In article <RbKuvxW.tymo...@delphi.com>,
TYMOTHY DJ <tymo...@delphi.com> wrote:
>I am a Amiga games collector and am looking for the following
> 256 color version of King's Quest VI
This one doesn't exist, but it has an interesting story behind it.
In the beginning, Sierra did all of their own Amiga ports in house.
Ironically, despite what everyone SEEMS to say, they actually did a
pretty good job with them.
The early ports of their 16-color-CGA games such as Space Quest I and II,
King's Quest I and II, Leisure Suit Larry I, etc. Were all spot on. Of
course, 16-color-CGA and 3 voices of square wave audio were not that
impressive on the Amiga as even the earliest Amiga games were better than
THAT (see note about Mindwalker below).
The reason they were so "awful" is that these games were originally
developed for MS/DOS compatibles (I may be wrong, see below).
16-color-CGA was chosen because that was the best MS/DOS compatible
display system that mortals could afford back then, and sound cards were
more or less nonexistent (see note about 16-color CGA, 3-voice square
wave, the Tandy and the PCJr below).
Later, as better display devices (EGA, then VGA) and better audio systems
(AdLib, Soundblaster, MIDI boxes) for MS/DOS compatibles became
affordable, Sierra coded their games to take advantage of them.
Contrary to what many Amiga users said at the time, Sierra actually took
a great deal of interest in the Amiga, and their programmers worked hard
to try to make good Amiga versions of their products. They consistently
upgraded the Amiga interpreter to support better display modes, and they
worked to create better audio drivers so that we could actually enjoy the
On the graphics front, EGA was no problem, we got it. VGA, on the other
hand, was difficult to convert as we had no 256-color displaymode back
then. So they did what they could, and dithered the graphics down to 32
colors as best as they could within the limits of their system. They
tried real hard to make it look good. Unfortunately, as I will explain
later, Sierra's Amiga programmers were by-the-book American code grinders,
who could produce well-organized and smartly-designed code, but had
difficulty grasping the fiddly tricks that were common on the Amiga. It
took them some time to figure out how to implement EHB support, a concept
that seems simple to us, and by the time they had finally got it working
upper management was moving to discontinue Amiga support. Only one
64-color port (Robin Hood) made it to release. In fact, shortly before
the end, they even attempted to bring the VGA graphics over more or less
intact by using HAM mode, but the necessary trickery to get something
like that to work without horrible fringing while still maintaining a
sane frame rate was firmly in the domain of hackers and beyond the skills
of CS-style code-theory American programmers
On the audio front, they also worked hard to give us the best that they
could. They gave us a choice of THREE audio drivers! The old 3-voice
square driver for people with slow disk access or small amounts of RAM, a
4-voice sample-based driver for most folk, *AND* a MIDI driver! To the
best of my knowledge, Sierra games are the ONLY games on the Amiga ever
to support MIDI. Admittedly, the driver was specifically for the MT-32,
so it can sound a little weird on other devices, but back then there
wasn't such a thing as General MIDI, so they tried to aim for a popular
device that sounded good. Back then people complained that the internal
Amiga sound was awful, but in reality it was pretty darn good considering
the circumstances (I'll elaborate below), and with a good MIDI device the
music was absolutely fantastic. Listening to King's Quest V on my
friend's A3000 w/Sound Canvas was exhilarating, despite the occasional
instrument mixups (usually involving sound effects, a chest closing
became a deep tuba sound, for instance).
Yet despite all of Sierra's efforts, most of the Amiga community scorned
their ports. To be certain, they had a point. Compared to European games
developed specifically for the Amiga, Sierra's offerings looked a bit
shoddy in comparison. The 32-color graphics didn't look as nice. By all
appearances they weren't optimizing their palette usage. The internal sound
sounded flat and disjointed compared to the incredible scores of the
European games. Furthermore, and this was the most common complaint, they
ran slowly on 7.14Mhz 68000 machines for apparently no good reason,
crawling somewhat even when only one object was crossing the screen.
