Shanghai.EXE

[Anonymous No.]

>Latest Release (0.504e5):
https://gitgud.io/SHNecro/shanghaiexe/-/releases/0503e4

>What this is:
Shanghai.EXE was a Touhou-themed Megaman Battle Network clone.
Then the original developer got a C&D from Capcom and shut everything down, just before it was completed.
Luckily, someone leaked a donor-wall demo build, and the whole thing was made in C#, which meant it could be decompiled.
This is a modified/translated version of that demo build, along with a handful of bugfixes.
The end goal is to finish the rest of the game, which off the top of my head involves the final battle, the postgame areas/bosses, some sidequests, and netplay.

Just curious about the progress since some stuff is getting updated.
My main question is about the areas that weren't developed yet and if they're accessible yet.

[Anonymous No.]

It still baffles me how it got C&D'd when other games that are basically complete rips get off scott free behind a paywall.

[Anonymous No.]

>>25748759
You made a thread asking about the progress while simultaneously posting a link to the project page that shows that there hasn't been any in 7 months?

[Anonymous No.]

>>25752270

No releases in 7 months but progress is being made. There are regular commits to the repository, the last a week or so ago.

[Anonymous No.]

>>25752270
Look at the commits.
Another anon already answered for me but changes or additions do not mean a new release. I can only wonder what is being changed as it has not been compiled for another release yet and the fact that I know nothing about the C languages and the process of compiling regardless of how simple it might be to someone else.

[Anonymous No.]

>>25748759
If it's any indication of how unreliable my estimates are, 7 months ago I thought I'd be able to knock this out by about May.
Some things completely out of my control also happened between then and now, but I still don't want to put out any estimates.

A lot of the work so far has been pretty major internal/non-story changes like rewriting the PA/chip system+all affected pages to be more flexible, re-adding the music player/editor sound preview, and moving out hardcoded stuff so the editor can edit them without having to recompile.
Right now I'm beating my head against the wall that is the render order, trying to replace whatever magic Koki found that did it all in 2 lines but gets it wrong 1/100 times.
I know it doesn't really help the release speed, but it does quash old, unnoticed bugs as well as prevent new ones from popping up, and I get a ton of experience out of it.

Story-wise, I'd say I'm about halfway through the stuff I have planned for the next release (finishing the 1st postgame area), including cutscenes, sprites, effects, etc.
I haven't touched creative writing or image editing since highschool so please adjust your expectations, but I think I'm also alright at telling when what I have is complete shit so the end result should at least be better than nothing, which is what I'm aiming for.

The major problem I've run into is the game design, in part because this is my first real experience with it, but also because I have zero idea what Koki was planning for the endgame.
For example, by the end of the story, comps are already pulling out V3 viruses, which leaves me without much room for the postgame encounters besides trying something creative with virus composition and pulling out the EX ranks which all have that fallback chip reward behavior of the V1-3 * and alt codes.
Similarly, before redesigning the PA/chip system there was a hard cap on chip number/order, not nearly enough for all the concept art etc. or PAs for all the navi chips.
HertzUp and CorePlus I've got a bit of a better handle on after writing those scripts to scrape all the map files, but the best I can do is fill in the gaps between the biggest Minus AddOn and AddOn requirements.
I think I've got something workable, and I'm sure people will find some game-breaking meta no matter what I do, but this whole thing feels like I'm flying blind.
On that topic, I saw someone dug out my commit message about nerfing BublBlst, all I did was add a tiny i-frame at the end of it so spamming A makes you miss the first shot, reducing overall damage. If you wait a handful of frames you can still sneak it in before most flinches end, but it now needs a bit of timing.

Apologies for the wall of text and the slow progress, but it's what I've got and you did ask for a progress update.
As always, if anybody has the time to contribute code, assets, ideas, or even some sort of project timeline/management, I'd be happy to work it in.

[Anonymous No.]

>>25748822
Because nodoby gives a shit about bad games.
Shanghai.exe, in other hand, good enough to make the original games look bad.

[Anonymous No.]

>>25777488
>if anybody has the time to contribute code, assets, ideas, or even some sort of project timeline/management

None of those things are in my wheelhouse but I can contribute time if you've got any grunt work you want to slide off your plate.

[Anonymous No.]

bump

[Anonymous No.]

>>25834331
This for a moment. I need to read whats above.

[Anonymous No.]

