NVIDIA Fermi: Architecture discussion

Groo The Wanderer said:
Edit: Forgot the NVIO chip. That could increase costs and lower assembled end product yields a bit. It is only tangentially relevant to the discussion though.

-Charlie
Click to expand...
Nvidia already explicitly said, there's not going to a an NVIO this time around. And all that's known about Fermi and what's deductible from that in a second step leads me to believe them, because it just wouldn't make sense any more.

L'Inq:
http://www.pcgameshardware.de/aid,6...ech-mit-Nvidias-Toni-Tamasi/Grafikkarte/News/

[Use online translator of choice as needed]

edit:
Ahhhh damn it. Too late.
 
Does anyone know what Mr. Bennett is talking about?

Actually from what I am hearing is that NVTTM is going away for the most part. NVIDIA will be building and selling the first few Fermis via Flextronics, but it is supposed to be small supplies this time. NV wants its AIBs to grow up.
smile.gif
This will change the face of retail card sales in North America greatly.
Click to expand...

http://www.hardforum.com/showpost.php?p=1035036711&postcount=19

Charlie also has an Nvidia article up and it's full of optimism. Must be the holiday cheer kicking in :D

http://www.semiaccurate.com/2009/12/10/fermi-a3-silicon-oven/

Nvidia has put the A3 stepping of Fermi in the oven, it happened some time last week. That means that you may see cards as early as February, depending on two or three other factors.
Click to expand...
 
Silus said:
Is he really using clock frequencies to compare Fermi and Cypress, even though Fermi is a new architecture ?

Why does anyone read/listen to this guy again ??
Click to expand...

I'd dare say his anti-nV bias is about equal to your pro-nV green tinted shades and matching knee pads, I'd go far to even say that at least Charlie knows half of what he's talking about.. ((and Yes OMG .. an almost Pro-nV post by Charlie.. has hell frozen over ?!)

And NO, I would not say Charlie is comparing Fermi v. Cypress based upon clock frequencies alone.. he is stating that when Fermi info was coming forth a few months back that the target clock rate was approx. 750 mhz and that supposedly current sillycon is yeilding 500Mhz.. and that at the projected 750 Mhz pro nV source were claiming that G100 would outpace 5870 by 40%. It's not THAT outlandish to come to the conclusion that at 66% projected (500 supposed / 750 projected) that performance would drop equally), unless you are claiming that a 50% increase in clock rate wont make a difference in regard to market placement and that there would be no such penalty imposed by a difference in clock rates.. really ?!
 
Last edited by a moderator:
FrameBuffer said:
I'd dare say his anti-nV bias is about equal to your pro-nV green tinted shades and matching knee pads, I'd go far to even say that at least Charlie knows half of what he's talking about.. ((and Yes OMG .. an almost Pro-nV post by Charlie.. has hell frozen over ?!)

And NO, I would not say Charlie is comparing Fermi v. Cypress based upon clock frequencies alone.. he is stating that when Fermi info was coming forth a few months back that the target clock rate was approx. 750 mhz and that supposedly current sillycon is yeilding 500Mhz.. and that at the projected 750 Mhz pro nV source were claiming that G100 would outpace 5870 by 40%. It's not THAT outlandish to come to the conclusion that at 66% projected (500 supposed / 750 projected) that performance would drop equally), unless you are claiming that a 50% increase in clock rate wont make a difference in regard to market placement and that there would be no such penalty imposed by a difference in clock rates.. really ?!
Click to expand...

I some how doubt he got any Green team insider to give him any kind of info, and if he did, it was propably BS to begin with just to make him look like a dolt when it launches.
 
XMAN26 said:
I some how doubt he got any Green team insider to give him any kind of info, and if he did, it was propably BS to begin with just to make him look like a dolt when it launches.
Click to expand...
I dunno, spreading rumours underselling Fermi at this point in time just doesn't seem a great idea to me unless they are going to go public with some real info soon.
 
digitalwanderer said:
And you didn't want to share it with US?!? :oops:

I am sooo hurt right now Neliz, truly. :(
Click to expand...

There's still people here that believe GT300 to be hardly bigger than Cypress, why would they believe me?

