PDA

View Full Version : Nvidia G80 (NV50) has around 500 million transistors


NV55
June 13th 06, 11:21 PM
http://www.beyond3d.com/forum/showthread.php?t=28551&page=9


QUOTE:

.................................................. .................................................. .................................
"Some G80 details from a JP Morgan investor conference on 5/24. It is ~
500 million transistors and will be done when its done, but they are
shooting for Septemberish."
.................................................. .................................................. .................................

nv55
June 14th 06, 05:57 AM
further proof that G80 / NV50 has over half a billion transistors


http://www.beyond3d.com/forum/showthread.php?p=775737#post775737


Q: When's your next gen part?


Mike Hara, VP Investor Relations on 5/24/2006:

A: "Well, you know GPUs are getting so large and so complex the timing
of it is almost one of those things that the timing of it is 'it comes
out when it comes out'. . .as much as you try to plan it for a certain
event it really is on a schedule that says 'when it's done it's done'.
.. .now the reason why that works, at least for the high-end. . .the
first high-end gpu that comes out. . .is because it is going to get
purchased by enthusiasts who really don't care if it is Christmas, the
middle of summer, or spring. If it has a discernible advantage over the
last gpu, they'll buy it. So our schedule right now on the next
generation GeForce [Hmm, that's the first time we've heard them confirm
it is still 'GeForce'?] is going to be second half, and the objective
is to hit it for 'back to school'. But fundamentally we're really just
targeting second half. Which means that the current GeForce 7 family,
with the exception of the high-end, is really the family you're going
to see in the back to school cycle. [So what I got out of that: if next
spin goes well, they hit 'back to school' for a high-end part. If not,
it'll be later. And no full family simultaneous launch.] But this one
is. . .we kind of describe it inside the company as. . . 'this is
probably the biggest architectural change in the company's history from
one generation to the next' [Hmm!] I'll give you a little bit of
insight. . .this device is going to be over half a billion transistors
large. It will, without doubt, be the most complex device being built
in the semi-conductor business today. So the current schedule right now
is to have it in the second half. And we'll do typical, which is a
'hard launch', which means we'll launch it when it's actually available
in retail."


__________________________________________________

AUDIO: http://www.megaupload.com/?d=ZRUFCLGL

Lorenzo Sandini
June 14th 06, 10:30 AM
> large. It will, without doubt, be the most complex device being built
> in the semi-conductor business today. ...

.... and one of the most expensive probably. Be ready to sell one kidney
and one cornea for the basic model, half your liver and some skin on top
of that for the crunchy version.

Lorenzo

Big Bird
June 14th 06, 12:00 PM
Lorenzo Sandini wrote:
>
>> large. It will, without doubt, be the most complex device being built
>> in the semi-conductor business today. ...
>
>
> ... and one of the most expensive probably. Be ready to sell one kidney
> and one cornea for the basic model, half your liver and some skin on top
> of that for the crunchy version.
>
> Lorenzo

theres no point making something that no one can afford.
It will cost probably the same as the 7900GTX did at release time.

Lorenzo Sandini
June 14th 06, 02:59 PM
Big Bird wrote:
> Lorenzo Sandini wrote:
>>
>>> large. It will, without doubt, be the most complex device being built
>>> in the semi-conductor business today. ...
>>
>>
>> ... and one of the most expensive probably. Be ready to sell one
>> kidney and one cornea for the basic model, half your liver and some
>> skin on top of that for the crunchy version.
>>
>> Lorenzo
>
> theres no point making something that no one can afford.
> It will cost probably the same as the 7900GTX did at release time.

In my experience the best cards nowadays cost more than the best card 2
years ago, which cost more than the best card 4 years ago, and so on. I
remember when my beginning-of-death Voodoo 2 12 MB and Matrox Millenium
8MB cost me 300 euros together, and that was a high price for a graphic
card (combination). Now some pay around 500 euros for theirs...

What's the point of making an Aston Martin, or a Bentley ? You'll
always find a buyer, whatever the price.

Lorenzo

Neil J Bundy
June 14th 06, 08:25 PM
In message >, Lorenzo Sandini
> writes
>
>> large. It will, without doubt, be the most complex device being built
>> in the semi-conductor business today. ...
>
>... and one of the most expensive probably. Be ready to sell one kidney
>and one cornea for the basic model, half your liver and some skin on
>top of that for the crunchy version.
>
>Lorenzo

Catch 22 - sell the cornea & you don't get to see the benefit of the
latest flashy graphics :-/ LOL

But... 3 months down the line and the next one is out so you can pick
that one up at a half decent price... Same story as always, it's a case
of when to jump on the bandwagon, and how high to jump!

--
Neil J Bundy

June 14th 06, 08:38 PM
On Wed, 14 Jun 2006 21:00:15 +1000, Big Bird
> wrote:

>Lorenzo Sandini wrote:
>>
>>> large. It will, without doubt, be the most complex device being built
>>> in the semi-conductor business today. ...
>>
>>
>> ... and one of the most expensive probably. Be ready to sell one kidney
>> and one cornea for the basic model, half your liver and some skin on top
>> of that for the crunchy version.
>>
>> Lorenzo
>
>theres no point making something that no one can afford.
>It will cost probably the same as the 7900GTX did at release time.

Oh good only 500.00+ bucks ! ;-(

First of One
June 15th 06, 12:11 AM
Just pimp out your sister for $90 a "trick" and you'll collect enough in no
time....

Jokes aside, I seem to recall the G80 will be on the 65 nm process, so
nVidia should be able to fit the same number of dies on a wafer, keeping
costs within reason. Of course, the GPU accounts for only half the cost. The
other half is typically paid to Samsung for ultra low-yield RAM modules.

I wouldn't be surprised if the G80 turns out to be not much faster than the
7900. Unified shaders would add much complexity, but not performance, in
today's games.

--
"War is the continuation of politics by other means.
It can therefore be said that politics is war without
bloodshed while war is politics with bloodshed."

"Lorenzo Sandini" > wrote in message
...
> ... and one of the most expensive probably. Be ready to sell one kidney
> and one cornea for the basic model, half your liver and some skin on top
> of that for the crunchy version.
>
> Lorenzo

/mel/
June 15th 06, 06:11 PM
"First of One" > wrote in message
...

> Jokes aside, I seem to recall the G80 will be on the 65 nm process, so
> nVidia should be able to fit the same number of dies on a wafer, keeping
> costs within reason. Of course, the GPU accounts for only half the cost.
> The other half is typically paid to Samsung for ultra low-yield RAM
> modules.

Same number of dies, but a more costly wafer, and a lower yield.

/mel/

John Lewis
June 15th 06, 07:34 PM
On Thu, 15 Jun 2006 18:11:01 +0100, "/mel/"
> wrote:

>"First of One" > wrote in message
...
>
>> Jokes aside, I seem to recall the G80 will be on the 65 nm process, so
>> nVidia should be able to fit the same number of dies on a wafer, keeping
>> costs within reason. Of course, the GPU accounts for only half the cost.
>> The other half is typically paid to Samsung for ultra low-yield RAM
>> modules.
>
>Same number of dies, but

initially

> a more costly wafer, and a lower

initial

> yield.

It takes 1-2 years for process maturity, at which time the
cost/wafer and yield ( for the same die area ) is the same or
less than that of the previous process at the same maturity
stage. A corollary to Moore's Law.

How long silicon can continue to shrink without serious
current leakage and/or voltage-breakdown issues is the
$64 question.

John Lewis

>
> /mel/
>
>