http://we.pcinlife.com/thread-929091-1-1.html
65nm process/ MC 512 1GB GDDR3 / 240 SP / 32 ROP/ 6+8Pin
Monster
65nm process/ MC 512 1GB GDDR3 / 240 SP / 32 ROP/ 6+8Pin
Monster
NVIDIA GT200 Rumours & Speculation Thread
- Thread starter Arun
- Start date
- Status
Well my rig is in my sig. My 750W PSU is powering all that fine. I forgot the 8800 Ultra wattage numbers but this 9900 seems significantly higher than that hence my comment. If the 750W works fine for me even better!
Single 8800 Ultra peaks 286W with the same machine as above
GeForce "Next"
2x GT200
1536MB VRAM
384-bit MC
http://www.computerbase.de/forum/showthread.php?t=416410
This is no BS source, the guy was in Singapore and saw them.
2x GT200
1536MB VRAM
384-bit MC
http://www.computerbase.de/forum/showthread.php?t=416410
This is no BS source, the guy was in Singapore and saw them.
Maybe GT200 is infact G90? i.e 55nm version of G80. Sounds more economical compared to rumours involving one big monolithic power hungry GPU.
384bit bus points to 24 ROPs but SP/TMU count could be higher unless they are relying on higher clock speed.
What im more curious about is whether or not nVIDIA is willing to "re-use" its G80 architecture and stick with DX10 or they've fiddled with the architecture abit to support DX10.1, DP etc
note that in the discussions, they refer to this GT200 as Geforce "XXXX" meaning that the naming scheme still hasn't been decided.
Wouldn't it be ironic if Nvidia also takes the multiple small die approach and surprises everybody...again. Same thing happened with DX10 and unified shaders - all the hype was on ATI but they got blown away in the end.
Whatever it is, I just hope it's something interesting to talk about. We don't have any games to play on it anyway.
OK, lets get it back on track
Do you think the GT200x2 in Quad-SLi will require LESS than 1000w?
PSU Thread: http://forum.beyond3d.com/showthread.php?t=47989
Enjoy.
Enjoy.
OMG!!
i am there too =P
--controversy follows me like an angry chihuahua
BUT, anyway, it appears that the reason this thread has calmed down is that there is nothing really new to talk about; it appears that all the NDA stuff that could be pried out was pried out of the knowledgeable ones and has been beaten to death and we are all apparently just waiting for r700 and then GT200's announcement
Quick question; has anyone really realized that there is really no "GT-100" ?
Has anyone realized that there is no G60?
cool .. with that old one i had to stand on my head to see it coming as 92/94b and not "100"
- glad we're on the same page .. i was behind for a whole week!
thanks!
Has anyone realized that there is no G60?
There is no NV60, but G60 = NV40.
Did NV ever use the "Gxx" nomenclature internally prior to G70, because it certainly wasn't discussed publicly. Heck, I even made an ass of myself on R3D prior to G70's launch when rumors of a name change from NV47 were floating around because I insisted that the change in nomenclature was senseless (and from an outsider's perspective at the time it was).
As for the G60 thing, I wanted to point out that from time to time, nVidia likes to change their codenames for no apparent reason. Which I believe is the case with GT200. It's supposed to make people believe it's gonna be a totally new architecture, but the same as with G70, it won't be.
=>aaronspink: You forgot to add the speaking for myself thing. Perhaps you'd want to put it into your signature to save you the trouble?
=>aaronspink: You forgot to add the speaking for myself thing. Perhaps you'd want to put it into your signature to save you the trouble?
So, you can confirm that he was really there and really saw what he - IMHO only - claims to have seen?
- Status
