The only way to replace Vanilla Engine (IMHO)

General discussion regarding the OpenMW project.
For technical support, please use the Support subforum.
TamrielCitizen
Posts: 22
Joined: 04 Dec 2018, 06:25

Re: The only way to replace Vanilla Engine (IMHO)

Post by TamrielCitizen »

To moderators: I just wanted to reply to a comment in the other thread, but while I was typing it got locked, so I thought to post my reply here as I think my reply is relevant to this thread. If you disagree and think this comment shouldn't be here, I don't mind if you move or even delete it (I hope you won't, though), I just hope you understand that I don't intend anything bad by it.
sonicboom12345 wrote: 06 Dec 2018, 04:04 I can download the Morrowind Overhaul, and in fifteen minutes, have a kitted-out Morrowind with 100+ mods, model and texture upgrades, and graphics and gameplay extension via MGE and MWSE. Unless I'm a hardcore Morrowind fan with an extensive knowledge of the user-driven landscape, I'm not going to recognize all of the ways OpenMW is superior, because most of OpenMW's improvements are under the hood. I'm probably just going to download the Morrowind Overhaul and stop at that because 1. it looks prettier and 2. it's the path of least resistance. That's what you need to overcome, and that's the reason I don't necessarily agree that "a one-to-one recreation of the original Morrowind" is necessarily the best goal post to strive for.
What if at some point in the future Morrowind Overhaul works in OpenMW? Obviously, someone would have to rewrite the MO's installer for OpenMW first, I don't think there is a way to avoid that, but after someone does, for the end-user it would be just a matter of downloading a different installer. As for why would the end-user spend additional effort to install/configure OpenMW, see my opening post in this thread, specifically my view on how to get rid of reason №5. TL;DR: If OpenMW offers a noticeable increase in performance (which is especially important for Morrowind Overhaul users), it would be worth the additional effort to install/configure it, even if the user in question doesn't care about what else OpenMW has to offer. Of course, I am speaking of distant future here, but I got the impression that you, too, are concerned about the eventual goal here.

P.S. I have no idea whether spending additional effort on PR, apart from maybe making some additions to the FAQ, could benefit the project. I am not very experienced with communication.
Also, I don't see what's wrong with OpenMW's website. Is it a matter of "modern fashion"? I think this is off-topic, though.
User avatar
lysol
Posts: 1513
Joined: 26 Mar 2013, 01:48
Location: Sweden

Re: The only way to replace Vanilla Engine (IMHO)

Post by lysol »

MGSO already works for the most part with the regular mods it contains. You just need to patch the fake bump mapped mods basically. Maybe something else too, haven't tried it myself, but several people have reported that it works except for shiny rocks and similar.

So what is left? Well, MGSO installs MGE XE and makes some settings for it. Do we need this? No! Because when OpenMW actually have the same and better graphical features as MGE XE, why would you ever want to download a mod to get these features? No, they will be easily accessible from within the ingame options menu.
davidcernat
Posts: 256
Joined: 19 Jul 2016, 01:02

Re: The only way to replace Vanilla Engine (IMHO)

Post by davidcernat »

TamrielCitizen wrote: 04 Dec 2018, 19:19 [*]Multiplayer (we are still in the future). This is actually a feature that could convince some of the people who were otherwise unconvinced. However, I'd argue that a lot of people don't care about multiplayer, and this is likely even more true among those who play Morrowind than among those who play games in general.
I'm not sure what "we are still in the future" is supposed to mean. It's been possible to play the entire game from start to finish in the multiplayer branch for the past year and a half, unless you suppose videos like this one were created by our collective imagination.

As for "I'd argue that a lot of people don't care about multiplayer," I'd argue that a lot of people do care about it too. The last time I checked, the TES3MP Discord server had 3189 members, which is more than twice as much as any other Discord server related to Morrowind and only 1000 less than the main Discord server for the entire Elder Scrolls series.
TamrielCitizen
Posts: 22
Joined: 04 Dec 2018, 06:25

Re: The only way to replace Vanilla Engine (IMHO)

Post by TamrielCitizen »

lysol wrote: 06 Dec 2018, 09:13 MGSO already works for the most part with the regular mods it contains. You just need to patch the fake bump mapped mods basically. Maybe something else too, haven't tried it myself, but several people have reported that it works except for shiny rocks and similar.

So what is left? Well, MGSO installs MGE XE and makes some settings for it. Do we need this? No! Because when OpenMW actually have the same and better graphical features as MGE XE, why would you ever want to download a mod to get these features? No, they will be easily accessible from within the ingame options menu.
Well, as I said, it's probably inevitable that someone would have to rewrite MGSO's installer for OpenMW, so even if those fake bump mapped mods can't be automatically patched by OpenMW (or can they?), they would probably be patched by that new installer. As for MGE XE settings, obviously the new installer would be able to just configure appropriate OpenMW settings instead.
TamrielCitizen
Posts: 22
Joined: 04 Dec 2018, 06:25

Re: The only way to replace Vanilla Engine (IMHO)

Post by TamrielCitizen »

davidcernat wrote: 06 Dec 2018, 09:21
TamrielCitizen wrote: 04 Dec 2018, 19:19 [*]Multiplayer (we are still in the future). This is actually a feature that could convince some of the people who were otherwise unconvinced. However, I'd argue that a lot of people don't care about multiplayer, and this is likely even more true among those who play Morrowind than among those who play games in general.
I'm not sure what "we are still in the future" is supposed to mean. It's been possible to play the entire game from start to finish in the multiplayer branch for the past year and a half, unless you suppose videos like this one were created by our collective imagination.

