Some notes about 020+


 Most of 020 cycletimes are same than on 030; haven't found any differenties.

;-------------------------
 ...
 clr.l -(a0)
 clr.l -(a0)
 ...
 is 25% faster than
 ...
 clr.l (a0)+
 clr.l (a0)+
 ... 
;-------------------------
 move.l (a0,d0.l)  is faster than
 move.l (a0,d0.w)
 same with adda.w <-> adda.l etc.
;-------------------------
 move.b d0,-(a7) will decrease a7 with 2!
;-------------------------

     Never use those silly #xx,([a0]) new modes, they are slower than
 move.l (a0),a0
 move.l #xx,(a0)
     ...case you have no extra spare registers, but... that's rare.

;-------------------------

 Time-optimizing on A1200 is more sparing BUS than CPU; the CHIP RAM
is just too slow.. Remember to adjust your WRITES to CHIP; Adjusting
memory READS wont do you no good. (Naturally, how could computer keep
going on without knowing have we read something we're using?)

;--------------------

 Interrupts are deadly slow, try to figure out something better.. like
Copper ;-) Just plain movem*2, Trap and RTE will take HUGE amount of
cycles.

 movem.l all,-(sp)
 movem.l (sp)+,all
 rte

 will take about 80 Cycles... okey, why use all regs ;-)

HTML Conversion by AG2HTML.pl V2.941126c, perl $RCSfile: notes020+.HTML,v $$Revision: 1.1 $$Date: 1999/09/14 21:14:11 $ Patch level: 36 & witbrock@cs.cmu.edu