Date   : Thu, 01 Apr 2004 18:52:47 +0100 (BST)
From   : Pete Turnbull <pete@...>
Subject: Re: Aspect ratio question

On Apr 1, 10:43, Richard_Talbot-Watkins@... wrote:
> Pete Turnbull wrote:
>
> > This is the calculation where you said "each scanline is 52 µs of
> > visible time, of which the Acorn machines use 40 µs"?
> >
> > Firstly: that's not strictly true.  That 52µs is the non-blanked
> > interval, not the visible part, because of overscan.

> No, Thomas is right.  The PAL horizontal period is 64us.  12us of
this
> contains the horizontal flyback (non-visible).  The remaining 52us is
the
> visible time, of which 40us is the actual memory-mapped screen, and
the
> rest is the left and right borders.
>
> There's a bit more on this at:
> http://www.howell1964.freeserve.co.uk/logic/video_clone.htm

I've read the reference you give (I'd seen it before) and I made the
statement above specifically to compare an overscanned PAL image (4:3
aspect ratio) to an underscanned 1024x768 display (also 4:3 aspect
ratio).  I was making the point that it's not a correct comparison
because the PAL image extends to the extreme edges of the raster, and
beyond the visible part of the screen (by 6µs I think I once read,
though that seems too much to me) so the 40µs is not 40/52 of the
displayed image, it's 40/52 of the *raster*.  That 52µs is the raster
width.

> > They are *designed* to be square, so that if you draw a circle with
> > a diameter of, say, 1000 OS units, it comes out visibly as wide as
> > it is high [...]

> I would say they are designed to be roughly square, but probably do
not
> display quite as such on some screens.  For example, on my TV set, a
24x24
> pixel block (a size I became very used to plotting from my games
writing
> days) looks slightly taller than it is wide.  I'd say calculations
are all
> very well, but there's no substitute for getting an appropriate image
on
> screen and then measuring its dimensions with a ruler!

Which I did :-)

You have to be a bit careful about that, though, because quite a small
error in either the setup of the monitor, or the measurements (which
are quite hard to make since the screen is not usually flat, and
there's a thick piece of glass betwen ruler and image) makes quite a
noticable difference to the ratio you end up with.

-- 
Pete                                           Peter Turnbull
                                               Network Manager
                                               University of York