Sign in / up

back to article Nvidia's 2015 Tegra ARM chip promises '100X' speed-up

Nvidia has fleshed out details about its next Tegra mobile processor, code-named Logan, and revealed that its long-running 64-bit ARM "Project Denver" effort will yield its first fruit from Logan's follow-on, code-named Parker. "Logan has something we've been dying to bring to the world for so long," Nvidia cofounder, …

COMMENTS

House rules Send corrections
This topic is closed for new posts.
  1. Eddy Ito

    Pity

    They're just going to call the next chip Logan? They could have named the it Logan 5 to be followed by Jessica 6 and Francis 7. Probably just as well since they were founded in 1993 it would only give them 10 more years before being "renewed".

    2 0
    1. Reading Your E-mail
      Reply Icon
      Go

      Re: Pity

      not Johnny 5, phone 6 and Shakin 7 ?

      0 0
    2. Anonymous Coward
      Reply Icon
      Anonymous Coward

      Re: Pity

      Given The previous development project codenames were "Kal El" (Superman) and "Wayne" (Batman) then "Logan" probably refers to X-Men character Wolverine and "Parker" to Spider-Man.

      1 0
  2. Tom 7

    Can they get someone over the age of 5

    to write the Linux drivers for this one please.

    1 1
  3. Zot
    Meh

    More parallel instead of 100X

    Parallel computing is a totally different way of coding, saying that it's X times faster at doing things is misleading.

    They say a ray-tracer took one day to getting running on it, but ray-tracing is THE easiest thing to convert to parallel as each pixel is independant.

    Having said that, this technology is definitely heading in spectacular direction. Although it will never be fast enough for me. : )

    2 0
    1. Frumious Bandersnatch
      Reply Icon

      Re: More parallel instead of 100X

      but ray-tracing is THE easiest thing to convert to parallel as each pixel is independant.

      It may be the easiest to make it work in parallel, but it won't be the most efficient since you need to access colour info from all over the scene. Memory will be the bottleneck, in other words, not computing power. Rendering fractals, on the other hand would be an application where pixels are truly independent.

      2 0
This topic is closed for new posts.

Other stories you might like