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.
davidcernat
Posts: 256
Joined: 19 Jul 2016, 01:02

Re: OpenMW's Roadmap and Future

Post by davidcernat »

akortunov wrote: 20 Jul 2020, 06:32 It depends on used assets. Arktwend, for example, is quite linear and has plenty of such situations.
It's one of few notable examples for Morrowind, because Morrowind's existing design has de facto highly encouraged open world style content that causes fewer problems in multiplayer by its nature compared to one-way trips – as will OpenMW's design – which is what the point is here.

Arktwend does, as expected, need manual adjustments for those situations.
akortunov wrote: 20 Jul 2020, 06:32 We have no control over such things on the engine level, especially if long-term goal is to turn OpenMW to the multi-purpose engine
We don't need "control" over anything. The simple fact remains that it's an engine recreating another engine with a very strong open world bias that mostly had open world content created for it.
akortunov wrote: 20 Jul 2020, 06:32 (IIRC, the whole Storm of Zehir is a single sandbox-based module with interactive world map).
Storm of Zehir isn't one of the 9 official campaigns for Neverwinter Nights 1, it's one of the 4 official campaigns for Neverwinter Nights 2 (and the only one with a more open world design out of those 4). It also still employs one-way teleportation via a magical portal to send you from one world region to another that would break the game in multiplayer if it wasn't specifically set to teleport all players.
akortunov wrote: 20 Jul 2020, 06:32 Even in long official campaigns chapters switching happens just a couple of times for the whole campaign.
One-way teleportation happens a whole, whole lot, and – in my experience playing through Neverwinter Nights 1's content in coop for more than half of my life – is the main source of issues in multiplayer.
akortunov wrote: 20 Jul 2020, 06:32 you try to decide for another people (content creators) which kind of content they will create and which kind of content they won't create, which looks like the "640KB ought to be enough for anybody" IMO. Probably content creators are smart enough to decide themselves which focus their content will have.
Again, here's the hubris causing you to employ mental shortcuts to completely misrepresent what other people say. I'm not deciding anything about what people can create, I'm saying the content created for Morrowind so far tends to work well in multiplayer because it has predominantly been open world-oriented instead of employing one-way-teleportation-to-various-places, and OpenMW content will follow a similar pattern for the simple reason that the engine's greatest appeal comes from the open world capabilities.
Greendogo wrote: 20 Jul 2020, 09:30 This discussion has a few people lobbing aggressive statements at each other and I don't think anyone else here wants to read that.
No, I'd say this is just another discussion where Akortunov wants to not get along with anyone. I suggest reading psi29a's post about this again so I don't have to go into my own extensive details on it.
User avatar
akortunov
Posts: 899
Joined: 13 Mar 2017, 13:49
Location: Samara, Russian Federation

Re: OpenMW's Roadmap and Future

Post by akortunov »

davidcernat wrote: 20 Jul 2020, 12:35 Morrowind's existing design has de facto highly encouraged open world style content that causes fewer problems in multiplayer by its nature compared to one-way trips – as will OpenMW's design
The main cause why OpenMW "encourages" it is the primitive scripting system from Morrowind. The second cause - an animation system, which does not support popular data formats.
As a result, it is quite hard to create complex scripted events or cutscenes, as in Enderal, for example.
From what I can tell, long-term goal is to eliminate these causes and give more freedom to content creators.
davidcernat wrote: 20 Jul 2020, 12:35 Storm of Zehir isn't one of the 9 official campaigns for Neverwinter Nights 1, it's one of the 4 official campaigns for Neverwinter Nights 2
Does it really matter? Electron engine is basically an Odissey/Aurora engine with an upgraded renderer. Both games use the same the same scripting language (NWScript), both have multiplayer, both are CRPG's from the same game series.
davidcernat wrote: 19 Jul 2020, 13:50 Vanilla quests and scripts were designed from a strictly singleplayer perspective, but were simple enough structurally that the vast majority of them have worked perfectly fine in multiplayer by just synchronizing quest stages and certain mwscript functions and variables across players
"Vast majority" of Mororwind's quest have the "bring N count of X object", "make GetDeadCount of X actor to be greater than X" or "talk to X NPC to get given journal entry" template. Issues start when content creator tries to do something more complex.
davidcernat wrote: 20 Jul 2020, 12:35 I'm saying the content created for Morrowind so far tends to work well in multiplayer because it has predominantly been open world-oriented instead of employing one-way-teleportation-to-various-places, and OpenMW content will follow a similar pattern for the simple reason that the engine's greatest appeal comes from the open world capabilities.
The beginning of conversation:
raevol wrote: 19 Jul 2020, 07:27 One thing that I do not understand at all is how a single-player-only mod developer could create a mod in the new multiplayer-aware Lua scripting system without having to code for multiplayer.
davidcernat wrote: 19 Jul 2020, 13:50 OpenMW content will have an open world focus instead of relying on frequent teleportation to isolated areas like in Neverwinter Nights, removing the main source of multiplayer-breaking bugs.
So the discussion from beginning was about new content, not about existing one.
Your point is that most of new content will be multiplayer-friendly because it is very hard to make multiplayer-unfriendly content for OpenMW (complex plots, scripts, cutscenes, teleportation, blocked cells, etc. as in SureAI mods) in its current state.
My point is that OpenMW will not always be in its current state, and our long-term task technically is to allow to create mentioned content easier.

