Yes, I know, the forum is named *Real* World Tech, but I'm thinking that perhaps

By: Montaray Jack (none.delete@this.none.org), January 2, 2019 7:26 am
Room: Moderated Discussions
Domaldel (moris.b.rubeksson.delete@this.gmail.com) on January 1, 2019 4:52 pm wrote:
> Domaldel (moris.b.rubeksson.delete@this.gmail.com) on January 1, 2019 4:44 pm wrote:
> > Maynard Handley (name99.delete@this.name99.org) on January 1, 2019 12:17 pm wrote:
> > > Simon Farnsworth (simon.delete@this.farnz.org.uk) on January 1, 2019 5:05 am wrote:
> > > > Domaldel (moris.b.rubeksson.delete@this.gmail.com) on December 31, 2018 11:40 pm wrote:
> > > > > David Hess (davidwhess.delete@this.gmail.com) on December 31, 2018 8:32 pm wrote:
> > > > > > Doug S (foo.delete@this.bar.bar) on December 31, 2018 9:26 am wrote:
> > > > > > > David Hess (davidwhess.delete@this.gmail.com) on December 30, 2018 6:26 pm wrote:
> > > > > > > >
> > > > > > > > The die price (size) problem has puzzled me for a while. For thermally limited designs which
> > > > > > > > have a die smaller than can be produced, couldn't a higher clock speed be achieved be deliberately
> > > > > > > > using a lower transistor density for a lower junction to case thermal resistance at the cost
> > > > > > > > of higher power? I do not know of anybody doing it though or at least admitting it.
> > > > > > >
> > > > > > > What's the cause of hitting that thermal limit, it is a lack of ability to remove heat from a specific sq
> > > > > > > mm or few sq mm, or is it the lack of ability to remove heat from the entire package? For the former what
> > > > > > > you suggest will work, for the latter you upgrade to water cooling etc. to remove heat more quickly.
> > > > > >
> > > > > > It is the former which is why thermal management tracks
> > > > > > power and temperature of different areas. One thing
> > > > > > which would help is better materials for heat spreaders but they are all much more expensive than copper or
> > > > > > even silver. A directly attached heat pipe would be ideal
> > > > > > but the power density is much greater than a heat
> > > > > > pipe will support except in low power applications some of which do use a directly attached heat pipe.
> > > > > >
> > > > > > I had not considered it before but a lot of the performance gain or rather lack of performance
> > > > > > loss with denser processes since integrated temperature sensing was first used comes from monitoring
> > > > > > temperature and power at many points instead of the average temperature which explains why so
> > > > > > much effort is put into it. It is the peak temperatures and hot spots which matter.
> > > > > >
> > > > >
> > > > > While not really viable for home build systems etc I imagine that you could
> > > > > probably deal with a 3D server processors hot spots using a combination of thermal
> > > > > planning and a built in cooling system *inside* the processor itself.
> > > > >
> > > >
> > > >
> > > > Do you mean something like IBM's ICECool research project? This is where IBM runs a
> > > > dielectric fluid through the chip, set up so that the chip boils the fluid internally
> > > > (cooling the chip) and it condenses again outside the chip ready to flow back round.
> > >
> > > The problem with a liquid to gas transition is that you get a huge accompanying volume change.
> > > For many purposes that's horribly undesirable. I don't see any pictures in the ICECool article,
> > > or actual numbers, but I expect this makes the tech a non-starter outside maybe data centers.
> > >
> > > More generally what you want is a transition from a very low to a very high entropy state. Liquid to
> > > gas is great --- but since the entropy increase DERIVES (mostly) from the volume increase, it's always
> > > going to be impractical. What we want is something like maximum entropy change/volume NOT maximum entropy
> > > change/mass. There are, of course, dozens of these - changes in magnetic state, changes in crystal
> > > structure, Mott transitions - but then we add in all the other constraints. I'm guessing a liquid is
> > > more practical than a solid, so most of the transitions that rely on crystal structure are out. Maybe
> > > a solid to liquid melting would do the job --- but again the entropy change is just not that large
> > > compared to boiling. (Enthalpy of vaporization tends to be about 10x that of fusion).
> > > Now there are other ways to slash entropy back and forth. You can use high structure molecules that switch
> > > between orientation phases (liquid crystals). You can use
> > > chemical reactions. But as far as I can tell these
> > > are all lousy numerically compared to the sweet sweet chaos of running around in a much larger volume.
