16 Bit register compare

[IndyUK]

Hi All,
I've been delving into Assembly leanguage, in particular Z80 (Amstrad fan). I've put together a program which moves a pixel around the screen (Mode 1) using the Arrow keys. This is all working fine however, I'm stuck on how to check the min/max x/y positions so the pixel doesn't disappear of the screen. In any high level language it would be a very simple case of using a typical x-pos>320 etc. However, that's not the case with Assembly. The pixel co-ordinates in my code are being stored in registers DE & HL because I'm using the firmware to plot the pixel (&BBEA). So, the problem is I don't know how to compare DE to a maximum allowed value i.e the screen max x pixels. Of course I have the same issue with the y-pos. I have a feeling I have to split the registers since there is no 16 bit comparison option.
I would be grateful for any advice/pointer.
Thanks.

[Docent]

Hi All,
I've been delving into Assembly leanguage, in particular Z80 (Amstrad fan). I've put together a program which moves a pixel around the screen (Mode 1) using the Arrow keys. This is all working fine however, I'm stuck on how to check the min/max x/y positions so the pixel doesn't disappear of the screen. In any high level language it would be a very simple case of using a typical x-pos>320 etc. However, that's not the case with Assembly. The pixel co-ordinates in my code are being stored in registers DE & HL because I'm using the firmware to plot the pixel (&BBEA). So, the problem is I don't know how to compare DE to a maximum allowed value i.e the screen max x pixels. Of course I have the same issue with the y-pos. I have a feeling I have to split the registers since there is no 16 bit comparison option.
I would be grateful for any advice/pointer.
Thanks.

You can use the following code:

Code Select Expand


ld bc, 320
or a
ex de, hl
sbc hl, bc
add hl, bc
ex de, hl
jr c, bc_higher
...


[Axelay]

An option for checking de>=320 if you can live with effectively only being able to check even coordinate positions, that might work for your situation:

Code Select Expand


ld a,d
rra
ld a,e
rra
cp a,160
jr c, de_on_screen
...

[IndyUK]

You can use the following code:

Code Select Expand


ld bc, 320
or a
ex de, hl
sbc hl, bc
add hl, bc
ex de, hl
jr c, bc_higher
...


Hi,
Thanks for the example. I added that into my code in all four key checks and, with a bit of adjustments, everything is working very nicely. Thank You!
Does your approach have any limitations? And also, what is the "or a" used for? I commented it out since I couldn't see what purpose it served and the code works fine.
Thanks

[dthrone]

"or a" resets the carry flag because that is included in the "sbc".

[IndyUK]

"or a" resets the carry flag because that is included in the "sbc".

Thanks for the clarification. I shall reinstate it.

[Docent]

Hi,
Thanks for the example. I added that into my code in all four key checks and, with a bit of adjustments, everything is working very nicely. Thank You!
Does your approach have any limitations? And also, what is the "or a" used for? I commented it out since I couldn't see what purpose it served and the code works fine.
Thanks

No problem.
There are two limitations I can think of - the execution time is quite long and you need to load one of values into hl due to add/sbc operating only on hl. There are also benefits - the code is compact and it doesn't modify any registers.

'or a' clears the carry flag.

[AMSDOS]

I've just grabbed these from my Assembly book (Assembly Language Programming For the Amstrad CPC 464,664 & 6128):


Example 9.3
Objective: branch to 'over' if 'number' in register pair = zero

Code Select Expand


LD BC,(number)
LD A,B
OR C
JR Z,over
- - - - - - -
- - - - - - -
- - - - - - -
over:
- - - - - - -



Example 9.4
Objective: branch to 'over' if 'number' in register pair = non zero.

Code Select Expand


LD BC,(number)
LD A,B
OR C
JR NZ,over
- - - - - - -
- - - - - - -
over:
- - - - - - -



Example 9.6
Objective: branch to label 'over' if 'first' double byte number = 'second' double byte number.

Code Select Expand


LD HL,(first)
LD DE,(second)
AND A
SBC HL,DE
JR Z,over
- - - - - - -
- - - - - - -
over:
- - - - - - -



Example 9.8
Objective: branch to 'over' if 'first' double byte number is not equal to, 'second double byte number.

Code Select Expand


LD HL,(first)
LD DE,(second)
AND A
SBC HL,DE
JR NZ,over
- - - - - - -
- - - - - - -
over:
- - - - - - -



Example 9.10
Objective: branch to 'over' if 'first' double byte number is greater than, or equal to, 'second' double byte number.

Code Select Expand


LD HL,(first)
LD DE,(second)
AND A
SBC HL,DE
JR NC,over
- - - - - - -
- - - - - - -
over:
- - - - - - -



Example 9.12
Objective: branch to 'over' if 'first' double byte number is less than 'second' double byte number.

Code Select Expand


LD HL,(first)
LD DE,(second)
AND A
SBC HL,DE
JR NC,over
- - - - - - -
- - - - - - -
over:
- - - - - - -



Example 9.14
Objective: branch to 'over' if 'first' double byte number is less than 'second' double byte number.

Code Select Expand


LD DE,(first)
LD HL,(second)
AND A
SBC HL,DE
JR C,over
- - - - - - -
- - - - - - -
over:
- - - - - - -



Example 9.16
Objective: branch to 'over' if 'first' double byte number is less than, or equal to, 'second' double byte number.

Code Select Expand


LD HL,(second)
LD DE,(first)
AND A
SBC HL,DE
JR NC,over
- - - - - - -
- - - - - - -
over:
- - - - - - -


[IndyUK]

Hi Guys
Thank you to all of you! You've been a great help.

AMSDOS - Thank you for the different examples. They will no doubt come very handy. The book you have referred to seems to be very useful. I have a book on Machine Code for the Amstrad but doesn't give examples like yours.
Thanks

[AMSDOS]

Hi Guys
Thank you to all of you! You've been a great help.

AMSDOS - Thank you for the different examples. They will no doubt come very handy. The book you have referred to seems to be very useful. I have a book on Machine Code for the Amstrad but doesn't give examples like yours.
Thanks

Some of the books which were available for the Amstrad seem to be dealing with routines which would normally be too slow to operate in BASIC. My book I feel correctly avoids using the term "Machine Code", which is essentially the final output of that Assembly Source Code, so by saying "Assembly Language", it's implying it's a book with Assembly Language examples. Some BASIC stuff has crept into it, though merely as example to deal with an Assembly routine. It's not a book with Advanced Assembly code and many of the routines make use of the Firmware, which is why those examples I posted present a more simplistic approach for comparing larger numbers, so like all things Assembly can also be refined or improved.


Another book which looks similar to mine is "Machine Code for Beginners on the Amstrad", it has a chapter on  Flags, Conditions and Decision Making. The Decision Making part of my book was where those examples were included, though my book continues on with Loops of all kinds.
The small Assembly game I made earlier in the year presented me with a new problem of creating an infinite loop which I hadn't done before. In that game, I had written a way to test where the Rocks were placed and return a colour to determine if the game continued or cease.

[GUNHED]

If you hack in hex codes, then this is machine code. Well, the world before MAXAM was all about to hack it in with your own fingers.