And basically an answer to raevol's question is that mentioned "single-player-only mod developer" is going to create multiplayer-compatible content automatically as far as he follows Morrowind's approach (mod is a set of locations with some mobs and simple quests, or just a replacer for something). If he starts to deviate from this approach and write a lot of complex scripts, mod may have unexpected issues in multiplayer, and content creator should become a multiplayer mod developer when it happens. If he tells "personally I do not use multiplayer, so do not ask me to support it", an another multiplayer mod developer later should adapt mod to multiplayer every time when new mod version is released. The same thing if content creator makes a quite linear story-driven adventure rather than sandbox.

Existing content, as it was mentioned above, is quite different as well - while vanilla and Tamriel Rebuilt should be (mostly?) OK, large mods with custom quest lines (such as Chaos Heart or Arktwend, for example) may be more problematic.
User avatar
AnyOldName3
Posts: 2668
Joined: 26 Nov 2015, 03:25

Re: OpenMW's Roadmap and Future

Post by AnyOldName3 »

By the time the engine lets you do things like cutscenes easily, multiplayer will be something already in people's minds. The documentation for our scripting system will point out gotchas, the scripting system will make multiplayer-compatible stuff similarly easy to create compared with single-player stuff, and there'll be enough users nagging mod authors to make things multiplayer-compatible that anything that isn't multiplayer-compatible will likely have a third-party patch. We don't need to worry about this kind of content not using the scripting features OpenMW provides to make this easy as it'll have to target OpenMW in the first place due to that kind of content being unfeasible in vanilla or MWSE.
davidcernat
Posts: 256
Joined: 19 Jul 2016, 01:02

Re: OpenMW's Roadmap and Future

Post by davidcernat »

akortunov wrote: 20 Jul 2020, 15:29
davidcernat wrote: 20 Jul 2020, 12:35 Morrowind's existing design has de facto highly encouraged open world style content that causes fewer problems in multiplayer by its nature compared to one-way trips – as will OpenMW's design
The main cause why OpenMW "encourages" it is the primitive scripting system from Morrowind.
It's starting to feel strange for me to have to explain why an open world-focused game engine encourages open world content. You can look at the content created for the other Bethesda games and also see only a small amount of one-way-teleportation there, but I feel like I'm going to run into your insert-automatic-disagreement procedure no matter what I say.
akortunov wrote: 20 Jul 2020, 15:29 From what I can tell, long-term goal is to eliminate these causes and give more freedom to content creators.
Here's the real freedom: having an engine that lets you experience linear highly scripted questlines, prebuilt open worlds, procedurally generated open worlds, coop servers, MMO-lite servers, roleplaying servers, deathmatch servers, capture the flag servers, survival servers, base-building servers, VR servers, and anything else you can come up with, and any flexible combination of the aforementioned.

