Recent Negativity Regarding OpenMW

General discussion regarding the OpenMW project.
For technical support, please use the Support subforum.
User avatar
psi29a
Posts: 5356
Joined: 29 Sep 2011, 10:13
Location: Belgium
Gitlab profile: https://gitlab.com/psi29a/
Contact:

Re: Recent Negativity Regarding OpenMW

Post by psi29a »

Progress is directly correlated to how many people are working on the project. Some times it's not very much and sometimes it comes in bursts. FOSS isn't like a game development studio where you have deadlines and people who are salleried to work on a project for 40~60 hours a weak with many others. That isn't exactly a fair comparison and people need to really have a reality check on exactly what OpenMW is and how it is developed.

It's not like OpenMW's development has ever stopped, or that any goal posts have been moved, it keeps being worked on regardless of anyone's time table by people willing to work on it... whenever they have the time to work on it. We all have lives outside of the project, we're not being paid to work on full time nor part-time so there is no other reason to work on OpenMW other than just the love of it. We're not a large team but OpenMW is a rather large project.

Have a look here: https://www.openhub.net/p/openmw/estimated_cost
OpenMW is the sum of about 61 man years, at an estimated cost of 3 million dollars. That is how much _our_ time working on the project is worth.

Every bit of work that has been put into OpenMW is freely and openly available for anyone.

You want to play Morrowind from begin to end, great, you're able to do so, right now! We've hit that goal a long time ago.

The real issue with 1.0 release isn't that OpenMW isn't done, but that OpenMW-CS is lagging behind development! We want to give modders and content creators the tools they need right out the box with 1.0 but so far, not many people have taken up the call to work on the editor. I hope that changes soon.

If you or anyone else wants to help OpenMW reach 1.0, then help. Many hands make short work.
User avatar
AnyOldName3
Posts: 2668
Joined: 26 Nov 2015, 03:25

Re: Recent Negativity Regarding OpenMW

Post by AnyOldName3 »

As if half the team isn't going to hang up their hats when 1.0 comes out, like what happens with so many of these type projects... :| Leaving behind a skeleton crew to attempt to enshrine MGE graphics in the code, a futile effort that will inevitably sputter and fail.
I'm pretty sure almost all of us developers are here specifically because of what we'll be able to do after 1.0 - Morrowind already works on Windows and under Wine, so there'd be no point in any of us being here if we weren't. The cool stuff only happens post-1.0, so that's likely to bring us more people rather than fewer.

Also, aiming for MGE graphics is not something we're doing - it's a pile of hacks and can't go much further than it already has. For example, except for 2-3 bugs, the shadows in my PR are leagues ahead of what MGE can do and MGE can't get much better than it already is - it's already a mountain of hacks trying to hijack Morrowind's 2001-style systems, and eventually you run out of things you can do with a hack. Besides, the only things MGE has that we don't are:
  • Shadows. A PR exists and is almost finished, so the wait is almost over.
  • Post-process shaders. Actually not too complicated to do, and one of the things I'm most likely to work on once shadows are merged.
  • Optional fancy water shaders. Drummyfish is working on improving what we've already got to add dynamic ripple simulation like vanilla Morrowind has, and then making prettier water should be fairly simple. In fact, because the GLSL source is included with OpenMW releases, anyone can switch out their water shader as if it was a mod, so anyone who prefers MGE XE's water shaders can pick their favourite and convert it from HLSL to GLSL and it'll just work.
  • Distant statics. MGE XE requires these to be pre-generated and requires a complicated list of every static that behaves weirdly, and even then, requires different lists and static regeneration for different stages of the game. A better solution is possible, but working out what it is will take some time and require a lot of discussion. In the meantime, as Akortunov recently added ways to limit the processing distance of AI and animation etc., it should be much faster and safer to use an increased ECLD with OpenMW than MGE XE (I'm not sure, but it might even be risk-free, but checking with Akortunov is the best bet).
  • PBR. So far three mods use MGE XE's PBR prototype, so it's not a priority yet, although it's something we'll eventually support. I've already written a PBR shader that wouldn't require much adaptation that we could use as a basis, for example. One of the PBR prototype's features - proper normal mapping - is something OpenMW had before MGE XE did, though.
We're not going to delay 1.0 to wait for the graphics to get better, but we're also very unlikely to hold off a graphics upgrade just because OpenMW isn't at 1.0 yet. If you think we're not up to par, you could come and help, or if you can't do that and you have spare money and nothing you need to do with it, give it to one of the OpenMW developers with a Patreon.
User avatar
silentthief
Posts: 456
Joined: 18 Apr 2013, 01:20
Location: Currently traversing the Ascadian Isles

