hidefuku / AnimeEffects

2D Animation Tool
GNU General Public License v3.0
755 stars 99 forks source link

AnimeEffects Development Hiatus | Sibling Fork Discussion #40

Open Jose-Moreno opened 4 years ago

Jose-Moreno commented 4 years ago

Hi, currently and as of August 2017 the state of development for AnimeEffects has been halted due to unknown circumstances. The original developer @hidefuku has been missing from this repository and any of his / her social media account. We sincerely hope they are doing well <秀福さん, 元気にやっていることを願ってるよ! > :pensive:

This thread has been created to discuss the future of this project as some developers and users are interested in continuing the development of AnimeEffects.

Part of this discussion has already taken place in #38 but to avoid derailing the original issue [DragonBones JSON Format Export] even further we have created this issue instead for all remaining developers and users to see and participate.

Previously Discussed Topics

  1. Forking AnimeEffects and continuing off in a different branch while we wait for Hidefuku-san to come back.
  2. Forking AnimeEffects and re-brand it as a new project (with proper credits to the original project and respecting licensing due to GPLv3)
    • In this aspect we're going to wait until Jan of 2020, to see if Hidefuku-san has issued any announcement on this topic.
  3. Create a new organization to democratize developer access and better regulate the influx of PR's
    • The name of the organization has been proposed to be AnimeEffects-Devs for the time being. Once the due date has passed we'll re-brand it to the agreed name that is to be discussed later.
  4. Broaden the application target market and promote it, not only for cut-out / modular animators, but also for game developers and game engines (e.g Godot, GDevelop, Xenko, Ct.js, Armory3D, etc) along runtime support integration.
  5. Roadmap creation and hosting platforms
  6. New companion website hosting using github-pages
  7. Creation of decentralized communication channels:

Additional development proposals

Re-branding Discussion

There was a brief discussion for names if a re-branding comes into effect, but these are not final and we'll welcome any other proposal.

The criteria used was:

  • Anime-effects name doesn't really communicate what it does. "Anime" is an art style in the west, but a legitimate word in the east (it means animation in Japan). "Effects" is something people associate with compositing vfx or sound effects.
  • AnimeEffects was probably meant to capture the Japanese market which depends a LOT in (Adobe) After Effects (particularly in studio settings and hobbyist circles). This app really seemed it was looking to fill the place that the FOSS DUIK plugins have in After Effects to provide skeletal character animation capabilities.
  • If we're going about names, consider something that is catchy and communicates the core function of the software (even if it is expanded to be more complex, or have more functionality later on).

Proposed Names

We hope that anyone interested in improving this excellent application will come forward. There is no small task when it comes to FOSS development and any kind assistance will prove to be of critical importance for the survival of this project moving forward.

Thank you in advance for your attention :bowing_man:

Special thanks to @blurymind, @herace, @Larpon for their initial interest and efforts to contribute to this renewal project.

larpon commented 4 years ago

... and thank you @Jose-Moreno for kicking this off! Let's all help AnimeEffects to the next level.

herace commented 4 years ago

Cool, thanks for creating the issue @Jose-Moreno :+1:

hellozee commented 4 years ago

Best of luck folks, :) psst, since it is a qt project, you are more than welcome to join the @KDE community, :)

larpon commented 4 years ago

@hellozee - follow our progress here: https://github.com/herace/AnimeEffects (we need stars, bells, whistles and luck) :stuck_out_tongue:

hellozee commented 4 years ago

@Larpon oo nice, :D

ghost commented 4 years ago

This is good, however I'd like to offer some input.

I hope we stick to the After Effects DUIK side of things rather than changing it to the Spine2D side. I'm not apposed to having exporting/features for game dev. I don't think it should be the main focus though.

When I first came across this a year ago I messed with it for a bit. I was like oh this tool would be great just for animating short films, etc. I've been interested ever since.

Another thing I think we should try to come up with a name that is more universal. All the names suggested make me only think of game dev.

