using idle() with grayscales - Page 1

I would like to use idle in a grayscale game
there is a loop in my code which waits for the AI1 to trigger in order to synchronize every frame (i think this loop takes 50% of each frame's time)
I tried this :

while(AI1_Counter < 5)
idle();
AI1_counter = 0;


But as it's writtem in tigcc help, it interfers with the grayscales until I press a key
After that, it works fine

Is there a way to prevent idle from interfering with graycales ?
or do you have any idea to reduce the power consumption in this loop ?

you should use the pokeIO function. I think Kevin has already gave the solution in the forum.

In some of my source files, I have that : you should verify it is correct :
pokeIO(0x600005,0b10111)

yes it is that:

//set the calc in low-power-mode (wakes cpu only up for grayscale, keypress, amstimer, clock)
pokeIO(0x600005, 0b10111);

Here are some precisions (taken from j89hw.txt) :

$600005 -W Turn off OSC1 (and the CPU), wake on int level 6 (ON key) and ...
:7-5 -
:4 ... int level 5 (programmable timer)
:3 ... int level 4 (link)
:2 ... int level 3 (OSC2/2^19)
:1 ... int level 2 (keyboard)
:0 ... int level 1 (OSC2/2^11)
Note: Make SURE int level 6 is acknowledged before writing $00 to this port...

Grayscales use the AI1, and AMS >= 2.07 uses the AI3 for the clock, and your program uses the AI5, so you should write 0b00010101. The code posted below waked up on AI2 too. I don't know if it's important or not...

Allright, I'll try that

but does the calculator fall asleep just after this instruction, or is there a small delay (as in _rowread) ?

I don't know, try to put some code just after the move.b #%10101,$60005 (something like move.w #$FFFF,$4C00).

but it will be executed when the AI1 triggers
I will try something else

Yes try #%10 (it should wait until a keypress).

could post the result of the test please, I'm curious to know the anwser

according to my tests, move.b #%0, $600005 stops instantly the calculator

Yes, only pressing "ON" will wake it up then. This is exactly as reported in hw89.txt, the bane of 68k programming. ./4 pretty much explains everything you need to know about this low-level "idling".

~d

If you are under Kernel Mode, kernel_idle(); does the job too.