OpenMW's Roadmap and Future

Anything related to PR, release planning and any other non-technical idea how to move the project forward should be discussed here.
User avatar
akortunov
Posts: 899
Joined: 13 Mar 2017, 13:49
Location: Samara, Russian Federation

Re: OpenMW's Roadmap and Future

Post by akortunov »

psi29a wrote: 22 Jul 2020, 08:54 horizontal banners
Well, if I remembered correctly, MGE did not have this issue when I used it in 2011, both TES CS and OpenMW-CS seem to render them fine as well.
I have a suspicion that for animated objects you need to take pose for time 0 from animation file in Object Paging, not pose from the mesh itself.
psi29a wrote: 22 Jul 2020, 08:54 We'll have to take care of those in a case-by-case basis.
Not a good thing, IMO:
1. "Case-by-case" means that we either need to spend additional manhours to discuss every case or you will need to handle every single case solely.
First approach can decrease development speed, the second one can lead people to frustration (for example, if you merge PR that recreates case A from vanilla engine, but do not merge PR which recreates case B from vanilla engine because you personally do not like this case, it is basically a discrimination).
2. End user can not tell if given deviation from original engine is a bug or intended behaviour, so he creates bugreport just for sure. Then go to 1.
3. Third-party developer which wants to implement feature X can not tell if X is going to be merged or it is outside of project scope, so he is forced to ask OpenMW team members about it. Then we go to the step 1 again.

I'd prefer to have more formal and consistent guidelines.
User avatar
AnyOldName3
Posts: 2668
Joined: 26 Nov 2015, 03:25

Re: OpenMW's Roadmap and Future

Post by AnyOldName3 »

Remember MGE XE comes with a huge text file blacklisting things from its distant land and altering the state of others. Banners looking right when you first see them could well just be because someone's manually made them look right.
CMAugust
Posts: 285
Joined: 10 Jan 2016, 00:13

Re: OpenMW's Roadmap and Future

Post by CMAugust »

akortunov wrote: 22 Jul 2020, 13:14I'd prefer to have more formal and consistent guidelines.
If the guidelines are to be updated and clarified - for example, that OpenMW will eventually ship with a .owmaddon for dehardcoded features - where is the best place for them to go? The wiki, I assume.
User avatar
psi29a
Posts: 5356
Joined: 29 Sep 2011, 10:13
Location: Belgium
Gitlab profile: https://gitlab.com/psi29a/
Contact:

Re: OpenMW's Roadmap and Future

Post by psi29a »

akortunov wrote: 22 Jul 2020, 13:14
psi29a wrote: 22 Jul 2020, 08:54 horizontal banners
Well, if I remembered correctly, MGE did not have this issue when I used it in 2011, both TES CS and OpenMW-CS seem to render them fine as well.
I have a suspicion that for animated objects you need to take pose for time 0 from animation file in Object Paging, not pose from the mesh itself.
Can you (or someone) verify that? It would be great if that was consistent through-out all animations. It is a possible work-around.

I agree that having a formal and consistent guideline is good but I think you are misreading what I mean by "case-by-case" though.

1) We do handle every single case of whether we implement, enhance or ignore something related to MCP, for example, here: https://wiki.openmw.org/index.php?title=MCP That is just something that has to be done. Do we implement it or not? If not, why? Is it something that can be done as a mod? The assumption is that we keep the wiki pages up to do date and refer to it when the topic comes up.

2) This is going to happen regardless of what we do, the only thing we can do to help is again refer people to our decisions, even if that means looking at the issue and posting the link. Take for example this page: https://wiki.openmw.org/index.php?title=Bugs These are Bugs, Exploits and other Brokenness in Morrowind to show how OpenMW deviates from Morrowind and why. Just saying "no" isn't good enough, we have to explain why. This might be where consistency breaks down a bit and the 'case-by-case' comes in, for example when it comes to gameplay mechanics where one person sees it as part of the Morrowind experience awhile others consider it an exploit. There is also nothing stopping a modder from creating a mod that changes the behaviour.

