128k memory paging, how can we do that? :S

[nitrofurano]

hi everyone!
based on the information i found at <!-- m --><a class="postlink" href="http://www.worldofspectrum.org/faq/reference/128kreference.htm">http://www.worldofspectrum.org/faq/refe ... erence.htm</a><!-- m -->

i tried this one:

Code:

cls
page1=0
a=(((peek($5B5C))band $F8)bor (page1 band 7))
poke $5B5C,a:out $7FFD,a
poke $C123,80
page1=1
a=(((peek($5B5C))band $F8)bor (page1 band 7))
poke $5B5C,a:out $7FFD,a
poke $C123,81
page1=0
a=(((peek($5B5C))band $F8)bor (page1 band 7))
poke $5B5C,a:out $7FFD,a
print peek($C123)
page1=1
a=(((peek($5B5C))band $F8)bor (page1 band 7))
poke $5B5C,a:out $7FFD,a
print peek($C123)
pause 0

and i can't figure out what is going wrong there...- any clue? :S

[nitrofurano]

now it seems weird why

Code:

cls
poke 23388,16:out 32765,16:poke 50000,16
poke 23388,17:out 32765,17:poke 50000,17
poke 23388,19:out 32765,19:poke 50000,19
poke 23388,20:out 32765,20:poke 50000,20
poke 23388,22:out 32765,22:poke 50000,22
poke 23388,23:out 32765,23:poke 50000,23
poke 23388,16:out 32765,16:print peek (50000)
poke 23388,17:out 32765,17:print peek (50000)
poke 23388,19:out 32765,19:print peek (50000)
poke 23388,20:out 32765,20:print peek (50000)
poke 23388,22:out 32765,22:print peek (50000)
poke 23388,23:out 32765,23:print peek (50000)
pause 0

works, and

Code:

sub setmemorypage(ta as ubyte)
  dim tb as ubyte
  tb=(ta band 7)bor 8
  poke 23388,tb:out 32765,tb
  end sub
cls
setmemorypage(0):poke 50000,80
setmemorypage(1):poke 50000,81
setmemorypage(3):poke 50000,83
setmemorypage(4):poke 50000,84
setmemorypage(6):poke 50000,86
setmemorypage(7):poke 50000,87
setmemorypage(0):print peek(50000)
setmemorypage(1):print peek(50000)
setmemorypage(3):print peek(50000)
setmemorypage(4):print peek(50000)
setmemorypage(6):print peek(50000)
setmemorypage(7):print peek(50000)
pause 0

isn't... :S

[nitrofurano]

fixed (i think...)

Code:

sub setmemorypage(ta as uinteger)
  tb=(ta band 7)bor 16
  poke 23388,tb:out 32765,tb
  end sub
cls
setmemorypage(0):poke 50000,80
setmemorypage(1):poke 50000,81
setmemorypage(3):poke 50000,83
setmemorypage(4):poke 50000,84
setmemorypage(6):poke 50000,86
setmemorypage(7):poke 50000,87
setmemorypage(0):print peek(50000)
setmemorypage(1):print peek(50000)
setmemorypage(3):print peek(50000)
setmemorypage(4):print peek(50000)
setmemorypage(6):print peek(50000)
setmemorypage(7):print peek(50000)
pause 0

[boriel]

fixed (i think...)

Code:

sub setmemorypage(ta as uinteger)
  tb=(ta band 7)bor 16
  poke 23388,tb:out 32765,tb
  end sub
cls
setmemorypage(0):poke 50000,80
setmemorypage(1):poke 50000,81
setmemorypage(3):poke 50000,83
setmemorypage(4):poke 50000,84
setmemorypage(6):poke 50000,86
setmemorypage(7):poke 50000,87
setmemorypage(0):print peek(50000)
setmemorypage(1):print peek(50000)
setmemorypage(3):print peek(50000)
setmemorypage(4):print peek(50000)
setmemorypage(6):print peek(50000)
setmemorypage(7):print peek(50000)
pause 0

