It's really important to have graphical parity with MGE.

Anything related to PR, release planning and any other non-technical idea how to move the project forward should be discussed here.
sonicboom12345
Posts: 10
Joined: 13 Sep 2016, 23:47

It's really important to have graphical parity with MGE.

Post by sonicboom12345 » 25 Apr 2018, 11:58

We all love and support you as you move towards 1.0, but this is important. OpenMW won't become the new gold standard for playing Morrowind unless it equals or exceeds MGE graphically. As janky as it is, MGE currently stands head and shoulders above OpenMW in visual appeal. Its solution for pre-rendering distant statics enables view distances many times greater than what OpenMW is capable of delivering while also maintaining performance. I'm worried because I get the impression the core developers have set their goals for 1.0 in stone, and those goals don't necessarily include graphical fidelity to what MGE already provides. The project philosophy seems to be that 1.0 should aspire to be a faithful recreation of the vanilla Morrowind engine minus any bugs, but not necessarily anything more than that, for fear of feature creep. However, nobody plays the version of Morrowind that the team is aspiring to recreate anymore, and nobody has played that version of Morrowind in over a decade. Effort seems to be concentrated on squashing bugs, and that's great. But there doesn't seem to be a lot of energy spent on the graphics side of the house, at least not from my perspective as an outside observer.

I'm worried that if the team doesn't bring the graphics to parity with MGE, the 1.0 release is going to land with a thud. It isn't going to generate the buzz in the gaming press that it needs, it isn't going to win over streamers on Twitch, it isn't going to get downloads. I've tracked the project for a long time and brought up my concerns here and there, and usually gotten pushback for it. There are things the 1.0 release could do that would really draw eyeballs from would-be players, like including a built-in menu toggle to turn off dice roll combat. But the graphical capacity of the engine seems like something non-negotiable to me. Maybe I'm altogether mistaken and my impressions of the project are totally off-base, but if I'm not, I hope you'll take graphics to heart as something that's really, really important for 1.0.

User avatar
lysol
Posts: 1069
Joined: 26 Mar 2013, 01:48
Location: Sweden

Re: It's really important to have graphical parity with MGE.

Post by lysol » 25 Apr 2018, 13:01

Why is it important that 1.0 includes the fancy graphical features when version 1.1 might include them? If we include them for version 1.0, it only means that 1.0 will be named 0.4x instead of 1.0, and that 1.1 will be named 1.0. No other difference. Why does it matter?

EDIT: That said, shadows are being worked on right now, and when they are finished, OpenMW will look a LOT better already.
Normal mapped texture replacers, exclusive for OpenMW:
My Nexus page

magamo
Posts: 105
Joined: 28 May 2016, 17:28

Re: It's really important to have graphical parity with MGE.

Post by magamo » 25 Apr 2018, 17:32

I hate this comparison. After all, OpenMW had real normal mapping before MGE did. It has dynamic water reflections, which MGE does not. It'll be capable of reflecting shadows, which to my knowledge, MGE XE cannot do. It has the capability of rendering distant terrain at ranges far surpassing MGE with almost no loss in performance. The five major things in terms of 'graphical fidelity' that OpenMW can't really do in mainline that MGE can are: Distant statics, dynamic Shadows, better fog rendering, Effect Compositing (and dynamic shader chains) and Physical Based Rendering. These are very ancillary things compared to OpenMW's 1.0 goals of feature parity with vanilla Morrowind.

Sorry, OP, not everyone plays Morrowind with the same options as you do. A lot of what you are saying is hyperbole, and really isn't terribly useful to the conversation. I'm rather certain that the developers are aware of where the OpenMW engine is lacking in terms of modern graphical techniques. It will gain these things, some may be before 1.0, some may be after. Right now, Shadows is being worked on actively. Better fog has been worked on from time to time, but work as far as I know has stalled for the moment. Effects Compositing has had some work, but to my knowledge is not actively in development right now.

Be patient. Likely as those features are developed, they will not attain parity, but will likely immediately attain superiority. OpenMW's distant terrain for example, needs no pre-caching or seperate generation step. It looks better than the distant terrain mesh that MGE generates in my opinion (Again, just my opinion) and runs a lot faster.

Give it time. Have patience. I know people have been having to 'have patience' for years. The developers will keep plugging away. I'll keep testing. If you have knowledge of C++, or a desire to learn it, and knowledge of Openscenegraph, and/or a desire to learn it, your help would be appreciated, and you would see those features you care about take shape even faster.

User avatar
Jemolk
Posts: 106
Joined: 20 Mar 2017, 02:12
Location: Sadrith Mora

Re: It's really important to have graphical parity with MGE.

