Nvidia GT300 core: Speculation

Status
Not open for further replies.
There's nothing to salvage if they have < 10% yields. They would make far more money selling GT200 than they would GT300. It wouldn't be any different to what ATI did with RV670. Give up the high-end and do what you can to survive until things turn around.

I meant salvaging your entire development. once you've taped out it's obviously not smart to call it a day afterwards. These "tests" with a limited number of wafers running a fast-track are not adding the most costs ofcourse.
AMD could've killed R600 when they knew about their leakage problems, but they didn't maintaining some kind of face/dignity by slotting it lower than it's intended competition.
With GT300 however, what is it's intended competition? if it's LRB, it's still in a hot-seat to be out of the gates on time, if and only if they manage to solve their problematic "power" issues.

So now, the "sudden" (ahem) resurrection of GT212 makes more sense, not an ambitious 384-bit design, but it looks like someone or something smacked some sense into the engineers, maybe it was the wall they were running into?
 
Hahahahahaha!!! :LOL:

Bing's translation: NVIDIA is GT300 engineering sample completed in September, fastest in December of this year to start of delivery. On the other hand, the AMD Radeon HD 5800 (RV870), in the latter half of September, and shipping, scheduled in October.Radeon HD 5800 is first of all that on Radeon HD 5870 and 5850 2 model.

Google's translation: NVIDIA GT300 is an Engineering sample of nine will be completed in January of this year's fastest 12 to begin shipping in January. Meanwhile, AMD is Radeon HD 5800 (RV870) was launched in late May to September, the shipments are scheduled for July 10. First Radeon HD 5800 and Radeon HD 5870 will launch two models in 5850.

Which one seems more accurate eh? Google says RV870 launched in late May and will ship on July10, Bing says latter half of September ;)

Has anyone really been far as decided to use even go want to do look more like?
 
I meant salvaging your entire development. once you've taped out it's obviously not smart to call it a day afterwards. These "tests" with a limited number of wafers running a fast-track are not adding the most costs ofcourse.
AMD could've killed R600 when they knew about their leakage problems, but they didn't maintaining some kind of face/dignity by slotting it lower than it's intended competition.
With GT300 however, what is it's intended competition? if it's LRB, it's still in a hot-seat to be out of the gates on time, if and only if they manage to solve their problematic "power" issues.

So now, the "sudden" (ahem) resurrection of GT212 makes more sense, not an ambitious 384-bit design, but it looks like someone or something smacked some sense into the engineers, maybe it was the wall they were running into?
I cant imagine a G200 on steroids with a 256b mem interface would be feasable (without being bandwidth starved), Nvidia prove me wrong please.
 
And couldn't it be that Nvidia's design is poorly tuned for the process? It's a real question, not a rhetorical one, I don't know much about that kind of stuff.
As their design would be based upon TSMC's libraries, just like ATI's is, ATI would be having the same problems.
 
As their design would be based upon TSMC's libraries, just like ATI's is, ATI would be having the same problems.

That's quite an assumption that two different architectures and design philosophies, much less different die sizes, would have the same problems

Not to mention that increased die size does not decrease yield linearly... it's exponentially
 
this site:
http://translation.babylon.com/Japanese/to-English/#

translate charlies source:
NVIDIAもAMDも2009年下半期にDirectX 11対応の最新GPUをローンチしようとしている。

NVIDIAはGT300のEngineering sampleを9月に完成させ、最速で今年の12月に出荷開始とする。
一方、AMDはRadeon HD 5800(RV870)を9月後半にローンチし、出荷を10月に予定している。Radeon HD 5800はまずRadeon HD 5870と5850の2モデルが投入される。
 
さらにAMDはフラッグシップモデルとなるDual-GPUカードを2009年第4四半期に予定している。このカードは“Hemlock”と呼ばれていたものでRadeon HD 4870X2の後継となる。製品名はRadeon HD 5870X2となると思われる。Radeon HD 5870X2の詳細は明かされていないが、2GBのGDDR5を搭載し、出力としてDisplayPort, HDMI, DVIを搭載するといわれている。

RV870とGT300については過去の情報とそれほど変化はありません。 GT300の方はまだ若干流動的要素がありますが、RV870は9月22日ローンチ・10月出荷の方向でだいぶ固まってきています。9月10日には RV870に関するイベントもあるようで、その時に新たな情報が明かされることが期待されます。
Hardware-InfosにはRadeon HD 5800シリーズの価格が掲載されており、これによるとRadeon HD 5850が$279~299、5870が$379~399、5870X2が$599となっています。

into:
Both NVIDIA and AMD are going to do raunch of latest GPU for DirectX11 in the second half of the year in 2009.

NVIDIA completes Engineeringsample of GT300 in September and I am fastest and do shipment with starting in December of this year.