That's what I've been working towards, and that's what I'm going to have with or without your help.
akortunov wrote: 20 Jul 2020, 15:29 Does it really matter? Electron engine is basically an Odissey/Aurora engine with an upgraded renderer. Both games use the same the same scripting language (NWScript), both have multiplayer, both are CRPG's from the same game series.
As noted, Storm of Zehir still has a one-way portal that teleports you from one game region to another, and you can't return to where you came from without progressing in the story enough to find and use another portal. Here's a YouTube video with the timestamp at the exact moment where the first portal is used. If this one-way teleportation wasn't adjusted for multiplayer, it would be a main source of multiplayer breakage – and it usually is in other content when campaign creators forget to keep the players together.
akortunov wrote: 20 Jul 2020, 15:29 "Vast majority" of Mororwind's quest have the "bring N count of X object", "make GetDeadCount of X actor to be greater than X" or "talk to X NPC to get given journal entry" template. Issues start when content creator tries to do something more complex.
There's no good reason why complex scripting in OpenMW can't just be multiplayer-friendly by default considering the project has advanced multiplayer already.
akortunov wrote: 20 Jul 2020, 15:29 Your point is that most of new content will be multiplayer-friendly because it is very hard to make multiplayer-unfriendly content for OpenMW (complex plots, scripts, cutscenes, teleportation, blocked cells, etc. as in SureAI mods) in its current state.
My point is that OpenMW will not always be in its current state, and our long-term task technically is to allow to create mentioned content easier.
No, my point isn't tied to OpenMW's current state, it's tied to the fact that OpenMW is always going to be appealing primarily for its open world capabilities, as well as the fact that content for all Bethesda Game Studios games has a huge bias towards open world gameplay.
akortunov wrote: 20 Jul 2020, 15:29 And basically an answer to raevol's question is that mentioned "single-player-only mod developer" is going to create multiplayer-compatible content automatically as far as he follows Morrowind's approach (mod is a set of locations with some mobs and simple quests, or just a replacer for something). If he starts to deviate from this approach and do a lot of complex scripts, he should become a multiplayer mod developer. If he tells "personally I do not use multiplayer, so do not ask me to support it", an another multiplayer mod developer later should adapt mod to multiplayer every time when new mod version is released.
Any complex scripting can work perfectly fine in multiplayer, as long as a few guidelines are followed:

1) When you need to teleport one player and keep them somewhere, teleport all the players
2) When you need to freeze one player in place or prevent them from interacting with something, do it for all players
3) Realize that players can split up unless you force them to stay together

The "multiplayer mod developers" coming up with fixes don't have a hard task if the scripting system itself is shared between singleplayer and multiplayer. You could even have a single multiplayer compatibility mod that solves most complex scripting problems by forcing players to follow their leader and only allowing the leader to interact with NPCs and activators.
User avatar
raevol
Posts: 3093
Joined: 07 Aug 2011, 01:12
Location: Caldera

Re: OpenMW's Roadmap and Future

Post by raevol »

So let's get back on track with getting TES3MP-Lua into OpenMW. Can we start looking over that code? @psi29a what do we need to do from your perspective to get this moving?
davidcernat
Posts: 256
Joined: 19 Jul 2016, 01:02

Re: OpenMW's Roadmap and Future

Post by davidcernat »

I need to release the version of TES3MP I've been working on for the past year. Then I'll make whatever changes psi29a wants me to make to it so it moves towards mergeability.
User avatar
AnyOldName3
Posts: 2668
Joined: 26 Nov 2015, 03:25

Re: OpenMW's Roadmap and Future

Post by AnyOldName3 »

Basically we just need to keep David alive until it's merged.
CMAugust
Posts: 285
Joined: 10 Jan 2016, 00:13

Re: OpenMW's Roadmap and Future

Post by CMAugust »

Meanwhile, does anyone have their own thoughts about the future of OpenMW engine and content mentioned earlier in the thread?

So far, the general OpenMW policy is "OpenMW is the engine project; we do not provide assets." But if we could wave a magic wand and have the fully dehardcoded OpenMW in front of us, what would it look like?

When dehardcoded, virtually everything will become an asset. Including a lot of existing features and improvements users have enjoyed over the years, currently handled in-engine. What happens to them? "They become outside of engine scope. We do not provide assets." It is easy to imagine those same features must now be looked for elsewhere as loose files, the user navigating possibly dozens of download pages on mod websites, just to enable present OpenMW feature set.

Does that sound like a good future? It doesn't to me. It sounds tedious. OpenMW is somewhat renowned as a great user experience out of the box, but soon there will be many more boxes to unpack.

Is this something users will "have to" do eventually? I don't think it is. I am all for dehardcoding the engine, but the line of what content is provided by OpenMW release must be less strict. At the very least, any features that already exist must not "go away" as engine logic is dehardcoded, but provided as (for example) a script package that is virtually identical to the original hardcoded feature as far as the user is concerned. Yes, a line must be drawn somewhere, but that ought to be debated on a case-by-case basis rather than eschewing assets entirely.