I suggest Animesh. "Ani" short for Animation and "mesh" being that the main feature is Free Form Deformation using a mesh.

larpon commented 4 years ago

@RyanJEC cool! come join us on Discord or IRC. I haven't tried either of the 2 you mention so I don't know about the sides things can go to.

I'm currently interested in keeping the unique features of AnimeEffects and improve upon @hidefuku 's great foundational work.

Mainly I think the workflow "that doesn't need a carefully thoughout plan" is one of the forces of the software. It's easy and intuitive to work with.

I'd really like the UI to get a little more familiar looking when compared to other animation software (keeping the current simplicity and ease of use in mind ofc).

I also think it shouldn't aim for any specific industry (gamedev, animated movies, etc.) unless we can make it as plugin friendly as Blender so it's just a matter of what features you enable 🙂

A name change or rebranding will hopefully come as a natural thing when/if development takes everything too far away from the original code base.

Jose-Moreno commented 4 years ago

@RyanJEC Hey thanks for commenting, glad to see more people interested! :smile:

I understand what you're saying for sure, but I think you shouldn't worry about the tool losing focus anytime soon.

AnimeEffects has always been first and foremost an animation tool first, and is primarily geared as a cutout / modular animation tool, so I don't think having modules /add-ons to export for game engines is going to change the functionality at all.

If anything we've asked for inclusion of more traditional features & settings such as timecode display and to set various FPS instead of constantly using 60fps for everything. However from an animation production pipeline perspective, there are still many elements missing and a lot of UX issues, which would have to be sorted before even considering "new features" though.

On the topic of DUIK vs Spine, in my experience the latter has no extraneous tools for a seasoned digital cut-out animator. You could perfectly use it for TV or film if you wished, It's just that no one use it for that because there is a marketing focus towards game dev pipelines and direct integration with animation pipeline software is non-existent (to my knowledge), however the animation editor itself can work just fine with those goals in mind. Same for Dragonbones. Note: Well the only issue i'm aware of, is that image import & canvas support is limited to 2K textures, so no True 4K export.

I personally feel Anime Effects is a tool that has a virtually unexplored market, at least for a FOSS tool, even though Synfig exists and it's active, it is still very niche. That said I honestly think Anie can have a definite impact in the FOSS animation market.

I agree it's paramount that the project strives to preserve a focused core vision, but this also means it has to extend the compatibility with other creative pipelines, so if Anie won't have extra drawing, compositing or editing features, there's a potential need for import plugins & export modules to be included for scriptable apps such as:

Import

Export

While it's a bit too soon to tell, maybe Anie could have some simple post-processing effects by implementing libraries such as G'MIC, in order to have simple vfx baked for exported images, videos and converted runtimes (e.g exported as GLSL shaders), but having any more than that would simply be daunting for the potential maintainers, contributors and devs.

Other than that as Larpon put it, we should try to retain the familiarity and push for intuitiveness and ease-of-use while encouraging modular extensibility through plug-ins or add-ons in such case so it's easier for both developers and advanced users to contribute and perhaps extend the software themselves if it needs to be (similar to all the aforementioned apps)

blurymind commented 4 years ago

If you do an exporter for gamedev pipeline, which is honestly my only interest in using this tool for, please do not target the spine2d json format. The runtimes for spine on different game engines all have a license clause that requires the dev to spend 300+$ for a spine2d license just to use them. Dragonbones runtimes on the other hand do not have that awful clause and are all just mit open source licensed runtimes. That is why coa tools extension for blender can export to dragonbones too. It's a friendlier format for indie devs with limited budgets.

It's worth noting that both spine and dragonbones runtimes on different game engines just use json files and an image atlas with the characters body parts. That is all an exporter needs to do.

It is also worth noting that as of now there is not a single free and open source animation editor that can export animation as json data that can be used by a game engine. At the same time there are many runtimes for many free and open source game engines for the dragonbones format. This includes Godot too.

