Re: Computer architects leaving Intel...

On 9/3/2024 8:46 AM, Terje Mathisen wrote:
 > Stephen Fuld wrote:
 >> On 9/2/2024 1:23 AM, Terje Mathisen wrote:
 >>> Stephen Fuld wrote:
 >>>> On 8/31/2024 2:14 PM, MitchAlsup1 wrote:
 >>>>> On Sat, 31 Aug 2024 21:01:54 +0000, Bernd Linsel wrote:
 >>>>>> You compare apples and peaches. Technical specifications for your
 >>>>>> pressure vessel result from the physical abilities of the chosen
 >>>>>> material, by keeping requirements as vessel border width, geometry
 >>>>>> etc.,
 >>>>>> while compiler writers are free in their search for optimization
 >>>>>> tricks
 >>>>>> that let them shine at SPEC benchmarks.
 >>>>>
 >>>>> A pressure vessel may actually be able to contain 2× the
 >>>>> pressure it
 >>>>> will be able to contain 20 after 20 years of service due to stress
 >>>>> and strain acting on the base materials.
 >>>>>
 >>>>> Then there are 3 kinds of metals {grey, white, yellow} with different
 >>>>> responses to stress and induced strain. There is no analogy in code--
 >>>>> If there were perhaps we would have better code today...
 >>>>
 >>>> Perhaps an analogy is code written in assembler, versus coed written
 >>>> in C versus code written in something like Ada or Rust.  Backing
 >>>> away now . . . :-)
 >>>
 >>> IMNSHO, code written in asm is generally more safe than code written
 >>> in C, because the author knows exactly what each line of code is
 >>> going to do.
 >>>
 >>> The problem is of course that it is harder to get 10x lines of
 >>> correct asm than to get 1x lines of correct C.
 >>>
 >>> BTW, I am also solidly in the grey hair group here, writing C code
 >>> that is very low-level, using explicit local variables for any loop
 >>> invariant, copying other stuff into temp vars in order to make it
 >>> really obvious that they cannot alias any globals or input/output
 >>> parameters.
 >>>
 >>> Anyway, that is all mostly moot since I'm using Rust for this kind of
 >>> programming now. :-)
 >>
 >> Can you talk about the advantages and disadvantages of Rust versus C?
 >
 > Q&D programming is still far faster for me in C, but using Rust I don't
 > have to worry about how well the compiler will be able to optimize my
 > code, it is pretty much always close to speed of light since the entire
 > aliasing issue goes away.
 >
 > Rust also gets rid of the horrible external library/configure/cmake mess
 > that kept me from successfully compiling the reference LAStools lidar
 > code for nearly 10 years.
 >
 > Using the Rust port I just tell cargo to add it to my project and that's
 > it.
Thank you.  I find it interesting that the main advantage of Rust as touted by its evangelists, memory safety, didn't make your list.
--
  - Stephen Fuld
(e-mail address disguised to prevent spam)