Sujet : Re: Baby X is bor nagain
De : bc (at) *nospam* freeuk.com (bart)
Groupes : comp.lang.cDate : 01. Jul 2024, 13:22:42
Autres entêtes
Organisation : A noiseless patient Spider
Message-ID : <v5u3i2$11kg9$1@dont-email.me>
References : 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
User-Agent : Mozilla Thunderbird
On 30/06/2024 16:47, David Brown wrote:
On 30/06/2024 12:48, bart wrote:
It opens up possibilities.
>
It opens possibilities for doing lots more manual work, making more mistakes, finding those mistakes later, and generally working in a way most people were glad to move away from a generation ago.
Fast compilation is fine and good in itself. And I can understand feeling that giant IDE's like Eclipse or MS Code are bigger than you want, and take time to learn. But it is incomprehensible to me that you'd /want/ to use such limited tools as you do.
I've been doing this stuff a long time. I've got to know what problems come up for me in programming and in development, and which features of editor, compiler and language would have helped.
Remember that I generally write all those tools and can do what I like.
Of most help have been micro-features in my non-C language, some of which made my efficient compilers possible. There I can't use external tools since no one else supports my language.
If I was to use C, then I'd lose those features, and HAVE to employ smarter tools to get over /some/ of the problems.
Since I started creating my tools, the machines they run on have gotten several magnitudes faster /per core/, while the individual binaries they have to produce might be typically 1-2 magnitudes bigger.
(In my Windows system folder, 90% of the DLLs and 93% of the EXEs are smaller than 1MB. 1MB is 1/8000th of the RAM of the PC. In 1985, a program 1/8000 the) size of 640KB would have been 80 bytes in size!
And yet, with all the computing power available, people are still compiling one 1000-line module at a time (about 10KB of code); they are still using techniques to avoid compilation where possible; they are having to parallelise via multiple cores; they're having to use 100-times bigger and more complex compilers to get a mere 2x speedup on top of that 1000x faster hardware.
What's gone wrong?