Re: A82: Random routine

To: assembly-82@lists.ticalc.org
Subject: Re: A82: Random routine
From: Jeremy Wazny <j.wazny@ugrad.unimelb.edu.au>
Date: Mon, 6 Dec 1999 13:57:35 +1100 (EST)
In-Reply-To: <001901bf3f49$371133e0$396a3d80@eastnet.gatech.edu>
Reply-To: assembly-82@lists.ticalc.org

Actually, the value of r is used as an offset into page zero of memory, so
the randomness depends on the distribution of data there. I think that's
where the zero you're talking about came from.
I think that probably even the simplest scheme for random number
generation would suffice in this case; who's going to be able to mentally
pick the next combination of cards from their current state? :) 

-Jeremy

On Sun, 5 Dec 1999, Harper Maddox wrote:

> 
> I looked at the previous scheme of shuffling a deck for O(n) performance.
> This is what i came up with.  It's going to be a much more random stack by
> getting a random number out of a seed rather than using "ld a,r"
> 
> 
> same_code:
> 
> shuffle:
>  ld hl,cards            ; start of 20 byte array
>  ld b,20
> shuffle_loop:
>  push bc
>  call getrand
>  ld de,cards
>  or %1111               ; between 0 and 15
>  ld b,a
>  srl a
>  srl a
>  add a,b                ; between 0 and 18
>  add a,e
>  ld e,a
>  ld b,(de)
>  ldi                    ; (de) = (hl) /*/  hl++ /*/ de++
>  ld (hl),b
>  pop bc
>  djnz shuffle_loop
> 
> getrand:                ; by Jason Todd
>  ld a,(randseed)
>  ld b,a
>  ld a,r
>  add a,b
>  sla b
>  sla b
>  add a,b
>  sla b
>  sla b
>  add a,b
>  add a,e
>  inc a
>  rrc a
>  ld (randseed),a
>  ret
> 
> randseed: .db %10101010    ; im guessing this is better to start it off
> weird than at zero
> 
>

Follow-Ups:

Re: A82: Random routine

From: "Harper Maddox" <gte172i@prism.gatech.edu>

References:

Re: A82: Random routine

From: "Harper Maddox" <gte172i@prism.gatech.edu>