replicators? -> Phoenix & TimeScapesource
Juha Pohjalainen
> Phoenix (if it's ever published... :)
I have tried, but I think it did not make it! Lets try again! And
now I can also attach here one version of my TimeScape warrior.
(But I can not give my Silk Warrior, because I had hard disk
break down and no backups from my warriors; BTW, that Phoenix
is also rebuilded one!)
Both warriors use basic silk copying style, but code is executed
only once, so they mightbe weaker than normal silk papers in some
ways, but they can also be faster/stronger in other ways. Feel free
to experiment some more if you want to.
[Yep, I know, I'm stuck with replicators ... ]
Jippo
;redcode-94 verbose
;name Phoenix
;author J. Pohjalainen
;assert CORESIZE==8000
;kill Phoenix
PSTEP equ 2720
NSTEP equ 1072
ESTEP equ 455
warrior
spl 1, <-100
spl 1, <-300
mov.i -2, 0
phoenix spl @phoenix, >PSTEP
mov.i }phoenix, >phoenix
nest spl @nest, >NSTEP
mov.i }nest, >nest
egg spl @egg, >ESTEP
mov.i }egg, >egg
dat.f <2667, <2*2667
end warrior
;redcode-94 verbose
;name TimeScape (0.9)
;author J. Pohjalainen
;assert CORESIZE==8000
;strategy I'm stuck with replicators! Here is _The Latest_ one!
;strategy \---------------------------\ ----------------------
;strategy / ts1 spl @ts1, }STEP1 / Phoenix/Cell warrior
;strategy \ mov.i }ts1, >ts1 \ body, 6+ processes to
;strategy / ts2 spl @ts2, }STEP2 / keep That Thing alive,
;strategy \ mov.i }ts2, >ts2 \ two of them working
;strategy / mov.i {ts2, <ts3 / together with proper
;strategy \ ts3 jmp @ts3, }STEP3 \ constants and you have
;strategy /___________________________/ found >>--> TimeScape!
;strategy \T I M E T O E S C A P E\ ----------------------
;kill TimeScape
TSTEP equ 1800
CSTEP equ 3740
NSTEP equ -1922
tim1 spl @tim1, }TSTEP
mov.i }tim1, >tim1
cel1 spl @cel1, }CSTEP
mov.i }cel1, >cel1
mov.i {cel1, <ncl1
ncl1 jmp @ncl1, >NSTEP
st for 84
dat.f 0, 0
rof
warrior
spl 1, <-200
mov.i -1, 0
spl 1, <-300
spl tim1, <-400
tim2 spl @tim2, }TSTEP
mov.i }tim2, >tim2
cel2 spl @cel2, }CSTEP
mov.i }cel2, >cel2
mov.i {cel2, <ncl2
ncl2 jmp @ncl2, >NSTEP
end warrior
Robert Macrae
> ;redcode-94 verbose
> ;name Phoenix
.
.
> phoenix spl @phoenix, >PSTEP
> mov.i }phoenix, >phoenix
> nest spl @nest, >NSTEP
> mov.i }nest, >nest
> egg spl @egg, >ESTEP
> mov.i }egg, >egg
> dat.f <2667, <2*2667
Drat. I wrote a warrior called tree (-> branch -> twig) which was
_identical_ bar
1) Use of SPL xxx, 0 in place of you (better) SPL @0, >xxx
2) Different step sizes.
The difference in scores on the hill, however, is immense; I think I made
about 110 or so, certainly failing to get on. I guess the improvement must
come late on, when there are so many processes that your SPLs start bombing
a cell long before the mov.i hits it...
Surprising what a difference such a small change makes, but I suppose when
you've only got 7 instructions they've got to work _hard_!
--
Robert Macrae
Randy Graham
: > ;redcode-94 verbose
: > ;name Phoenix
: > phoenix spl @phoenix, >PSTEP
: > mov.i }phoenix, >phoenix
: > nest spl @nest, >NSTEP
: > mov.i }nest, >nest
: > egg spl @egg, >ESTEP
: > mov.i }egg, >egg
: > dat.f <2667, <2*2667
me or put it here again.
Randy