How CPU's work (was Re: Hi, I'm new...)

Henk Gooijen henk.gooijen at hotmail.com
Mon Aug 7 14:39:29 CDT 2006


>From: "Roy J. Tellason" <rtellason at verizon.net>
>Date: Mon, 07 Aug 2006 15:14:15 -0400
>
>On Monday 07 August 2006 02:58 pm, Bryan Pope wrote:
> > And thusly were the wise words spake by Roy J. Tellason
> >
> > > Wasn't the 68010 supposed to fix this?  I have one of those,  too...
> >
> > I remember the 68010 was all the rage for Amiga owners when it came 
>out..
> > All you had to do was replace your 68000 and BAM! you got almost close
> > to a 10% speed increase...  :)
>
>Well,  that one is a 10 MHz part while the 68000 I have is I think an 8 MHz
>part,  which is a bigger increase than that,  but the clock driving the 
>thing
>would need to be changed too,  and I don't think they have exactly the same
>pinout,  though it's been a really long time since I looked at that 
>material.

Yes, it's been a while for me too, but AFAIR both the 68000 and 68010
can be obtained at higher speed ratings. My StarShip had a self-developed
68000 board, and I upgraded it to a 68010. Hardware-wise that is simple,
as they are pin-compatible. Software-wise it depends what you are doing.
In my case, I had a tiny operating system to schedule tasks, a time-wait
queue, etc., it was trickier, because the *stack frame* that a 68000 and a
68010 push onto the hardware stack when an exception occurs is *different*.
My OS used exceptions to return from user to supervisor mode (which is BTW
the only method).
But the 68010 was a little faster, especial in tiny (3 instruction IIRC) 
loops,
which where "cached".
I am not sure there where much extra fancy instructions as as soon after
that upgrade got a 68020 (at 30 MHz) running, and that one had a few
instructions that I really missed on the 68000.
Yes, I am a "Moto-man", not an "Intel-chap" :-)
- Henk, PA8PDP.

_________________________________________________________________
Eindeloos zoeken naar dat ene document is nu voorbij!  http://desktop.msn.nl




More information about the cctech mailing list