3) This is really expected and encouraged behaviour: someone wants to work on something on OpenMW's issue tracker then they're at least on a good path because it can be discussed in the ticket. Should someone new have a question, they'll ask it or even create an issue and it can be discussed there. If someone comes to the project with their merge request without any previous discussion, then it will have to be reviewed with the potential of it being rejected. Regardless of which path is chosen, a developer will have to step in, spend time on it and take responsibility.

I think the wiki can be useful here to help lighten the burden, if that is something that developers feel is a current problem. We can make it clear about where Morrowind and OpenMW diverge and why the decisions have been made.

So for starters we have:
https://wiki.openmw.org/index.php?title=Bugs
https://wiki.openmw.org/index.php?title=MGE_XE
https://wiki.openmw.org/index.php?title=MCP

These can be reorganised and expanded upon.

I'm also open to suggestions as to how it could be done better though.
User avatar
akortunov
Posts: 899
Joined: 13 Mar 2017, 13:49
Location: Samara, Russian Federation

Re: OpenMW's Roadmap and Future

Post by akortunov »

psi29a wrote: 22 Jul 2020, 15:07 Can you (or someone) verify that?
I got a confirmation from Colt17 that vanilla banner meshes work fine with MGE's Distant Land - they are rotated vertically. No special treatment (e.g. blacklists) was found.
I can not tell which workaround it exactly uses, though. Maybe it would be better to ask Hrnchamd about it.
CMAugust
Posts: 285
Joined: 10 Jan 2016, 00:13

Re: OpenMW's Roadmap and Future

Post by CMAugust »

Another consideration for the future - existing dehardcoded and new features will be handled by packaged .omwaddon file. Many of these features will also be handled with simple Lua scripts. Some are already packaged with tes3mp's installation, albeit as loose files. However, .esp and .omwaddon are also the intended home of scripts, specifically mwscript, and I suppose that's where lua scripts will be housed too. Is there any disagreement on this?

And if lua scripts are to become part of .omwaddon, when should this take place?
Keeping in mind this change will break the soft compatibility with .esp files, as was planned post-1.0 anyway, though perhaps earlier than expected. Should this happen at the same time as Lua scripting is introduced for OpenMW, or should they instead be kept as loose files until after 1.0?
User avatar
AnyOldName3
Posts: 2668
Joined: 26 Nov 2015, 03:25

Re: OpenMW's Roadmap and Future

Post by AnyOldName3 »

I think it's probably better to let Lua scripts be loose, at least optionally (although they'll still need to be linked into the omwaddon). That way, modders can use their editor of choice instead of being stuck with whatever we build into the CS. There should be an option to bundle them into a BSA or the omwaddon itself, though.

This is pretty much how later games like Skyrim's PEX/PSC scripts work.
User avatar
psi29a
Posts: 5356
Joined: 29 Sep 2011, 10:13
Location: Belgium
Gitlab profile: https://gitlab.com/psi29a/
Contact:

Re: OpenMW's Roadmap and Future

Post by psi29a »

I'd be interested if there was some kind of embedded LUA Editor or IDE that we could integrate into OpenMW-CS? If there isn't any, we can then perhaps recommend a preferred editor?
User avatar
AnyOldName3
Posts: 2668
Joined: 26 Nov 2015, 03:25

Re: OpenMW's Roadmap and Future

Post by AnyOldName3 »

I think it would be better to provide plugins for other popular editors than to try and roll our own. I imagine Lua supports something similar to stubs or headers, so if we supply those for our API, that's most editors supported as thoroughly as most people need, and the only extras we might conceivably want are things like auto-suggestion of reference IDs in scripts, which add less utility than being able to use your regular editor.
User avatar
psi29a
Posts: 5356
Joined: 29 Sep 2011, 10:13
Location: Belgium
Gitlab profile: https://gitlab.com/psi29a/
Contact:

Re: OpenMW's Roadmap and Future

Post by psi29a »

Linking in this thread: viewtopic.php?f=6&t=7251
Post Reply