Re: [stella] tutorial progress

Subject: Re: [stella] tutorial progress
From: KirkIsrael@xxxxxxxxxxxxx
Date: 28 Jun 2002 11:59:36 -0000
> You have 22 + 53 which is 75 cycles, not 76.
> Wouldn't it be 54 cycles?

Good question. The thing is, the numbers don't work out quite right.

228 Clock Counts / 3 = 76 Machine Cycles
But according to the SPG, the 228 is split between 68 and 160.
I thought it was safer to round down for both (hence 22 and 53)
but if rounded normally, it would be 23 and 53, not 22 and 54.
Perhaps some hardcore cycle counter has an opinion on which one
of those is more accurate, or if my conservative rounding down
strategy is the best bet.  

> I also think it might be good to have an alternate diagram that is more 
> WYSIWYG, in other words, superimposed onto a television screen.  The reason 
> being that on a real TV, horizontal blank expresses itself as being 
> horizontal overscan on both the left and right side of the screen.  As far 
> as 2600 coding goes, it looks like all the overscan is on the left which 
> isn't true in a visual sense.

I might make a footnote of it.  I don't want to change the main diagam
(which isn't what I think you're suggesting anyway), since this method
is simpler, what the SPG has, and usually "good enough", but it might be
interesting to note.
> Nobody has ever really made any detailed charts that show where the beam 
> really is and when.
> For instance, in a typical program, the beam is not quite at the right hand 
> side when you issue a WSYNC.  So to visualize this you could show a 
> staggered set of scanlines to indicate the timing for when each scanline's 
> code ends and the TIA's WSYNC function kicks in.

Yeah, it's not the easiest diagram. I have to balance strict accuracy with
usefulness and understandability.
> Also, you should visually chart out safe zones for when it's safe to store 
> and rewrite each of the playfield registers in both reflected and repeated 
> modes.

Maybe...I'm deferring playfield talk for now to a later page, 
with the idea that if most newbies are like me, the idea of sprites and 
missiles is much more interesting than playfields. (And I do admit
I have a lot to learn about playfieds...and player graphics...and 
cycle counting...and pretty much everything. The tutorial is as much
a learning as a teaching project for me, which is why I welcome the
feedback I've been provided here so much.)

Anyway, no updates this weekend 'cause the wife and I are going
away for our 1 year anniversary...but i'll drag along the SPG
(and "assembly in one step") and if I finish my book, I'll redive into

"If I can't be free, at least I'll be cheap"--some guy on the net

Archives (includes files) at
Unsub & more at

Current Thread