Re: Recent Negativity Regarding OpenMW

Post by silentthief »

There was a negative nancy who posted about the slowing down of releases saying something along the lines of "OpenMW is dead", and then like 2 weeks later we had the largest bug fix and feature update in the history of the project (This was like at version 42 I think). The reason that I am saying this is that things like this take time. I am impatient about it too because I am a huge fan of Morrowind. The devs feel the passion for this game as strongly as you do. You only need to go back in this very topic and see the heated exchanges about this.

The game engine should be version 1.0 or like just pre-1.0 but like was mentioned earlier in this conversation the CS is lagging behind. The devs don't want people to have to use Bethesda's CS or even 3rd party tools which is why the delay.

Unfortunately the development of the project is based upon fans doing this as a hobby outside of family/jobs/real life. If you want to help speed it up, then you can help with coding (either yourself or recommend devs if you know any, or as a bug/feature tester with new releases).

ST the lagging
charlieg
Posts: 50
Joined: 30 Jun 2015, 14:17

Re: Recent Negativity Regarding OpenMW

Post by charlieg »

If OpenMW is dead then there's some lit necromancers around here!
natinnet
Posts: 4
Joined: 12 Oct 2018, 21:53

Re: Recent Negativity Regarding OpenMW

Post by natinnet »

A bit of opinion from an end-user. You are free to consider it as negativity.

First, each time when you, developers answer on feature request "this feature may be implemented in post 1.0 version" we, users, hear "NEVER". Really, OpenMW 1.0 sounds even worse, then GIMP 3.0. So it looks like as plain ignoring of user's requests. And then you are surprised with "NEGATIVITY", heh.
AnyOldName3 wrote: 18 Nov 2018, 23:05
  • Shadows. A PR exists and is almost finished, so the wait is almost over.
Yes, the absence of shadows is a problem. Actually, it's a huge problem for your... reputation as developers, because you are promising to add this feature in the next version for years. Now it's in the same status, as I see.
AnyOldName3 wrote: 18 Nov 2018, 23:05
  • PBR. So far three mods use MGE XE's PBR prototype, so it's not a priority yet, although it's something we'll eventually support. I've already written a PBR shader that wouldn't require much adaptation that we could use as a basis, for example. One of the PBR prototype's features - proper normal mapping - is something OpenMW had before MGE XE did, though.
The main feature of PBR pipeline is not more realistic render, honestly. Most people can't see the difference between PBR and legacy shaders, if the shaders were made by a professional. But making good shaders in PBR pipeline is much more simple and intuitive process for the beginners.

Normal maps isn't a part of "PBR brend", legacy pipeline has them too. You are wrong with it, I just don't know any modern engine which can't work with normals.

Now, my personal reasons for discontent. Till closer acquaintance with your engine I wanted to make a character model for my portfolio in your engine AND Marmoset Toolbag. Now I'm going to make this char ONLY for Marmoset. The last screenshot in OpenMW:
Image

As you may see, I've made a high-poly sculpt for the body, then retopology and rig. I was beginning to make textures, but...

So, the list, no, THE LIST:

1. Shadows, again. As a 3d-artist I want a nice render. Without shadows... You understand, yes? Just try to find a modern engine without shadows.

2. Subsurface scattering. Really, human skin material without it looks not so good.

3. And, last, but not least. Actually, I thought it was already implemented. So, at first I thought it was another my mistake. You may see, I already have a problem even with usual normal maps... I tried different order of postfixes, I tried different file formats, then I went to your wiki, found nothig, went forum, searched keywords... And then I found this topic...