Post by Jemolk » 25 Apr 2018, 18:48

magamo wrote:
25 Apr 2018, 17:32
I hate this comparison. After all, OpenMW had real normal mapping before MGE did. It has dynamic water reflections, which MGE does not. It'll be capable of reflecting shadows, which to my knowledge, MGE XE cannot do. It has the capability of rendering distant terrain at ranges far surpassing MGE with almost no loss in performance. The five major things in terms of 'graphical fidelity' that OpenMW can't really do in mainline that MGE can are: Distant statics, dynamic Shadows, better fog rendering, Effect Compositing (and dynamic shader chains) and Physical Based Rendering. These are very ancillary things compared to OpenMW's 1.0 goals of feature parity with vanilla Morrowind.
Emphasis mine. While I broadly agree with you, distant statics is HUGE. Not having distant statics makes using distant terrain largely useless. This in particular needs to be resolved before 1.0. I'm fine with not using distant terrain at the moment myself, as are most people here, but plenty of humans in our current society are incredibly shallow and will insist on that at the very least. Releasing 1.0 without this is therefore a bad idea. We also need to be able to work with MWSE mods ASAP, even if it means converting them, though such a thing should be made as easy as possible to do. Because again, as the OP said, the vanilla engine is not used without extensions anymore, for better or worse. I've encountered modders on the Nexus who said there's no point in using OpenMW because we can't support MWSE functions or similar. This is a problem.

User avatar
lysol
Posts: 1069
Joined: 26 Mar 2013, 01:48
Location: Sweden

Re: It's really important to have graphical parity with MGE.

Post by lysol » 25 Apr 2018, 19:10

Jemolk wrote:
25 Apr 2018, 18:48
Releasing 1.0 without this is therefore a bad idea.
Still wondering why the number 1.0 is such a big deal. For OpenMW, reaching 1.0 has always (at least under Zini's lead) meant "Morrowind on OpenMW is done, time to work on cool stuff". Reaching 1.0 is therefore still a huge deal, but we need to focus on getting to this point before anything else. This goal is the number one reason OpenMW still exists really. Not having a clear goal like this could break any project.

Yes, there are already cool features implemented before 1.0. Why? Someone developed them because they felt like it and/or had the skills to do it. Want ambient occlusion before 1.0? Start developing it. I don't want to sound harsh, but it's the reality.
Normal mapped texture replacers, exclusive for OpenMW:
My Nexus page

User avatar
Jemolk
Posts: 106
Joined: 20 Mar 2017, 02:12
Location: Sadrith Mora

Re: It's really important to have graphical parity with MGE.

Post by Jemolk » 25 Apr 2018, 20:16

lysol wrote:
25 Apr 2018, 19:10
Jemolk wrote:
25 Apr 2018, 18:48
Releasing 1.0 without this is therefore a bad idea.
Still wondering why the number 1.0 is such a big deal. For OpenMW, reaching 1.0 has always (at least under Zini's lead) meant "Morrowind on OpenMW is done, time to work on cool stuff". Reaching 1.0 is therefore still a huge deal, but we need to focus on getting to this point before anything else. This goal is the number one reason OpenMW still exists really. Not having a clear goal like this could break any project.

Yes, there are already cool features implemented before 1.0. Why? Someone developed them because they felt like it and/or had the skills to do it. Want ambient occlusion before 1.0? Start developing it. I don't want to sound harsh, but it's the reality.
I'd love to, I just don't have the skills myself, not yet. And yes, the number is arbitrary. You recognize that. I recognize that. Tons of idiots out there will probably not recognize that. Too many people put too much importance on arbitrary numbers. That's the crux of the issue, really. The perpetual problem with programming (and any other task at the moment, really) is that you have to appeal to dumbasses who have no idea what they're doing but think their ignorance should carry as much weight as another's expertise. At least in the US, there's always plenty of numbnuts who think they can reasonably second-guess experts when they have no knowledge at all. They are the ones who will place arbitrary importance on the number. That's my issue with this. Now, should we have to care? No, of course not, but unless you have a better idea for spreading the word later, we'll have to deal with this inanity sooner or later.

User avatar
AnyOldName3
Posts: 1129
Joined: 26 Nov 2015, 03:25

Re: It's really important to have graphical parity with MGE.

Post by AnyOldName3 » 25 Apr 2018, 21:10

Before starting work on OpenMW, I had minimal experience with C++ (although quite a lot of experience in C# and Java), minimal experience with OpenGL, and no experience with OpenSceneGraph, but I saw that OpenMW needed something doing, and I learned the necessary skills to do it. In retrospect, I should maybe not have done that at the expense of generating an income, but it's a better approach than complaining to a group of people who are already investing a huge portion of their free time. I'm pretty sure that most developers here would really be very keen on having graphical parity with MGE at our 1.0 release, but it's an issue of manpower and time as lots of stuff requires people who know a lot about complicated things to be willing and able to make stuff work.

Open Source software thrives when passionate users get cross enough to become competant developers and retain their passion for long enough to make things work.
AnyOldName3, Master of Shadows

jmelesky
Posts: 28
Joined: 02 Mar 2017, 20:52

Re: It's really important to have graphical parity with MGE.

Post by jmelesky » 26 Apr 2018, 21:21

Slightly different take: OpenMW already has an audience, and that audience will use OpenMW even if it does not have graphical parity with MGE.