For what I read in the speccy.org forum, you need to disable interruptions (DI) before bankswitching. If an interrupt enters during bankswitching, the computer might hang, or have unpredictable results. :?: :?:

[LCD]

Why not use inline ASM? Not tested:

Code:

sub setmemorypage(ta as ubyte)
   ASM
      di
      ld a,(ix+5)
      and 7
      or 8
      ld bc,32765
      out (c),a
      ld (23388),a
      ei
   END ASM
end sub

[boriel]

Why not use inline ASM? Not tested:

Code:

sub setmemorypage(ta as ubyte)
   ASM
      di
      ld a,(ix+5)
      and 7
      or 8
      ld bc,32765
      out (c),a
      ld (23388),a
      ei
   END ASM
end sub

This is a perfect case for FASTCALL

Code:

sub FASTCALL setmemorypage(ta as ubyte)
   ASM
      di
      ; ld a,(ix+5) ; Not needed already in A
      and 7
      or 8
      ld bc,32765
      out (c),a
      ld (23388),a
      ei
   END ASM
end sub

[LCD]

Why not use inline ASM? Not tested:

Code:

sub setmemorypage(ta as ubyte)
   ASM
      di
      ld a,(ix+5)
      and 7
      or 8
      ld bc,32765
      out (c),a
      ld (23388),a
      ei
   END ASM
end sub

This is a perfect case for FASTCALL

Code:

sub FASTCALL setmemorypage(ta as ubyte)
   ASM
      di
      ; ld a,(ix+5) ; Not needed already in A
      and 7
      or 8
      ld bc,32765
      out (c),a
      ld (23388),a
      ei
   END ASM
end sub

Ooops, you are right, I overseen it. Thanks! Generally I prefer 128K USR0 mode because interrupts are not harmful in this mode.

[boriel]

Ooops, you are right, I overseen it. Thanks! Generally I prefer 128K USR0 mode because interrupts are not harmful in this mode.

128K USR0 :?: What is this mode?
(I'm still learning this maybe I can do something for it in the compiler).

[LCD]

Ooops, you are right, I overseen it. Thanks! Generally I prefer 128K USR0 mode because interrupts are not harmful in this mode.

128K USR0 :?: What is this mode?
(I'm still learning this maybe I can do something for it in the compiler).

USR0 mode is if you enter in 128 BASIC the command USR0. This resets the machine into 48K ROM but with enabled 128K features like AY or bank switching+second screen.
You cannot use +3 DOS or RAMDISC commands, but you also don't need to POKE 23388 to switch banks, OUT 32765 is enough. DivIDE also works in this mode on Spectrum 128K machines.

[boriel]

Ooops, you are right, I overseen it. Thanks! Generally I prefer 128K USR0 mode because interrupts are not harmful in this mode.

128K USR0 :?: What is this mode?
(I'm still learning this maybe I can do something for it in the compiler).

USR0 mode is if you enter in 128 BASIC the command USR0. This resets the machine into 48K ROM but with enabled 128K features like AY or bank switching+second screen.
You cannot use +3 DOS or RAMDISC commands, but you also don't need to POKE 23388 to switch banks, OUT 32765 is enough. DivIDE also works in this mode on Spectrum 128K machines.

Interesting!!!!
Could it be possible to start this mode without resetting the RAM? I mean: Executing a program in 128K mode in ZXBASIC and the program manages in some way to enter that mode an continue execution? Seems a good start.
Another question: I guess +2 ROM won't be available. Will it be?

[LCD]

Interesting!!!!
Could it be possible to start this mode without resetting the RAM? I mean: Executing a program in 128K mode in ZXBASIC and the program manages in some way to enter that mode an continue execution? Seems a good start.

Maybe switching to the 48K ROM and jumping in the 48K Editor routine will do the job... I think, the topic was discussed on WOS before.

Another q#uestion: I guess +2 ROM won't be available. Will it be?

It won't be available unless you page it in.

[RandomiserUsr]

So it's 2021 and I wonder if 
1) The below worked ?

2) If there is any further news on opening up ZXB to 128K ?

Thanks 
Ken