(C) Deadie / HorrorSoft / CPU
Generally I say the following. While I kodil,
as if he were going HorrorFasTest, my
seditious thought crept into my head about
autoconfig'e for everyone there multicolor'ov
and border'nyh effects. For me so
remains a mystery why demomaker'y
such fixed effects only (mostly)
under two models of the Spectrum (Pentagon and
Scorpion). As many know, for creating something on the boarder
or multicolor need to know the number of rows (lines) of the
screen / tpl / and the number of cycles per line (line) / lps
/. Naturally, based on of these values and constants are
defined for procedures to display. Here are the most
values can be obtained by using protsedurki stated below. I
note that the true value you will get only if off turbo mode.
For a start, give procedure that determines enabled
or off turbo mode:
ATTENTION!!
On the authorship of the following procedures
Do not pretend!
, Identification of the state turbo mode
; Input: -
; O: A = 0 - disabled, A = 1 - on
TURBO CALL INT_TRB
LD A, B
CP # 00
JR Z, TURBOFF
CP # 02
JR NC, TURBON
LD A, C
CP # 41
JR NC, TURBON
TURBOFF XOR A
RET
TURBON LD A, 1
RET
TRB_INT DI
LD A, # C3
LD (# BCBC), A
LD HL, INT_TR1
LD (# BCBD), HL
LD A, # BA
LD I, A
IM 2
LD DE, # 0000
EI
HALT
LD HL, INT_TR2
LD (# BCBD), HL
TRB_IN3 INC DE
LD B, # 0F
DJNZ $
LD A, R
JR INT_TR3
TRB_IN1 EI
RET
TRB_IN2 POP HL
LD C, E
LD B, D
IM 1
EI
RET
And now the most important thing - the procedure
calculation of lps and tpl:
; Calculating count of rows of the raster
, (The lines in the screen)
; Input: -
; O: DE - lps
CLC_LPS CALL IM2SET
LD A, # C3
LD (# FEFE), A
LD HL, LPS2
LD (# FEFF), HL
LD DE, 0
EI
HALT
LD HL, LPS3
LD (# FEFF), HL
LPS1 INC DE
LD B, # 0F
DJNZ $
LD A, R
JR LPS1
LPS2 EI
RET
LPS3 POP HL
CALL IM1SET
EI
RET
Almost all done with Spectrum 312
or 320 lines (for example: in the company
312, at the Pentagon - 320). Extend:
; Calculating count cycles in a row ruling
, (Line screen)
; Input: -
; O: A - tpl
CLC_TPL CALL IM2SET
LD A, # C3
LD (# FEFE), A
LD HL, TPL1
LD (# FEFF), HL
LD DE, 0
LD HL, TPL2
EI
HALT
TPL1 LD (# FEFF), HL
LD HL, TPL3
EI
HALT
TPL2 LD (# FEFF), HL
NOP
NOP
NOP
INC HL
EI
INC DE
JP $ -1
TPL3 POP HL
POP HL
EX DE, HL
ADD HL, HL
ADD HL, HL
ADD HL, HL
LD A, N; N = 0 -312, N <> 0 -320 lines
OR A; ^ here puts the number of
LD BC, 156
JR Z, TPL4; 156 (312) and 160 (320)
LD BC, 160
XOR A
TPL4 OR A
SBC HL, BC
INC A
JR NC, TPL4
CALL IM1SET
EI
RET
; Protsedurki install interrupt
IM2SET DI
LD HL, # FE00
LD A, H
LD B, L
IM2SETL LD (HL), A
DEC HL
DJNZ IM2SETL
LD (HL), A
LD A, H
LD I, A
IM 2
RET
IM1SET DI
LD A, # 3F
LD I, A
IM 1
RET
A length INT'a in measures calculated even simpler: tpl *
lps (only noWAIT SPEC) and all!
lps 312 320 tpl tpl
216 67392 69120
220 68640 70400
224 69888 71680
228 71136 72960
232 72384 74240
Actually, for the set up
INT'a know the position with respect to the
frame, but more on that in the next episode ...
Instead, PS: I'm not a coder, so my
speculation about the need autoconfig'a
taken as a theorem, which must still be proved. I hope that my
guesses all the same and become an axiom, and this information
you are somehow going to help.
