KQ6AGI - Complete after 18 years of development!

[MusicallyInspired]

Brandon Kouri has finished his KQ6AGI project and it's a beast of an AGI game! You can download it here:
https://kq6agi.com/

He also put let's play videos up on his YT channel showing it off:

[pabuslem]

I just saw it today. This is beautiful.

I wish more AGI games were made and released nowadays. Thanks for the heads-up!

Cheers!

[lskovlun]

AGI: The little engine that could. Yeah, this is amazing! Kudos!

[pmkelly]

Damn, that's some serious dedication! This is probably the largest game ever done in AGI. Nice work!

[doomlazer]

The readme.txt says:

Code: [Select]

It is highly recommended that this game be run using
the interpreter that is supplied with it. KQ6AGI is very
resource demanding, and has been tailor-made to operate using
the supplied interpreter.
It would be interesting to know what's been modified in the interpreter.

[lance.ewing]

This is truly amazing! It's great to see so many AGI labours of love being released over the past year or so. I was aware of the project from way back, but didn't realise that it was close to release. I've just watched through the whole video and it looks great. I'll give the game a go in AGILE this evening to see how it works there. It will be interesting to see. AGILE (both the C# and web versions) wouldn't have the resource limitations that the original interpreter had, so hopefully the customisations to the interpreter won't be an issue.

Watching the video has made me realise that I played the original KQ6 so long ago now that this is like a new story to me. I have images of the original graphics in my mind from back then, but the story line feels almost new. I really should play this through, and also KQ5 at some point. It's been a while. I've been so focused on AGI over the years that my memories of SCI games are getting vague

[AGKorson]

The readme.txt says:

Code: [Select]

It is highly recommended that this game be run using
the interpreter that is supplied with it. KQ6AGI is very
resource demanding, and has been tailor-made to operate using
the supplied interpreter.
It would be interesting to know what's been modified in the interpreter.

There are no modifications. It's v2.917 (from KQ1). He just renamed it from agi.exe to kq6agi.exe.

It is big, but not the biggest. It's got 1,485,268 bytes of resources in two VOL files, uncompressed. Goldrush had 1,652,765 bytes of resources (compressed to 1,185,381 bytes as a V3 game).

[doomlazer]

There are no modifications. It's v2.917 (from KQ1). He just renamed it from agi.exe to kq6agi.exe.

It is big, but not the biggest. It's got 1,485,268 bytes of resources in two VOL files, uncompressed. Goldrush had 1,652,765 bytes of resources (compressed to 1,185,381 bytes as a V3 game).

Oh, I guess he just meant it was coded against 2.917; my bad. Strange they didn't use the most recent interpreter version.

[AGKorson]

Strange they didn't use the most recent interpreter version.

The amount of free memory (heap space) is actually slightly smaller in v2.936 and all thee v3 interpreters (except v3.002.149, which gained a few bytes by dropping string storage from 24 to 12). So unless you really need the few commands added in v2.936 and/or v3, you actually get a few more bytes of available heap space by using 2.917. In AGI, 'a few bytes' can be a LOT of memory.

v3 interpreters are less desirable because you have to modify the executable file with your game's id. If you don't mind that hassle, then 3.002.149 is the best choice for maximum heap size, if you don't mind the smaller string space.

[lskovlun]

v3 interpreters are less desirable because you have to modify the executable file with your game's id.

That's a very small problem given that we now have Sierra's original tool (setstr.exe) for doing that. It only has to be done once.