🚀 go-pugleaf

In <jd-wong-1008950341330001@berlin-15.slip.uiuc.edu> jd-wong@uiuc.edu (Jim Wong) writes:

>In article <danpop.808015471@rscernix>, Dan.Pop@mail.cern.ch (Dan Pop) wrote:
>> Nope.  Not even Motorola claimed this.  A cpu with a 16-bit data bus
>> and a 16-bit ALU is by no means a 32-bit processor.
>
>If depends on how you look at it.  To quote Tanenbaum:
>
>"If you are ever trying to determine if someone is a hardware person or a
>software person, ask him if the 68000 is a 16- or 32-bit machine.  A
>hardware engineer will say 16, due to the 16-bit data bus.  A programmer
>will say 32, because the registers are all 32 bits wide."

Being a software (assembly language) person, I used to care about the
number of cpu cycles required by the instructions used in time-critical
sequences of code.  And 32-bit arithmetic instructions took more cpu
cycles than 16-bit ones, because the ALU was 16-bit only.  The high level
language programmer couldn't care less about the bitness of the underlying
hardware.

So, take Tannenbaum's remark with a grain of salt.

Dan
--
Dan Pop
CERN, CN Division
Email: Dan.Pop@mail.cern.ch 
Mail:  CERN - PPE, Bat. 31 R-004, CH-1211 Geneve 23, Switzerland