> Want a good reason to teach ASM in uni's? Remind people that
> games will
> become the biggest entertainment industry in the US within
> the next 10
> years and we need ASM programmers.
You may need assembly level programmers, but your
entire industry is unlikely to employ them in
sufficient numbers to be anything other than a
blip on the radar.
You also have the problem that your hardware is
quite specialised and likely to diverge even more
as time goes on from the general purpose CPUs
that the rest of the world runs its code on.
You presumably need people who understand the
hardware at a sufficiently deep level. I'm not
sure that machine level programming is enough
for that (although I guess it *might* be a
useful first step).
I know that I started out with a bit of 8080,
fortunately followed quite swiftly by Z80
(and CP/M). Then PDP-11 and VAX MACRO languages.
So a trend of constantly increasing programmer
friendliness.
Then I had to pick up Alpha (PALcode, lovely
stuff), MIPS (R5K, no branch delay slots that
I recall), PowerPC, Solaris and more PPC. I
learnt to be grateful for compiler's then. I
still had to write machine code, but only to
provide the basic underpinnings for the OS
and maybe a few performance-critical bits.
The rest of the time I'm either trying to
make sense of debug info that spits out a
bunch of registers or trying to squeeze the
hardware by having it do multiple things at
once.
But if I look back on all of this, I still
think I've spent 95% of my coding time
reading and writing code in a HLL (I'm
counting BLISS and C as HLL for the
purpose of this message :-))
If the games industry in the UK pays
well (and if it employs more than a
few dozen programmers!) these days,
do let me know.
Antonio
--
---------------
Antonio Carlini arcarlini_at_iee.org
Received on Tue Jun 22 2004 - 16:50:10 BST