Skip to content

Open-source and renderer changes

bladeshriekbladeshriek Member Posts: 16
I'll try to keep this short. I'm one of those fans that was somewhat disappointed by the cancellation of the new renderer. I understood it to be a series of changes that would port NWN to either DirectX 11 or DirectX 12, allowing it to look significantly better with the addition of modestly improved textures and shaders (likely to come from the community instead of Beamdog). However, I understand Beamdog is a small shop juggling a variety of efforts and changes this significant may not necessarily be worth the ROR. I respect the decision and kind of appreciate the news being disclosed transparently.

With that said, I also know Beamdog is not your usual corporate-influenced and hype-driven game developer. As a company there is no doubt that you employ gamers who take pride in their work and are passionate about the industry, including the D&D CRPG niche.

I think Beamdog should consider open-sourcing parts or all of Neverwinter Nights. Open-sourcing enough of the code to allow for major graphical updates, while providing a fork that is an already in-progress effort towards modernizing NWN with respect to DirectX and/or Vulkan, would go a long way towards keeping the game popular for several more years. I am pretty confident that OSS programmers would eventually finish what was started. If you want a case study in modding potential, look no further than Skyrim. Freelancing modders in the Skyrim community have advanced the game more than Bethesda ever could. We aren't talking about just the addition of new content here, but rather fundamental graphics and engine improvements. While Bethesda has not open-sourced parts of Skyrim, they have participated in efforts for community developers (under NDA) to improve the game.

While I don't think NWN could reach the level of modding popularity shown in TES, the point is by open-sourcing a fraction of the game you could punt the effort toward improving the graphics over to the community while ensuring that the game stays popular and remains a potential source of revenue with respect to new DLC. Since Beamdog isn't as tied down with corporate strategy and stuck in a legal morass as many other game developers are, I implore you guys to consider it as an option towards keeping NWN popular and advancing CRPGs in general. By far the weakest aspect of NWNEE currently is its outdated graphics and DirectX model. That wide demographic of gamers generally don't want to play games that look dated.

Finally, if Beamdog or anyone else reads this and thinks "that's crazy, no game developer could ever open-source their game": EA- arguably the epitome of commercialized game development- has allowed the open sourcing of the remastered Command & Conquer. While there are other precedents for open-source in games, this recent example is one of the most high profile ones to date.

Comments

  • JuliusBorisovJuliusBorisov Member, Administrator, Moderator, Developer Posts: 22,714
    We're working with the rightsholder based on the license. So open-sourcing the game is not an option. And we're far from being done with NWN:EE, so this question is off the table for the time being.
  • bladeshriekbladeshriek Member Posts: 16
    edited June 2020
    We're working with the rightsholder based on the license. So open-sourcing the game is not an option. And we're far from being done with NWN:EE, so this question is off the table for the time being.

    I don't think keeping NWN on OpenGL and DirectX 9 ensures a bright future for the game, as I discussed in my post. Many negative reviews for NWN:EE cite the poor (relatively speaking) graphics, and the community here obviously has waited a long time for a "makeover". I appreciate the response in that it shows you guys are listening, but I don't really have confidence that Beamdog is committed to the game after abandoning a feature that would likely have attracted way more players than continued backend fixes. Considering these improvements have been in the works for years, you definitely can't expect the community here to accept virtually no roadmap and cancellation of a planned feature that originated close to day 1.

    Regarding licensing, I didn't realize that Beamdog did not have full control over the IP. In that case, as you said, open-sourcing is not possible. But concerning the second point- the goal of open-sourcing anything would be to allow the community to give the game the enhancements Beamdog doesn't have the resources to commit to. C&C Remastered/EA actually cites this as a reason for open-sourcing their game less than a month after it launched.
  • FreshLemonBunFreshLemonBun Member Posts: 909
    OpenGL is modern though, all buzz words aside, and open source just doesn't work in this context even if it was possible.

    For what it's worth there is already an open source emulator for all Aurora-style games from NWN to Witcher, and NWN2. It's been around more than five years. There was also discussion in some places about people being able to use it to modernize the game and let it do amazing new things never thought possible before. The issue as you can probably guess is that none of the people talking had any intention of doing the work, they expected the mythical someone else to do it all.

    What's missing from demands like this is people talking about what they personally will do and what they personally can do for the product. It's easy to talk when you're expecting someone else will do it.

    Some open source projects are useful because they have broad use over various industries and raise $100,000 in development funds every month. Not really an option in this timeline.

