REGISTER | LOGIN
Breaking News
Comments
Newest First | Oldest First | Threaded View
<<   <   Page 2 / 2
Tom Murphy
User Rank
Author
Re: Nvidia's retort
Tom Murphy   7/31/2013 10:51:09 AM
NO RATINGS
Rick and Peter raise an interested question:  just what is the tipping point between real performance increase and the cost of going custom?

Peter Clarke
User Rank
Author
Re: Nvidia's retort
Peter Clarke   7/31/2013 10:36:48 AM
NO RATINGS
yes but is that betraying an emphasis on computational performance.

ARM maybe putting out lots of cores and POPs but maybe 15 percent better power efficiency justifies going custom?

rick merritt
User Rank
Author
Nvidia's retort
rick merritt   7/31/2013 10:22:52 AM
NO RATINGS
In the past, Nvidia defended itself saying Qcomm did not get significant performance benefits from the work of doing a custom core design (less than a 15% boost they claimed). They also argued ARM is putting out so many cores and interconnects so fast these days there's no real time to market advantage doing it yourself either.

Applied Micro may find the latter point is true by the time it gets out its 64-bit server SoCs.

<<   <   Page 2 / 2


Most Recent Comments
David Ashton
 
Smartalix
 
Sang-Yun
 
Sang-Yun
 
realjjj
 
DMcCunney
 
David Ashton
 
rick merritt
 
rick merritt
Like Us on Facebook