> > >
> > > You can look at this from a different direction. There are
> > > technologies today that generate a large amount of
> > > heat in a very small volume. (Eg quantum cascade lasers
> > > are generating 20W or more in a volume the size of a
> > > small rice grain.) These are cooled not by phase transitions
> > > by active cooling -- TE Coolers using Peltier effect.
> > > Of course these come with their own headaches -- they're expensive, finicky to mount, and large...
> > >
> > > And remember WHATEVER you do, all you're doing is moving heat from where it's generated to somewhere
> > > else where it can be radiated. You're not actually "eating up" the heat. So what's the radiation plan
> > > once the heat gets moved from inside the chip to the chamber where the boiled vapor collects?
> > >
> > > It's theoretically POSSIBLE that 3D architecture could generate lots of heat deep inside a structure,
> > > and that it's advantageous to stream that out of the system as fast as possible using a liquid heat
> > > transport medium (basically capillaries running through your 3D chip). Now the figure of merit is
> > > not how much heat can be absorbed at a fixed temperature (ie phase change issues); it's how much
> > > heat can be moved at a certain speed (so a combination of heat capacity and viscosity).
> > > But while this is theoretically possible, that's very different from saying it's likely. It's probably
> > > VASTLY more efficient to try to solve this problem via other means first --- better 3D planning, keep the
> > > circuits as power efficient as possible, create CVD diamond VIAs to wick heat, things like that...
> > >
> > > I think the bottom line is that you have to clarify what's the market you have in mind, and so how
> > > much hassle can it put up with? But whatever you want, physics is not really on your side :-(
> > >
> > > - Mainstream is mobile, laptops, NUCs, thin AIOs --- all devices where
> > > large volumes to hold gas+radiators, or even a TEC, are not feasible.
> > > - Data centers have enough space, but generally aren't interested in paying a lot more (and
> > > giving up a fair bit of space) to buy a small amount of additional latency reduction.
> > > - FINANCIAL data centers are probably already doing this, but they can just dump their boards in fluorinert
> > > or something equally insane and crazy expensive --- nothing special needed in the CPU beyond the ability
> > > to function (for at least a month or two before it dies) at the highest voltage you can pump into it; and
> > > what they do is mostly decoupled from and has no effect on what the rest of the world is doing.
> > >
> >
> > The discussion came out of the whole eager execution thread to begin with.
> > Let's just say that we're *not* talking mobile here. ;-)
> >
> > But yeah, thanks for your input. =)
> >
> > I'm not working in the industry, so right now I'm essentially "theorycrafting" trying
> > to imagine some ways that a high end processor in the future might need if maximum
> > performance is the design goal with priority over all other considerations.
> >
> > In essence if at every chance where there's a trade off between latency optimization
> > and anything else be that power consumption, processor cost or whatever, you consistently
> > go for latency optimizations then how would that processor end up looking?
> >
> > So I'm thinking semi-eager execution (at some point you're forced to turn to specualtion but you
> > can minimize the amount of branch misspredictions by increasing the amount of multi-pathing).
> >
> > And so one and so forth.
> >
> > Think a design that's not just a few cm square but perhaps
> > the size of a whole wafer *and* 3D (meaning multiple
> > wafers pr package) with built in redundancy to cope with
> > flaws during the production, and possibly also quite
> > a bit of binning, so perhaps 1 in 200 wafers are accepted as a layer in a say 50 layer design?
> >
> > I don't know.
> > Point is just if you take things to extremes and do everything
> > (non-quantum) that current and near future technology
> > is capable of what is possible to achieve with binary technology in terms of latency reductions?
> >
> > Mind you, with as much of those optimizations as possible being hardware side and
> > not compiler side and with the design being general purpose and not an ASIC.
> >
> > Think a custom made super computer where hand placing individual atomes on tens of thousands of
> > wafers is a genuine option if that's required, in other words money is not a concern at all.
>
> Yes, I know, the forum is named Real World Tech, but I'm thinking that perhaps
> speculating about some fictional possibilities might inspire us a bit and help
> us come up with something real long term even if it's not directly usefull.
>
> I'd love to make a home brew processor of some kind that's got
> features not found anywhere else that might still be usefull.

The complexity of making a single chip is around the same as making all of the masks for all the layers. Modern equipment has around a 6 -8 hour exposure time in the SEM, multiply that by the number of layers, say 25, so 200 hours in the electron microscope. You would need to etch the wafer between each step, deposit metals and the other materials at necessary stages, and realign the wafer at each stage with less alignment error than your feature size.

It's not impossible, but would be a considerable undertaking.
