Re: MASM Expert needed immediately

Betov wrote:

That Code level Optimization is usefull?

Once you're done with algorithmic optimization, yes. Never before that,
though - "premature [instruction level] optimization is the root of all
evil". (of course the generalization of the term is a bit bull, you need to
consider data structures while designing).

That SEE is usefull?

Yes, to reach higher throughput without buying more expensive hardware with
a higher clock frequency. SSE/2/3 is useful on the x86, just like AltiVec
was on the Mac, and like dedicated DSP chips are.

Then, admitting the SSE routine would be really faster,
respecting all conditions that is, first, that the
normal way would also have been optimize to the max,
and that this speed-up would keep true on all SEE able
Processors, the point would still have many issues:

Of course you shouldn't do instruction-level optimization before you have
the best algorithm - that's pretty much a DUH point. But even if you stuck
with the same algorith and optimized both the "plain x86" and the SSE code
to the max, I'd say it's interesting to see which code performs better.

1) The FASM programmers who do such things are often
times wery good experts (One among other reason
why it does not have any interrest).

I would have thought them being "good experts" *would* have made it
interesting? O_o

2) The old computers would never run this.

Old computers aren't used for heavy sound/graphics/scientific calculations.

3) Take a Processor twice faster for the normal way,
and it will be twice faster.

Sure. Get out your wallet and buy a bigger processor, that's the solution. I
thought that kind of statement would be an offense in an assembly language
newsgroup. "Write bad code, the customer can always buy a faster system".
Yay.


.

Relevant Pages

  • Re: Numerical Recipes (Fortran) Usenet ??
    ... algorithm, see if you can multiply all input cooefficients and values ... I've looked at different optimization libraries (netlib, Neumaier, ... Miller, Burkardt, and others) with no luck in identifying a Fortran ... outside the box as those values on the box boundaries. ...
    (comp.lang.fortran)
  • Re: Behaviour of FMINCON - question.
    ... The large-scale algorithm is a subspace trust region method and is ... Medium-Scale Optimization ... Use equality constraints and the medium-scale method instead. ...
    (comp.soft-sys.matlab)
  • Re: Numerical Recipes (Fortran) Usenet ??
    ... algorithm, see if you can multiply all input cooefficients and values ... I've looked at different optimization libraries (netlib, Neumaier, ... The Fortran bobyqa optimization algorithm is a derivative-free ... outside the box as those values on the box boundaries. ...
    (comp.lang.fortran)
  • Re: Numerical Recipes (Fortran) Usenet ??
    ... algorithm, see if you can multiply all input cooefficients and values ... I've looked at different optimization libraries (netlib, Neumaier, ... Miller, Burkardt, and others) with no luck in identifying a Fortran ... outside the box as those values on the box boundaries. ...
    (comp.lang.fortran)
  • Re: Population count in SSE2, again
    ... This is a crude version of the well-known code in AMD's optimization ... and is not as fast as the AMD code. ... oneself that the algorithm is mathematically sound. ... Yes I'm sure you are right, but a typo in assembler is easily made. ...
    (comp.lang.asm.x86)