    The C&C remaster was apparently done by old C&C developers, who would have guessed. Beamdog already has that, so moving on.

    In comparison to Skyrim you'll find NWN has more modding potential and most content is available in an open way through scripts and open file formats, 2das, and rule change files. Besides the built-in ability to create and host entire interlinked custom worlds of your own design, but I might have missed where Skyrim does that.

    Now if you're talking about noticeable graphical improvements then Beamdog already released shader technology for NWN two and a half years ago with demonstrations for modders to pick apart and study. So if the argument is that people will surely work hard to improve and modernize the game if only they had the chance, well, we're actually not seeing a lot of that.

    There are already plenty of ways to get involved if you're serious about improving the game, rather than ask Beamdog to do something they never had the rights to do.
  • bladeshriekbladeshriek Member Posts: 16
    edited June 2020
    OpenGL is modern though, all buzz words aside, and open source just doesn't work in this context even if it was possible.

    For what it's worth there is already an open source emulator for all Aurora-style games from NWN to Witcher, and NWN2. It's been around more than five years. There was also discussion in some places about people being able to use it to modernize the game and let it do amazing new things never thought possible before. The issue as you can probably guess is that none of the people talking had any intention of doing the work, they expected the mythical someone else to do it all.

    What's missing from demands like this is people talking about what they personally will do and what they personally can do for the product. It's easy to talk when you're expecting someone else will do it.

    Some open source projects are useful because they have broad use over various industries and raise $100,000 in development funds every month. Not really an option in this timeline.

    The C&C remaster was apparently done by old C&C developers, who would have guessed. Beamdog already has that, so moving on.

    In comparison to Skyrim you'll find NWN has more modding potential and most content is available in an open way through scripts and open file formats, 2das, and rule change files. Besides the built-in ability to create and host entire interlinked custom worlds of your own design, but I might have missed where Skyrim does that.

    Now if you're talking about noticeable graphical improvements then Beamdog already released shader technology for NWN two and a half years ago with demonstrations for modders to pick apart and study. So if the argument is that people will surely work hard to improve and modernize the game if only they had the chance, well, we're actually not seeing a lot of that.

    There are already plenty of ways to get involved if you're serious about improving the game, rather than ask Beamdog to do something they never had the rights to do.

    I'm not some legendary modder for Skyrim, but I have participated in its community. I've also developed a couple C++ plugins using SKSE, which some very talented developers have worked on in tandem with Bethesda under NDA. Much of your response is veiled ad hominem and an appeal to authority, suggesting that opinions about NWNEE are not valid unless they come from someone who has spent hard time "improving" the game. Unfortunately, I spent most of my time modding in Skyrim's world.

    OpenGL is legacy technology and is being replaced by Vulkan. Apple has marked OpenGL applications for deprecation in favour of its own API. I'm sure you have some technical knowledge yourself, so tell me- does Beamdog even plan on providing long-term support for an OpenGL-based NWN on iOS? Apple has indicated they will "pull the plug" at some point. And in terms of new-ish games being released, most of them aren't using OpenGL at this point for many perfectly valid technical and compatibility reasons.

    The best singular thing Beamdog could do for NWN at this point is modernize the game's rendering technology to either Vulkan or DX11, while leaving it to the community to provide new assets and shaders in the long-term because what we see now is simply not impressive. And the reason it's not impressive is that the rendering back-end is ancient. I'm a realist and I sympathize with the many comments on the game not being "enhanced" in any way- visually. Because most potential customers don't understand persistent world enhancements. They want to see a game that doesn't look basically the same as it did in 2004. With the news of the renderer being cancelled, we have seen a confirmation that Beamdog has no plans to either 1) modernize the game's graphics or 2) mitigate the problem by providing new assets.

    Honestly, your post can easily be shifted around to say the exact same thing about Beamdog. If they have no intention to improve one of the game's worst problems, why should I or any modder try? At least Skyrim:SE and C&C Remastered were released with entirely new rendering back-ends because...that makes sense? That's one of the checkboxes for "enhancing" a game, regardless of what you think.
  • FreshLemonBunFreshLemonBun Member Posts: 909
    Honestly my response is not in any way an ad hominem attack on you. The reality is that someone must do the work you want done. We are not having a philosophical debate about the abstract qualities of open source versus proprietary software. If you are offended by a statement of the reality of the situation then perhaps you should have a long hard think.

