Subject:
|
Re: Graphics Programming on the Apple II (was: Wow! This guy is good!)
|
Newsgroups:
|
lugnet.off-topic.geek
|
Date:
|
Wed, 7 Jun 2000 04:22:21 GMT
|
Viewed:
|
655 times
|
| |
|
|
I'm not a TV engineer, but I recall the Apple ][ color distribution was made
possible by a trick that used the luminance signal to gate an oscillating chroma
signal. The transition from 1 to 0 (or from 0 to 1) was the key that produced a
color besides black or white. So it took two bits of control to form such a
transition.
You can observe any black/white transition on a TV screen producing a tinge of
color. On more capable video controllers like TI's 9918 with the potential for
neighboring pixels to be any of 16 colors, there was color distortion on a TV,
because some transitions were abrupt and they created new, unpleasant hues. The
hollow box shaped cursor would take on a faint aura of yellow, pink, or blue
depending on what column it was in, due to the varying position of its leading
black single-pixel width bar. This was also apparent with Apple ][ white text on
a black background.
When you designed graphics you had to take these things into account. You would
tweak a pixel here and there to escape the artifacts if you could, or choose
your neighboring colors very wisely.
To a lesser extent these artefacts are still with us on RGB monitors.
The reference to Microsoft's ClearType trick applies only to LCD displays, where
it is possible to make use of the unequal number of color elements in a pixel.
It's a kind of anti-aliasing where you plan on obtaining a slightly distorted
hue, based on the physical properties of the LCD, so you pre-adjust the pixel
yourself to best advantage. When used, this property will allow a smaller point
size to look better on an LCD.
I don't think LCDs have the same kind of abrupt transition artefacts, it's more
a matter of the uneven distribution of color elements.
Yes, there are quite a few of us still around who cut our teeth on "home
computers" (yet like our coffee beans already ground...)
In lugnet.off-topic.geek, John J. Ladasky, Jr. writes:
> In lugnet.general, Matthew Miller writes:
> > Bruce Schlickbernd <corsair@schlickbernd.org> wrote:
> > > more acceptable here, but boy, those 16 non-definable colors drove me nuts
> > > (let's not mention the Apple ][ and the horrors it imposed). Anyway, I'd
> >
> > Ohh, lets. No drawing one-pixel wide white lines -- one lone pixel is either
> > blue or yellow, depending if it's in an even or odd column. (This is due to
> > the techniques Microsoft invented 20 years later for their ClearType
> > technology...)
>
> Mein Gott! Someone who actually remembers graphics programming on the Apple II!
>
> But I remember things a little differently.
>
> A pixel was defined in a rather slippery fashion in Apple II "high-resolution"
> graphics. A pixel was either one bit or two, depending on the values of
> adjacent bits. The four basic colors were blue, violet, green, and orange (no
> yellow!). You could only get white by having two adjacent bits turned on. Two
> adjacent bits turned off = black.
>
> Why was it done this way? It had nothing to do with Microsoft. Several years
> ago I heard an interview with one of the original employees of Apple (not Jobs
> or Wozniak) on a local public radio station. He was discussing the quirky Apple
> II graphics. He said that the reason that the graphics were implemented the way
> they were had to do with saving money -- but not actually very much! Jobs and
> Wozniak set out to make the Apple II for under $1,000, out-of-pocket. The logic
> required to give each pixel two independent bits would have cost them -- get
> ready -- an extra six dollars.
>
> Probably more significant, at two bits per pixel, a page of high-res graphics
> would have occupied twice the memory space. The high-res page, as implemented,
> occupied 16K of RAM. If high-res graphics were implemented in a way that
> required 32K of RAM, two things would have happened: 1) at a time when RAM was
> really, really expensive, no one would ever use high-res graphics; 2) the 6502
> could only address 64K of RAM, and the "page-flipping" approach to displaying
> flicker-free high-res graphics would have required all the addressable memory
> space of the CPU. Oops.
>
> In 1982, I got my Apple II. The only reason that I got 64K of RAM for it was
> that my Dad worked at National Semiconductor, and was able to obtain "cosmetic
> reject" 16K RAM chips from the factory floor. They looked ugly, but they
> worked.
>
> It's all *so* much easier now -- you just have to understand the Windows API!
> :^P
>
> --
> John J. Ladasky Jr., Ph.D.
> Department of Structural Biology
> Stanford University Medical Center
> Stanford, CA 94305
|
|
Message is in Reply To:
34 Messages in This Thread:
  
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
