S-L-I
all the way baby!!! Plus Nvidia just signed a mult-billion dollar deal with Intel to make chipsets for the P4 man. They are loaded man.
S-L-I
all the way baby!!! Plus Nvidia just signed a mult-billion dollar deal with Intel to make chipsets for the P4 man. They are loaded man.
--===== proud owner of new razor tarantula gaming keyboard =====--
loaded enough to stop making new vpus? i dont think soOriginally Posted by Kanavit
i never had a problem with a 2slot cooler, the only thing that annoyed me was the noise of it. now most mobo makers leave the room next to the agp slot unpopulated sto evena 2 slot cooler doesnt block a pci slot, and the ati dual slot coolers arent noisy, so thats probably why some people changed their mind about it.Originally Posted by quicksilverXP
define what you mean with "in development" lolOriginally Posted by DudeMiester
if you mean have a piece of paper with basic idears about the architecture, then your right, but thats it. detailed specs about the architecture change all the time, so they only exist for the next gen cards, maybe 2 generations ahead in some situations, but thats very rough stuff afaik.
and i think nv50 was cancelled and instead of it they are making another nv40 based cardsm3.0 is all they need and nv40 is running great, so why risk anything with a new architecture?
i highly doubt that... whatever they planned for nv60, you can impossibly manufacture it at this moment, especially not at a low enough price so it would beat a nv50 card in price/performence ratings.Originally Posted by Geforce4ti4200
nv60 has at least 400 million transitors, its simply impossible to manufacture such a card atm, it would even be tough to build it in 90nm!
yeah, then they started to become one of the two big players... now via and xgi are the underdogs.Originally Posted by Crankster
gentlemen... i think i have the answer to our questions
nv4x
why bother with a 16 pipe 110nm shift (nv48) if you can make a 32P 90nm shift? you end up with a fast high end product (20/24P) and can sell the damaged chips as 16P/20P/24P/28P products!
this means the low end parts with pipelines disabled would cost them slightly more to make, BUT they already have a design to compete with r520!![]()
I doubt it, sure maybe for the NV80 it's all theory and ideas, but with the NV60 they've got to start working out how it will be done in hardware, because they've only got a 1.5 or so years before it's planned release due to the product cycles. With > 222 million transistors it takes a lot of time to perfect! There's a huge amount of work involved, to they have to have detailed designs, and not all up in the air about it. However, I'm not an expert in the feild and afaik neither are you, so we could both be wrong.Originally Posted by saaya
How is it a risk? Obviously they are not braindead enough to release something worse then before! They can benchmark their products, determine this, and toss the bad ideas that way. Either way, the current architecture of the NV40 isn't that great. There are a lot of major improvements and radical changes that will have to be made in order for Nvidia to have products fully complient with OpenGL 2.x, DX10 and Windows Longhorn. It would be crazy for them not to revamp their architecture.Originally Posted by saaya
Forums are the Opiate of the Masses
yhea! DX 10 promises to be a hoot :p so why not?
and about dual slot coolers, I wish they indeed flipped the cards over and casemanufacturers began making room for a dual slot cooler there!.
that'd be sweet, I personally have no problem with dual slots but I ask myself why GPU's "hang" at the moment isn't this like letting something sidder in it's own heat? I mean if you'd flip it wouldn't that alone be MUCH better for temperatures and so on?
the heat'd be able to get away much quicker that way
Bookmarks