Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


Messages - EricOakford

Pages: [1] 2 3 ... 17
1
Okay, now it's working. I've updated my AGI3 remasters of Residence 44 Quest, Tonight the Shrieking Corpses Bleed, and all the Tex McPhilip games. I should play-test them again to make sure nothing broke.

2
I've installed it. Unfortunately, trying to run it gives the error "This app can't run on your PC", the same error that appears when trying to run a 16-bit program in 64-bit Windows. I think something's wrong with the installer or the executable.

3
Everything-Else / Re: What's up peeps!
« on: January 28, 2026, 04:01:02 PM »
What I'd been doing is AGI3 remasters of certain fan games. This had been going on for a while, but now I'm ready to share updates on that.
At this point, the Tex McPhilip games are all set and optimized. with various quality of life improvements. I tested them to completion with no known issues. What's cool is that I merged the first two games into one. The third game was too big, and different enough to remain its own game, but I did quite a bit of cleanup and optimization for that game too. Same for Tonight the Shrieking Corpses Bleed and Residence 44 Quest.

In all of these cases, the games have been re-coded to use the local/dynamic flags/variables system that's common in the Sierra games. I also did proto-SCI room scripts that were common in KQ3 and PQ1. That freed up a lot of globals.

When rearranging the flags, I discovered that flag 16 was often being used for various things. Problem is, if this flag is set, you get no confirmation if you want to restart the game. Obviously, during the AGI Studio days, it wasn't known that this flag was being used by the interpreter. Now we know (thanks to discovery of original source) that flags/vars 0-29 are all in the interpreter range. I rearranged things with this in mind.

I even tested the remastered games with really low CPU speed. 2000 cycles works well for most AGI games, but Tex McPhilip 3, with its larger views, chokes with anything lower than 3000. The games clearly weren't optimized with these slower speeds in mind. I took care of that.

So there you go. That's what I'd been doing all this time.

4
At this point, I've given up trying to get SQ0 running on the original AGI interpreter. The game is simply too advanced. It's better off just getting an SCI0 remake, which I'd started work on a while ago. Maybe if the AGI pics can be converted to SCI format, I could easily adapt the AGI logic code to SCI script code.

On the plus side, I've had much better luck getting other fan games ported to AGI3.

5
Do you have an account for the AGI Wiki?

No, I don't.

6
I'd like to note that the third Tex McPhilip game actually was completed and released. I was able to download it here. For that matter, newer versions of Goodman were released here (note the version labeled 4.0T; it's more suited for developers!) I even reported these releases to ScummVM, so they'll work on that just fine.

For the sake of preservation, I've attached the games in question.

Finally, I've been working on doing AGI3 remasters of some of the fan games. You already know about SQ:TLC, for example. I've also discovered that the real reason SQ0 doesn't work on the original interpreter is because the game is so inefficiently coded!
What I've been doing for the AGI3 remasters is designating 220-239 for local flags/variables, and 240-255 for dynamic flags/variables (which are automatically cleared, thanks to a special logic that is executed on room change), similarly to the original Sierra games. I can understand why the original AGI template (on which these games are based) didn't implement this trick; it wasn't well-known at the time.
For that matter, I've been trimming each room down so that redundant code has been moved into new dynamic logics. While SQ:TLC had barely any room for dynamic logics, SQ0 has much more space available.

7
SCI Development Tools / Re: SCI11+ now has working pathfinding
« on: April 14, 2023, 12:18:14 PM »
Great! I'll have to go through SCIQuest and the QFG1VGA upgrade with the new interpreter. I shouldn't have to worry about fragmentation now.

8
SCI Development Tools / Re: SCI1.0 Template Game
« on: March 03, 2023, 10:58:17 AM »
And now I've looked through the code for foreign versions, and added the control panel views for Spanish, French, German, and Italian languages, as well as code to switch between English and one of the other languages.
The standard fonts and INTERP.ERR for each language are included in their own directories.
I'll be adding the updated script code and views to the EGA version soon.

9
SCI Development Tools / Re: SCI1.0 Template Game
« on: February 25, 2023, 08:09:02 AM »
And here it is, Circus Quest in SCI1.0!

