NVIDIA Kepler speculation thread

Kaotik

Drunk Member
Legend
Supporter
Time for next gen nVidia speculation then

nvroadmap.png



edit:
mod could add tags, I forgot
 
My speculation is it will be late.
 
Damn, you're fast -- I was about to fire up that thread for "Kepler". :D

I guess it's too early for "Maxwell". :LOL:
 
Damn, you're fast -- I was about to fire up that thread for "Kepler". :D

I guess it's too early for "Maxwell". :LOL:

Fast? I even went via photoshop in between :p
(and noticed afterwards that I had accidently zoomed out the video tab in ie9 so it's smaller than it should)
 
Time for next gen nVidia speculation then

nvroadmap.png



edit:
mod could add tags, I forgot

I'm happy to learn that Fermi was released in 2009, and that Tesla was released in 2007. Kind of makes me think about how reliable that "2011" will turn out to be for Kepler…
 
I could be wrong but the graph seems to show super linear growth in DP flops/W. I guess 2013 is when last call for ff hw goes out.

Also, source?
 
I could be wrong but the graph seems to show super linear growth in DP flops/W. I guess 2013 is when last call for ff hw goes out.

Also, source?

Jen-Hsun's keynote

The growth is extremely similar to what AMD has had so far in "SP flops/w"
 
FWIW:
http://en.wikipedia.org/wiki/Kepler
"Kepler also incorporated religious arguments and reasoning into his work, motivated by the religious conviction that God had created the world according to an intelligible plan that is accessible through the natural light of reason." and "As he indicated in the title, Kepler thought he had revealed God’s geometrical plan for the universe."

Anyone else hearing "Big dies make good GPUs?" ;)

"Kepler claimed to have had an epiphany on July 19, 1595, while teaching in Graz, demonstrating the periodic conjunction of Saturn and Jupiter in the zodiac; he realized that regular polygons bound one inscribed and one circumscribed circle at definite ratios, which, he reasoned, might be the geometrical basis of the universe. After failing to find a unique arrangement of polygons that fit known astronomical observations (even with extra planets added to the system), Kepler began experimenting with 3-dimensional polyhedra."

Even more geometry? A Polymorph engine for each ALU perhaps? =)
 
Back
Top