WTF, DEVS, WHY IS IT NOT DESCRIBED PROPERLY IN YOUR WIKI?

NEGATIVITY!

NEGATIVITY!

NEGATIVITY!


Actually, I'd spent a half of an hour before I found out why I should not use your engine. :D Look, at their site Marmoset publish a plugin, which allows users to work with 4 different sets of detail maps at one object. In the same time, you can't managed properly even one set of detail maps, fie. Moreover, you think thats it's completely OK. And the feature will be never implemented. Sorry, you wrote "post 1.0", but it's the same.

Here you can see why this feature is so important for such materials as skin and fabric, so it's not a my whim, it's a real problem: without detail normal&gloss maps I can't make a nice close-up.

So, AnyOldName3, it was you, who answered that the situation is perfectly normal :D That's why:

SHAME ON YOU!

NEGATIVITY!

NEGATIVITY!

NEGATIVITY!


And now look at your Example Suite forum section. You could easy solve the issue#3185 with my mesh, at least partially. I'm not going to make money with the model, and I could share it, maybe... But there is no sense for me to make a model for an engine, which is obsolete even before it's own release. And I don't think that there are lot of 3d-artists, who think different. So good luck with learning of 3d-modeling!

P.S. Untextured mesh of Vivec looks at you with

NEGATIVITY!

P.P.S. Feel free to ban me for the

NEGATIVITY!

P.P.P.S. Maybe, I'll need my account "post 1.0 version", so it looks like

NEVER!

P.P.P.P.S. Actually, add the description of the behavior of your so called "detail maps" to this page. It was not so fun to search the answer at the forum.
User avatar
psi29a
Posts: 5356
Joined: 29 Sep 2011, 10:13
Location: Belgium
Gitlab profile: https://gitlab.com/psi29a/
Contact:

Re: Recent Negativity Regarding OpenMW

Post by psi29a »

Thanks for the feedback! :) No, we're not going to ban you for giving criticism on bits we can work on it. Some of the things you have pointed out are things you could help with by updating the wiki and posting about how you got things to work. Even creating an issues on our tracker, attaching any assets you are willing to part with allowing us to support it. Going at it by yourself is hard, make use of the tools we provide to help us help you. :)

That being said, there are some bits that we really can't help and it's sad that things out of our control are then seen negatively.

For example the comparison of OpenMW with Marmoset [1], Marmoset devs actually get paid for their work and consider it their primary source of income. So I perfectly understand the sentiment behind the fact that people are looking around in comparison and saying that X is better than OpenMW because of feature Y. Here's the deal with Marmoset, you have to pay for a license or if you're an academic or freelancer, you get a 'free' license to use it but you don't have access to the source and are beholden to Marmoset. More or less the same deal with Unity and UE. I mean, these guys are professionals who work on this stuff _and_ get paid. :)

The problem is that OpenMW can only move as fast as people who are willing to work on it. OpenMW can't be all things to everyone, especially with so small a development team of volunteer developers who have full time jobs elsewhere. That is just a fact. If people want the project to move faster then help or help us find people who would be willing to help, it's really that simple.

Since the goal of 1.0 is to be 1:1 with the original Morrowind with official expansions, all the extra bells and whistles is icing on the cake. So if you are looking for more than that, then yeah, as you have done, you go with something else. This is by no means me saying: piss off. Just that we're still a work-in-progress and can't compete with other proprietary engines and tools without similar resources.

[1] https://marmoset.co/toolbag/
User avatar
lysol
Posts: 1513
Joined: 26 Mar 2013, 01:48
Location: Sweden

Re: Recent Negativity Regarding OpenMW

Post by lysol »

Does anyone even realize how dumb it is to expand massively on cool stuff before even the foundation is done? We'd NEVER see a finished OpenMW this way.

