RE: [hatari-devel] No sound in the Pacemaker Demo

["Konador, Cyprian" <cyprian.konador@xxxxxx>]

> -----Original Message-----
> From: Laurent Sallafranque 
> Sent: Friday, June 06, 2014 12:04 AM
>
> > I see you noted this in the source code for future reference.
> > (I would expect the mask to be shifted 16 bits, not rotated.)
>
>
> According to the doc I had when I wrote this code, the mask was rotating 
> and the data was shifted.
> If I remenber well, the mask must be rotating, because one can test when 
> the transfer to the LMC is over by testing the mask value (ie when the 
> mask value returns to the original value, the sending is finished).
> But my doc may not be complete/accurate.
>

That’s true, Microwire Mask register is rotated and Data register is shifted. Sometimes ago I check that with Blitter and CPU.
I posted test application MCRW2 and a dump from it in thread "RE: LMC1992/Microwire test WAS: [hatari-devel] No sound in the Pacemaker Demo"

And in this case Hatari behaves like real STE

Here is a dump from STE Mask register, done every 8 cycles for value $c1ff:
$c1ff	$83ff	$0ffe	$1ffc	$7ff0	$ffe0	$ff83	$ff07	$fc1f	$f83f	$e0ff	$c1ff

Data register for value $80e2:
$80e2	$0388	$0710	$1c40	$3880	$e200	$c400	$1000	$2000	$8000	$0000