🚀 go-pugleaf

In article <BtLG5o.Fq3@watserv1.uwaterloo.ca> bee@watarts.uwaterloo.ca (Bill E.  
Eickmeier) writes:
> In article <1992Aug25.201428.14395@trlabs.CA> macg@trlabs.CA (Mike MacGregor)  
writes:
> >
> >I'd like to get X running on my Turbo (so that rules out McGill X).
> 
> Could someone in the know explain why McGill X doesn't work on a Turbo
> but does on a NeXTstation - what's the fundamental difference here?
> 
> 	Bill

Well, my guess would be that the McGill X doesn't take into account that the  
frame buffer width changed on the turbos.  If you look at the "video.h" file in  
NS 3.0, you will see:

#define	 VIDEO_MW	((dma_chip == 313)? 1152 : 1120)	/* Actual pixels per  
scanline */

So it looks like that a dma chip that talks to the frame buffer was changed.   
This means that 25 mHz machines have a frame buffer width of 1152 while the  
turbos have a frame buffer width equal to the visible width, 1120.

I assume (I've never seen it) that when McGill X runs on the Turbos, the scan  
line accellerator (frame width) is hard coded to 1152 and everything comes out  
skewed.

(BTW, don't go looking for dma_chip to be defined anywhere.  It's an internal  
kernel variable.)

Chris Chauvin
Product Manager - co-Xist
Pencom Software
chauvin@pencom.com
(512) 343-1111