And seriously. You can build shadows yourself already if you want to try them out. Think AnyOldName should be ashamed for working too slow? Well give him some money on patreon, contribute yourself to help him out, or stop saying he should be ashamed.
And the feature will be never implemented. Sorry, you wrote "post 1.0", but it's the same.
See my first paragraph. Also, how many lines of code did you contribute? None? Did you donate anything? No? Then why are you behaving like this?
P.P.P.P.S. Actually, add the description of the behavior of your so called "detail maps" to this page. It was not so fun to search the answer at the forum.
This being an open source project, you can actually do this yourself. Also, I'm quiiite sure this behavior just mimics Morrowinds behavior, because, you know, it basically has to.

Also, this thread was started quite a long time ago, and was relevant to a discussion we had back then and a major dev leaving the project. We're not in that situation anymore. Can we just let this thread die? If you have comments about OpenMW sucking too hard, start your own thread. My personal opinion.

Also, don't mind me. For a polite reply, see psi29a above.

EDIT: Sorry for getting a bit upset. I edited some stuff.
User avatar
l1lartur0
Posts: 11
Joined: 06 May 2018, 07:34

Re: Recent Negativity Regarding OpenMW

Post by l1lartur0 »

psi29a wrote: 22 Nov 2018, 09:35The problem is that OpenMW can only move as fast as people who are willing to work on it. OpenMW can't be all things to everyone, especially with so small a development team of volunteer developers who have full time jobs elsewhere. That is just a fact. If people want the project to move faster then help or help us find people who would be willing to help, it's really that simple.

Since the goal of 1.0 is to be 1:1 with the original Morrowind with official expansions, all the extra bells and whistles is icing on the cake. So if you are looking for more than that, then yeah, as you have done, you go with something else. This is by no means me saying: piss off. Just that we're still a work-in-progress and can't compete with other proprietary engines and tools.

[1] https://marmoset.co/toolbag/
lysol wrote: 22 Nov 2018, 09:57 Does anyone even realize how dumb it is to expand massively on cool stuff before even the foundation is done? We'd NEVER see a finished OpenMW this way.

And seriously. You can build shadows yourself already if you want to try them out. Think AnyOldName should be ashamed for working too slow? Well give him some money on patreon, contribute yourself to help him out, or stop saying he should be ashamed.
Indeed.

Well, for whatever it's worth, this end user appreciates the work that has been put into this project by everyone involved, and is pleased as Punch with OpenMW in its current state (actually since 0.43, when I decided to give Morrowind a go again). I'm utilizing around 150 mods and haven't had one hiccup in regards to performance or stability... something which I've never experienced with the "bog-standard" Morrowind, Oblivion, or Skyrim (both 32 and 64 bit regarding the latter). Hope that you all keep doing what you're doing- when you can, obviously- and thanks for sharing the fruits of your labour!
User avatar
psi29a
Posts: 5356
Joined: 29 Sep 2011, 10:13
Location: Belgium
Gitlab profile: https://gitlab.com/psi29a/
Contact:

Re: Recent Negativity Regarding OpenMW

Post by psi29a »

I don't mind this being the 'shit on OpenMW' thread, so long as something good comes out of it:
* allow players, users, content creators the ability to vent their frustrations without fear of being banned, so long as they are respectful [1]
* address issues that people have about the engine
* refine what OpenMW is, can be and won't be
* find any bugs that have might been overlooked
* give an indication to developers where to focus their energy

As I said, just be respectful. It's obvious that people have deep seated emotions about this project, otherwise they wouldn't take the time to post about it. OpenMW should be receptive to criticism, but with all things, it is hard to not take things personal. Something everyone needs to work on.

[1] trolling behavior is not in good faith and will be banned
User avatar
AnyOldName3
Posts: 2668
Joined: 26 Nov 2015, 03:25

Re: Recent Negativity Regarding OpenMW

Post by AnyOldName3 »

I think a lot of this stems from you being mistaken about what OpenMW actually is. First and foremost, it's a recreation of the engine from Bethesda's 2002 RPG Morrowind. Getting to a point of feature parity with that engine has always been our primary goal. Once that's been achieved, we can expand our goals and scope to include things such as:
  • Being a platform to mod the bejeezus out of Morrowind until it's unrecognisable, going far beyond what can currently be done with MGE XE, MWSE, ESPs and asset replacements.
  • Being an engine for people to make new Morrowind-like games.
  • Eventually, maybe allowing the same opportunities for other Bethesda games, such as Oblivion and Fallout 4.
Having graphics that rival UE4 is one small part of the first bullet point, but it's definitely not something that takes priority while there are still parts of the vanilla Morrowind experience that we can't replicate.
natinnet wrote: 22 Nov 2018, 01:30 A bit of opinion from an end-user. You are free to consider it as negativity.