    OpenGL is not legacy technology, it is still being used, and switching to something else wont change how the game looks in any way. That's the reality and why the suggestion is so silly, because it wont make people suddenly create new content. Vulkan would probably be better if you were starting out to make a new game today. There are a lot of RPGMaker clone developers excited about using Vulkan to make 16-bit looking 2D tiled games. Which is honestly fine but the idea that switching will instantly change the fortunes of a game is obviously silly. There are plenty of amateur developers that get stuck on technical nerdery and forget to make an actual game.

    You are always welcome to start developing the content you want to see, or to work on the open source emulator and show us all what you can do with Vulkan and how much better it would be. I hope to see your results and prove me wrong, I really do. After all you went on so much about how skilled you are, so I hope there was a good reason for that. I just don't expect any of us will ever see more than demands for someone else to do it, and weaseling out of doing it yourself.

    Responding that I'm just insulting you is honestly your prerogative but it just means you wont do the work and you want someone else to be blamed for that. Arguing for the sake of pride is pointless.
  • bladeshriekbladeshriek Member Posts: 16
    edited June 2020
    OpenGL is not legacy technology, it is still being used, and switching to something else wont change how the game looks in any way.

    My post was intellectual in nature, and your responses have been more philosophical while being littered with fallacies and appeals to emotion. Just want to hone in on this one point, which is factually and categorically incorrect. This comment per upgrading rendering is actually totally untrue. Just by upgrading to Vulkan and/or DX11+ you get a basket of features right out of the box that improve the visual fidelity of a game with virtually no effort towards modernization of assets. While you won't get a radically improved game by going from OpenGL/DX9 to DX11, you'll definitely get out of the box visual improvements.

    To be clear, I was playing Neverwinter Nights in 2002 at release for hundreds of hours. I bought multiple copies of NWN:EE just to support Beamdog. I'm a fan. Yet, the game obviously hasn't performed that well in the market and - in my humble opinion - it largely has to do graphics. You failed to address that, nor did you address my point about support on iOS. Launching a game with an API that was deprecated by Apple over a year ago is patently insane. It is more telling in that it reveals that long-term support is not really a consideration for NWN:EE.

    I delivered my points as best I could, and I appreciate your responses, but they boil down to: your standard for quality is too high, and if you aren't willing to do the work yourself then your point isn't valid nor worth addressing. I've encountered this line of reasoning used before when discussing NWN:EE actually, and other games with mod support throughout the years. It's a poor one.

    If you can't address concerns and provide constructive rebuttals without suggesting that the naysayers should be the ones radically improving a game (not the developers themselves), then you aren't really addressing anything at all. It's fine to think that the status quo is OK, but you need to back that up with actual constructive points instead of sophomoric drivel. This isn't a sense of pride manifesting but rather a passion for the game. It causes me pain to see NWN:EE at a fraction of the popularity of NWN circa 2000s. We have only a few servers filling themselves up at prime time, instead of dozens and dozens at all times back in the day. It makes me incredibly sad. So your last sentence could not be farther from the truth.
  • FreshLemonBunFreshLemonBun Member Posts: 909
    No your post was just a basic demand for a product.

    There is nothing intellectual about it.

    Let me be clear nobody wants to read an argument about your internet reputation or any other reputation for that matter. Are you going to make it or not? Basically. There is no fallacy there.


    Now arguably Beamdog already wasted a lot of time fussing over technical nerdery (with questionable results) given higher priority over real practical concerns that actually affect real players. I'm failing to see how these pie in the sky suggestions are any different.

    Then again since you don't have a working prototype it's hard to make a credible case. I also notice the people saying 64-bit would be "a game changer that justifies the (inordinate) development time" have become rather inconspicuous.
  • bladeshriekbladeshriek Member Posts: 16
    edited June 2020
    No your post was just a basic demand for a product.

    There is nothing intellectual about it.

    Let me be clear nobody wants to read an argument about your internet reputation or any other reputation for that matter. Are you going to make it or not? Basically. There is no fallacy there.


    Now arguably Beamdog already wasted a lot of time fussing over technical nerdery (with questionable results) given higher priority over real practical concerns that actually affect real players. I'm failing to see how these pie in the sky suggestions are any different.

    Then again since you don't have a working prototype it's hard to make a credible case. I also notice the people saying 64-bit would be "a game changer that justifies the (inordinate) development time" have become rather inconspicuous.

    Huh. You know basically every internet forum representing a video game community has fans demanding or advocating for resources to be allocated to features X and Y right? There's nothing special here. But I should have known that the first response to suggesting the open-sourcing of NWN would be bogged down with reductions to absurdity and hyperbole. So you're saying that there's no reason to have an intellectual debate on NWN's rendering engine because the developers don't have the manpower to implement it and I myself don't have a working prototype?