It may be that the graphical improvements (particularly distant statics, as noted earlier in the thread) are important to expanding that audience, but that doesn't mean it's required for 1.0 (or any version). 0.44 will have plenty of users, 1.0 will have plenty of users, etc, even if progress on graphics stalls out.

If nothing else, OpenMW is a vastly better way of playing on non-Windows platforms. And, honestly, it's incredibly useful for mod debugging -- not just the error outputs, but also the ability to toggle mods on and off easily to narrow down where a weird problem is coming from. There's no amount of improvements to MGE that can take that away.

Of course, I'd love it if the graphical improvements kept coming. I just don't think that should block and releases, regardless of the size of the version number.

sonicboom12345
Posts: 10
Joined: 13 Sep 2016, 23:47

Re: It's really important to have graphical parity with MGE.

Post by sonicboom12345 » 28 Apr 2018, 05:37

Why is it important that 1.0 includes the fancy graphical features when version 1.1 might include them? If we include them for version 1.0, it only means that 1.0 will be named 0.4x instead of 1.0, and that 1.1 will be named 1.0. No other difference. Why does it matter?

Meaning no disrespect to you, but these are the words of someone who's 100% in the mindset of a programmer and either doesn't understand or doesn't give a toss about marketing. :roll: The number 1.0 *means* something to people. When OpenMW hits 1.0, it will get a thousand times more eyeballs than it will when it hits 1.1, or 1.2, or 1.3. Assuming your public relations and media outreach team is on the ball (OpenMW does have an outreach team, right?) you will get featured articles on major gaming websites like Kotaku. You will get featured videos from YouTubers like LGR. You will get interest from streamers on Twitch. Those articles and videos and streams are going to bring your project even more eyeballs. And each and every one of those eyeballs are going to assess your project and determine if it's something that's worth their time to play. If the assessment is favorable, OpenMW 1.0 could bring tens of thousands of new players to Morrowind, revitalizing the community, including the modding community, and ushering in a new renaissance for the game. If the assessment is unfavorable, none of that can possibly happen. You will have squandered an incredible opportunity, and your product will get a fraction as many downloads as it could have.

Which is why I'm begging you, PLEASE don't think like a programmer, JUST THIS ONCE. Think about the opportunity you have to do enormous good for the game, *beyond* just releasing a competent recreation of the game's vanilla engine. I am not telling you to delay 1.0, I'm only telling you the project *needs* to think beyond its overly-rigid roadmap and give ample weight to other graphical and gameplay aspects that people really, really care about, which are going to turn them off from downloading 1.0 and cause too much hard work to go to waste.

User avatar
drummyfish
Posts: 154
Joined: 22 Oct 2017, 10:13
Github profile: https://gitlab.com/drummyfish
Contact:

Re: It's really important to have graphical parity with MGE.

Post by drummyfish » 28 Apr 2018, 06:30

These are my personal views:
sonicboom12345 wrote:
25 Apr 2018, 11:58
and nobody has played that version of Morrowind in over a decade.
I have. I didn't even use the unofficial patch or any unofficial mods. I enjoyed it.
sonicboom12345 wrote:
25 Apr 2018, 11:58
The project philosophy seems to be that 1.0 should aspire to be a faithful recreation of the vanilla Morrowind engine minus any bugs, but not necessarily anything more than that
Yes, that's what it should be and that's how it usually works in open-source community. There's a base project and then there are forks. First things are done first, not all at once. People used to proprietary games don't usually get this because proprietary games don't allow this beautiful feature called forking.
sonicboom12345 wrote:
25 Apr 2018, 11:58
I get the impression the core developers have set their goals for 1.0 in stone, and those goals don't necessarily include graphical fidelity to what MGE already provides.
I think it's good to stick to your goal, and I think that goal is good.
sonicboom12345 wrote:
25 Apr 2018, 11:58
But there doesn't seem to be a lot of energy spent on the graphics side of the house, at least not from my perspective as an outside observer.
There is a lot of energy spent on graphics, just take a look at the shadow pull request thread at GitHub, it's just that there's a very limited number of devs and they have to choose what to focus on first. We only need standard graphics now. What needs to be worked on is the game behavior, bugs, the editor etc.
sonicboom12345 wrote:
25 Apr 2018, 11:58
It isn't going to generate the buzz in the gaming press that it needs,
I think it has just enough buzz. Of all the open source games, this is one of the most active communities. If the project focused on buzz, it wouldn't be a free and open-source hobbyist remake of decades old engine.

Don't get me wrong, I'd love to have a beautiful graphics too, and I've been suggesting things like postprocessing to be implemented, but I really think it's the lower priority right now and can even wait for after 1.0. It seems like it's a huge and obvious issue to you, but it may not be so to everyone. At least I'd still play it even if it was a text-based game, and I think I am not alone.

Post Reply

Who is online

Users browsing this forum: No registered users and 6 guests