ARM vs. x86 is irrelevant from an energy efficiency POV

Article: AMD's Mobile Strategy
By: Exophase (exophase.delete@this.gmail.com), December 22, 2011 9:27 pm
Room: Moderated Discussions
David Kanter (dkanter@realworldtech.com) on 12/22/11 wrote:
---------------------------
>I think it's tough to discuss only a subset of the vector extensions. It's true
>that AVX might be irrelevant for phones, today. But I doubt it will stay that way
>forever and you might see direct comparisons in tablets.

It's true in the kind of tablets most people are interested in buying (not those running CPUs with > 10W TDP). It's not a foregone conclusion that AVX will ever hit these CPUs, but even if it is it's pretty moot. We don't know what'll be out to compare it with when it finally does. I'd probably give it more consideration if Atom sported SSE4.x which first appeared years ago.

>>If we want to keep it to just load/store behavior: NEON in ARMv7a can load or store
>>1-4 64-bit registers. Loads can perform de-interleaving and you can load single
>>elements (lanes in NEON parlance) instead of entire >vectors.
>

>Do you mask out the lanes?

You can load to single lanes or store single lanes. You can't a variable set of lanes against a dynamic mask, which would be a very nice feature and one I hope ARM adopts. Given how poor main RAM performance is on a lot of platforms you can spend a lot of extra time doing an RMW to do a partial update.

>You're still using a single address right? It looks like they probably handle cache line crossing loads, which is nice.

Yes, there's no gather/scatter support. There are 32-index byte shuffles ala XOP, although without all the extra features. And on Cortex-A8/A9 they're pretty slow at that.

You can specify aligned or unaligned for vector memory accesses. Accesses specified as aligned that aren't will fault. On Cortex-A8/A9 unaligned NEON accesses are handled by splitting them into two partial accesses. Which shows the hardware CAN support partial stores, or at least missing chunks from either end if not necessarily both. I don't think the penalties for crossing cache lines go beyond what you'd go for two distinct accesses.

One annoying gotcha I found on these processors is that NEON has no store to load forwarding (main integer unit does). Normally this isn't a huge deal if you're careful, but it hits you unknowingly if you perform an unaligned store following by an unaligned load to the address immediately after it, thanks to the partial accesses overlapping. If you're performing an RMW loop you pay for it pretty dearly, if you can't keep the loads ahead of the stores.

>I seem to recall from talking with one of the architects at ARM that they really
>got rid of all >2 cycle load/stores. I think the main motivation was simplifying
>exceptions, pipeline control and consistency.
>
>That's probably OK because ISTR they now have 128b registers to deal with double precision.

It already has 128-bit registers. And while floating point ops are 2x32-bit per cycle on A8/A9 the more common integer ops are 128-bit/cycle. Loads and stores are too, although I've hit some other bottlenecks with them (but on A8 the dcache interface for NEON is certifiably not 64-bit. Don't know about A9.)

So you can perform 2 128-bit loads/stores (hence 4 64-bit) in 2 cycles. But it has to be aligned, or it's three cycles. And some of the more complex operations with them cost a fourth cycle.

For some reason I thought the v8 overview didn't make things clear, but I looked at it again. 4-vector loads and stores are still present, and have been extended to allow 64 or 128-bit registers. So unless the v8 supporting processor the architect was referring to has a 256-bit or multiple 128-bit dcache interfaces it's going to take more than two cycles to do those instructions.

Other than that it looks like they're mostly the same - somewhat bizarrely including the alternating register options. Possibly motivated by compatibility, although with everything else they'll be breaking with NEON it's hard to see that being a huge factor..

They also stick to the highly limiting memory address modes. Strictly [reg] with a sizeof increment or an increment by another reg. That's it. It's pretty understandable since they really don't have encode space to play with but it's still a pretty big pain (fortunately Cortex-A8 tends to be at its best when doing ARM and NEON code simultaneously so it doesn't tend to hurt much from ARM generating addresses.. no idea about Cortex-A9)

>
>It's all a matter of degrees. AVX will eventually hit low power chips, it's a question
>of when. Also, you will see VFP/Neon chips that are hitting higher power levels.
>So I think the comparison is quite informative.
>
>I'd liken it to Bobcat vs. Atom. They are really different designs, aimed at different
>markets. They overlap in some areas, thus there is room for comparisons. But we
>all know that Atom doesn't hit 20W and Bobcat doesn't go under 4.5W. So they each have 'unique' areas.
>

But with Atom and Bobcat Intel actually is selling Atoms with much higher TDP than the lowest power Bobcats. Bobcat goes a little higher and Atom goes much lower but there's a lot of overlap, and more significantly, those lower power Atoms aren't really being used anywhere right now. So it's a pretty meaningful comparison.