As for "I'd argue that a lot of people don't care about multiplayer," I'd argue that a lot of people do care about it too. The last time I checked, the TES3MP Discord server had 3189 members, which is more than twice as much as any other Discord server related to Morrowind and only 1000 less than the main Discord server for the entire Elder Scrolls series.
Well, I was talking about OpenMW's (future) main release the entire time, wasn't I? As I understand, merging the TES3MP's branch is a tentative goal for OpenMW 2.0? I know that TES3MP already works quite well (apart from some remaining bugs/issues that are being worked on), and thank you very much for your continuing work on it! :D

Of course, there are a lot of people who care. However, there are also a lot of people who don't, it's not mutually exclusive. And naturally, those who actively participate in online communication related to Morrowind, such as on Discord, are very likely to also be those who care about multiplayer (and maybe some of them even joined Discord because of TES3MP).
Sagacity
Posts: 31
Joined: 05 Mar 2019, 12:58

Re: The only way to replace Vanilla Engine (IMHO)

Post by Sagacity »

With regard to Xenuria's points on 64-bit and multi-core support, there's a lot to unpack with that.

Generally, Xenuria's complaint is that, since a large enough portion of OpenMW's code is not multithreaded, it may as well not be multithreaded at all. He makes an identical argument for 64-bit support. This is a non-insignificant point to be made, as it is true that most of the application should support these features in order for them to be worth advertising. The issue with his complaint however is that he is criticizing incomplete features of the program, specifically physics and the graphical backend, both of which are non-trivial to make multithreaded.

Specifically OpenMW does not package its own graphical APIs or its own physics backend. It uses OpenSceneGraph (which in turn uses OpenGL) and there's still ongoing talks about which physics engine to use, and whether to jump to VulkanSceneGraph. OpenSceneGraph, by virtue of using OpenGL, is not very multithreaded, and the most taxing part of the game at the moment is the graphical backend, and as such most of the game's load will land on a single core, due to this imbalance.

64-bit support has the same issue; most features of the game/engine simply would never need more than 32-bit values or registries, and as such, supporting them is superfluous. Many, in fact, are just fine with standard 4-byte or 8-byte values, and as such very little expansive 64-bit support is needed. The fact that OpenMW is a 64-bit aware application, however, is enough to make it 64-bit. The complaint that "not enough things utilize said 64-bit address space" is a superfluous and useless complaint. They don't need to.

With that said, OpenMW could use a lot of improvement in these areas. Especially the graphical portion of the engine. This is no one's fault really, since it would seem there's really only one dev working on it actively, and that's bound to cause some problems just by pure virtue of overworking him. Then there's the fact that there's no real good solution for the physics. No good open source physics engines are out there, but god help us if we start to write our own.
User avatar
AnyOldName3
Posts: 2667
Joined: 26 Nov 2015, 03:25

Re: The only way to replace Vanilla Engine (IMHO)

Post by AnyOldName3 »

Most of Xenuria's points were based on a very flawed understanding of a very stupid problem Second Life has because its developers didn't want to pay licence fees for the 64-bit version of some middleware they were using, so implemented a bonkers system that let the 64-bit client use the 32-bit version of the middleware. This caused lots of problems for obvious reasons. Xenuria, not fully understanding the issues, basically conflated slowness with his impression of that can of worms, saw our engine is sometimes slower than it ideally would be, and assumed it was for the same reasons, and when we disagreed, decided it was because of a big coverup. It took a little while for us to get to the bottom of this because there were enough phrases that sounded kind of right that we initially thought it was just another of your bog-standard end-user misunderstandings, not an advanced end-user misunderstanding.

Some of what you've said is correct, but it's actually completely unrelated to anything Xenuria said.
Sagacity
Posts: 31
Joined: 05 Mar 2019, 12:58

Re: The only way to replace Vanilla Engine (IMHO)

Post by Sagacity »

AnyOldName3 wrote: 08 Sep 2019, 20:24 Most of Xenuria's points were based on a very flawed understanding of a very stupid problem Second Life has because its developers didn't want to pay licence fees for the 64-bit version of some middleware they were using, so implemented a bonkers system that let the 64-bit client use the 32-bit version of the middleware. This caused lots of problems for obvious reasons. Xenuria, not fully understanding the issues, basically conflated slowness with his impression of that can of worms, saw our engine is sometimes slower than it ideally would be, and assumed it was for the same reasons, and when we disagreed, decided it was because of a big coverup. It took a little while for us to get to the bottom of this because there were enough phrases that sounded kind of right that we initially thought it was just another of your bog-standard end-user misunderstandings, not an advanced end-user misunderstanding.

Some of what you've said is correct, but it's actually completely unrelated to anything Xenuria said.
Thank you for the clarification. It was a crazy drama fest that I didn't follow.
Post Reply