[hatari-devel] DSP Instruction cycle count

[Eero Tamminen]

Hi,

On Tuesday 07 July 2009, Laurent Sallafranque wrote:
> I've just uploaded the dsp instruction cycle count.
> Now, 68030 and DSP are accurately synchronized.
>
> I've done tests with many programs, demos, games.
> It seems that nothing regressed.
>
> I've noticed that BadMood seems to run much faster.

Which version?  BadMood 2.14 seems to work fine, but
BadMood 3.x graphics look broken.


> Dsp sound seems to be clearer (but still not perfect).

Is it normal that when programs (demos) exit, there's still some
background sound?


> Do you notice any speed improvement ?

I think some more complex things are faster, and simpler ones are slower.
I think both the addition of breakpoints and cycle counting to the DSP
mainloop will slow down things a bit for things that tax DSP only lightly.


> emulation improvement ? 

"Atari UK demos" and "Les Dinosaures" demo version work now.


> Tell me if you detect regression with some programs that were running
> before.

"Built-in Obsolescence" earlier run a bit, now it stops at startup.  Hatari
outputs a DSP stack overflow warning.

Similar issue with with the "Are you sitting comfortably?" demo from RG.

And when selecting PT (protracker?) mode in FlexTracker.


> I'll try yo improve a bit more the tuning of the cycle counting, but I
> think I'm close to the reality now.


	- Eero

Btw.  Have you tried the conditional breakpoints yet & do they work OK for 
DSP? :-)