10
SCI Development Tools / Re: SCI1.0 Template Game
« on: February 21, 2023, 08:01:44 PM »
It's done! The EGA version of the SCI1.0 template is now here! SCI0 views and pics are fully compatible, and SCICompanion's importer works better than Sierra's original tools. The script code between the EGA and VGA versions is identical - the game checks the video driver's capabilities, and sets the appropriate global colors.

I think this can replace the old SCI0 point-and-click template, since this is MUCH more elegantly designed. I've even been working to bring Circus Quest to the new engine as a proof of concept, but the code is a huge mess that I'll need to clean up. Good thing there's only four real rooms.

11
SCI Development Tools / Re: SCI1.0 Template Game
« on: February 18, 2023, 11:39:46 PM »
Here are all the specialized EGA versions I own: https://drive.google.com/drive/folders/0B5j-_ZMS8_UoUzAyVEg0NnZKUVE?resourcekey=0-0YAJ6GVxwOeQIn0RPOJ3nw&usp=share_link

You can find Conquests of the Longbow EGA in there as well.

Thanks. I'll be putting out an EGA version when I find the time to.

12
SCI Development Tools / Re: SCI1.0 Template Game
« on: February 18, 2023, 11:36:13 AM »
So I've managed to do over the SCI1.0 template, using the interpreter and drivers from the Mixed-Up Fairy Tales demo. That game supports separate music and audio drivers, so GMBLAST is not needed.
Next step is to do an EGA version. The closest compatible interpreter and drivers are from the EGA version of Conquests of the Longbow. Unfortunately, I haven't been able to track that one down.
I know that MAKEP16 and MAKEV16 were the original tools used to convert the VGA graphics to EGA.

13
Everything-Else / Re: LSL1 AGI Source Code Released
« on: January 09, 2023, 11:22:47 AM »
Hopefully, the SCI games will follow soon. I've been wanting to get more original procedure and global names, as well as earlier versions of the header files.

No interpreter code? That code doesn't interest me. I'm more interested in the scripts and headers.

What do you know? More source was released this Christmas. LSL5 here. I'm most interested in the system script directory. Here's more here..

14
Well, it turns out the synonyms block is no longer supported from SCI01 on. LSL2 and SQ3 didn't use it much, if at all, so it doesn't affect them. KQ4 and LSL3, on the other hand, use it quite often, so I'll have to adjust the said specs to not use it.

By the way, I've managed to test LSL3 to completion using the S.old.010 interpreter. Next step is to port to version 1.000.172. As with SQ3, there is actually a German Amiga version of LSL3 that uses SCI1, and has SCI1 sound resources... but they have no Adlib tracks. Nothing Sounder can't fix.

The only things that don't work are rm140's RoomScript (the age quiz) and rm340's ComicScript (the Comedy Hut's inputting of ethnic groups). For the former, I just couldn't decompile it ("Couldn't find DUP source: 20"). For the latter, I had to do some tricks to generate workable code, but it doesn't work right (fortunately, the sequence is still completable, so you still can get the full 4000 points)

15
I've done something impressive - I've ported KQ4, LSL2, and SQ3 to the SCI1 engine (1.000.172, from the Seasoned Professional). KQ4 and LSL2 have the digital audio from their Amiga versions. Ever wonder why you can't save or bring up the inventory in KQ4's organ room? It's because there's not enough memory to do so. The new engine's improved memory management fixes this.

LSL2 uses the codebase for the Amiga version 1.003, which overhauled all the scripts.

Also, SQ3 1.018 (the version included in collections and digital distribution) actually has a bug where the scripts refer to the wrong said specs. That was fixed for ScummVM., and I've applied that fix to my upgrade.

Now the main issue is actually distributing it. For the RESOURCE.001 and RESOURCE.MAP files, they'll have to be in binary patch form. The drivers and interpreter were from an officially freely available Christmas card, so no issues there.

My next goal is LSL3, with its Amiga-exclusive sound effects and slight script revisions. Maybe grab some decompiled code from the French and German versions as well...

Pages: [1] 2 3 ... 17

SMF 2.0.19 | SMF © 2021, Simple Machines
Simple Audio Video Embedder

Page created in 0.052 seconds with 19 queries.