First, each time when you, developers answer on feature request "this feature may be implemented in post 1.0 version" we, users, hear "NEVER". Really, OpenMW 1.0 sounds even worse, then GIMP 3.0. So it looks like as plain ignoring of user's requests. And then you are surprised with "NEGATIVITY", heh.
Some people might think post-1.0 means never, but we're not the thought police. We can't stop people thinking wrong things.
AnyOldName3 wrote: 18 Nov 2018, 23:05
  • Shadows. A PR exists and is almost finished, so the wait is almost over.
Yes, the absence of shadows is a problem. Actually, it's a huge problem for your... reputation as developers, because you are promising to add this feature in the next version for years. Now it's in the same status, as I see.
I've only been working on shadows for a year, so how on earth could we have been promising they're coming in the next version for years? You might have read something someone that isn't us wrote, but they were either mistaken or lying. It's bad that it hurts our reputation, but again, without being the thought police, we can't stop people thinking and saying wrong things.
AnyOldName3 wrote: 18 Nov 2018, 23:05
  • PBR. So far three mods use MGE XE's PBR prototype, so it's not a priority yet, although it's something we'll eventually support. I've already written a PBR shader that wouldn't require much adaptation that we could use as a basis, for example. One of the PBR prototype's features - proper normal mapping - is something OpenMW had before MGE XE did, though.
The main feature of PBR pipeline is not more realistic render, honestly. Most people can't see the difference between PBR and legacy shaders, if the shaders were made by a professional. But making good shaders in PBR pipeline is much more simple and intuitive process for the beginners.
You might want to go and see an optometrist. You also might want to go and look up what a shader is, as I think you're getting confused with materials, the input data that shaders use. Non-PBR games have all been using the same Blinn-Phong shader for decades, so while you can add extra little effects on top of it, the inherent problems with that approach are always going to be there, and unless you want to redo all of your textures and material parameters for every situation an asset will get used in, normal people will be able to see that there's a difference, even if they can't highlight what, specifically, the difference is.

Normal maps isn't a part of "PBR brend", legacy pipeline has them too. You are wrong with it, I just don't know any modern engine which can't work with normals.
The vanilla Morrowind engine outright doesn't support normal mapping. MGE XE adds partial support, in that normal maps are used for the environment map reflections, but not for the basic lighting. MGE XE's PBR prototype adds proper normal map support for all effects. If you try and compare us to modern engines, you're not going to have a great time as we've mostly got to maintain compatibility with a not-very-good engine from 2002.

Now, my personal reasons for discontent. Till closer acquaintance with your engine I wanted to make a character model for my portfolio in your engine AND Marmoset Toolbag. Now I'm going to make this char ONLY for Marmoset. The last screenshot in OpenMW:
Image

As you may see, I've made a high-poly sculpt for the body, then retopology and rig. I was beginning to make textures, but...

So, the list, no, THE LIST:

1. Shadows, again. As a 3d-artist I want a nice render. Without shadows... You understand, yes? Just try to find a modern engine without shadows.
That's fair, but they're just not quite done yet.
2. Subsurface scattering. Really, human skin material without it looks not so good.
No Elder Scrolls or Fallout games have subsurface scattering without mods. We need to remain compatible with Morrowind's assets, and they don't have this at all.
3. And, last, but not least. Actually, I thought it was already implemented. So, at first I thought it was another my mistake. You may see, I already have a problem even with usual normal maps... I tried different order of postfixes, I tried different file formats, then I went to your wiki, found nothig, went forum, searched keywords... And then I found this topic...
I'm not 100% sure what you're saying didn't work as you've said some fixes you attempted, but not the basic goal. If you were trying to use a standard RGB-as-xyz normal map, provided you've not done something silly, it should 'just work'. If you were trying to use an RGBA-as-xyz-plus-parallax-height normal map, that would also 'just work'. Both of these are the standard formats in Morrowind's sequels, so while UE4 or Unity 5 might do something different, we're not doing anything stupid.

