NVIDIA GF100 & Friends speculation
- Thread starter Arty
- Start date
Who's the other very good source ?
Neliz has been saying 576mm² for a long time, Charlie's been saying ~550mm² for a little while now, and Fudo seems to have similar information.
Rys => when you say "methinks" or "I'm pretty convinced" do you mean that you have information you're not 100% sure of, or that you're guessing?
Charlie is a "very good source" now ?
And neliz saying 576 mm2 seems a bit odd, since that's the same size as GT200.
Also, other rumors have floated around with the < 500 mm2 die size. People just choose to believe whatever suits them best.
aaronspink
Veteran
Charlie is a "very good source" now ?
And neliz saying 576 mm2 seems a bit odd, since that's the same size as GT200.
Also, other rumors have floated around with the < 500 mm2 die size. People just choose to believe whatever suits them best.
actually IIRC Charlie said it was ~23x23.
Solid information from duplicate sources at NV that I'm currently relying on, but haven't confirmed independently yet (mostly because I don't have a board). I'd lean on it more heavily if Charlie and related goons hadn't pooped in the Fermi waters so much. Plus, I didn't get the graphics spec completely right either, so I'm naturally cautious these days
Charlie said 23.8x23.8 if I remember correctly.
Makes you wonder who's right about the size..
People that can tell you precise tape-out and production dates and from that, it's a small step to the size of the chip if you have the correct sources.
Or the people that said that Fermi is much smaller and on display at Computex 2009 running without problems and will be launched in Q3/Q42009, January 2010 at the latest? When Rys wrote Jan'10 in his TechReport Article, NV wasn't even sure what to tell it's partners yet as they still left Q1 and Q2 open.
Now, I don't know anything for sure, but I trust someone when he says it's the same size as GT200 over anyone doing damage control on NV's behalf.
Last edited by a moderator:
itsmydamnation
Veteran
if true then god damn
http://www.semiaccurate.com/2010/03/02/nvidia-threatens-its-partners-cebit/
/runs
http://www.semiaccurate.com/2010/03/02/nvidia-threatens-its-partners-cebit/
/runs
DegustatoR
Legend
So you trust someone who's doing damage on AMD's behalf instead of trusting Rys? LOL ok.
Why would an nvidia partner like to do damage on AMD's behalve?
Find someone at EVGA to talk, tell happily tell you if they trust you.I don't quite understand why this is so damaging to nVidia. Their launch is coming up, and they want it to have as large an impact as possible.if true then god damn
http://www.semiaccurate.com/2010/03/02/nvidia-threatens-its-partners-cebit/
/runs
I don't see anything wrong with that list. It seems pretty reasonable considering the products aren't launching at Cebit. Not to mention there's still a month or so of driver work to get in before launch. As far as threats go Charlie's probably just trolling for page hits as usual.
Solid information from duplicate sources at NV that I'm currently relying on, but haven't confirmed independently yet (mostly because I don't have a board). I'd lean on it more heavily if Charlie and related goons hadn't pooped in the Fermi waters so much. Plus, I didn't get the graphics spec completely right either, so I'm naturally cautious these days
OK, thanks. It's indeed quite strange to have such conflicting information so close to launch, paperish as it may be. Maybe it's just me, but it seems both AMD and Nvidia are getting better at keeping information confidential.
Still, it must be pretty frustrating for partners not to be able to show anything at a major trade show, days away from launch.
So now you think I'm doing damage control for NV?
Well I've hung my source out (it's the company that designed the chip!), which is more than Charlie tends to do. As far as January went, it was at SC'09 in silicon, so it was a reasonable thing to think at the time (and I think they were genuinely pushing for it). The reason it went back for A3 was unfortunate, although the more recent delays are definitely more their own doing.
As for "others", who were they? Track record counts for a lot here, unless you're inclined to believe any old person throwing data out there
As for "others", who were they? Track record counts for a lot here, unless you're inclined to believe any old person throwing data out there
Oh, you haven't heard? the chip is going back to TSMC after the "launch" for it's first full re-spin!
Hence the no-to-limited availability etc. until June/July. It also is the basis for Charlies "handful" of chips for partners, A3 will be here for launch and B1 will be the "shipping" product.
Similar threads
- Deleted member 2197
- Architecture and Products