There is a HUGE untapped potential here for this tool to become quite a popular tool amongst indie game devs, just by being able to export the data as json. This tool is perfect for that exactly because of its focus on cut out and deformation style animation, which is what alot of modern 2d games use, because it is more efficient and flexible than baked spritesheets

On Tue, Dec 3, 2019, 12:23 AM Morr notifications@github.com wrote:

@RyanJEC https://github.com/RyanJEC Hey thanks for commenting, glad to see more people interested! 😄

I understand what you're saying for sure, but I think you shouldn't worry about the tool losing focus anytime soon.

AnimeEffects has always been first and foremost an animation tool first, and is primarily geared as a cutout / modular animation tool, so I don't think having modules /add-ons to export for game engines is going to change the functionality at all.

If anything we've asked for inclusion of more traditional features & settings such as timecode display and to set various FPS instead of constantly using 60fps for everything. However from an animation production pipeline perspective, there are still many elements missing and a lot of UX issues, which would have to be sorted before even considering "new features" though.

On the topic of DUIK vs Spine, in my experience the latter has no extraneous tools for a seasoned digital cut-out animator. You could perfectly use it for TV or film if you wished, It's just that no one use it for that because there is a marketing focus towards game dev pipelines and direct integration with animation pipeline software is non-existent (to my knowledge), however the animation editor itself can work just fine with those goals in mind. Same for Dragonbones. Note: Well the only issue i'm aware of, is that image import & canvas support is limited to 2K textures, so no True 4K export.

I personally feel Anime Effects is a tool that has a virtually unexplored market, at least for a FOSS tool, even though Synfig exists and it's active, it is still very niche. That said I honestly think Anie can have a definite impact in the FOSS animation market.

I agree it's paramount that the project strives to preserve a focused core vision, but this also means it has to extend the compatibility with other creative pipelines, so if Anie won't have extra drawing, compositing or editing features, there's a potential need for import plugins & export modules to be included for scriptable apps such as:

Import

  • Photoshop (JSX API)
  • Krita (Python API)
  • GIMP (Python)

Export

  • After Effects (JSX)
  • Nuke (Python)
  • Natron (Python)
  • Blender (Python)
  • Fusion (Python)

While it's a bit too soon to tell, maybe Anie could have some simple post-processing effects by implementing libraries such as G'MIC, in order to have simple vfx baked for exported images, videos and converted runtimes (e.g exported as GLSL shaders), but having any more than that would simply be daunting for the potential maintainers, contributors and devs.

Other than that as Larpon put it, we should try to retain the familiarity and push for intuitiveness and ease-of-use while encouraging modular extensibility through plug-ins or add-ons in such case so it's easier for both developers and advanced users to contribute and perhaps extend the software themselves if it needs to be (similar to all the aforementioned apps)

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/hidefuku/AnimeEffects/issues/40?email_source=notifications&email_token=ABRRWVOAAGKIO4AJVHN25KTQWWRJJA5CNFSM4I545X32YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEFXUTRA#issuecomment-560941508, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABRRWVJSHBIXRDGMUNR2U7TQWWRJJANCNFSM4I545X3Q .

larpon commented 4 years ago

@blurymind - and even more popular if we provide an easy plug'n'play runtime for said format :smile: - but that's not where we are at the moment.

AnimeEffects is a unique animation tool - and anyone involved in the project should strive to keep that "uniqueness" IMO.

I'd welcome any contributed exporters/importers to the code base as long as it's free, fairly well-written and unencumbered code. No matter the format they import or export to. Importers and exporters (usually) won't mess with/require any usability changes. Because of this I consider it fairly "safe" to extend AnimeEffects in this direction. The optimal would probably be to have a huge set of importers/exporters for users to choose from :slightly_smiling_face:

That said, it's my personal preference to support the most free (as in freedom) toolchains/formats with higher priority.

blurymind commented 4 years ago

any news on the forked project? It's almost a year later now :)

Jose-Moreno commented 4 years ago

@blurymind Hey. There's been some slow progress which has been committed towards @herace 's fork. There were various contributions made by @larpon Most discussion has happened on discord, however due to personal reasons and the quarantine, it's been extremely difficult to commit further work. This week however @azagaya 's was planning on submitting a PR related to continuous integration with Travis-ci

blurymind commented 4 years ago

Do you guys still consider splitting it into a new project, so as to focus attention to one active master branch? Right now this app very much looks like abandonware due to its development not happening on the master branch, but on various forks. There is no obvious central branch that the open source community can identify.

