Difference between revisions of "Z80 Good Programming Practices"

From WikiTI
Jump to: navigation, search
(added alternative to LUTs for if few options and sequential values of a)
(Ix thing was painfully wrong, lut code wasn't the best still isn't.)
Line 1: Line 1:
== Using IX ==
 
 
If you have objects represented by adjacent chunks of data in memory, you can use IX to easily manage them.
 
 
<table border="1" cellpadding="2">
 
<tr><th>Without</th><th>With</th></tr>
 
<tr>
 
<td>
 
<nowiki>
 
ld b,10
 
ld hl,SpritesData
 
DisplaySpritesLoop
 
ld b,(hl) ; coordx
 
inc hl
 
ld c,(hl) ; coordy
 
inc hl
 
ld d,(hl) ; first part of address
 
inc hl
 
ld e,(hl) ; end of address
 
inc hl
 
call DisplaySprite
 
djnz DisplaySpritesLoop
 
</nowiki>
 
</td>
 
<td>
 
<nowiki>
 
COORDX .equ 0
 
COORDY .equ 1
 
ADDR1 .equ 2
 
ADDR2 .equ 3
 
 
ld b,10
 
ld ix,SpritesData
 
DisplaySpritesLoop
 
ld b,(ix+COORDX)
 
ld c,(ix+COORDY)
 
ld d,(ix+ADDR1)
 
ld e,(ix+ADDR2)
 
call DisplaySprite
 
ld hl,4
 
add ix,hl
 
djnz DisplaySpritesLoop
 
</nowiki>
 
</td>
 
</tr>
 
</table>
 
 
Defining constants for the offsets of each field of your "objects" makes the code more understandable. However this particular example is not the best, being that the original code is faster and smaller.  Typically in sequential access using the HL register gives better performance, but if random access of different objects and different elements in an object is required throughout a particular iteration, then ix would be the better choice.
 
 
 
 
 
== Lookup table ==
 
== Lookup table ==
  
Line 78: Line 27:
 
  <nowiki>
 
  <nowiki>
 
  ld a,(SpriteNumber)
 
  ld a,(SpriteNumber)
 +
cp 0
 +
jp z,ChooseSprite0
 
  cp 1
 
  cp 1
 
  jp z,ChooseSprite1
 
  jp z,ChooseSprite1
Line 86: Line 37:
 
  cp 4
 
  cp 4
 
  jp z,ChooseSprite4
 
  jp z,ChooseSprite4
 +
cp 5
 +
jp z,ChooseSprite5
 
...
 
...
 +
ChooseSprite0
 +
ld hl,Sprite0
 +
jp DisplaySprite
 
ChooseSprite1
 
ChooseSprite1
 
  ld hl,Sprite1
 
  ld hl,Sprite1
Line 98: Line 54:
 
ChooseSprite4
 
ChooseSprite4
 
  ld hl,Sprite4
 
  ld hl,Sprite4
 +
jp DisplaySprite
 +
ChooseSprite5
 +
ld hl,Sprite5
 
  jp DisplaySprite
 
  jp DisplaySprite
 
...
 
...
Line 108: Line 67:
 
  <nowiki>
 
  <nowiki>
 
  ld a,(SpriteNumber)
 
  ld a,(SpriteNumber)
  add a,a ; a*2
+
  add a,a   ; a*2 (limits SpriteNumber to 128)
 
  ld h,0  
 
  ld h,0  
 
  ld l,a  
 
  ld l,a  
Line 118: Line 77:
 
  ld l,a
 
  ld l,a
 
  ld bc,(coordinates)
 
  ld bc,(coordinates)
  call SpriteRoutine
+
  jp SpriteRoutine
 
...
 
...
 
SpriteAddressLUT
 
SpriteAddressLUT
 +
.dw Sprite0
 
  .dw Sprite1
 
  .dw Sprite1
 
  .dw Sprite2
 
  .dw Sprite2
 
  .dw Sprite3
 
  .dw Sprite3
 
  .dw Sprite4
 
  .dw Sprite4
 +
.dw Sprite5
 
  </nowiki>
 
  </nowiki>
 
</td>
 
</td>
Line 137: Line 98:
 
  <nowiki>
 
  <nowiki>
 
  ld a,(MenuChoice)
 
  ld a,(MenuChoice)
 +
cp 0
 +
jp z,Choice0
 
  cp 1
 
  cp 1
 
  jp z,Choice1
 
  jp z,Choice1
Line 145: Line 108:
 
  cp 4
 
  cp 4
 
  jp z,Choice4
 
  jp z,Choice4
 +
cp 5
 +
jp z,Choice5
 +
cp 6
 +
jp z,Choice6
 +
cp 7
 +
jp z,Choice7
 
...
 
