Official AGI Documentation

[Kawa]

Again, I am reminded of the whole SCUMM/SPUTM stuff.

[lance.ewing]

Would development of the port have anything to do with it? Perhaps from the tools for Apple?

I wondered whether it was something to do with the relationship with Walt Disney. Perhaps they used a different name for the system in relation to the games they built that were associated with Disney, such as Black Cauldron and Donald Duck's playground. I also wondered whether maybe the AGDS name was a wider external name for the system, for use with any external partner, i.e. not just for Disney, and AGI was perhaps the internal name. - Just some thoughts I had. - I guess Al Lowe would remember. He was the Disney dev guy, right? Might send him an email.

Omer, without necessarily revealing your sources, can you tell us if there is any suggestion that the AGDS document was associated with a Disney game or Disney games?

Edit:

Email just sent to Al. We'll see what he remembers. I asked him about GAL as well.

[lance.ewing]

Reference to AGDS by a former Sierra On-Line employee:

http://www.sierranet.net/~pgr/resume.htm

...where Greg Rowland (a developer of the original Kings Quest game) says:

Helped design and produce graphics and animation utilities, game logic compiler, later to become known as Adventure game development system AGDS.

So there we go. Both the official documentation and a former employee refer to the graphics and animations utilities, and game logic compiler, as AGDS.

[lance.ewing]

The more I look, the more I see it. It would appear that it's not only in Black Cauldron the Apple II version, but also the PC version of BC. Greg Rowland mentioned King's Quest 1, as we saw in his resume above, but I can't find it mentioned in the game. LSL1, PQ1, and KQ3 all refer to Adventure Game Development System though. I just hadn't noticed it was so prominent before:

http://policequest.wikia.com/wiki/PQ1EGA_transcript

I've been searching the net for "Adventure Game Development System" and turned up the above page. It's a dump of messages from PQ1. I then looked a bit closer at a few other games and can see it in LSL1 and KQ3. Still looking...    Also in Donald Duck's Playground.

http://sarien.net/policequest

http://sarien.net/kingsquest3

I guess whenever I saw that in the past, I probably thought of it more as a description of the system rather than its name. I always thought it was called AGI, and certainly there are games that say "Adventure Game Interpreter" in the credits and opening screens, and there are files names that use AGI and packaging that mentioned AGI, and books, etc., etc. But AGDS also seems to have been a name that they used for it.

[NewRisingSun]

I would say that any 2-D game development system would need to include:

• Something to edit or convert graphics, including backgrounds and sprites
• Something to edit or convert music and sound effects
• Something to edit or convert/assemble/compile game logic
• Something for the end user to play the finished game

In the case of the "Adventure Game Development System", there's the "Picture Editor" and "Object Editor" for 1., the "Sound Editor" for 2., "CG - the game compiler" for 3., and "AGI", the "Adventure Game Interpreter" for 4. Therefore, the whole thing is the AGDS, and AGI is Part 4 of it. 

[lance.ewing]

Yeah, you're probably right, in that that is probably how it started out. AGDS was probably the whole development package of tools, and AGI the interpreter.

More links to easily verifiable screens that mention "Adventure Game Development System":

https://archive.org/details/msdos_donald_ducks_playground_1984_sierra

https://archive.org/details/msdos_Leisure_Suit_Larry_1_-_Land_of_the_Lounge_Lizards_1987

https://archive.org/details/msdos_Black_Cauldron_The_1986

(Those runnable DOS games on archive.org are quite cool really).

[NewRisingSun]

(Those runnable DOS games on archive.org are quite cool really).

I find them to be nightmares, as the underlying emulators are badly configured, the game versions mislabeled, and the aspect of openly offering thousands of games for free stretching the meaning of "preservation" to a ludicrous level. But maybe that's just me. 

[lance.ewing]

I don't disagree with any of that. I was thinking more of the underlying technology of running DOS games in a browser using Javascript, at seemingly full speed.

[Kawa]

I don't like it.

Sure, it's impressive, but that's just not what webpages are for, know what I'm saying? Webpages are for downloading a DOSBox from and the games to run on it with a nice description on how to run it 'n shit, not for hosting a DOSBox in-place holy shit no wonder people^[who?] say web browsers are operating systems onto their own. It's shit like JS having grown faster than Tetsuo on crack that makes all these web browsers basically the same two or three cores in different coats of user-unfriendly paint.

