Guys, stop spreading FUD.
Kyle from Bennett from HardOCP has this to say about Fudzilla's garbage:
http://hardforum.com/showpost.php?p=1036392508&postcount=16
Also, remember that Fudzilla was the one spewing garbage about how Barts was delayed till November. When was Barts launched? Hmm? Hmm?
Fudzilla = incompetent at best, hired gun at worst
That 61% was MOBILE discrete -figure.
Nvidia's Desktop Cards Shipments Skyrocket as Mobile Chips Sales Collapse
http://www.xbitlabs.com/news/video/..._on_Discrete_GPU_Market_Mercury_Research.html
Did you seriously just link to The Inquirer?
Oh God, help us all...
TPU says the 580 is 520mm^2 (Nvidias words to him).
So its 9mm^2 smaller than the 480.
but its going to compete agasint a 6970 that is expected to be 360mm^2.
So chip for chip... the 580 probably costs 45% more to make (IF the costs scaled where linar, their not.. so it ll cost more than 45% (becuase bigger = less yeild %'s ontop of the added cost of just more waffer used pr chip)).
So nvidia probably wont get rich with their 580 if it goes into a price war with the 6970's.
they use 5000MHz gddr5 just like ATI but the controller was the bottleneck.The thing is though, none of us actually know the margins on these things Nvidia or AMD actually claim.
We can assume that the price cuts are not entirely absorbed by the partners, as there would be none left if this were the case. However, all we can say is the GPU is more expensive.
For all we know the margins on GPUs are healthy enough that both companies have room to move around and the cost of the GPU is relatively insignificant compared to the other costs (memory and board) and markups.
I have no idea how much the GDDR5 costs, but Nvidia clearly uses far slower memory (whether they use cheaper chips on purpose or are just saddled by their memory controller I don't know). Simply not enough to go on.
I suppose if they absolutely refuse to budge on price to compete we could make an educated guess that that is the profit point... but shrug.
TPU says the 580 is 520mm^2 (Nvidias words to him).
So its 9mm^2 smaller than the 480.
but its going to compete agasint a 6970 that is expected to be 360mm^2.
So chip for chip... the 580 probably costs 45% more to make (IF the costs scaled where linar, their not.. so it ll cost more than 45% (becuase bigger = less yeild %'s ontop of the added cost of just more waffer used pr chip)).
So nvidia probably wont get rich with their 580 if it goes into a price war with the 6970's.
The entire Fermi architecture is a financial disaster for Nvidia so far but they don't have a choice - it's either (non-x86) CUDA or die and gamers with GeForces good for making up the R&D costs, nothing else.
Do we accept your statement as gospel like your total 230w power draw w\ OC i7 & 1000 core 5850...?...LMAO, I think not!
AMD has a HUGE lead in market share, over 61% now. It's a fact, not something debatable, period.
Do we accept your statement as gospel like your total 230w power draw w\ OC i7 & 1000 core 5850...?...LMAO, I think not!
PS: i7 @150+ Watts? Where, at 4.5GHz?
oc 5850 @ 1000 core w/voltage uses 200 watts alone.
oc i7 uses another 150+ watts.
The overall discrete market share is ~ 50.40% AMD / 49.60% for NV, whereby AMD is leading in the mobile sector, but NV has a 59% market share for desktop discrete graphics.
Sol, you have T2k confused with MrK6. ^_^
Sol, you have T2k confused with MrK6
6970 is late. 580 is the happening scene. this thread is now about apple pie
Rumors for the 6970's are saying 360mm^2
The 58xx cards are 334mm^2 (from 1year ago)
So them makeing a chip thats 26mm^2 bigger is gonna give them problems with the 40nm?
Nvidia are doing 529mm^2.
360mm^2 is tiny compaired to that, and like I said amd has had more time with the 40nm proccess. They didnt have any problems with the 334mm^2 they did a year ago, I dont think today with even more understanding of 40nm they have problems with a 360mm^2.
Much larger and complex?
Wasn't that fudzilla too?
From an interview with Eric Demers (AMD's GPU CTO), http://pc.watch.impress.co.jp/docs/column/kaigai/20101101_403541.html
Seems to be under 400mm^2 and only slightly larger than Cypress.
On the other hand, we all know that AMD is planning to unveil its high-end Cayman design before the end of the year—and there’s not a ton of time between then and now. Regardless of how Cayman performs, it’s a fairly safe bet that prices will shift to reflect the relationship between that that card and this one. There's a twist, though. Back when the Radeon HD 6870/6850 launched, we were given a date for the Cayman debut. As we inch closer to it and nobody anywhere knows anything about it (board partners haven't seen cards, system builders are still in the dark), it begins to look like there may be delays. From what I've heard, AMD won't even be briefing its partners until after that original embargo date passes.