< Previous Post in ThreadNext Post in Thread >
TopicPosted ByDate
Independent micro threadsBrett2018/12/30 01:55 PM
  Independent micro threadsTravis Downs2018/12/30 04:18 PM
    Independent micro threadsBrett2018/12/30 07:16 PM
      Independent micro threadsTravis Downs2018/12/30 07:34 PM
        Independent micro threadsBrett2018/12/30 07:48 PM
          Independent micro threadsTravis Downs2018/12/30 09:06 PM
            Independent micro threadsBrett2018/12/30 10:57 PM
              Independent micro threadsTravis Downs2018/12/31 12:42 AM
                Independent micro threadsBrett2019/01/01 07:02 PM
                  Independent micro threadsMichael S2019/01/02 02:01 AM
                    Independent micro threadsMaynard Handley2019/01/02 03:29 PM
              Independent micro threadsMontaray Jack2019/01/01 02:12 AM
  Independent micro threadsanon2018/12/30 04:34 PM
  Independent micro threadsPaul A. Clayton2018/12/30 05:15 PM
  Independent micro threadsDavid Hess2018/12/30 06:47 PM
    Independent micro threadsDomaldel2018/12/30 07:06 PM
      Independent micro threadsDavid Hess2018/12/30 07:26 PM
        Independent micro threadsDoug S2018/12/31 10:26 AM
          Independent micro threadsDavid Hess2018/12/31 09:32 PM
            Independent micro threadsDoug S2019/01/01 12:40 AM
              Independent micro threadsDavid Hess2019/01/01 11:41 AM
            Overcomming thermal limits of a high dencity 3D arcitecture (Formerly "Independent micro threads")Domaldel2019/01/01 12:40 AM
              Overcomming thermal limits of a high dencity 3D arcitecture (Formerly "Independent micro threads")Domaldel2019/01/01 12:44 AM
                Overcomming thermal limits of a high dencity 3D arcitecture (Formerly "Independent micro threads")Domaldel2019/01/01 12:49 AM
                  Overcomming thermal limits of a high dencity 3D arcitecture (Formerly "Independent micro threads")Domaldel2019/01/01 12:51 AM
              Overcomming thermal limits of a high dencity 3D arcitecture (Formerly "Independent micro threads")Simon Farnsworth2019/01/01 06:05 AM
                Overcomming thermal limits of a high dencity 3D arcitecture (Formerly "Independent micro threads")Domaldel2019/01/01 08:01 AM
                Overcomming thermal limits of a high dencity 3D arcitecture (Formerly "Independent micro threads")Maynard Handley2019/01/01 01:17 PM
                  Taking things to extremes.Domaldel2019/01/01 05:44 PM
                    Yes, I know, the forum is named *Real* World Tech, but I'm thinking that perhapsDomaldel2019/01/01 05:52 PM
                      Yes, I know, the forum is named *Real* World Tech, but I'm thinking that perhapsMontaray Jack2019/01/02 07:26 AM
                        Yes, I know, the forum is named *Real* World Tech, but I'm thinking that perhapsMontaray Jack2019/01/02 08:21 AM
                    Taking things to extremes.Maynard Handley2019/01/01 06:55 PM
                    Taking things to extremes.Kevin G2019/01/04 08:57 AM
              Overcomming thermal limits of a high dencity 3D arcitecture (Formerly "Independent micro threads")David Hess2019/01/01 11:36 AM
    Independent micro threadsTravis Downs2018/12/30 07:38 PM
    Independent micro threadsBrett2018/12/30 07:41 PM
  Independent micro threadsanon2018/12/30 08:20 PM
    Independent micro threadsBrett2018/12/30 08:51 PM
      Independent micro threadsTravis Downs2018/12/30 09:48 PM
        Mill and Independent micro threadsBrett2019/01/01 07:39 PM
          No. Mill does not get the hit because it does not get the benefit even when correctly predicted.Heikki Kultala2019/01/02 12:29 AM
            No. Mill does not get the hit because it does not get the benefit even when correctly predicted.Brett2019/01/02 01:15 AM
              No. Mill does not get the hit because it does not get the benefit even when correctly predicted.Heikki Kultala2019/01/02 02:22 AM
                No. Mill does not get the hit because it does not get the benefit even when correctly predicted.Brett2019/01/03 01:13 AM
                  "Leaf branch" is not a commonly used termHeikki Kultala2019/01/03 03:48 AM
                    "Leaf branch" is not a commonly used termBrett2019/01/03 04:35 AM
                      You lack parallelism than OoOE givesHeikki Kultala2019/01/03 07:38 AM
                        You lack parallelism than OoOE givesBrett2019/01/04 02:41 AM
                          You lack parallelism than OoOE givesBrett2019/01/04 04:10 PM
                            You lack parallelism than OoOE givesBrett2019/01/05 08:29 PM
                              Mill speculates, more parallelism than OoOE givesBrett2019/01/05 08:31 PM
                                Mill *is* a speculationEric Bron2019/01/06 05:04 AM
                                  Mill *is* a speculationMichael S2019/01/06 05:53 AM
                                    Mill *is* a speculationBrett2019/01/06 09:03 PM
