NTSC minimg question, test on a real Amiga required !

[FrenchShark]

Hello,

I am new on EAB.
I do have a question that I asked on A.org for an ECS/AGA minimig implementation I am working on.

I am wondering how the long frame line count during interlace mode is done on a real amiga.

In order to save logic, my guess is that on the real stuff, the vertical beam counter is set ($1FF) at the end of a long frame instead of being cleared.

To summarize,
the vertical beam counter values at the end of a PAL short frame would be:
...
310
311 <- last line
0 <- new frame (counter cleared)
1
...
and for a long frame:
310
311
511 <- last line (counter set)
0 <- new frame (counter wraps around)
1
...
Can someone write a short ASM program that:
- Activate interlace mode
- Disable interrupts
- In a loop:
* Read continuously VPOSR + VHPOSR ($DFF004 - $DFF006)
* Wait for V10-V0 to be equal to 311 and LOF = 1
* Wait for V10-V0 to be different from 311 -> exit the loop
- Put the V10-V0 value in a register so we can read the result (312 or something else ?)

PS: this is for a PAL display, if you do it with an NTSC one, replace 311 by 261 !

Regards,

Frederic

[Shoonay]

Hello and welcome to EAB!

Minimig with AGA would be DZEE SHEET!!!
You're the original Minimig constructor?

[DrF]

Individual Computers claimed to be doing a AGA clone a while ago, strangly around the time MiniMig was first mentioned, heard or seen anything more I have not

Someone else claimed to be doing a high end Amiga clone (020+/AGA/Fast Ram and some other stuff) I think its all in the Wikipedia if anyone cares to look

[RedskullDC]

Hi Frederic,

Originally Posted by FrenchShark

Hello,

- Activate interlace mode
- Disable interrupts
- In a loop:
* Read continuously VPOSR + VHPOSR ($DFF004 - $DFF006)
* Wait for V10-V0 to be equal to 311 and LOF = 1
* Wait for V10-V0 to be different from 311 -> exit the loop
- Put the V10-V0 value in a register so we can read the result (312 or something else ?)

Regards,
Frederic

Just out of curiosity, I tried this under WinUAE using PAL LORES interlace but got some fairly meaningless values

Will try it on my real 500 and let u know.

Red

[Toni Wilen]

Originally Posted by RedskullDC

Just out of curiosity, I tried this under WinUAE using PAL LORES interlace but got some fairly meaningless values

Do not use JIT?

[FrenchShark]

Originally Posted by RedskullDC

Hi Frederic,



Just out of curiosity, I tried this under WinUAE using PAL LORES interlace but got some fairly meaningless values

Will try it on my real 500 and let u know.

Red

You have to isolate bits V10 to V0:

move.l #$8007FF00,D1
move.l #$00013700,D2
.Wait1
move.l $DFF004,D0 ;Read VHPOSR and VPOSR
and.l D1,D0 ;Keep LOF and V10-V0 bits
bclr #31,D0 ;Test and clear LOF bit
beq.b .Wait1 ;Short frame, keep waiting
cmp.l D2,D0 ;Equal to line 311 ?
bne.b .Wait1 ;No, keep waiting

bclr #31,D1 ;LOF not needed anymore
.Wait2
move.l $DFF004,D0 ;Read VHPOSR and VPOSR
and.l D1,D0 ;Keep V10-V0 bits
cmp.l D2,D0 ;Still equal to line 311 ?
bne.b .Wait2 ;Yes, keep waiting
lsr.l #8,D0 ;D0 contains the line # after line 311

I should have posted this code snippet earlier.

Regards and thank you.