...
 
  </nowiki>
 
  </nowiki>
Line 163: Line 132:
 
  jp (hl)
 
  jp (hl)
 
...
 
...
CodeBranchLUT
+
CodeBranchLUT:
 +
.dw Choice0
 
  .dw Choice1
 
  .dw Choice1
 
  .dw Choice2
 
  .dw Choice2
 
  .dw Choice3
 
  .dw Choice3
 
  .dw Choice4
 
  .dw Choice4
 +
.dw Choice5
 +
.dw Choice6
 +
.dw Choice7
 
  </nowiki>
 
  </nowiki>
 
</td>
 
</td>

Revision as of 00:26, 27 December 2006

Lookup table

If you have a place in your code where a value is tested to choose between a lot of things, like subroutines or data, it can be a good idea to use lookup tables instead of a series of tests. It makes the code more readable, concise and extensible.

In terms optimisation though it should be used when the data is not sequentially ordered or when the objects being pointed to are not the same size. For example, using LUTs (Look Up Tables) to find a tile in a block of memory that is only tiles would both slower and cost more memory. Using LUTs to find a particular string would be quicker but would waste more memory than a linear search. Using LUTs as a jump table to different code blocks located through out a program would be faster and smaller compared to the alternative.

However, if there aren't many jumps and many of the values of a are sequential, it would be more efficient to do something like:

 ld a,(Number)
 or a
 jp z,A_is_0
 dec a
 jp z,A_is_1
 dec a
 jp z,A_is_2
 sub 2
 jp z,A_is_4
 dec a
 jp z,A_is_5
 

Examples:

WithoutWith
 ld a,(SpriteNumber)
 cp 0
 jp z,ChooseSprite0
 cp 1
 jp z,ChooseSprite1
 cp 2
 jp z,ChooseSprite2
 cp 3
 jp z,ChooseSprite3
 cp 4
 jp z,ChooseSprite4
 cp 5
 jp z,ChooseSprite5
...
ChooseSprite0
 ld hl,Sprite0
 jp DisplaySprite
ChooseSprite1
 ld hl,Sprite1
 jp DisplaySprite
ChooseSprite2
 ld hl,Sprite2
 jp DisplaySprite
ChooseSprite3
 ld hl,Sprite3
 jp DisplaySprite
ChooseSprite4
 ld hl,Sprite4
 jp DisplaySprite
ChooseSprite5
 ld hl,Sprite5
 jp DisplaySprite
...
DisplaySprite
 ld bc,(coordinates)
 call SpriteRoutine

 ld a,(SpriteNumber)
 add a,a   ; a*2 (limits SpriteNumber to 128) 
 ld h,0 
 ld l,a 
 ld de,SpriteAddressLUT
 add hl,de
 ld a,(hl)
 inc hl
 ld h,(hl)
 ld l,a
 ld bc,(coordinates)
 jp SpriteRoutine
...
SpriteAddressLUT
 .dw Sprite0
 .dw Sprite1
 .dw Sprite2
 .dw Sprite3
 .dw Sprite4
 .dw Sprite5
 

And this one :

WithoutWith
 ld a,(MenuChoice)
 cp 0
 jp z,Choice0
 cp 1
 jp z,Choice1
 cp 2
 jp z,Choice2
 cp 3
 jp z,Choice3
 cp 4
 jp z,Choice4
 cp 5
 jp z,Choice5
 cp 6
 jp z,Choice6
 cp 7
 jp z,Choice7
...
 
 ld a,(MenuChoice)
 add a,a ; a*2
 ld h,0
 ld l,a
 ld de,CodeBranchLUT
 add hl,de
 ld a,(hl)
 inc hl
 ld h,(hl)
 ld l,a
 jp (hl)
...
CodeBranchLUT:
 .dw Choice0
 .dw Choice1
 .dw Choice2
 .dw Choice3
 .dw Choice4
 .dw Choice5
 .dw Choice6
 .dw Choice7
 
This article is a stub. You can help WikiTI by expanding it.