I think NV actually presented Fermi last week, Tuesday they updated everyone that "needs to know"
 
FrameBuffer said:
I'd dare say his anti-nV bias is about equal to your pro-nV green tinted shades and matching knee pads, I'd go far to even say that at least Charlie knows half of what he's talking about.. ((and Yes OMG .. an almost Pro-nV post by Charlie.. has hell frozen over ?!)
Click to expand...

pro-NV green ? Yeah, ok...my "news site" is filled with AMD hate...

I would rather call it anti-FUD and thus anti-Charlie, but let's stay on topic :)

FrameBuffer said:
And NO, I would not say Charlie is comparing Fermi v. Cypress based upon clock frequencies alone.. he is stating that when Fermi info was coming forth a few months back that the target clock rate was approx. 750 mhz and that supposedly current sillycon is yeilding 500Mhz.. and that at the projected 750 Mhz pro nV source were claiming that G100 would outpace 5870 by 40%. It's not THAT outlandish to come to the conclusion that at 66% projected (500 supposed / 750 projected) that performance would drop equally), unless you are claiming that a 50% increase in clock rate wont make a difference in regard to market placement and that there would be no such penalty imposed by a difference in clock rates.. really ?!
Click to expand...

Yet, it's exactly what he did. And of course it makes a difference, but in a new architecture, we don't know how much.

Also, what you're saying is that he's using unconfirmed data to support his conclusions of Fermi is X % faster than Cypress. Does that mean that the "pro nV source" was correct then ? With a target of 750 Mhz as the clock frequency, 40% faster was accurate ? So if 750 Mhz is achieved, Fermi will be 40% faster than Cypress ?
But since this is Charlie, obviously NVIDIA can't make it to that target and so that 40% number is wrong, which is why I'm not getting why some people are seeing this as "optimistic" :)
 
fellix said:
NVTTM == Nvidia Time To Market ?!
Click to expand...

Yeah but what's up with that crack about AIB's growing up? Are they pissing and moaning about not having cards to sell so Nvidia is gonna teach them a lesson? Or are they holding Fermi close to their chest to avoid the typical AIB leaks before launch? The latter would make sense if Fermi performance is a big surprise - especially in the underwhelming direction.
 
I really don't want to have to see another round of debate and speculation on Charlie himself.

Unless he's a future Fermi instantiation sent back from the future to kill John Connor, we don't need to go down this road again (again).

Sure we may debate briefly on whether given claims or sources may or may not pan out.
Going back to the personal jab-fest for pages is a waste of bytes.
 
trinibwoy said:
Yeah but what's up with that crack about AIB's growing up? Are they pissing and moaning about not having cards to sell so Nvidia is gonna teach them a lesson? Or are they holding Fermi close to their chest to avoid the typical AIB leaks before launch? The latter would make sense if Fermi performance is a big surprise - especially in the underwhelming direction.
Click to expand...

I think they will earlier allow their partners to produce custom designs like GT210/GT220/GT240. I never saw this route for high-end cards since years. nVidia will produces the first batches of cards and then they will give it up to the partners.
 
Do you folks mind getting back on topic?

They were reports from chiphell in the past that A1 ran at 495/1100 but with quite immature drivers also. Take that as a lead and speculate from that point an onwards and with just anything you might consider usefull or useless (depending on POV) from each newsblurb, but for heaven's sake enough with that Charlie this and Charlie that crap. See even my baby daughter is crying....*runs to feed her*
 
You must log in or register to reply here.

Similar threads

hoom
AMD: Zen 4, Speculation, Rumours and Discussion
8 9 10
Replies
199
Views
39K
MfA
M
R
nVidia Denver discussion
2 3
Replies
56
Views
31K
A1xLLcqAgt0qc2RyMz0y
A
J
GPGPU running on FPGA
Replies
16
Views
10K
silent_guy
S
W
Xbox One Architecture Panel *Main Console HW Points* Transcript
3 4 5
Replies
89
Views
34K
Deleted member 11852
D
T
Is Quadro's SP count more Marketing than performance?
Replies
5
Views
5K
trinibwoy
T
Back
Top