                                  Mill *is* a speculationjuanrga2019/01/06 06:10 AM
                                    probably ~2 (NT)Michael S2019/01/06 06:51 AM
                                  Mill *is* a speculationBrett2019/01/06 01:18 PM
                                    Mill *is* a speculationEric Bron2019/01/06 03:36 PM
                                      Mill *is* a speculationBrett2019/01/06 08:47 PM
                                        Mill *is* a speculationJacob Marley2019/01/06 10:29 PM
                                          Mill *is* a speculationBrett2019/01/07 04:24 AM
                                            Mill *is* a speculationMichael S2019/01/07 05:23 AM
                                            Mill *is* a speculationEric Bron2019/01/07 05:36 AM
                                              Mill *is* a speculationBrett2019/01/07 03:40 PM
                                                Mill *is* a speculationEric Bron2019/01/07 05:32 PM
                                            Mill is something you don't understandHeikki Kultala2019/01/08 04:19 AM
                                              Mill is something you don't understandMichael S2019/01/08 07:44 AM
                                                Itanium and static vs dynamicHeikki Kultala2019/01/09 03:14 AM
                                                  Itanium and static vs dynamicPaul A. Clayton2019/01/09 08:51 AM
                                        Mill *is* a speculationEric Bron2019/01/07 05:27 AM
                                          Mill *is* a speculationEric Bron2019/01/07 06:23 AM
                                      Mill *is* a speculationanon2019/01/07 06:24 AM
                                        Mill *is* a speculationEric Bron2019/01/07 06:52 AM
                                          Mill *is* a speculationanon2019/01/07 08:36 AM
                                            Mill *is* a speculationEric Bron2019/01/07 09:20 AM
                                        Mill *is* a speculationjuanrga2019/01/07 10:22 AM
                                          Mill *is* a speculationanon2019/01/07 01:16 PM
                                        Mill *is* a speculationanon2019/01/07 09:46 PM
                                          Mill *is* a speculationanon2019/01/08 01:56 AM
                                            Mill *is* a speculationanon2019/01/08 03:39 AM
                                              Mill *is* a speculationMichael S2019/01/08 03:52 AM
                                                Mill *is* a speculationanon2019/01/08 10:10 PM
                                                  Wasted width not wasted work.Brett2019/01/09 11:44 AM
                                                    No such thing was declared. (NT)anon2019/01/09 03:41 PM
                                                    Very simple test for new uarch ideassomeone2019/01/10 07:03 AM
                                                      Very simple test for new uarch ideasdmcq2019/01/10 07:21 AM
                                                        Very simple test for new uarch ideasDoug S2019/01/10 10:01 AM
                                                          Very simple test for new uarch ideasDan Fay2019/01/10 01:13 PM
                                                      Very simple test for new uarch ideasanonymous22019/01/10 11:03 AM
                                                        Very simple test for new uarch ideasAlberto2019/01/10 11:32 AM
                                                      Very simple test for new uarch ideasEtienne2019/01/11 03:03 AM
                                                        Very simple test for new uarch ideasFoo_2019/01/11 04:31 AM
                                                          Very simple test for new uarch ideasEtienne2019/01/11 05:51 AM
                                                            Very simple test for new uarch ideasFoo_2019/01/11 05:53 AM
                                                              Very simple test for new uarch ideasdmcq2019/01/11 06:08 AM
                                                              Very simple test for new uarch ideasEtienne2019/01/11 06:13 AM
                                                                Very simple test for new uarch ideasFoo_2019/01/11 06:54 AM
                                                                  Very simple test for new uarch ideasEtienne2019/01/11 07:32 AM
                                                                    Very simple test for new uarch ideasBrett2019/01/11 10:25 AM
                                                                      Very simple test for new uarch ideasMegol2019/01/12 06:29 AM
                                                                        Very simple test for new uarch ideasMichael S2019/01/12 09:21 AM
