy makes a lot of sense if they can offer a more seamless, transparent offering than traditional AFR offerings of today.
The limitations of AFR from load balancing; to scaling issues; from titles not working at all; to the dreaded wait for official profiles as the only viable choice for the high-end consumer is troubling to me.
The choice of AFR is a sound one over-all; the choice of only AFR at the high-end is not enough but that's just a view -- one of many!
The limitations of AFR from load balancing; to scaling issues; from titles not working at all; to the dreaded wait for official profiles as the only viable choice for the high-end consumer is troubling to me.
The choice of AFR is a sound one over-all; the choice of only AFR at the high-end is not enough but that's just a view -- one of many!
One could flip that argument and say that all NVIDIA "TWIMTBP" titles have built in NVIDIA optimisations. Thus creating inherit performance disadvantages for ATI cards.
That's all well and good, AMD, but everyone has moved onto GT200. -_-
Well they're positioning the RV770 against the G92b ..
Also interesting to note that they expect 9800GT to come in a $199.
That's not what JPR is saying about the quarters after the G92 and RV670 launches. Quite the opposite.
In fact, Nvidia has had several billion dollar quarters in this time frame, while AMD stagnated somewhat (which i could only attempt to explain by the fact that they sell more low-end/low-margin products like IGP's, HD24xx/HD34xx, etc-, while NV prefers to concentrate on higher margins, even if it means less market share than they could have had otherwise).
See, that's what i meant by disliking dual-GPU cards vs single-GPU models.
If you rely on a single mainstream chip, and that chip turns out to be the underdog, then the IHV will be in trouble because a dual-GPU card is a clear desperate measure in the eyes of consumers, reviewers and even developers.
Nvidia did it twice (7950 GX2 against X1950 XTX; 9800 GX2 against the previously released HD3870 X2), AMD will also do it a second time with the HD4870 X2.
This and the fact that Nvidia was able to counter the RV670 with the G92 and the G94 cores put significant pressure on AMD prices.
Now imagine if AMD didn't drop the ball and had released a true high end chip last Fall with, say, 480 stream processors and a 512bit GDDR4 bus on 55nm. Wouldn't that be preferable to a HD3870 X2 and significantly lift the bar for the competition ?
It would also command higher selling prices (not to mention the halo effect on the rest of the line), and that's what a cash-strapped company should do, seek more dollars.
One could flip that argument and say that all NVIDIA "TWIMTBP" titles have built in NVIDIA optimisations. Thus creating inherit performance disadvantages for ATI cards.
They did it because AMD's similar GITG program didn't do so well.
That's what a company without money to seed developers gets into.
AMD really needs to be back in the high-end market.
Last edited by a moderator:
ATI and NVIDIA are now on different release schedules ... NVIDIA starts at the high end, ATI resigns themselves mid end using mid end chips in "SLI" to have some high end reach (which as long as they have a hard time competing at the high end is probably their safest bet). GT200 isn't really what they are competing against ...
Funny how AMD and Nvidia are holding back on announcing the final clocks till the very last minute.
With respect to AMD's new strategy let's not forget that R600 was a 420mm^2 die at 80nm that was released only a year ago. It most certainly was an attempt to grab the single-GPU crown. How much of this new multi-GPU strategy is really a reaction to that part's inability to compete at the high-end of the market?
With respect to AMD's new strategy let's not forget that R600 was a 420mm^2 die at 80nm that was released only a year ago. It most certainly was an attempt to grab the single-GPU crown. How much of this new multi-GPU strategy is really a reaction to that part's inability to compete at the high-end of the market?
Unknown Soldier
Veteran
serenity, remove Philip Scholz's e-mail address please. First, it's a good way to get spammed, second, you can get into trouble. Rather put his name as you did with the second.
US
US
Last edited by a moderator:
[rant]Game devs insist they have to render each frame from scratch and it is taken from granted that somebody has to pull a miracle out of silicon that lets them continue to do exactly that.
I'm all for progress and additional performance, but IMHO it is hypocritical to declare AMD a loser just because they didn't build a monolithic billion+ transistor chip to speed up the rendering of sunshafts.[rant]
I'm all for progress and additional performance, but IMHO it is hypocritical to declare AMD a loser just because they didn't build a monolithic billion+ transistor chip to speed up the rendering of sunshafts.[rant]
So if that's such a big part of R700... Why isn't anyone talking about it? Why is no rumour mentioning it? Why is no 'leak' from ATI docs heralding it as the next best thing since sliced bread? I'm not saying it's not true; I'm just skeptical, and wouldn't be surprised if that rumour was, let us say, outdated.
And Lux_, uhhhh, I have no idea why you think it's even theoretically possible NOT to render his frame from scratch? (at least for the main framebuffer)
- Status
