[Coco] How does the SCS line / GIME Setting Work?

[Gene Heskett]

On Sunday 20 June 2010, jdaggett at gate.net wrote:
>On 19 Jun 2010 at 23:26, Gene Heskett wrote:
>> James, Kevin has been wrong before, and in fact that was the reason I
>> put on my bulletproof stuff and started on RBF.MN all those years ago,
>> even before the 6309 was discovered.  The level2 version was not doing
>> exactly as the docs said it should, so I fixed that code in 7 or 8
>> different locations before I got to Ed34 and thought I was done.  But
>> that one had a land mine in its FD.SEG(48) handling that I put in :(, so
>> now we've had 2 or 3 more versions since.
>
>Gene
>
>I will say this. It was not my intent to say anyone is right or wrong even
> though it may well have come out like that.

Trying to find ways to make the truth palatable has always been one of my 
downfalls.  Sorry.

> Like I said I defered to his
> measurements until otherwise I can prove myself either right or wrong.
> For now it makes little or no difference to me.

James, I'd never start an argument with you.  Not on a SCS or GIME point.  
You have been much closer to that hardware than I will ever be.  As I have 
the scope which not everyone here has, I could, with a bit of disassembly to 
get to it, confirm the behavior, but I'm not inclined to check until 
something that should work doesn't.

Let me word it this way.  In level 1 versions of os9, the earlier ones had 
code that in one case repeated itself, but generally did it right.  It was 
after he gave us the 'Christmas presents' that a couple of lug nuts came 
loose.  At one point I had it all broken out into individual functions, and 
when I put it back together, the dups got left out, and some of the stuff he 
took out in the Christmas presents got put back in, mostly getstt/setstt 
details but it made it complete and only grew about 20 bytes.

Then when the 6309 arrived, I ran through it again and made the megaread 
time 2 seconds better, from 13 secs to 11 secs, but I also put the backar 
bit in, and that, because it was sensitive to big files and no one setting 
SAS up to a reasonable value, several got hurt, including me BTW.

>The GIME chip does not do everything the same as the SAM chip. There are a
> lot of similarities and there are differemces. I will leave at that.

I think they/you made it only the same in just enough places to be backwards 
compatible with existing SW.  That was done very well.  For my part, faster 
rgb outputs would have been a major blessing, but I suppose there was a 
power budget to meet.

-- 
Cheers, Gene
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
You will always get the greatest recognition for the job you least like.