On the flip side, we have no idea just how long it'll take until we see convergence between power usage of AVX CPUs and NEON CPUs. But I definitely think it'll come with IB or maybe Haswell first, not Atom.

>Brands don't matter much, price points do. There are SNBs available for around $100. I'd expect that to drop with IVB.
>

That's far, far too high of a price point to enter this discussion. I'm going to wait and see on IB, current rumor is same price as SB for higher end (which seems like a good move)

>Besides, with Jen-Hsun Huang talking about how Kal-El is faster than Core2...it's
>not crazy to compare it against Sandy Bridge : P
>

Let's keep nVidia's fraudulent marketing out of this ;)

>It's unrealistic to expect ALL of Intel's products to have AVX 1 year after the
>first SNB. There are still folks who want less expensive parts.
>
>David

This isn't about expecting all of Intel's products to have AVX, it's about expecting Intel's CPUs with AVX to have it enabled. Intel's actions are not only unreasonable but completely unprecedented. Coppermine-128 had SSE, P4 Celerons had SSE2, Prescott Celerons had SSE3, etc, even though these were low end and not well performing chips - and they had them before the much better competition from AMD did (or maybe that had something to do with it?).

This generation Intel went crazy with market segmentation feature fusing. I understand how it works and why it works in general and I'm not arguing against performing any feature removal or underclocking or whatever to make other parts look better and sell for more. But some of the decisions seem pulled from a hat. Fusing off instruction sets that are not special purpose is a bad idea because it discourages adoption which is needed to make the instruction set useful.