It would be good to get more opinions on this matter, sooner rather than later. Dehardcoding may seem like a distant dream, but lua scripting is coming, and that is like a dehardcoding in many ways.
User avatar
AnyOldName3
Posts: 2668
Joined: 26 Nov 2015, 03:25

Re: OpenMW's Roadmap and Future

Post by AnyOldName3 »

We're pretty obviously going to have to provide a Morrowind.omwgame which includes all the stuff that Morrowind relies on that's no longer hardcoded into the engine. Optional stuff, like what's now in the launcher's advanced tab, is less clear, though. Maybe we could provide a bunch of omwaddons that implement what we used to provide, maybe we could let modders take over, or maybe we could let mods have settings a bit like the MCM modders implemented into later games and expose everything we used to provide through that.
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 »

AnyOldName3 wrote: 21 Jul 2020, 03:06 We're pretty obviously going to have to provide a Morrowind.omwgame which includes all the stuff that Morrowind relies on that's no longer hardcoded into the engine. Optional stuff, like what's now in the launcher's advanced tab, is less clear, though. Maybe we could provide a bunch of omwaddons that implement what we used to provide, maybe we could let modders take over, or maybe we could let mods have settings a bit like the MCM modders implemented into later games and expose everything we used to provide through that.
This was also implied/alluded to when dealing with distant land (Object Paging) in that there is vanilla content that have an initial state and the scripts are triggered until you entered the cell. The fog of the original engine would keep this hidden to you: horizontal banners, villages that shouldn't exist (yet) and other anomalies.

The correct way to deal with this is have our own Morrowind.omwgame (or omwaddon) that cover these issues. This is something that can be done now and eventually carry over to post 1.0 de-hardcoding when we push Morrowind specific things out of the engine.

When our openmw-lua comes online, we'll have to take a hard look at additional QoL things to see if they can't pushed into omwaddons as well. These are things that OpenMW should provide by default and left up to the user to see if they want. It also means a decision about what is content and what is engine related.

This however is putting the cart in front of the horse; for anyone looking for "Hey, how do we get OpenMW to 1.0" then have a look at the issue tracker. We still have quit a few blockers to tackle first before we can consider making a milestone release. Along the way we'll end up getting more issues being made from concerned and well meaning users: "That's not my Morrowind". We'll have to take care of those in a case-by-case basis. We can't please everyone and we shouldn't because it will end up turning OpenMW into a game of 'wack-a-mole' trying to cover every issue in the short term while introducing subtle bugs in the long time. That also includes well meaning contributors wanting to add something (think MWSE 1.0 extensions or a fix for a Morrowind exploit that we don't want to replicate). We can and should be able to push back and be able to say "no" to users and at minimal to give a reason as to why.

For actual OpenMW regressions and bugs, those will also likely be blockers for 1.0; but there will be others that come up where we have to make a judgement call for inclusion or not.

We've been pretty bad in letting PR and MRs hang around without making a decision to pull in or not; some people have left after not showing interest. I've had to contact some of these people in private to motivate them to continue their work because I see merit in it and it would be a pity to see all that work wasted. I do see that developers are great at helping and guiding other new developers, which is great! However that last extra kilometer seems to fall short like we're waiting for something or even just that it is up to me to hit the merge button when you are also capable of doing so.

How do you guys (developers) feel about the above? Is there something else going on perhaps? Those with merge access, I trust to have the common sense to know what is good for OpenMW and that' we're all pulling on the same rope. Perhaps we should be more explicit about why there is no merge (yet) to MRs of new developers. It also isn't always clear to me when my input is needed unless someone tags/pings me with @psi29a asking for guidance.

What happens sometimes in a MR is a question is asked, it is discussed, then there is no clear indication to the author what the best way forward is. Sometimes they ask for: "what can I do to make my MR better" but we should get into the mindset automatically. That at the end of a technical discussion that a 'best way forward' for the MR's author.

As for TES3MP merging, that is still post 1.0 but the openmw-lua part can go in before that. That is not a blocker but a nice to have and I trust David to also have OpenMW's best interest at heart. We communicate quite a bit and we've also synced with Zini on a few things. The genera idea is to make sure that the single-player experience does not change even though under the hood it is running a local server and this also means keeping all platforms in mind while doing so, meaning that the overhead of such a setup is minimal. Before this can be undertaken though, we have to bring openmw-lua in first and make sure that it works both single-player as we know it now and in tes3mp, that they are using the same subsystem.

I think we could all do with being nicer to each other. There seems to be quite a bit of frustration when there shouldn't be any and it comes out as aggression or even just perceived aggression because English is the first language. We live in rather stressful times and I think it couldn't hurt to be nicer and more forgiving towards each other and not short. Again, if there are particular issues, bring them up as a new topic to be talked about or perhaps a group chat in private. Burn-out is a real thing. If necessary, perhaps take a step back from OpenMW and focus on something else for awhile might help?
Post Reply