wizzmilk / mist-board

Automatically exported from code.google.com/p/mist-board
0 stars 0 forks source link

Regressions in minimig AGA core from january 2015 (r763) #35

Closed GoogleCodeExporter closed 8 years ago

GoogleCodeExporter commented 8 years ago
It looks like there are efficiency problems with this core (aga core from 
november 2014 works fine). Problems can be observed even in OCS and ECS 
chipsets, so they are not connected directly with AGA.

What steps will reproduce the problem?

Run Kefrens demo (desert dream - http://ada.untergrund.net/?p=demo&i=60)
This is old demo for A500 (OCS chipset - nothing extraordinary).
You have to wait quite long, because at the beginning everything works fine. 
First problems you can notice on green-white-orange scroll - it looks like some 
frames are skipped, then some effects works more or less fine (sometimes with 
frame skips) and then demo reaches "sine dots part with 12000 dots" and it 
cannot go any further - music plays and thats it.

I've also notices simillar problems with other demos (I don't remember names). 
Mainly you can notice frame skips - it looks like simulation is slower than 
original amiga (A500) in some cases !!!

I use MIST with newest firmware, so it shouldn't be connected with firmware 
version.

Previous version of AGA-enabled core still works fine, so it looks like 
"enhancements by Alastair M. Robinson" added this.

Original issue reported on code.google.com by acid.ma...@gmail.com on 23 Apr 2015 at 11:27

GoogleCodeExporter commented 8 years ago
Are there still any regressions in the new minimig-mist-1_0 release?

Original comment by rok.kra...@gmail.com on 30 Jun 2015 at 12:54

GoogleCodeExporter commented 8 years ago
New core version available which fixes this, and in any case I will not fix 
older versions.

Original comment by rok.kra...@gmail.com on 10 Jul 2015 at 10:41