Processors like Pentium and even Celeron are higher margin than processors like Atom - or at least Atom's going to have to become lower margin to survive. So long as Intel sees AVX as only being worthy of paying highest margin for I can't see them putting it in its lowest margin chips. Won't people wonder why Atoms have it if Pentiums don't?
< Previous Post in ThreadNext Post in Thread >
TopicPosted ByDate
AMD's Mobile StrategyMark Pillsbury2011/12/14 06:58 PM
  ARM vs. x86 is irrelevant from an energy efficiency POV Daniel Bizo2011/12/15 05:30 AM
    ARM vs. x86 is irrelevant from an energy efficiency POV Foo_2011/12/15 06:08 AM
    ARM vs. x86 is irrelevant from an energy efficiency POV an2011/12/15 06:25 AM
      ARM vs. x86 is irrelevant from an energy efficiency POV Azazel2011/12/15 09:50 PM
        ARM vs. x86 is irrelevant from an energy efficiency POV Mike C2011/12/22 02:14 PM
          ARM vs. x86 is irrelevant from an energy efficiency POV Michael S2011/12/23 07:06 AM
        ARM vs. x86 is irrelevant from an energy efficiency POV Mike C2011/12/23 08:07 PM
    ARM vs. x86 is irrelevant from an energy efficiency POV Wilco2011/12/15 07:46 AM
      ARM vs. x86 is irrelevant from an energy efficiency POV Linus Torvalds2011/12/15 10:32 AM
        ARM vs. x86 is irrelevant from an energy efficiency POV a person2011/12/15 10:59 AM
        ARM vs. x86 is irrelevant from an energy efficiency POV Michael S2011/12/15 11:18 AM
          ARM vs. x86 is irrelevant from an energy efficiency POV anon2011/12/16 07:22 PM
            ARM vs. x86 is irrelevant from an energy efficiency POV Wilco2011/12/17 06:56 AM
              Thumb vs ARM instruction countsnone2011/12/17 11:15 AM
                Thumb vs ARM instruction countsWilco2011/12/22 07:50 AM
                  Thumb vs ARM instruction countsExophase2011/12/22 08:56 AM
        ARM vs. x86 is irrelevant from an energy efficiency POV Exophase2011/12/15 11:44 AM
          ARM vs. x86 is irrelevant from an energy efficiency POV Linus Torvalds2011/12/15 03:24 PM
            ARM vs. x86 is irrelevant from an energy efficiency POV Exophase2011/12/15 04:15 PM
              ARM vs. x86 is irrelevant from an energy efficiency POV Linus Torvalds2011/12/15 04:51 PM
                ARM vs. x86 is irrelevant from an energy efficiency POV anon2011/12/15 05:32 PM
                  ARM vs. x86 is irrelevant from an energy efficiency POV David Kanter2011/12/15 06:57 PM
                    ARM vs. x86 is irrelevant from an energy efficiency POV anon2011/12/15 08:58 PM
                      ARM vs. x86 is irrelevant from an energy efficiency POV Wilco2011/12/16 12:27 AM
                    Is scalar x86 decode that easy?Paul A. Clayton2011/12/15 09:20 PM
                      Is scalar x86 decode that easy?Linus Torvalds2011/12/16 12:09 PM
                        Is scalar x86 decode that easy?EduardoS2011/12/17 10:03 AM
                        Is scalar x86 decode that easy?Ricardo B2011/12/17 02:21 PM
                          Is scalar x86 decode that easy?Megol2011/12/18 05:23 AM
                            Is scalar x86 decode that easy?rwessel2011/12/18 10:04 AM
                              Is scalar x86 decode that easy?Megol2011/12/18 10:38 AM
                              Is scalar x86 decode that easy?Linus Torvalds2011/12/18 12:15 PM
                                Is scalar x86 decode that easy?Ricardo B2011/12/18 01:57 PM
                                  Is scalar x86 decode that easy?Linus Torvalds2011/12/18 02:49 PM
                    ARM vs. x86 is irrelevant from an energy efficiency POV Wilco2011/12/16 12:00 AM
                    ARM vs. x86 is irrelevant from an energy efficiency POV anon2011/12/17 12:13 PM
                      Conditional comparesMichael S2011/12/17 01:05 PM
                        Conditional comparesEduardoS2011/12/17 01:32 PM
                ARM vs. x86 is irrelevant from an energy efficiency POV Exophase2011/12/15 05:56 PM
                  ARM free to innovate?Paul A. Clayton2011/12/15 09:14 PM
                  ARM vs. x86 is irrelevant from an energy efficiency POV Linus Torvalds2011/12/16 10:01 AM
                    ARM vs. x86 is irrelevant from an energy efficiency POV Exophase2011/12/16 02:33 PM
                      ARM vs. x86 is irrelevant from an energy efficiency POV Linus Torvalds2011/12/16 06:30 PM
                        ARM vs. x86 is irrelevant from an energy efficiency POV rwessel2011/12/17 12:51 AM
                          ARM vs. x86 is irrelevant from an energy efficiency POV Octoploid2011/12/17 07:46 AM
                          ARM vs. x86 is irrelevant from an energy efficiency POV Linus Torvalds2011/12/17 10:56 AM
                        ARM vs. x86 is irrelevant from an energy efficiency POV Wilco2011/12/17 07:45 AM
                          ARM vs. x86 is irrelevant from an energy efficiency POV Linus Torvalds2011/12/17 11:31 AM