>>25777488
>Koki found that did it all in 2 lines but gets it wrong 1/100 times.
Example?

[Anonymous No.]

>>25777488
>Apologies for the wall of text and the slow progress, but it's what I've got and you did ask for a progress update.
No problem dude. It's all good. Take it one step at a time.

[Anonymous No.]

>>25777488
Keep up the good work, your continuation is a godsend.

[Anonymous No.]

>>25777488
keep up the good work anon

[Anonymous No.]

>>25835668
I think I'm starting to piece it together while trying to explain what I don't get about it, but as I've been placing new objects for the postgame it's felt like trial and error whether things would render in the right order, especially for long or tall objects.
I'm not sure if it's always been like this, or if the decompilation or an earlier bugfix messed things up, but despite 99% of the game having no issues there were occasional bugs like the table in the image, where I had to cheat and extend the collision box so the player would never be "behind" it.
I figured it would be a quick fix, but I think I took a wrong turn somewhere with googling isometric-world sprite rendering and made things way too complicated, leading to the previous picture.
At the moment I think the code in the image might be fine, and it's a problem with the sorting algorithm instead, so I might be getting close to a solution.

[Anonymous No.]

>>25797684
Playtesting's always good, and it's been too long since I've played anything but my debug folder so my sense of easy/hard is all screwed up.
I don't think I messed with the random encounter format, so these should still work:
https://pastebin.com/33wqQrya
Open up the map editor to whatever map you like, paste those into the Encounters tab like the picture, and you should be able to fight the encounters.

[Anonymous No.]

>>25842474
Ok, thanks for all of your hard work

[Anonymous No.]

>>25842281
Well, good luck is all I can say. You know more than I do at this moment.

[Anonymous No.]

>>25842474

So I've spent a couple of hours playing with these encounters and I'd say that on the whole they aren't *too* far off what I'd expect difficulty wise but mostly slightly too easy.

My folder is fairly strong but quite simple and "nooby" if that makes sense, and I'd probably say that I'm an average player, perhaps a little above average. With these battles I have tended to be able to make a plan with what I'm given from my folder then as long as it all goes to plan the encounters are fairly easy, if however I get a bad hand or miss something that I needed to connect it's fairly easy to get overwhelmed and loose a shitload of HP before you can delete them.

The exception to this is the DunBeetle family, those fuckers are evil.

[Anonymous No.]

>>25862089
I'll give it a try as well and see what happens

[Anonymous No.]

>>25842474
>https://pastebin.com/33wqQrya
It's been a while since I messed with map editor. Was there a text option before? I messed with the enemies but i don't remember a direct edit. Also, when adding another layer or floor above something like Shanghai's little hub with the navis that can be fought, is there anything I should look out for? I remember messing something up but I don't remember what since it was little over a year back.
I know some stuff requires code to make whatever changes you make on the surface operable, I'm just wondering if it's the same here.

[Anonymous No.]

>>25897720
>It's been a while since I messed with map editor. Was there a text option before?

You see that pencil symbol circled in the picture? Click that.

[Anonymous No.]

>>25862089
Seems to be a trend in the stuff I make, DruidMan had the same issue of having a "right" way that trivialized the fight, while shifting all the difficulty to pile on if you slip up.
I'll bump one or two EX viruses down to V3, balanced by adding in some wildcards to keep you on your toes, hopefully that'll help.
While I'm on that I'll take another look at DruidMan too, see if I can freshen things up.
Thanks for the feedback, it really helps.

>>25897720
Been there since V1, it's a handy fallback for when the UI design isn't all there and it's easier to just edit the raw values.
Case in point, editing the map dimensions (width/height/levels) is all fucked up and has been for a while, I've only just had to change those and realized it was broken the whole time.
I haven't even fixed it on source yet, for that I'd recommend exporting to raw-text .shd, editing it in notepad, and re-importing it.
The release version of the editor will also "corrupt" a map file if you save a file with an error (ex. missing text string) in it, since it'll refuse to load because of the error and there's no option to decode the game-format .she without trying to load it (which fails). At that point you'd need a build tool compiled from source to recover it.
I've made major improvements to the editor since I'm using it more especially for errors, but you'd have to build the source yourself unless until I make a release.

>>25843901
Definitely arguable, but at least I figured out the original was doing it wrong too.
Now to dig through tutorials about how 8-bit games managed to figure it all out 40 years ago.