Very short background info: Several years ago I obtained code for Elma's physics engine through reverse engineering. The physics code has not been public so far.
I had a brief discussion with Daniel a couple of days ago. I asked whether he'd be okay with if I published the physics code in some form, e.g. as part of Elmanager (SLE) which would make it possible to test levs in the editor itself.
In summary, he said:
1) He doesn't see this hurting Elma at all
2) He won't take legal action or ask for a takedown if I release it
3) Future consequences are unknown if the situation changes (e.g. rights are transferred or a larger publisher gets involved)
There's another data point to consider: I discovered that someone (with GitHub nick "kryptan") has already reverse engineered and published the engine (and the game is also somewhat playable). I've actually known about that repository for a while, but only recently (May/June) it seems like the author finished reversing/implementing the physics.
After looking at the code and testing it a little (it works if you have Rust installed, adjust the lev + lgr paths in main.rs and just "cargo run" it), it does seem to be the real deal (only missing some minor unimplemented stuff like taking apples and checking head collision).
All things considered, it seems like it would not make a practical difference if I released my version of the engine.
Thoughts?
On the publicity of the physics engine
Moderator: Moporators
Re: On the publicity of the physics engine
Are there any cons to releasing it? At least with Daniel's (kalmard, the new owner of Elma for those who don't know) permission I'd say go for it it there aren't any.
Team MiE - MiE Cup 1
Prestigious member of 14.6x Tutor14 club
Prestigious member of 14.6x Tutor14 club
Re: On the publicity of the physics engine
with kalmard agreeing, ez gogo imo
- insane guy
- Kuski
- Posts: 1673
- Joined: 22 May 2002, 20:53
- Contact:
Re: On the publicity of the physics engine
Say no more! This would probably make me switch to SLE entirely so I can finally use all the handy tools and hotkeys and omg
Thanks.
"Every night, me go to sleep, me have wet dream..."
Re: On the publicity of the physics engine
Quite sick other-elma program. I like holding space to auto pipe.
- analcactus
- Kuski
- Posts: 421
- Joined: 7 Dec 2010, 12:54
- Location: St. Petersburg, Russia
- Contact:
Re: On the publicity of the physics engine
super thisinsane guy wrote: ↑7 Jul 2020, 22:04Say no more! This would probably make me switch to SLE entirely so I can finally use all the handy tools and hotkeys and omg
Thanks.
had the conversation about editors with TTechnik just a few days ago and the thing stopping me from switching to SLE is alt-tab test orcaing
Re: On the publicity of the physics engine
After thinking about this lately, I don't feel comfortable enough to publish the physics code, specifically because of the 3rd point mentioned in the first post. Sure, it seems pretty unlikely that something like that would happen and cause trouble for me, but these days I want to take safe paths. If/when Elma is officially open-sourced, I will be able to do it as well.
The good news is, this doesn't mean I can't integrate the physics to SLE. In fact, I've already done that, and I will release a beta version soon.
So there will be 2 versions of Elmanager:
* The open source GitHub version which contains all other features except for the physics engine.
* The "enhanced" version which has the physics engine integrated. The download link to this version will not point to GitHub (because it would be bad practice to upload GitHub assets that don't correspond to the source code) but to my site (m-l.fi). The auto-updater will download either from GitHub or m-l depending on which version is being used.
To be clear, the GitHub version will also contain the code that needed to be written for integrating the physics engine. That way the source code diff between the two versions will stay minimal.
The good news is, this doesn't mean I can't integrate the physics to SLE. In fact, I've already done that, and I will release a beta version soon.
So there will be 2 versions of Elmanager:
* The open source GitHub version which contains all other features except for the physics engine.
* The "enhanced" version which has the physics engine integrated. The download link to this version will not point to GitHub (because it would be bad practice to upload GitHub assets that don't correspond to the source code) but to my site (m-l.fi). The auto-updater will download either from GitHub or m-l depending on which version is being used.
To be clear, the GitHub version will also contain the code that needed to be written for integrating the physics engine. That way the source code diff between the two versions will stay minimal.
Re: On the publicity of the physics engine
Very cool new version of SLE Smibu.
IMO, open source would make it possible to add awesome things to the game not seen in 20 years without breaking any backwards compatibility or nullifying any previous driven records, and I think that would be a huge bonus overall.
I hope one day it will be open source but perhaps there are drawbacks. Would it potentially spread out the player base or make it possible for people to cheat more easily?
For me it's an amazing game but so many cool features could be added with very little work. Can't imagine collision detection for non-static polygons to be an issue in 2021 for example. Some of the new battle type ideas could become reality. I hope this is a possibility for elma in the future.
Your 3rd point in the first post however, is a very good point, so I understand for now.
IMO, open source would make it possible to add awesome things to the game not seen in 20 years without breaking any backwards compatibility or nullifying any previous driven records, and I think that would be a huge bonus overall.
I hope one day it will be open source but perhaps there are drawbacks. Would it potentially spread out the player base or make it possible for people to cheat more easily?
For me it's an amazing game but so many cool features could be added with very little work. Can't imagine collision detection for non-static polygons to be an issue in 2021 for example. Some of the new battle type ideas could become reality. I hope this is a possibility for elma in the future.
Your 3rd point in the first post however, is a very good point, so I understand for now.