                            ARM vs. x86 is irrelevant from an energy efficiency POV anon2011/12/21 09:12 AM
                          ARM vs. x86 is irrelevant from an energy efficiency POV David Kanter2011/12/17 02:16 PM
                            ARM vs. x86 is irrelevant from an energy efficiency POV Linus Torvalds2011/12/17 03:30 PM
                            ARM vs. x86 is irrelevant from an energy efficiency POV none2011/12/17 05:00 PM
                              ARM vs. x86 is irrelevant from an energy efficiency POV Wilco2011/12/17 05:29 PM
                            ARM vs. x86 is irrelevant from an energy efficiency POV Wilco2011/12/17 05:45 PM
                        ARM vs. x86 is irrelevant from an energy efficiency POV Exophase2011/12/20 10:24 AM
                          ARM vs. x86 is irrelevant from an energy efficiency POV Linus Torvalds2011/12/20 01:07 PM
                            ARM vs. x86 is irrelevant from an energy efficiency POV Exophase2011/12/20 07:19 PM
                              ARM vs. x86 is irrelevant from an energy efficiency POV EduardoS2011/12/20 07:40 PM
                                ARM vs. x86 is irrelevant from an energy efficiency POV Exophase2011/12/20 08:02 PM
                                  ARM vs. x86 is irrelevant from an energy efficiency POV Linus Torvalds2011/12/20 08:49 PM
                                    ARM vs. x86 is irrelevant from an energy efficiency POV Exophase2011/12/20 08:59 PM
                                      ARM vs. x86 is irrelevant from an energy efficiency POV Michael S2011/12/21 03:49 AM
                                        ARM vs. x86 is irrelevant from an energy efficiency POV Exophase2011/12/21 10:10 AM
                                      ARM vs. x86 is irrelevant from an energy efficiency POV gallier22011/12/21 06:41 AM
                                        ARM vs. x86 is irrelevant from an energy efficiency POV Wilco2011/12/21 07:43 AM
                                          ARM vs. x86 is irrelevant from an energy efficiency POV Seni2011/12/21 09:28 AM
                                            ARM vs. x86 is irrelevant from an energy efficiency POV Wilco2011/12/21 11:59 AM
                                              Cortex A9 latenciesS. Rao2011/12/21 09:12 PM
                                                Cortex A9 latenciesWilco2011/12/22 04:59 AM
                                            ARM vs. x86 is irrelevant from an energy efficiency POV gallier22011/12/22 01:25 AM
                                        ARM vs. x86 is irrelevant from an energy efficiency POV Exophase2011/12/21 09:42 AM
                                          ARM vs. x86 is irrelevant from an energy efficiency POV gallier22011/12/22 02:02 AM
                                            ARM vs. x86 is irrelevant from an energy efficiency POV Wilco2011/12/22 05:44 AM
                                              ARM vs. x86 is irrelevant from an energy efficiency POV gallier22011/12/22 07:57 AM
                                    (in)visibility of ISA featureshobold2011/12/21 06:27 AM
                              ARM vs. x86 is irrelevant from an energy efficiency POV Linus Torvalds2011/12/20 07:51 PM
                                ARM vs. x86 is irrelevant from an energy efficiency POV Exophase2011/12/20 08:13 PM
                                  ARM vs. x86 is irrelevant from an energy efficiency POV EduardoS2011/12/20 08:41 PM
                                    second to that (NT)Michael S2011/12/21 02:37 AM
                                  ARM vs. x86 is irrelevant from an energy efficiency POV Linus Torvalds2011/12/20 09:03 PM
                                    ARM vs. x86 is irrelevant from an energy efficiency POV Exophase2011/12/20 11:05 PM
                                      ARM vs. x86 is irrelevant from an energy efficiency POV Seni2011/12/21 02:22 AM
                                        ARM vs. x86 is irrelevant from an energy efficiency POV Exophase2011/12/21 09:28 AM
                                          ARM vs. x86 is irrelevant from an energy efficiency POV Seni2011/12/21 10:06 AM
                                            ARM vs. x86 is irrelevant from an energy efficiency POV Exophase2011/12/21 10:33 AM
                                              ARM vs. x86 is irrelevant from an energy efficiency POV Seni2011/12/21 11:03 AM
                                                ARM vs. x86 is irrelevant from an energy efficiency POV Exophase2011/12/21 11:33 AM
                                                  Adding modes for flexible imm. sizesPaul A. Clayton2011/12/21 04:08 PM
                                                  ARM vs. x86 is irrelevant from an energy efficiency POV anonymous2011/12/21 10:57 PM
