Ok, what do you want to understand? The Geforce 8800 is the name of the next series of Nvidia cards.
Where the first 8 represents the 8th generation Geforce card, while the second letter 8 represents the high end status of the line.
Ok, what do you want to understand? The Geforce 8800 is the name of the next series of Nvidia cards.
Where the first 8 represents the 8th generation Geforce card, while the second letter 8 represents the high end status of the line.
And what are you trying to say with this, not to mention how is this even relevant?
Face it the calculations you made were off, mine are accurate, not only do you have figures that are off for the initally started values, but your 80nm value is way off, the maximum shrinkage level the 80nm process can give, without removing transistors from the die itself, is 20%. Your figure is something like 37% smaller then the 90nm die which is a physical impossibility for the 80nm process.
Algebra is not needed here to calculate the optical shrink values, all you need is multiplication and division.
Nice deflection by the way, posting some algebra junk.
isn't 8800 one of those ancient ATi bastards? the numbering system keeps making it sound like the old days all over again.
Do you know how to calculate surface area or transistor density?
are all dies perfectly square?
I agree the 80nm value is low, the 90nm and 65nm are dead on, just redid the math to appease you.
You nitpicking my math doesnt say a damn thing about the point i was making.
90nm and 80nm are too big to achieve the quoted speeds.
Yeah, it does say something about a person if someone is sloppy on their work, they are less likely to be taken seriously.
And I already said showed you a link in the above post how the enthusiast and performance is a significant percentage of Nvidia and ATi's income, having a die size of even 4x larger then the mainstream is possible, as they are making 8-9 times, the revenue.
~ 400mm2 or ~ 500mm2 dpending on the 80nm or 90nm respectively, are big but feasible, if Nvidia sticks to the MSRP's listed,
The wording was "scalable to 1.5GHZ" I believe, so you can't really draw that much of an conclusion of what the clock frequency of G80 is from that. It could mean the architecture itself is scalable to 1.5GHZ, or as said it could be talking about memory frequency, it's very vague to say the least.
Surface area is calculated from transistor density, while transistor density is derived from the current transistor density of G71. So yes we have that information.
GPU dies on the high end tend to be close to a square, but rarely are they perfectly square, doesn't effect the calculations at all though.
The 1.5GHZ figure is intriguing and it's one part of the listed specs that piques my interest. But like I said, 65nm is not an option for high end SKU's at the current time. It is a completely untested process for both Nvidia and ATI, no SKU's have been built thus far on it, it doens't matter if 90nm and 80nm come out to ~500mm2 or ~400mm2 respectively, if 700 Million is the correct figure, that is where the die size are going to be.
Well ATI never had an 8800 Card, on the desktop to my knowledge they had a the Radeon 8500, and if using pass nvidia naming nomencleture as a guideline, then the Geforce 8500 will be a value DX10 card, but it's likely going to be faster then the original Radeon 8500.
The 7300 GT for sure is quicker then the Radeon 8500.
We will see, lets leave it at that.
This thread will come back to haunt you in november, i promise.
lol I hope so, but look, you can still pick up the ATi model here for only $300. WOOT!:
ATi Fire GL 8800 128MB 4x DDR AGP Multi Monitor Graphics Card!
After all, those caynaydayans always produce kick ass IQ!
Yeah, though like I said there isn't any desktop graphic card with the 8800 moniker on it, as the FireGL line is targeted at the workstation class of users.
Yeha I sure hope so considering how large their die sizes are this generation, I hope for their sakes it's worth it.
Originally posted by: Pabster
As is the 7300GS though that isn't saying much.
Yeah I just wanted to be on the safe side though, I wasn't too sure though it should be as it has twice the shader power and more, not to mention it's much more advanced in terms of the shader units.
The low end budget cards take ages to get substantially faster it seems.
yeah for price/performance it's usually best to go with low-high or mid range cards that are at least 6 months old.
Hmm, so things turned out mostly of what I thought, it's a 90nm chip, and according to this picture
http://www.hothardware.com/image_popup....ge=big_g80_wafer.jpg&articleid=903&t=a
It's also a monolithic single die design.
As well we have the Shader Clock running at Core Clock x2 + 200MHZ. So in excess of 1GHZ on both 8800 GTX and 8800 GTS.
http://www.hothardware.com/image_popup....ge=big_g80_wafer.jpg&articleid=903&t=a
It's also a monolithic single die design.
As well we have the Shader Clock running at Core Clock x2 + 200MHZ. So in excess of 1GHZ on both 8800 GTX and 8800 GTS.
TRENDING THREADS
-
Discussion Zen 5 Speculation (EPYC Turin and Strix Point/Granite Ridge - Ryzen 9000)- Started by DisEnchantment
- Replies: 25K
-
News NVIDIA and Intel to Develop AI Infrastructure and Personal Computing Products- Started by poke01
- Replies: 347
-
Discussion Intel Meteor, Arrow, Lunar & Panther Lakes + WCL Discussion Threads
- Started by Tigerick
- Replies: 22K
-
Discussion Intel current and future Lakes & Rapids thread- Started by TheF34RChannel
- Replies: 23K
Facebook
Twitter