Structure and Interpretation of Computer Programmers

I make it easier and faster for you to write high-quality software.

Thursday, February 5, 2015

Dark Silicon

About 10 years ago, we decided that the performance gains in single-core processors that come “for free” with advancing semiconductor processes were slowing down. Many chip makers switched to scaling the number of cores on a die, and promoted parallel programming for their products.

Today I learned that the free multicore lunch is over, too. You can no longer turn on all of the transistors in a single chip, so you can no longer get 2× the threads running by doubling the number of cores in your processor.

posted by Graham at 09:05  


  1. Yes, it is a bit daunting.
    I noticed it quite prominently on a Galaxy Note4.
    For the first two seconds a great fps, but then plummets, as thermal alarms throttle gpu and cpu.
    Same on Galaxy S5, but iPhone appears not to do this.

    Comment by bram stolk — 2015-02-05 @ 18:37

  2. There’s that, and there’s the garbage collector which occasionally stops Android’s world. I think there’s an unaddressed gap between “the GC stops me having to worry about memory” and “I need to tune the GC” which Google need to close with Android.

    Comment by Graham — 2015-02-05 @ 19:48

  3. A friend of mine posted about an ACM paper not mentioned in the Wikipedia article. It includes a 6 min summary video:

    Comment by JJ Kress — 2015-02-06 @ 10:14

  4. “by the time we get to 8nm transistors, over half the chip is unused due to power limitations” :-O

    Comment by Graham — 2015-02-11 @ 11:34

RSS feed for comments on this post. TrackBack URI

Leave a comment

Powered by WordPress