                                                    ARM vs. x86 is irrelevant from an energy efficiency POV Exophase2011/12/21 11:12 PM
                                                      ARM vs. x86 is irrelevant from an energy efficiency POV anonymous2011/12/22 02:27 AM
                                                        ARM vs. x86 is irrelevant from an energy efficiency POV Exophase2011/12/22 08:27 AM
                                                ARM vs. x86 is irrelevant from an energy efficiency POV Brett2011/12/21 11:36 PM
                                            ARM vs. x86 is irrelevant from an energy efficiency POV Wilco2011/12/21 05:34 PM
                                              ARM vs. x86 is irrelevant from an energy efficiency POV Seni2011/12/21 06:53 PM
                                                ARM vs. x86 is irrelevant from an energy efficiency POV Wilco2011/12/22 06:42 AM
                                                  ARM vs. x86 is irrelevant from an energy efficiency POV David Kanter2011/12/22 05:29 PM
                                                    ARM vs. x86 is irrelevant from an energy efficiency POV Exophase2011/12/22 07:17 PM
                                                      ARM vs. x86 is irrelevant from an energy efficiency POV David Kanter2011/12/22 08:23 PM
                                                        ARM vs. x86 is irrelevant from an energy efficiency POV Exophase2011/12/22 09:27 PM
                                                          Confused by AVX....David Kanter2011/12/23 12:05 AM
                                                            Celeron, PentiumMichael S2011/12/23 01:15 AM
                                                            Confused by AVX....S. Rao2011/12/23 01:27 AM
                                                            Confused by AVX....Gabriele Svelto2011/12/23 01:33 AM
                                                              Confused by AVX....David Kanter2011/12/23 12:58 PM
                                                                Confused by AVX....Gabriele Svelto2011/12/23 02:44 PM
                                      Load Single Register (pc-relative, literal load)Michael S2011/12/21 02:55 AM
                                        Load Single Register (pc-relative, literal load)Wilco2011/12/21 08:35 AM
                                          Load Single Register (pc-relative, literal load)Brett2011/12/22 12:20 AM
                                            Load Single Register (pc-relative, literal load)Exophase2011/12/22 08:32 AM
                                              Load Single Register (pc-relative, literal load)iz2011/12/22 05:45 PM
                                        Load Single Register (pc-relative, literal load)Exophase2011/12/21 09:34 AM
                                          Load Single Register (pc-relative, literal load)Wilco2011/12/21 04:29 PM
                                            Load Single Register (pc-relative, literal load)Exophase2011/12/22 08:37 AM
                                              Load Single Register (pc-relative, literal load)Brett2011/12/22 11:59 PM
                                          Load Single Register (pc-relative, literal load)Konrad Schwarz2011/12/27 05:07 AM
                                        Load Single Register (pc-relative, literal load)stubar2011/12/21 03:00 PM
                                          Load Single Register (pc-relative, literal load)Exophase2011/12/21 03:25 PM
                                            Load Single Register (pc-relative, literal load)Wilco2011/12/21 04:30 PM
                                              Load Single Register (pc-relative, literal load)anon2011/12/21 09:20 PM
                                                Load Single Register (pc-relative, literal load)Wilco2011/12/22 05:50 AM
                                                  Load Single Register (pc-relative, literal load)anon2011/12/22 06:11 AM
                                                    Load Single Register (pc-relative, literal load)Wilco2011/12/22 07:25 AM
                                      ARM vs. x86 is irrelevant from an energy efficiency POV Linus Torvalds2011/12/21 08:18 PM
                                        ARM vs. x86 is irrelevant from an energy efficiency POV Wilco2011/12/22 06:07 AM
                                          ARM vs. x86 is irrelevant from an energy efficiency POV Linus Torvalds2011/12/22 10:56 AM
                                            idle philosophizing about localityhobold2011/12/23 08:26 AM
                                              Streaming needs regularity, not locality (NT)Rohit2011/12/23 01:32 PM
                                                Streaming needs regularity, not localityiz2011/12/24 12:38 AM
                                                Streaming needs regularity, not localityhobold2011/12/24 04:49 AM
