How much faster?

Article: What's Next for Moore's Law? For Intel, III+V = 10nm QWFETs
By: Joel (joel.hruska.delete@this.gmail.com), April 29, 2015 2:12 pm
Room: Moderated Discussions
Klimax (danklima.delete@this.gmail.com) on April 29, 2015 12:58 pm wrote:
> Joel (joel.hruska.delete@this.gmail.com) on April 29, 2015 12:26 pm wrote:
> > David Hess (davidwhess.delete@this.gmail.com) on April 29, 2015 8:38 am wrote:
> > > Pierre (Boutoukoat.delete@this.yahoo.fr) on April 29, 2015 4:38 am wrote:
> > > >
> > > > Dissipated power is O(frequency) and O(squared voltage). The article mention the voltage could be reduced
> > > > by 200 millivolts, e.g. go from 0.9 to 0.7 volts. This mean a reduction of power of 40%, and in layman
> > > > terms, 40% more battery life on a processor 4 times cheaper. Reasoning at constant power and ignoring
> > > > o a lot of details, frequency could increase by 40% if voltage goes from 0.9V to 0.7V. However, most of
> > > > the perceived performance out of a processor comes from memory speed (unchanged ...) , parallelism and
> > > > integration of co-processors like GPU and NICs (more transistors), and software benchmarks.
> > >
> > > Power density is also becoming a limiting factor. Heat pipes helped here however they have their own
> > > hard power density limits and now require heat spreaders between the CPU and evaporator. If you reduce
> > > the power by 40% but have the same number of transistors in 1/4 of the area, then the power density
> > > increases by 160%. Raising the frequency by 40% and keeping the power constant would increase the power
> > > density by 300%. Either the junction to case thermal resistance has to be lowered (diamond?) or a higher
> > > junction temperature is needed which will adversely affect reliability and leakage.
> > >
> > > > For the last 10 years, frequency of desktops for gamers has not increased (peak around 4 Ghz on
> > > > the most expensive Intel CPUs). It looks unlikely to change in the near future, and the trend
> > > > to add more cores will continue. I hope this helps to answer your obsolete question "how faster
> > > > ?" typical from the 90's. Right questions now are "how cheaper ?" , "how longer ?" ...
> > >
> > > AMDs current CPUs top out in the 4 GHz range as well. Adding more cores to take advantage of higher
> > > density processes is going to run up against power density limits and most applications including
> > > games cannot take advantage of more cores anyway. Do any games make use of more than 2 cores?
> > >
> >
> > Yes. There are many modern titles that can make some use of up to four cores, though
> > not much scales past that. DX12 is expected to vastly improve this scenario; it
> > allows for effective multi-threading in ways that DX11 never supported.
> >
> > Early DX12 games and applications can take full advantage of all eight
> > cores on a Haswell-E system and see performance gains from doing so.
>
> There won't be likely any real jump. Those early examples tend to be badly written benchmarks
> at best intended to present severely contrived case, which in practice won't exist.
> (Not to mention that current engines don't make use of half of DX11 anyway) Also DX12
> and co makes very bad trade offs regarding future changes in GPUs. (IMO)


Yeah, every developer I've spoken to -- and I've spoken to more than a few -- disagrees with that.

Mantle provides real evidence as well. Average performance uptick for AMD APUs using Mantle is about 13% with another 13% improvement in performance per watt. DX12 allows for asynchronous compute (sharing compute workloads between two vastly different GPUs) as well as a far more diverse set of multi-GPU configurations. For the first time, workloads don't *have* to be duplicated across VRAM.

No, I don't expect the first DX12 titles to be 50% faster than DX11 games, but low-overhead APIs allow a lot of options that previous APIs didn't.