                                                                          Word salad AI fundamentaliy brokenBrett2019/01/12 01:59 PM
                                                                          Very simple test for new uarch ideasMegol2019/01/13 11:51 AM
                                              Mill *is* a speculationanon2019/01/08 08:50 AM
                                                Mill *is* a speculationEric Bron2019/01/08 09:03 AM
                                                  Mill *is* a speculationanon2019/01/08 09:21 AM
                      "Leaf branch" is not a commonly used termMichael S2019/01/03 07:57 AM
                        "Leaf branch" is not a commonly used termBrett2019/01/04 03:29 AM
                  Calls are not needed for speculation for mill if there are no side effect,and dont help if there areHeikki Kultala2019/01/08 04:28 AM
              No. Mill does not get the hit because it does not get the benefit even when correctly predicted.anon2019/01/02 03:05 AM
              No. Mill does not get the hit because it does not get the benefit even when correctly predicted.Doug S2019/01/02 11:38 AM
                No. Mill does not get the hit because it does not get the benefit even when correctly predicted.rwessel2019/01/02 05:53 PM
                  No. Mill does not get the hit because it does not get the benefit even when correctly predicted.anon2019/01/02 08:56 PM
                    itanicBrett2019/01/03 12:41 AM
                      itanicanon2019/01/03 03:12 AM
                      itanicDavid Hess2019/01/03 08:06 AM
                    No. Mill does not get the hit because it does not get the benefit even when correctly predicted.rwessel2019/01/03 09:18 AM
                      No. Mill does not get the hit because it does not get the benefit even when correctly predicted.anon2019/01/04 05:25 AM
                    Itanium could have been RISC or CISC - same outcomesomeone2019/01/04 07:45 AM
                      Itanium could have been RISC or CISC - same outcomeDoug S2019/01/04 12:39 PM
                        Itanium could have been RISC or CISC - same outcomeJan Olšan2019/01/04 01:58 PM
                          "fluffyRISC" has a namevvid2019/01/04 03:48 PM
                        Itanium could have been RISC or CISC - same outcomeBrett2019/01/04 03:43 PM
                      Itanium could have been RISC or CISC - same outcomeanonymou52019/01/04 12:41 PM
                  No. Mill does not get the hit because it does not get the benefit even when correctly predicted.David Hess2019/01/03 08:15 AM
                No. Mill does not get the hit because it does not get the benefit even when correctly predicted.Maynard Handley2019/01/03 12:24 PM
                  No. Mill does not get the hit because it does not get the benefit even when correctly predicted.Maynard Handley2019/01/03 12:27 PM
                    No. Mill does not get the hit because it does not get the benefit even when correctly predicted.dmcq2019/01/04 01:59 AM
                  EPIC target marketsFoo_2019/01/04 06:29 AM
                    EPIC target marketsDoug S2019/01/04 12:42 PM
                      Lack of future visionDoug S2019/01/04 12:57 PM
                        Lack of future visionBrett2019/01/04 02:59 PM
                          Lack of future visionDoug S2019/01/04 04:25 PM
                            Lack of future visionBrett2019/01/04 05:18 PM
                              Lack of future visionDoug S2019/01/05 12:47 AM
                                Lack of future visionBrett2019/01/05 02:06 PM
                                  Lack of future visiondmcq2019/01/05 02:22 PM
                                  Lack of future visionanon2019/01/05 03:01 PM
                                    Lack of future visionMichael S2019/01/05 04:18 PM
                                      Lack of future visionanon2019/01/05 06:14 PM
                                        Lack of future visionMichael S2019/01/06 02:01 AM
                                          Lack of future visionanon2019/01/06 03:23 AM
                                          Mitch Alsup's MY66000 uses IF-like predication (I think) (NT)Paul A. Clayton2019/01/06 04:54 PM
                                            ??? (NT)Michael S2019/01/07 05:25 AM
                                            88K ? (NT)anonymous22019/01/07 04:20 PM
                                          Modestly expanded response: MY66000 predicate shadowPaul A. Clayton2019/01/07 11:53 AM
                      Thanks for the correction (NT)Foo_2019/01/04 04:31 PM
              No. Mill does not get the hit because it does not get the benefit even when correctly predicted.sdrc2019/01/04 07:36 AM
          Mill and Independent micro threadsMichael S2019/01/02 02:32 AM
Reply to this Topic
Name:
Email:
Topic:
Body: No Text
How do you spell green?