The author has gone completely quiet. I tried to find if he/she has activity elsewhere without success

larpon commented 4 years ago

Our efforts and "official" fork is now located at https://github.com/herace/AnimeEffects We hang out and discuss on Discord and on IRC (freenode #animeeffects-dev)

IMO we need further and bigger advancement, plus way more (developer) activity before a rebranding makes sense.

Come join us and maybe we can lift AE off

Jose-Moreno commented 4 years ago

@blurymind I understand the importance of revisiting this question but the main problem that we considered is that somewhere in Japan hidefuku is still paying for the animeeffects.org domain and hosting. Maybe he forgot to turn off the server but if the site is still up under that domain, there should still be hope to reconcile the forks.

Of course I trust we all know the dangers of forking projects like these by now, dividing the developers and the userbase is never a good move, but even if we branch out and excise completely the anime effects brand we'll still need people to work on it, so unfortunately it also comes down to a question of conviction vs money and with the current pandemic emergency it's not easy to ask people to trust in a project that's been abandoned and the forks have low commit activity.

With that in mind I agree with @Larpon , until there is significant progress to show and a somewhat predictable developer activity, we shouldn't think of doing a huge overhaul to the branding and will probably need to leech off the main repository for a while to help with developer "discoverability".

That said this is unfortunately a time sensitive matter. If we could at least secure 2 more developers willing to dirty their hands every weekend or at nights, after putting their kids to bed, so we can start seeing some progress, then we might be able to reconsider as a community how to at least establish an organization account to host the fork and work from there.

I personally would rather start over all the social media, repositories, etc and advertise it from scratch while mentioning it's an off shoot of the original project as well, but for that we need results both from development so users can create artwork with it and start advertising via word of mouth.

The technology stack is not that difficult to setup for compilation, This code base uses C++11 and Qt 5 framework SDK which is surprisingly a very benevolent combination even for novice programmers to get everything running with a simple guide. Also since the same tech is shared among similar FOSS projects (e.g Pencil2D, Krita, Opentoonz, etc) there might be potential interest for cross-team development or one-off contributions.

Other than the existing contributors, any of you reading this can help the project by advertising it to developers with potential interest in the technology

Even a single bugfix or feature contributions can and will make a difference

blurymind commented 4 years ago

I would be interested to study its code base but my experience in c++ is zero. I am a Js/python programmer. If I was to fix a bug or submit a pr who is going to review it and guide me through the code base? Maybe I could try to open and build it in qt creator over the weekend but I have no idea where to even begin. Most of my development so far has happened in vscode and chrome web browser :)

It would really help of someone wrote down build instructions for Linux and wrote some sort of a code structure documentation that tells new devs what folder houses what.

In any case if we can't use this repo to do that, it would help updating the readme at the fork

On Mon, May 4, 2020, 10:47 PM Morr notifications@github.com wrote:

@blurymind https://github.com/blurymind I understand the importance of revisiting this question but the main problem that we considered is that somewhere in Japan hidefuku is still paying for the animeeffects.org domain and hosting. Maybe he forgot to turn off the server but if the site is still up under that domain, there should still be hope to reconcile the forks.

Of course I trust we all know the dangers of forking projects like these by now, dividing the developers and the userbase is never a good move, but even if we branch out and excise completely the anime effects brand we'll still need people to work on it, so unfortunately it also comes down to a question of conviction vs money and with the current pandemic emergency it's not easy to ask people to trust in a project that's been abandoned and the forks have low commit activity.

With that in mind I agree with @Larpon https://github.com/Larpon , until there is significant progress to show and a somewhat predictable developer activity, we shouldn't think of doing a huge overhaul to the branding and will probably need to leech off the main repository for a while to help with developer "discoverability".

