Design Con 2015
Breaking News
News & Analysis

Nvidia Flexes Custom 64-Bit ARM

8/12/2014 08:25 AM EDT
2 comments
NO RATINGS
More Related Links
View Comments: Newest First | Oldest First | Threaded View
rick merritt
User Rank
Author
Re: Overheads
rick merritt   8/12/2014 11:29:20 AM
NO RATINGS
@Maltanar: Thanks for making the Transmeta connection. Reportedly, Denver was initially going t sue Transmeta tech to create a virtual x86, but I guess that wadsn't true or didn't work out.

It will be interesting to see this benchmarked against the A57 and other custom ARM V8 cores.

I'm told Nvidia canceled its server versions of Denver...anyone heard about that?

maltanar
User Rank
Rookie
Overheads
maltanar   8/12/2014 9:15:46 AM
NO RATINGS
Interesting approach for extracting ILP -- though I guess it's been tried before (first commenter to name Transmeta wins a...a rubber duck!) and it doesn't seem to have caught mainstream attention. The forecasted end of Moore's Law may change things there, though.

What immediately came to mind was the optimization overheads. Although the optimization cache (the 128MB chunk of main memory) should ameliorate the overhead for frequently used applications, infrequent applications or those that show very input-dependent behavior may not get as large benefits.

Also, I expect benchmarking this and comparing against the competition will be non-trivial.

 

 

 

Flash Poll
Like Us on Facebook

Datasheets.com Parts Search

185 million searchable parts
(please enter a part number or hit search to begin)
EE Times on Twitter
EE Times Twitter Feed
Top Comments of the Week