'Yeeeeehah! - 50MHz 16C5X clones ! (getting [OT], t'
William Chops Westfield
>As processors get faster in general, applications DON'T get faster
YES for sure if Billy is involved - see the Windows threads.
software must get more complex to require new hardware
to give more profit to the desktop computer industry :(
NO in some other cases.
Actually, I made that comment coming mainly from a "small mainframe"
world. The DEC-20's I used to program (10 to 20 years ago) would support
about 50 to 100 users doing email and editting and programming classwork,
using dumb terminals connected via a network or to hardwired terminal
ports, a couple 700MB drives the size of washing machines, and about 20MB
of memory. (4 Mwords, 36 bit words.) I think they ran at about 1 MIP.
Today, the same 100 users, doing approximately the same thing, require
a multi-processor sparcserver of some kind with several hundred mips,
several 12Gbyte disk drives, and 256Meg of memory. Of course, now the
network terminals are X terminals, and so on... "real" SW development
requires a different machine, of course.
[doing it in SW on Atmel AVR]
all those things run in the background and can be considered as virtual
hardware a concept we feel Parallax and Scenix are targetting too.
On the small AVRs, I'd worry about code space consumed as well...
[home DIY system would]
Cold boot in 4 seconds maximum.
Man, it was SO depressing when I upgraded by aging MacIIci (25 MHz
68030) to a new Mac (8500 - 120MHz powerPC with theoretically faster
disks) and it didn't boot any faster :-(
More... (looser matching)
- Last day of these posts
- In 1997
, 1998 only
- New search...