That said this is unfortunately a time sensitive matter. If we could at least secure 2 more developers willing to dirty their hands every weekend or at nights, after putting their kids to bed, so we can start seeing some progress, then we might be able to reconsider as a community how to at least establish an organization account to host the fork and work from there.

I personally would rather start over all the social media, repositories, etc and advertise it from scratch while mentioning it's an off shoot of the original project as well, but for that we need results both from development so users can create artwork with it and start advertising via word of mouth.

The technology stack is not that difficult to setup for compilation, This code base uses C++11 and Qt 5 framework SDK which is surprisingly a very benevolent combination even for novice programmers to get everything running with a simple guide. Also since the same tech is shared among similar FOSS projects (e.g Pencil2D, Krita, Opentoonz, etc) there might be potential interest for cross-team development or one-off contributions.

Other than the existing contributors, any of you reading this can help the project by advertising it to developers with potential interest in the technology

Even a single bugfix or feature contributions can and will make a difference

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/hidefuku/AnimeEffects/issues/40#issuecomment-623725291, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABRRWVNB364RNCY4AWYA6RDRP4ZWBANCNFSM4I545X3Q .

azagaya commented 4 years ago

@blurymind for compiling in linux, its really easy even if you dont have experience with qt. Just donwload the Qt online installer and install Qt creator and the toolchain (the wizard will guide you). Open the AnimeEffects.pro file in Qt creator and click run. If you dont want to install the whole Qt, you can install needed Qt packages from your system repositories. What linux distro are you using? About someone guiding people through the code or writting what does what, the project itself is quite big, and idk if one of the current contributors knows the code enough. I tried to get familiar with the project some times but i couldnt really progress much..

blurymind commented 4 years ago

It could really help in attracting contributors to update the forks readme with better getting started documentation :) Some stress test example files can also be really useful. Both for getting more attention to this app and for ourselves to use when working on it. Maybe we can convert some of the dragonbones examples to anime effects?

@azagaya thanks for explaining. I am on manjaro btw. Already got qt creator from its repos. Any good courses you could recommend on c++? I mean for someone coming from a dynamic type language like myself.

From experience I find that having a maintainer to review prs and guide developers makes a huge difference to an open source project. Someone who knows the codebase and features well enough to catch regressions

azagaya commented 4 years ago

I could make another PR updating the readme with compiling instructions

hellozee commented 4 years ago

@Jose-Moreno if it is feasible, KDE will be happy to incubate this project. Even though I expect the transitioning would be a bit difficult, but nevertheless if things change, do give this a look.

larpon commented 4 years ago

@hellozee - good to see you here! We have you guys in mind :)

blurymind commented 4 years ago

@Jose-Moreno if it is feasible, KDE will be happy to incubate this project. Even though I expect the transitioning would be a bit difficult, but nevertheless if things change, do give this a look.

That will really help the project get more exposure in the open source community :+1: fantastic idea!

larpon commented 4 years ago

@blurymind - no matter what, we need progress and activity for the project to live on. I hope to get time to push my theme update soon - so we don't fry our eyes while testing :stuck_out_tongue:

blurymind commented 4 years ago

I opened an issue here https://github.com/hidefuku/AnimeEffects/issues/42

I am failing to build and run it. Looks like I'm not alone https://twitter.com/lgworld/status/1257675554712666112

One advantage of having control over the repository is the ability to set up automatic builds :)

azagaya commented 4 years ago

@blurymind my PR sets up Travis CI and builds an appimage, which is then uploaded to github releases. Something like that?

blurymind commented 4 years ago

yes exactly :) can you do a PR to @herace 's branch, which is the main one to use to progress this?

azagaya commented 4 years ago

@blurymind i made the PR to @herace 's branch. herace/AnimeEffects#15

blurymind commented 4 years ago

An examples repository where we include community made files would really help us with catching feature regressions. Something like https://github.com/DragonBones/Demos

in that regard, we will need to make a call to all the users on twitter and other places to submit to us things they have made with the app.

Files that demonstrate what it can do and even stress test it to set the bar of the performance we aspire to have for heavier files

edit: opened a new issue here #43