Anyone of you have a good idea on how to create a random number from 1 to 3 in ASM?

<TABLE BORDER=0 ALIGN=CENTER WIDTH=85%><TR><TD><font class="mrc-small">Code:</font><HR></TD></TR><TR><TD><FONT class="mrc-small"><PRE>
random:
ld a,r
ld b,8
loop:
rra
rl c
djnz loop

ld hl,0
add hl,bc
add hl,bc
add hl,bc
ld a,h
inc a
ret
</PRE></FONT></TD></TR><TR><TD><HR></TD></TR></TABLE>

i already have something like this but it doesn't quite do the trick...

any help is welcome

You can use the refresh register to get a "random" number

the source states a LD A,R already

a simple and tiny one is this:

loop: ld a,r
and 3
jr z,loop

Not all bits are used in the r register
Don't now if it is the lowest or the highest bit that is unused

r u sure about that MSD? can't remember having that problem

from 1 to 3 eh..

<TABLE BORDER=0 ALIGN=CENTER WIDTH=85%><TR><TD><font class="mrc-small">Code:</font><HR></TD></TR><TR><TD><FONT class="mrc-small"><PRE>
loop:
ld a,r
and 3
jr z,loop
ret
</PRE></FONT></TD></TR><TR><TD><HR></TD></TR></TABLE>

will generat 0, 1,2 or 3. if it's zero, it'll try again.

what and why wanna use it for might m?

bit 7 of the R register remains untouched.
If you store 0 in R, it will count from 0 to 127. If you store 128 in R, it will count from 128 to 255.
In other words, you can store a whole extra bit in the R register! Might be cool for optimizing storage of carry-flag or something

Other R trivia:
On R800 the R register is simulated, it has no connection with the memory refresh anymore.
(So loop: XOR A : LD R,A : JP LOOP doesn't 'damage' the memory)
On Z380 the R register is untouched and can be used as an extra 8-bit register!

Quote:

the source states a LD A,R already a simple and tiny one is this: loop: ld a,r and 3 jr z,loop

Since the R register isn't a random one, but rather one that is just incremented all the time (see below) there's a small problem with the above code: when the result of the and 3 is zero then the result of the next iteration of the loop will always result in the same value. I think it's 3, but I might have miscounted something. This would mean that the value 3 has a 50% chance of occuring and 1 and 2 have a 25% each.

Things can get even worse of course if this random-routine is used inside a loop. Let's for example assume this random-routine is being used to fill a piece of memory:

ld hl,#4000
ld bc,#100
fillloop:
randomloop:
ld a,r
and 3
jr z,randomloop
ld (hl),a
inc hl
dec bc
ld a,b
or c
jp nz, fillloop

As you can see, all the other code in the loop is non-conditional which means it'll always execute in the same number of cycles and R will be increased by the same value.
So, every time the and 3 gives a 1, 2 or 3 you can calculate what the result of the and 3 in the next iteration will be and for 1 it will always be the same as it is for 2 and for 3 (and as 0 -> 3 also for 0).
This means you either get the same value every time:
1,1,1,1,1,1,1 or
2,2,2,2,2,2,2 or
3,3,3,3,3,3,3
or you get a repetetive series:
...1,2,3,1,2,3,1,2,3,1,2,3... or
...1,3,2,1,3,2,1,3,2...
Neither of which is desirable.

To quote from "The undocumented Z80 documented"

Quote:

The R register is increased at every first machine cycle (M1). Bit 7 of the register is never changed by this; only the lower 7 bits are included in the addition. So bit 7 stays the same, but it can be changed using the LD R,A instruction. Instructions without a prefix increase R by one. Instructions with an ED, CB, DD, FD prefix, increase R by two, and so do the DDCBxxxx and FDCBxxxx instructions (weird enough). Just a stray DD or FD increases the R by one. LD A,R and LD R,A access the R register after it is increased (by the instruction itself).

hmm, funny. didn't know that (anymore?!)

Quote:

i already have something like this but it doesn't quite do the trick... any help is welcome

Take a look at members.lycos.co.uk/leeedavison/z80/prng/
This will give you an 8-bit random value in A (and if you also take the carry into account you could view it as a 9-bit random value) from which you then get to a value of 1-3....

sorry guys for the late reply, but I can only connect to the internet at home... so here it goes:

I'm working on a little project (can't tell more) but I need a random number from 1 to 3...

I already tried the short sourcy (ld a,r + and with mask 3)... It just isn't random enough... some number pops up more often than the others...

Use a bigger table with 128 entries and do something like the following:

xor a
ld r,a (to mantain R between 0 and 127)

....
ld a,r
ld b,0
ld c,a
ld hl,TABLE
add hl,de
ld a,[hl]
....

TABLE: db 0,1,2,3,1,2,3,0,2,3,0,1,3,0,1,2,... Really random, please!


I know this is really unoptimized, but I am a Gameboy freak, I do not use IX or IY. And I prefer 16-bit arithmetic.

Kind regards,

Ed Robsy

Google helps...

<TABLE BORDER=0 ALIGN=CENTER WIDTH=85%><TR><TD><font class="mrc-small">Code:</font><HR></TD></TR><TR><TD><FONT class="mrc-small"><PRE>
Rand: ld de,0
seed: equ $-2
ld a,d
ld h,e
ld l,253
or a
sbc hl,de
sbc a,0
sbc hl,de
ld d,0
sbc a,d
ld e,a
sbc hl,de
jr nc,Rndjmp
inc hl
Rndjmp: ld (seed),hl
ld a,l
and 3
jr z,Rand
ret
</PRE></FONT></TD></TR><TR><TD><HR></TD></TR></TABLE>

A variation of the one I use in Realfun to randomize a FM voice:

random13:
ld a,r
ld c,a
ld a,(bc)
ld hl,seed
add a,(hl)
add a,c
ld (hl),a
and 3
jr z,random13
ret
seed:
byte 0