                                                  Streaming needs regularity, not localityRohit2011/12/24 08:52 AM
                                                  Streaming needs regularity, not localityEBFE2011/12/25 06:30 PM
                                                    Streaming needs regularity, not localityRohit2011/12/25 09:42 PM
                                                      Streaming needs regularity, not localityEBFE2011/12/25 11:49 PM
                                                        Streaming needs regularity, not localityRohit2011/12/26 09:53 AM
                                    ARM vs. x86 is irrelevant from an energy efficiency POV mpx2011/12/21 02:57 AM
                                      ARM vs. x86 is irrelevant from an energy efficiency POV John Yates2011/12/26 10:46 AM
                                        Apollo Computer's DN10KBrett2011/12/26 06:56 PM
                                          Apollo Computer's DN10Kpeter huisken2011/12/27 07:40 AM
                                    ARM vs. x86 is irrelevant from an energy efficiency POV gallier22011/12/21 06:56 AM
                              ALU with shift; post-adjust, memcpyMichael S2011/12/21 03:22 AM
                                ALU with shift; post-adjust, memcpyExophase2011/12/21 09:02 AM
                                  memcpyMichael S2011/12/21 10:36 AM
                                    memcpyExophase2011/12/21 11:15 AM
                                    memcpyWilco2011/12/21 06:06 PM
                                      memcpyanon2011/12/21 09:15 PM
                                      memcpyMichael S2011/12/22 12:12 AM
                                        memcpyWilco2011/12/22 06:20 AM
                          Other x86 uop tricksDavid Kanter2011/12/20 07:21 PM
                            Other x86 uop tricksanon2011/12/20 07:35 PM
                              Other x86 uop tricksMichael S2011/12/21 03:55 AM
                            Other x86 uop tricksanon2011/12/20 07:53 PM
                            Other x86 uop tricksExophase2011/12/20 07:54 PM
                      stwxMichael S2011/12/17 10:46 AM
                        stwxWilco2011/12/17 11:33 AM
                          stwxMichael S2011/12/17 01:12 PM
            ARM vs. x86 is irrelevant from an energy efficiency POV mpx2011/12/16 11:06 AM
              ARM vs. x86 is irrelevant from an energy efficiency POV Linus Torvalds2011/12/16 11:17 AM
                Reason for shared decoderPaul A. Clayton2011/12/16 11:45 AM
                  Reason for shared decoderanon2011/12/16 06:37 PM
                    Reason for shared decodermpx2011/12/17 06:20 AM
                      Reason for shared decoderanon2011/12/17 04:19 PM
                        Reason for shared decodermpx2011/12/18 07:21 AM
                          Reason for shared decoderanon2011/12/18 08:30 AM
                          Reason for shared decoderKira2011/12/18 09:00 PM
                ARM vs. x86 is irrelevant from an energy efficiency POV mpx2011/12/17 05:40 AM
                  ARM vs. x86 is irrelevant from an energy efficiency POV Linus Torvalds2011/12/17 12:06 PM
                    ARM vs. x86 is irrelevant from an energy efficiency POV mpx2011/12/17 03:17 PM
        ARM vs. x86 is irrelevant from an energy efficiency POV Wilco2011/12/16 12:10 AM
          ARM vs. x86 is irrelevant from an energy efficiency POV David Kanter2011/12/16 04:19 AM
          ARM vs. x86 is irrelevant from an energy efficiency POV Linus Torvalds2011/12/16 10:14 AM
          ARM vs. x86 is irrelevant from an energy efficiency POV Eugene Nalimov2011/12/16 11:07 AM
            ARM vs. x86 is irrelevant from an energy efficiency POV Michael S2011/12/18 07:25 AM
              ARM vs. x86 is irrelevant from an energy efficiency POV Linus Torvalds2011/12/18 12:48 PM
                ARM vs. x86 is irrelevant from an energy efficiency POV Michael S2011/12/18 04:01 PM
                  ARM vs. x86 is irrelevant from an energy efficiency POV Linus Torvalds2011/12/18 05:21 PM
                    ARM vs. x86 is irrelevant from an energy efficiency POV Linus Torvalds2011/12/18 07:56 PM
      x86 compatibility is not as important as Microsoft compatibilityBrett2011/12/15 08:13 PM
      You're completley missing the pictureDaniel Bizo2011/12/16 08:12 AM
        Missed referencesDaniel Bizo2011/12/16 08:16 AM
          Missed referencesnone2011/12/16 09:17 AM
            Missed referencesDaniel Bizo2011/12/16 09:29 AM
    ARM vs. x86 is irrelevant from an energy efficiency POV Gabriele Svelto2011/12/15 12:55 PM
    ARM vs. x86 is irrelevant from an energy efficiency POV someone2011/12/15 01:19 PM
  AMD's Mobile StrategyDavid Kanter2011/12/15 06:13 PM
    AMD's Mobile StrategyMark Pillsbury2011/12/15 08:14 PM
      AMD's Mobile Strategybakaneko2011/12/15 08:57 PM