If this wasn't the case, why would *everyone* -- Apple, Khronos, and Microsoft -- have cumulatively leapt for low-overhead API designs? If AMD didn't have the core of a good idea with Mantle, nobody would've paid attention to it. Instead we get Vulkhan based on Mantle and DX12 offering essentially identical features alongside Apple's Metal.
< Previous Post in ThreadNext Post in Thread >
TopicPosted ByDate
RWT Article: What’s Next for Moore’s Law? For Intel, III+V = 10nm QWFETsDavid Kanter2015/04/21 01:53 PM
  RWT Article: What’s Next for Moore’s Law? For Intel, III+V = 10nm QWFETsExophase2015/04/21 02:42 PM
  RWT Article: What’s Next for Moore’s Law? For Intel, III+V = 10nm QWFETsJoe2015/04/21 05:06 PM
  RWT Article: What’s Next for Moore’s Law? For Intel, III+V = 10nm QWFETsMaynard Handley2015/04/21 07:57 PM
    RWT Article: What’s Next for Moore’s Law? For Intel, III+V = 10nm QWFETsExophase2015/04/21 08:25 PM
      RWT Article: What’s Next for Moore’s Law? For Intel, III+V = 10nm QWFETsanon2015/04/21 10:24 PM
        RWT Article: What’s Next for Moore’s Law? For Intel, III+V = 10nm QWFETsExophase2015/04/21 10:53 PM
          RWT Article: What’s Next for Moore’s Law? For Intel, III+V = 10nm QWFETsanon2015/04/21 11:14 PM
      RWT Article: What’s Next for Moore’s Law? For Intel, III+V = 10nm QWFETsDavid Ball2015/04/22 10:55 AM
    RWT Article: What’s Next for Moore’s Law? For Intel, III+V = 10nm QWFETsKurt Marko2015/04/21 10:34 PM
      RWT Article: What’s Next for Moore’s Law? For Intel, III+V = 10nm QWFETsDavid Kanter2015/04/22 08:11 AM
        How much faster?Mark Roulo2015/04/22 12:50 PM
          How much faster?Pierre2015/04/29 03:38 AM
            How much faster?David Hess2015/04/29 07:38 AM
              How much faster?Joel2015/04/29 11:26 AM
                How much faster?Klimax2015/04/29 11:58 AM
                  How much faster?Joel2015/04/29 02:12 PM
                    How much faster?Klimax2015/04/29 03:55 PM
            How much faster?rwessel2015/04/29 10:02 AM
              How much faster?abraidwood2015/04/29 10:27 AM
                How much faster?rwessel2015/04/29 02:00 PM
                  How much faster?abraidwood2015/04/29 10:48 PM
                    How much faster?rwessel2015/04/30 01:15 AM
    RWT Article: What’s Next for Moore’s Law? For Intel, III+V = 10nm QWFETsDavid Kanter2015/04/22 08:25 AM
  What is the quantum effect?Wouter Tinus2015/04/21 10:08 PM
    III-V Tailored by Taylor?native gaas2015/04/22 03:27 AM
      III-V Tailored by Taylor?Joel2015/04/22 08:54 AM
        III-V Tailored by Taylor?Rob VanHooren2015/04/22 01:13 PM
          III-V Tailored by Taylor?Joel2015/04/22 02:36 PM
            III-V Tailored by Taylor?Rob VanHooren2015/04/23 04:43 AM
              Continue and you will be bannedDavid Kanter2015/04/23 09:15 AM
      III-V Tailored by Taylor?David Kanter2015/04/23 01:35 PM
        III-V Tailored by Taylor?Gabriele Svelto2015/04/24 02:03 AM
          III-V Tailored by Taylor?Joel2015/04/24 07:13 AM
          III-V Tailored by Taylor?David Kanter2015/04/24 01:56 PM
    What is the quantum effect?David Kanter2015/04/23 09:19 AM
  RWT Article: What’s Next for Moore’s Law? For Intel, III+V = 10nm QWFETsOtis2015/04/22 06:48 AM
