TOP3 blunders

By: Potatoswatter (potswa_m.delete@this.c.com), September 26, 2009 1:38 pm
Room: Moderated Discussions
? (0xe2.0x9a.0x9b@gmail.com) on 9/26/09 wrote:
---------------------------
>- Instructions with operands which have arbitrary data layout. For example, suppose
>you have determined that the optimal bit-length you need is 11 bits per input and
>12 bits per output. No problem: you can extract the 5 11-bit numbers from a 64-bit
>register in parallel, do some 5-way computation and then pack the 5 12-bit results into a 64-bit register.

Don't underestimate the work in routing data arbitrarily. You're talking about adding shifters between the register file and arithmetic. That will add a lot of latency and only be useful occasionally, and not much faster than a shift instruction.

PowerPC adds a mask operation to shifts (useful for unpack operations), and it turned out to be a bitch to implement without commensurate utility.

For small-operand addition and multiplication, you can use scalar registers as vectors just by, for example, placing each 12-bit operand in a 24-bit field. So a 64-bit packed word unpacks to two half-empty 64-bit registers. I used this to write a vectorized transparency engine on the PowerPC 601. (Operands being 5-bit color channels.)

Of course this only allows vector + vector addition and vector * scalar mult. But you won't get vector * vector multiplication on an arbitrary number of arbitrary bitfields without a ludicrous amount of circuitry.

>- Let the inputs and outputs of CEs performing floating-point computations have
>3 separate parts per operand: sign, mantissa, exponent. Assuming we have such CEs,
>we can now try using them in video encoding/decoding applications, games and rendering. No freaking GP-GPU necessary.

What? All the SIMD ISAs put FP and integer values in the same registers. You can pack and unpack them to your heart's content. The size of the mantissa and exponent aren't going to change without making the FPU slower.

How do you propose to make this useful for graphics? How does GPGPU relate to graphics? (That would seem a contradiction!)

GPGPU has more to do with multithreading than arithmetic.

---

After that, I'm not reading in depth but it looks like an involved discussion of extended-precision arithmetic. Trust me, every ISA provides a carry bit which is the best way to do that. Don't reinvent the bitslice ;v) .
< Previous Post in ThreadNext Post in Thread >
TopicPosted ByDate
TOP3 blundersanon2009/09/25 07:38 PM
  TOP3 blundersrwessel2009/09/25 07:58 PM
  TOP3 blundersPotatoswatter2009/09/26 02:12 AM
    TOP3 blundersJouni Osmala2009/09/26 08:47 AM
    TOP3 blundersWilco2009/09/26 10:51 AM
      TOP3 blundersPotatoswatter2009/09/26 01:04 PM
        TOP3 blundersRagingDragon2009/09/28 02:58 PM
  TOP3 blunders?2009/09/26 03:44 AM
    TOP3 blundersanon2009/09/26 06:55 AM
      TOP3 blunders?2009/09/26 10:49 AM
        TOP3 blundersPotatoswatter2009/09/26 01:38 PM
        TOP3 blundersJouni Osmala2009/09/26 09:52 PM
          TOP3 blunders?2009/09/27 01:08 AM
            implementationAM2009/09/27 01:31 AM
          the generic mechanismAM2009/09/27 01:15 AM
            the generic mechanism?2009/09/27 02:30 AM
          TOP3 blunderssomeone2009/09/27 07:49 AM
            TOP3 blunders?2009/09/27 08:33 AM
              TOP3 blundersBlistering blue barnacles2009/09/28 08:44 PM
              TOP3 blundersslacker2009/09/29 07:28 AM
                (12nJ/cycle)*3GHz = 36W (NT)Michael S2009/09/29 08:31 AM
                  That's what I get for posting so early. (NT)slacker2009/09/29 06:17 PM
          TOP3 blunderskoby m.2009/09/29 07:26 AM
            TOP3 blundersJouni Osmala2009/09/29 11:05 PM
              TOP3 blunders?2009/09/30 02:26 AM
    TOP3 blundersPotatoswatter2009/09/26 07:11 AM
      TOP3 blunderssomeone2009/09/26 09:26 AM
    ISA extensions - tensilica?David Kanter2009/09/26 10:17 AM
      ISA extensions - StretchWes Felter2009/09/26 11:24 AM
        ISA extensions - Stretch?2009/09/26 01:14 PM
          ISA extensions - StretchGabriele Svelto2009/09/28 03:14 AM
  TOP3 blundersanonymous2009/09/26 06:23 AM
  TOP3 smart movesanonib2009/09/26 07:11 PM
    TOP3 smart movesDavid W. Hess2009/09/26 07:55 PM
      TOP3 smart movesRichard Cownie2009/09/27 01:57 PM
        TOP3 smart movesRagingDragon2009/09/28 03:46 PM
    TOP3 smart movesa reader2009/09/27 08:44 AM
    TOP3 smart movesRichard Cownie2009/09/27 12:12 PM
      TOP2 probably illegal smart movesRichard Cownie2009/09/27 01:43 PM
        TOP2 probably illegal smart movesnn2009/09/27 07:27 PM
        TOP2 probably illegal smart movesDavid Kanter2009/09/29 01:48 AM
          TOP2 probably illegal smart movesRichard Cownie2009/09/29 10:31 AM
            TOP2 probably illegal smart movesMichael S2009/09/29 11:04 AM
              TOP2 probably illegal smart movesDavid Kanter2009/09/29 01:35 PM
              TOP2 probably illegal smart movesRichard Cownie2009/09/29 04:28 PM
              TOP2 probably illegal smart movesJouni Osmala2009/09/29 11:08 PM
              Link to good analysis on the matter.Jouni Osmala2009/09/29 11:22 PM
                Link to good analysis on the matter.Richard Cownie2009/09/30 09:59 AM
  TOP3 blunder of todayMichael S2009/09/29 04:31 AM
Reply to this Topic
Name:
Email:
Topic:
Body: No Text
How do you spell green?