On Feb 16, 5:55 am, pauldkl...@[EMAIL PROTECTED]
wrote:
> Part III - pdQscan - A Working Example
>
> Below is a working example of a dense qscan. At the top you see the
> variables qM-qStep, which are taken from a list generated by program
> "qScansearch" that I described in Part II. Both the list and
> qScansearch will be posted shortly. I tested pdQscan against six
> qscan 4.5 programs with these results (Win% = pdQscan's win
> percentage):
>
> Opponent Win% Loss%
> The Human Machine 12.5 10.9
> paper(paper(paper(clear))) 12.0 10.0
> Soldier of Silkland 11.4 11.0
> Maelstrom 13.1 12.0
> unheard-of 7.4 13.5
> Return of the Fugitive 6.2 8.0
>
> pdQscan's attack is a standard forward-bombing run with an increment
> of 6. If a backward-bombing run is used, much better results are
> obtained against Return of the Fugitive. unheard-of uses a fast boot
> to escape, pdQscan rarely catches it in time.
>
> Given that pdQscan is much more exposed due to its size, these numbers
> are significant. I tried commenting out some of the scan and the
> results were always worse, so there is a positive effect to having a
> longer scan.
>
> Here it is, enjoy!
>
> ;redcode-94
> ;name pdQscan
> ;author P.Kline
> ;assert CORESIZE == 8000
>
> qM equ (6249+1)
> qMod equ 4249
> ptScan equ 8
> ptDecode equ 51
> ptLaunch equ 81
> qStep equ (250/2-1)
>
> qA equ ((qMod*(tA-qPtr))%CORESIZE+1)
> qB equ ((qMod*(tB-qPtr))%CORESIZE+1)
> qC equ ((qMod*(tC-qPtr))%CORESIZE+1)
> qD equ ((qMod*(tD-qPtr))%CORESIZE+1)
> qE equ ((qMod*(tE-qPtr))%CORESIZE+1)
> qF equ ((qMod*(tF-qPtr))%CORESIZE+1)
> qG equ ((qMod*(tG-qPtr))%CORESIZE+1)
> qH equ ((qMod*(tH-qPtr))%CORESIZE+1)
> qI equ ((qMod*(tI-qPtr))%CORESIZE+1)
> qJ equ ((qMod*(tJ-qPtr))%CORESIZE+1)
>
> tA dat tE ,qA
> tB dat 1 ,qB
> tC dat tF ,qC
> tD dat tJ ,qD
> bM dat 0 ,-67
> for ptScan-CURLINE
> dat 0,0
> rof
> qGo sne qPtr+qA*qM ,qPtr+qA*qM+qStep
> seq <tA ,qPtr+(qA-1)*qM+qStep
> djn.a decode ,{tB
> sne qPtr+qB*qM ,qPtr+qB*qM+qStep
> seq <tB ,qPtr+(qB-1)*qM+qStep
> jmp decode ,{tB
> sne qPtr+qC*qM ,qPtr+qC*qM+qStep
> seq <tC ,qPtr+(qC-1)*qM+qStep
> tE jmp decode ,qE
> sne qPtr+qD*qM ,qPtr+qD*qM+qStep
> seq <tD ,qPtr+(qD-1)*qM+qStep
> jmp decode ,}tB
>
> sne qPtr+qE*qM ,qPtr+qE*qM+qStep
> seq <tE ,qPtr+(qE-1)*qM+qStep
> jmp decode ,{decode
> sne qPtr+qF*qM ,qPtr+qF*qM+qStep
> seq <tF ,qPtr+(qF-1)*qM+qStep
> jJ jmp decode ,}decode
>
> sne qPtr+qG*qM ,qPtr+qG*qM+qStep
> seq <tG ,qPtr+(qG-1)*qM+qStep
> jmp decode-1 ,{pH
> sne qPtr+qH*qM ,qPtr+qH*qM+qStep
> pH seq <tH ,qPtr+(qH-1)*qM+qStep
> tF jmp decode-1 ,qF
> sne qPtr+qI*qM ,qPtr+qI*qM+qStep
> seq <tI ,qPtr+(qI-1)*qM+qStep
> jmp decode-1 ,}pH
>
> sne qPtr+qJ*qM ,qPtr+qJ*qM+qStep
> seq <tJ ,qPtr+(qJ-1)*qM+qStep
> jmp jJ ,}decode
> tJ jmp pGo ,qJ
> for ptDecode-CURLINE
> dat 0,0
> rof
> mov.a #pH-decode ,decode
> decode mul.b *tB ,qPtr
> sne null ,@[EMAIL PROTECTED]
> add.ab #qStep ,qPtr
> mov bM ,@[EMAIL PROTECTED]
> qPtr mov bM ,@[EMAIL PROTECTED]
> add.ab #6 ,qPtr
> djn.b -3 ,#16
> jmp pGo
> for ptLaunch-CURLINE
> dat 0,0
> rof
> pStep equ (7*3754)
> pStep2 equ (1303+7*1021)
> pGo mov <pCopy ,{pCopy
> tG spl 2 ,qG
> tH spl 1 ,qH
> tI spl 1 ,qI
> mov <pCopy ,{pCopy
> pCopy spl 1+7+4000 ,1+7
> mov <1 ,{1
> spl 2000+7 ,7
> pPap spl @[EMAIL PROTECTED]
,>pStep
> mov }pPap ,>pPap
> mov {pPap ,<1
> spl @[EMAIL PROTECTED]
,>pStep2
> mov.i #6000-1-2667 ,}2667
> mov.i >0 ,}0
>
> null dat 0 ,0
> end qGo
>
> P. Kline
Nice :) I wish i have time to test this.
|