If you were trying to use a normal map as a detail map, because the original Morrowind engine doesn't support normal maps in the first place, it's not going to support them when used as a detail map. You should probably familiarise yourself with how the original engine expects assets to be formatted before trying to add OpenMW-specific enhancements.
WTF, DEVS, WHY IS IT NOT DESCRIBED PROPERLY IN YOUR WIKI?
Our Wiki and ReadTheDocs describe OpenMW-specific behaviour. We handle things like detail maps exactly the same as the original Morrowind engine does, and a lot of our documentation assumes that the reader has at least a vague idea of what the original engine does.

NEGATIVITY!

NEGATIVITY!

NEGATIVITY!


Actually, I'd spent a half of an hour before I found out why I should not use your engine. :D Look, at their site Marmoset publish a plugin, which allows users to work with 4 different sets of detail maps at one object. In the same time, you can't managed properly even one set of detail maps, fie. Moreover, you think thats it's completely OK.
So your issue is just that you want a game engine for Morrowind to support fairly esoteric features that can't be done in any of Bethesda Game Studio's other games today and that aren't even supported natively (i.e. without a plugin) by the tool you're heralding as what we should be aiming for parity with.
And the feature will be never implemented. Sorry, you wrote "post 1.0", but it's the same.
I suppose that there's a non-zero chance that you'll get a Darwin Award before it's done...

Note: this may be on the wrong side of Wheaton's Law, so I can remove this bit if needed
Here you can see why this feature is so important for such materials as skin and fabric, so it's not a my whim, it's a real problem: without detail normal&gloss maps I can't make a nice close-up.
I think we all agree that having normal and specular components of detail maps would be a nice thing to have eventually, but as it would require abandoning support for loads of hardware which OpenMW currently works on and there are other, more pressing graphical features we've not got yet that are actually supported by things like MGE XE or Fallout 4, those are the priorities.
So, AnyOldName3, it was you, who answered that the situation is perfectly normal :D That's why:

SHAME ON YOU!

NEGATIVITY!

NEGATIVITY!

NEGATIVITY!


And now look at your Example Suite forum section. You could easy solve the issue#3185 with my mesh, at least partially. I'm not going to make money with the model, and I could share it, maybe... But there is no sense for me to make a model for an engine, which is obsolete even before it's own release. And I don't think that there are lot of 3d-artists, who think different. So good luck with learning of 3d-modeling!
Most things don't work properly before they're released. That's why they're not released. You get the basics (in this case, being able to play Morrowind with at least parity to the original engine) sorted out and only work on the extras if they're especially easy or the basics are already done. There's absolutely no point in having fancy features without the basics working as the basics are the most important part. You wouldn't use a modelling and animation package that bragged about its ability to use machine learning to create realistic walking and running animations if it wasn't capable of loading, saving or creating a skeleton to use the animations with yet, and people wanting to play Morrowind aren't going to switch to OpenMW before it can do Morrowind at least as well as the vanilla engine can.
P.S. Untextured mesh of Vivec looks at you with

NEGATIVITY!

P.P.S. Feel free to ban me for the

NEGATIVITY!

P.P.P.S. Maybe, I'll need my account "post 1.0 version", so it looks like

NEVER!

P.P.P.P.S. Actually, add the description of the behavior of your so called "detail maps" to this page. It was not so fun to search the answer at the forum.
That section covers OpenMW-specific graphics stuff, but our detail maps are exactly the same as vanilla Morrowind (and vanilla Oblivion, but I think they got rid of them for Fallout 3, New Vegas, Skyrim and Fallout 4 & 76). There are plenty of other resources on how to make Morrowind-compatible assets that have existed for years and probably do a just as good a job of explaining things as we would. We might benefit from adding a link, though.
Locked