Compared to games like Shadow of the Beast, with multiple parallax layers
and objects zooming across the screen at 50 frames per second on even the
slowest Amiga, Sierra ports looked like shoddy programming. In fact, the
programming was quite good. In order to understand why Sierra games
performed on such a relatively mediocre level, you have to understand how
the games themselves work.
Sierra games were not written in traditional computer languages, rather
they were written for an abstract graphic adventure-oriented game engine
using an interpreted language. This was done for several reasons.
Firstly, because the language was specifically designed to facilitate the
creation of graphic adventures, the actual coding of the game would be
less of a burden, and more time and effort could be spent into the actual
design of the adventure. This is the same reason why some of us prefer to
code mostly in C instead of assembly.
Second, because the language worked on a more abstract level, the bonds
of hardware restrictions were loosened, giving the creativity of the
author precedence over a given hardware configuration. While hard limits
like display system limitations still remained, the author no longer had
to constantly worry about how big his objects would have to be, etc.
Also, one only had to create one score of music, instead of one per
possible output device; the audio drivers would see that the user's
hardware made the appropriate sounds.
Third, not only could the same adventure code be used for machines with
different peripherals, it could also be used without a great deal of
modification on other platforms. In fact, outside of the color reduction
of VGA screens and the addition of samples for the Amiga's purely
sample-based audio system, there were almost no changes in the actual
adventure code that was brought over. Were it not for this ease of
conversion, the Amiga would probably not have received the plethora of
Sierra games that it did.
Now, considering how generalized Sierra's interpreters were, the amount
of work that they had to do to even move the character around the screen
was immense. Everything happened on a higher level of abstraction. So in
reality, Sierra's coding wasn't utterly slow and awful. Rather, it was
well thought out, well organized, and flexible. It ran at a quite
reasonable speed for the features that it provided. Despite appearances,
the code really had been optimized, and performed quite well next to its
MS/DOS-compatible counterpart on an equivalent processor.
Certainly, the system had its drawbacks. It meant that every display
couldn't have a 100% optimized palette, because that wouldn't leave any
flexibility for more objects that might appear onscreen. It meant that
playback on systems not designed for generalized score-based audio
information would not sound as good as MOD-style hardware-oriented music
routines. But despite these drawbacks it also meant that good games could
be made in greater numbers and for more systems than would otherwise be
Sierra's Amiga programmers weren't the fanciest in the world. They
weren't any good at assembly-level optimization. They didn't know how to
tweak 110% out of the Amiga's custom chipset. They just weren't the type.
They were, however, very good at what they did know how to do. Their
software used intuition, they knew how to properly access files, they
didn't shut down multitasking, their software almost never crashed. In other
words, their software followed the rules, and still played reasonably
well. They did their best to support the Amiga's internal audio, and went
even further and supported alternative audio devices when noone else
would. They did their best to support the Amiga's graphics system,
yielding good displays without sacrificing other parts of the game. They
mastered and supported EHB, something which most games should
have but didn't. They even worked towards using HAM (and may well have
succeeded if they'd been given the opportunity to study it more). They
not only supported but recommended hard drives in an era when most Amiga
games came on bizarre copy protected disks that left you screwed if they
Futhermore, because they followed the rules, their software continues to
work without the need to cripple your system. Every time I upgrade, I'm
pleased to see that the software just performs better and better. I can
chat on IRC with a fellow adventurer while I try to determine how to
rescue the Two Guys From Andromeda from their prison of lime Jell-o, or
gloat at yet another hop in the sack in Leisure Suit Larry. It likes my
copyback cache, it doesn't puke with a nonzero VBR, it doesn't grind my
floppy drives into powder, it drags, it multitasks, it mode promotes.
All that's left is to try running it under CyberGraphX or on a Draco.
The reason I wrote this admittedly long and impromptu history of software
from a company that has abandoned us in response to a simple question
that needed nothing more than a simple answer is difficult to explain.
Maybe it's because I was one of the foolish people who rashly jumped to
criticize Sierra's Amiga productions, eventually driving them out of the
market. Maybe this is just my way of apologizing to Sierra and the Amiga
read more »