</rant>

Still, gotta love availability and hidden gems.

[lance.ewing]

I've written a few emulators, and a couple of HTML5 games, ....so I love machine emulation, and I love HTML5, and DOS was the best thing Microsoft has done in a while. So running LSL3 in the browser at full screen without any issues definitely ticks all my boxes. I like it when something just works in the browser. I'm a Java developer with 18 years experience and I spent an hour earlier today tearing my hair out trying to work out how to get my Firefox installation to run a Java applet so that I could run an Apple II emulator in the browser. HTML5 is the way of the future. Java is so broken and dead in browsers these days, its embarrassing.   

[lskovlun]

Again, I am reminded of the whole SCUMM/SPUTM stuff.

And SCI, what with the language being called Script, the compiler being called SC, the interpreter being called SCI(V) etc.

[OmerMor]

On a related note, here's what Robert (Bob) Heitman wrote to me once:

Whenever I hear of people writing a modern version of SCI I ask myself a fundamental question and that question is: Just exactly what do they think SCI was?
 
SCI was a marketing invented "name" for a generation of Sierra's development environment that consisted of a proprietary scripting language (.SC files), several proprietary asset file formats (view, pic, sound, font), and a executable file that functioned both as the interpreter of the script files but also provided all of the graphics, input, sound, and resource management functions required to "play" the game.
 
The idea of someone needing to do anything to "reverse engineer" the engine assumes that there is a complicated master data structure underneath all of it. There is not. A SCI game is written completely in a procedural manner. There is no SCUMM-like data being interpreted. At best there is core set of low level functions to handle the heavy listing but everything else was glued together in the scripting language itself.
 
Given the existence of all sorts of DOS emulators, and the fact that you can still run Sierra games today (as long as you can accommodate the faster clock speeds), it just doesn't seem like that much of a challenge.
 
So, what am I missing?
 
It seems like the real challenge of the day is to recreate ANY development environment that empowers fans of the old fashioned graphics adventure games to create new games in a seamless manner.
 
For the record it took Jeff and myself approximately 8 months to create the core of the development environment and get to first scripting of the 16 color version. He wrote the language and the language interpreter and I wrote the animation editor (VE), the picture editor (PE), the resource manager, the initial text parser, and all I/O routines. I only mention this as it seems like a perfect project for a slightly larger group to have knocked out in a couple months.

Omer, without necessarily revealing your sources, can you tell us if there is any suggestion that the AGDS document was associated with a Disney game or Disney games?

I got the AGDS document with some files related to Donald Duck's Playground.

As for the AGDS/AGI question - I agree with NewRisingSun's assertion of System vs. Interpreter.

[lskovlun]

On a related note, here's what Robert (Bob) Heitman wrote to me once:

For the record it took Jeff and myself approximately 8 months to create the core of the development environment and get to first scripting of the 16 color version. He wrote the language and the language interpreter and I wrote the animation editor (VE), the picture editor (PE), the resource manager, the initial text parser, and all I/O routines.

And Pablo Ghenis wrote the parser, and Stuart Goldstein wrote the music system.

[lance.ewing]

Email just sent to Al. We'll see what he remembers. I asked him about GAL as well.

I received a reply back from Al Lowe earlier today. He says he has never heard of GAL, and had never heard of AGDS either, that is until he heard of the Russian AGDS tools much later on, in the 90s. He says that right back when he was working on Black Cauldron in 1983, they always called it AGI. I've responded by pointing him at Greg Rowland's resume, and the text on all those AGI game title screens. Waiting to hear back what he says. We obviously know from that AGDS.DOC file that there was at least someone at Sierra that called it AGDS. Perhaps the author of that particular document? We should try to track down Greg Rowland if we can.

[lskovlun]

We should try to track down Greg Rowland if we can.

He was the graphic artist according to the title screen, not a programmer. The programmers were (again, according to the title screen) Charles Tingley, Ken McNeill and Chris Iden. I've only ever heard of Tingley in the context of this game, oddly. Chris Iden does not appear in the credits for the PCjr version, so one might assume he was the one who ported the game to the ordinary PC?