    RWT Article: What’s Next for Moore’s Law? For Intel, III+V = 10nm QWFETsPaul2015/04/22 04:20 PM
      RWT Article: What’s Next for Moore’s Law? For Intel, III+V = 10nm QWFETstarlinian2015/04/22 04:28 PM
      RWT Article: What’s Next for Moore’s Law? For Intel, III+V = 10nm QWFETsDavid Kanter2015/04/23 11:28 AM
        RWT Article: What’s Next for Moore’s Law? For Intel, III+V = 10nm QWFETstarlinian2015/04/23 11:54 AM
    RWT Article: What’s Next for Moore’s Law? For Intel, III+V = 10nm QWFETsDavid Kanter2015/04/22 10:48 PM
      RWT Article: What’s Next for Moore’s Law? For Intel, III+V = 10nm QWFETsOtis2015/04/27 08:24 PM
        RWT Article: What’s Next for Moore’s Law? For Intel, III+V = 10nm QWFETsDavid Kanter2015/04/28 09:07 AM
  Group IV vs III-V combination championcarop2015/04/22 09:57 AM
    Group IV vs III-V combination championDavid Kanter2015/04/23 01:41 PM
  RWT Article: What’s Next for Moore’s Law? For Intel, III+V = 10nm QWFETsAryan2015/04/22 08:56 PM
  How would NTV play into this? Joel2015/04/23 11:38 AM
  RWT Article: What’s Next for Moore’s Law? For Intel, III+V = 10nm QWFETsJohn H2015/04/23 01:28 PM
  RWT Article: What’s Next for Moore’s Law? For Intel, III+V = 10nm QWFETsEric2015/04/29 08:54 AM
  RWT Article: What’s Next for Moore’s Law? For Intel, III+V = 10nm QWFETsThomas2015/05/13 12:50 PM
  Comparison of 14nm Broadwell and 22nm HaswellMatt2015/07/08 03:20 PM
    Comparison of 14nm Broadwell and 22nm Haswelljuanrga2015/07/08 11:06 PM
      Comparison of 14nm Broadwell and 22nm HaswellDoug S2015/07/09 08:01 AM
        Comparison of 14nm Broadwell and 22nm HaswellIan Ameline2015/07/09 10:21 AM
        Comparison of 14nm Broadwell and 22nm HaswellOops2015/09/25 06:21 AM
      Comparison of 14nm Broadwell and 22nm HaswellAlberto2015/07/11 12:59 AM
        Comparison of 14nm Broadwell and 22nm HaswellDoug S2015/07/11 10:45 AM
          Comparison of 14nm Broadwell and 22nm HaswellDoug S2015/07/11 10:48 AM
            Comparison of 14nm Broadwell and 22nm HaswellMaynard Handley2015/07/11 11:32 AM
              Comparison of 14nm Broadwell and 22nm HaswellAlberto2015/07/14 02:22 PM
                Comparison of 14nm Broadwell and 22nm HaswellAnon2015/07/16 12:12 AM
                  Comparison of 14nm Broadwell and 22nm HaswellAlberto2015/07/16 05:03 AM
                    Comparison of 14nm Broadwell and 22nm HaswellMaynard Handley2015/07/16 08:00 AM
                      Comparison of 14nm Broadwell and 22nm HaswellAlberto2015/07/16 01:06 PM
                        Comparison of 14nm Broadwell and 22nm HaswellMaynard Handley2015/07/17 11:37 AM
                          Comparison of 14nm Broadwell and 22nm HaswellAlberto2015/07/18 01:07 AM
                    Comparison of 14nm Broadwell and 22nm HaswellAnon2015/07/16 06:28 PM
                      Comparison of 14nm Broadwell and 22nm Haswelljuanrga2015/07/17 02:05 AM
                        Comparison of 14nm Broadwell and 22nm HaswellAlberto2015/07/17 02:22 AM
                    Comparison of 14nm Broadwell and 22nm HaswellMichael S2015/07/17 06:41 AM
                      Comparison of 14nm Broadwell and 22nm HaswellAlberto2015/07/17 12:22 PM
    Comparison of 14nm Broadwell and 22nm HaswellAlberto2015/07/11 01:07 AM
Reply to this Topic
Name:
Email:
Topic:
Body: No Text
How do you spell green?