      AMD's Mobile StrategyDavid Kanter2011/12/15 10:53 PM
    AMD's new GPU ISA less efficient?Paul A. Clayton2011/12/15 08:33 PM
      AMD's new GPU ISA less efficient?bakaneko2011/12/15 09:03 PM
      AMD's new GPU ISA less efficient?Rohit2011/12/15 09:09 PM
  AMD's Mobile StrategyIan Ollmann2011/12/16 09:21 AM
    AMD's Mobile StrategyMegol2011/12/16 09:33 AM
    AMD's Mobile StrategyDan Fay2011/12/16 07:02 PM
      AMD's Mobile StrategyMichael S2011/12/17 09:22 AM
        AMD's Mobile StrategyMegol2011/12/17 10:10 AM
          AMD's Mobile StrategyWilco2011/12/17 11:18 AM
        AMD's Mobile StrategyDan Fay2011/12/17 11:45 AM
    AMD's Mobile StrategyEduardoS2011/12/17 07:43 AM
      AMD's Mobile StrategyMichael S2011/12/17 11:55 AM
  AMD's Mobile StrategyBill Henkel2011/12/16 06:54 PM
    6 GHz 2-core water-cooled processorTom P2011/12/19 07:48 PM
      6 GHz 2-core water-cooled processorBill Henkel2011/12/19 08:21 PM
        6 GHz 2-core water-cooled processorgallier22011/12/20 01:31 AM
          "Benefit" of smaller market sharePaul A. Clayton2011/12/20 09:45 AM
            scalability of flash SSDSeni2011/12/20 11:22 AM
              USB drive price-point practical?Paul A. Clayton2011/12/20 12:49 PM
                USB drive price-point practical?Seni2011/12/20 02:02 PM
                  A minor comment on utilityPaul A. Clayton2011/12/20 03:19 PM
                    A minor comment on utilitySeni2011/12/20 03:35 PM
                    A minor comment on utilityFoo_2011/12/21 07:06 AM
                      100" Big screen.pH72011/12/26 12:37 PM
                USB drive price-point practical?David Hess2011/12/20 05:25 PM
                  USB drive price-point practical?iz2011/12/20 06:08 PM
                    USB drive price-point practical?David Hess2011/12/21 07:54 AM
                      USB drive price-point practical?iz2011/12/21 11:21 PM
                  USB drive price-point practical?Ungo2011/12/20 06:36 PM
                    USB drive price-point practical?David Hess2011/12/21 07:48 AM
              scalability of flash SSDEduardoS2011/12/20 03:50 PM
                scalability of flash SSDSeni2011/12/20 06:41 PM
                  scalability of flash SSDEduardoS2011/12/20 06:50 PM
                    scalability of flash SSDSeni2011/12/20 07:24 PM
                      scalability of flash SSDEduardoS2011/12/20 07:51 PM
                        scalability of flash SSDanon2011/12/21 04:03 AM
                          scalability of flash SSDanon2011/12/21 05:21 AM
                            scalability of flash SSDEduardoS2011/12/21 01:32 PM
                              scalability of flash SSDanon2011/12/21 03:49 PM
                                scalability of flash SSDEduardoS2011/12/21 04:58 PM
                                  scalability of flash SSDanon2011/12/21 09:00 PM
                                    scalability of flash SSDEduardoS2011/12/22 02:58 AM
                                    scalability of flash SSDRicardo B2011/12/22 08:36 AM
              scalability of flash SSDWes Felter2011/12/20 05:53 PM
                scalability of flash SSDSeni2011/12/20 06:24 PM
                scalability of flash SSDiz2011/12/20 06:29 PM
        6 GHz 2-core water-cooled processorExophase2011/12/20 09:05 AM
          6 GHz 2-core water-cooled processorEduardoS2011/12/20 03:58 PM
            6 GHz 2-core water-cooled processorExophase2011/12/20 06:13 PM
              6 GHz 2-core water-cooled processorEduardoS2011/12/20 06:43 PM
                6 GHz 2-core water-cooled processorExophase2011/12/21 11:36 AM
                  6 GHz 2-core water-cooled processorSeni2011/12/21 12:11 PM
                    6 GHz 2-core water-cooled processorExophase2011/12/21 12:26 PM
                  Diatomic Helium? (NT)Paul A. Clayton2011/12/21 03:50 PM
                    Sorry, should have read further in thread! (NT)Paul A. Clayton2011/12/21 03:52 PM
      6 GHz 2-core water-cooled processorsomeone2011/12/20 10:09 AM
        Per-core licensing fees.mpx2011/12/21 03:07 AM
    AMD's Mobile StrategyDoug Siebert2011/12/20 06:39 PM
  AMD's Mobile Strategy (NT)yundy2011/12/16 09:19 PM
  ARM transition is irrelevant from an economic POV Azazel2011/12/18 12:23 AM
  AMD's Mobile Strategypk2011/12/21 07:51 AM
    AMD's Mobile StrategyDavid Kanter2011/12/21 10:55 AM
      Samsung's fab technologyMark Roulo2011/12/21 11:34 AM
      AMD's Mobile StrategyAzazel2011/12/21 09:45 PM
        AMD's Mobile StrategyEduardoS2011/12/22 02:59 AM
          AMD's Mobile StrategyAzazel2011/12/22 10:03 PM
            AMD's Mobile StrategyEduardoS2011/12/22 11:51 PM
            AMD's Mobile StrategyVincent Diepeveen2011/12/24 11:15 AM
  AMD's Mobile StrategyAndrew Maizels2011/12/21 11:30 AM
  AMD's Mobile StrategyMr. Camel2011/12/27 09:50 AM
    AMD's Mobile StrategyMike C2011/12/27 11:01 PM