On the other hand, AMD does raunch of RadeonHD5800 (RV870) in the latter half of September and plans shipment in October. At first, as for RadeonHD5800, RadeonHD5870 and 5850-2 model are spent.

Furthermore, AMD schedules a Dual-GPU card becoming the flagship model for a fourth-quarter in 2009. This card becomes the succession of RadeonHD4870X2 with a thing called "Hemlock". It seems that the product name becomes RadeonHD5870X2. The details of RadeonHD5870X2 aren't revealed, but it carries GDDR5 of 2GB and is said to carry DisplayPort,HDMI,DVI as the output. There isn't the change with past information about RV870 and GT300 that much. The one of GT300 still has a few style dynamic element, but RV870 hardens considerably in the direction of .10 raunch Tsukide load on September 22. We seem to have the event about RV870 on September 10, and it is hoped that the sometimes new information is revealed. A price of the RadeonHD5800 series is printed in Hardware-Infos, and, according to this, $ 379-399, 5870X2 become $ 599 $ 279-299, 5870 RadeonHD5850.
 
Interesting. I thought that G300 was fucked up beyond the ability to launch in the next 9 months.
Or am I misundertood something, neliz?
 
I dont believe for a moment the 1.67% yield thing, and it is hard for me to digest even the 10% number, but yeah, a 10% yield is too fucked up to launch in my book.
 
Interesting. I thought that G300 was fucked up beyond the ability to launch in the next 9 months.
Or am I misundertood something, neliz?

Well now, Charlie thinks that blog is a rock-solid source due to the conveniently bad translation that falls in line with his wildest dreams. But that doesn't mean the proper translation means anything either - could be some guy just pissing in the wind as usual.
 
If they're process errors, then they're hardly nVidia's fault. They would be nVidia's problem, of course, but not their fault.
TSMC does not exist in a vacuum. They work with their partners and said partners leverage the strengths, and minimize the limitations. Now if for example Nvidia decides to push the outer boundaries of what TSMC is capable of on a certain node, then yields will be poor. And if Nvidia is not able to fine tune their designs/layout as well as AMD, then AMD gains an advantage, and of course the reverse is true.

It's not like baking cookies, where you can put the same dough into just about any oven and get tasty every time. Well it's sorta like baking cookies, the oven is not exactly perfect, the cookies on the edge of the baking sheets are never as good, the bigger the cookie the harder it is to get it perfect, and the oven will burn the cookies to a crisp if you don't get the recipe right.
 
Well now, Charlie thinks that blog is a rock-solid source due to the conveniently bad translation that falls in line with his wildest dreams. But that doesn't mean the proper translation means anything either - could be some guy just pissing in the wind as usual.

Too be fair, he also said he did more digging to back up the blog. Of course, some want to yell that he obviously didn't do any more digging and it's all fake, but you know.

Charlie seems to gaining credibility because he's right a lot lately. His dripping hate of Nvidia aside. Of course a lot of people are going to disect my statement, but that's not important.
 
Interesting. I thought that G300 was fucked up beyond the ability to launch in the next 9 months.
Or am I misundertood something, neliz?

The guy can't even get the simplest product names right.
If my mother blogged that AMD would relaunch the 9700 Pro, would it make it true then?
 
The guy can't even get the simplest product names right.
If my mother blogged that AMD would relaunch the 9700 Pro, would it make it true then?

to be fair, with the constantly changing codenames and rebadging, it's a suprise other than a handful of the people can get product names right..
 
So are you guys done yet translating every piece of crap you can possibly find around various websites?

Who's been right on what exactly? WTF is the supposed "GTX280 1.5GB", the "GT212" or any other phoenix that's supposed to step out of any ashes again?

It's bleedingly obvious that D12U is facing a delay. Now the debate is about what exactly whether its a huge or a small delay? It'll be "huge" for an ATI fanb** even if it's just a week and vice versa. It's definitely not 9 months and there is no such thing as a >GT215 in the pipe either.

I'd say breathe in and concentrate what such a chip could bring to the table of innovation (or lack thereof). Each and everyone's agenda and ulterior motives leave me fairly cold.
 
The guy can't even get the simplest product names right.
If my mother blogged that AMD would relaunch the 9700 Pro, would it make it true then?
I'm having a bit of a translation crisis. Who are you talking about? (Remember, English isn't my native language so I don't sence sarcasm as well as you probably think.)
 
I'm having a bit of a translation crisis. Who are you talking about? (Remember, English isn't my native language so I don't sence sarcasm as well as you probably think.)

Me neither ;)

There are some errors in his product naming, and thus, reliability for that piece (translated or not) is very close to 0%.
 
Status
Not open for further replies.
Back
Top