home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
Stolen Data 3
/
Stolen Data 3.adf
/
MAG-TEXT
/
CODE7b
< prev
next >
Wrap
Text File
|
1989-01-01
|
1KB
|
31 lines
1) The standard technique for speeding
up a routine is to remove all loops,
this is best achieved here by
generating your sine-scroll routines as
part of the setup routine.
2) Generally your sine-waves have
constant seperation, that is the amount
you add to your sine-pointer between
lines doesn't change during the course
of the scroller. Therefore it is
quicker to calculate these offsets
before you run the routine, in the
setup. If you combine this with the
previous idea, and use Indirect
addressing with displacement, with the
displacements as the offsets just
mentioned a saving is made in speed
terms.
3) This method only works if you use
techniques 1 and 2, it only makes a
small saving but if you rearrange the
order of plotting the sine-scroller so
that all the segments using the same
BLTAFWM value are plotted sequentially
then you only need to set BLTAFWM at
the start of each block.
4) Of course it goes without saying
always use Indirect addressing modes
for every operation. In my fastest
sine scroller, I used every address