  Need for DifferentiationBill Henkel2011/12/29 12:22 PM
    Need for DifferentiationBrett2011/12/29 09:53 PM
      Need for DifferentiationVincent Diepeveen2011/12/30 11:29 AM
      Requested software features already in productCharles2011/12/30 09:35 PM
        Requested software features already in productSanjay B2012/01/11 04:45 PM
          Requested software features already in productBrett2012/01/11 10:18 PM
    Need for DifferentiationDavid Kanter2011/12/30 01:23 AM
      Need for DifferentiationVincent Diepeveen2011/12/30 11:17 AM
      Need for DifferentiationBill Henkel2012/01/03 07:01 PM
        Need for DifferentiationDavid Kanter2012/01/03 08:34 PM
    Need for DifferentiationVincent Diepeveen2011/12/30 11:09 AM
    L4 CacheBill Henkel2011/12/30 05:55 PM
      L4 CacheDavid Kanter2011/12/30 09:34 PM
        L4 CacheEduardoS2011/12/31 03:42 PM
          L4 CacheExophase2011/12/31 05:37 PM
      Need for speedDoug Siebert2011/12/31 01:07 PM
        Need for speedAlex M2012/01/06 03:38 PM
          Need for speedDoug Siebert2012/01/06 07:51 PM
        Need for speedTom J2012/01/10 04:44 PM
          Need for speedrwessel2012/01/10 05:30 PM
      L4 CacheRicardo B2011/12/31 02:25 PM
        L4 CacheBill Henkel2012/01/03 06:33 PM
          L4 CacheDavid Kanter2012/01/03 08:54 PM
            L4 Cache--not quite that bad?Paul A. Clayton2012/01/04 01:10 PM
              L4 Cache--not quite that bad?David Kanter2012/01/04 04:59 PM
                L4 Cache--not so bad (for Intel/IBM perhaps)Paul A. Clayton2012/01/04 08:01 PM
                L4 CacheBill Henkel2012/01/05 05:50 PM
                  Integration class price/performance tradeoffs?Paul A. Clayton2012/01/06 03:26 PM
                    Integration class price/performance tradeoffs?David Kanter2012/01/06 06:29 PM
                      InterposersMark Roulo2012/01/09 09:43 AM
                        InterposersDavid Kanter2012/01/10 12:13 PM
                    Integration class price/performance tradeoffs?Bill Henkel2012/01/06 06:52 PM
                      Beware of invention-stealing time travelers :-)Paul A. Clayton2012/01/06 08:31 PM
                        AMD PrototypePoindexter2012/01/07 06:20 AM
                    L4 Cache with Silicon InterposerTom J2012/01/08 05:50 PM
              L4 Cache--not quite that bad?rwessel2012/01/04 07:20 PM
                Temptation for expensive optimizationsPaul A. Clayton2012/01/04 08:20 PM
            L4 CacheDan Fay2012/01/04 05:51 PM
              L4 CacheRicardo B2012/01/04 06:39 PM
                L4 CacheDan Fay2012/01/04 06:57 PM
              L4 CacheKevin G2012/01/06 09:02 AM
          L4 CacheRicardo B2012/01/04 06:29 PM
            L4 CacheMichael S2012/01/05 07:21 AM
              L4 CacheRicardo B2012/01/05 11:01 AM
              L4 CacheRicardo B2012/01/06 06:06 AM
                L4 CacheMichael S2012/01/06 07:35 AM
                  Non-latency advantagesPaul A. Clayton2012/01/06 03:04 PM
                    Non-latency advantagesRicardo B2012/01/06 06:12 PM
                      Tag check delay??Paul A. Clayton2012/01/06 07:05 PM
                        Tag check delay??Ricardo B2012/01/06 08:18 PM
                          Thanks for sharingPaul A. Clayton2012/01/06 08:46 PM
            L4 Cachesomeone2012/01/08 05:06 PM
              L4 CacheRicardo B2012/01/08 06:28 PM
      L4 CacheVincent Diepeveen2012/01/08 02:11 PM
        L4 CacheJouni Osmala2012/01/09 03:24 AM
  Value of x86 in ServersGeorge Baker2012/01/04 10:55 AM
    Value of x86 in ServersDavid Kanter2012/01/05 02:40 AM
      Value of x86 in Serversrwessel2012/01/05 03:27 AM
        Value of x86 in ServersDavid Kanter2012/01/05 04:49 AM
      Value of x86 in ServersKira2012/01/09 05:22 PM
        Value of x86 in ServersJoel2012/01/09 09:20 PM
    Value of x86 in ServersDan Fay2012/01/05 09:03 AM
    Value of x86 in ServersAlex M2012/01/06 04:23 PM
    Value of x86 in ServersTom J2012/01/10 04:25 PM
      Value of x86 in Serversanon2012/01/11 12:44 PM
        Value of x86 in ServersGregory2012/01/12 06:11 PM
          x86 complexity or high-performance complexity?Paul A. Clayton2012/01/12 06:52 PM
          Value of x86 in ServersMichael S2012/01/13 06:39 AM
            Value of x86 in ServersRicardo B2012/01/13 03:14 PM
              Value of x86 in ServersMichael S2012/01/14 12:22 PM
  Heterogeneous ComputingAlex M2012/01/06 05:10 AM
    Thoughput != SIMDPaul A. Clayton2012/01/06 05:10 PM
  AMD's Processor StrategyJason Chan2012/01/10 06:06 AM
    AMD's Processor StrategyMark Roulo2012/01/10 06:21 PM
      AMD's Processor StrategyMegol2012/01/11 03:51 AM
        AMD's Processor StrategyDavid Hess2012/01/11 10:46 AM
    AMD's Processor StrategyAzazel2012/01/12 09:46 PM
      AMD's Processor StrategyVincent Diepeveen2012/01/13 06:28 AM
Reply to this Topic
Name:
Email:
Topic:
Body: No Text
How do you spell purple?