    The emphasis on 64-bit is more about being future-proof for mobile platforms and console. I'd guess that mobile platforms will be the first to rid themselves entirely of 32-bit game deployments. BD would probably want their apps to just be compatible right off the bat instead of being flagged. If they were modernizing the game graphically it would also be critical to allow for fancier assets.

    I honestly didn't really expect this kind of reception, but I guess you're probably not representative of the whole community. You gave up trying to provide any sort of thoughtful opinions that would make me or anyone else who reads this feel differently if they're on the fence about NWN:EE's future. Here are a few samples from critical reviews on Metacritic:
    Likewise, this game doesn't get the score it could have deserved, because the development company Beamdog couldn't improve the interface or the graphics, so it's an adaptation that has absolutely nothing to do with remastering.
    Graphics aren’t everything, but dated appearance married to a frustrating UI and even some missing features are asking a lot to be forgiven. Extremely tolerant gamers willing to overlook these flaws and seek the under-the-surface quality will find it, but playing a true remake or graphically made-over game would be a more inviting prospect
    Neverwinter Nights remains a fantastic RPG, but the Enhanced Edition on PlayStation 4 fails to recapture that magic. If you can look past the incredibly dated graphics, there's tons of classic role playing to get stuck into, but playing it with a controller and clunky UI is not particularly pleasant.

    So I'll be off now but I wanted to end with the point that a top criticism for NWN:EE has been its weak graphics. It's not just something originating locally on these forums from disgruntled users such as myself. Scrapping the new renderer and committing to OpenGL makes it impossible (or at least incredibly unlikely) for Beamdog and the community to address the concerns of gamers and critics, and likely will not contribute towards NWNEE going mainstream (again). In fact, the prospect of a massively popular NWNEE seems quite hopeless now. It won't stop me from playing through the obscene amount of custom single-player content that arrived since I set the game down in the mid 2000s, but it does make me pretty sad. I apologize for my responses being a bit harsh. I just don't think your approach to assuaging concerns about Beamdog is the right one and honestly I was hoping for a better discussion.
  • FreshLemonBunFreshLemonBun Member Posts: 909
    Ah yes every post that doesn't agree with the basic assumptions of the idea is absurdist and hyperbole. Again, what is the end game here exactly?

    Had you actually committed to work on your own suggestions then we would have had something interesting to talk about. We could have discussed all the pros and cons of various design decisions, we could have maybe even discussed the implications of various technologies. If you and others had been serious and credible we could have even discussed and organized a community driven spiritual successor to NWN as a new product entirely.

    Since that obviously wasn't the case all we have is posts about fallacy this fallacy that, everything is a fallacy to be mean about this great idea. Let me be clear this is not an academic setting and there is no academic value to this discussion either.

    Alas we're stuck with this timeline. Maybe someone will demand open sourcing NWN for raytracing next week.

    It's honestly tiring.
  • ZwerkulesZwerkules Member Posts: 112
    Yes! NWN needs ray tracing! ;-)
  • TarotRedhandTarotRedhand Member Posts: 1,481
    Not at the current price of nvidia rtx (the ones with ray-tracing built-in) boards! They are double the price of equivalent nvidia rtx ones here in the UK.

    TR
  • R_TEAMR_TEAM Member Posts: 35
    Lol - just the second line in the top post make me laught ...
    Beamdog have barely enough dev power to finish the switch 32>64bit engine and half-way new shaders (from today perspective of "new" ..) - and he demand an switch to DX architecture - an full different engine ... show clearly his "knowing" ....
  • JFKJFK Member Posts: 214
    I myself see some advancement in the graphics of the old gal. But, some of them aren't completed and in the retail version at this time. That doesn't mean they don't exist. I think some shots from the devs of upcoming/in progress advances would be cool, but they know best when/if to do that sort of thing.
  • FreshLemonBunFreshLemonBun Member Posts: 909
    Honestly we get a much better bang for the buck out of art assets that take advantage of the new options already available. Unfortunately there isn't a whole lot of community energy directed towards that either. Part of it is probably also due to the lack of good documentation.
    bladeshriek
  • bladeshriekbladeshriek Member Posts: 16
    edited June 2020
    JFK wrote: »
    I myself see some advancement in the graphics of the old gal. But, some of them aren't completed and in the retail version at this time. That doesn't mean they don't exist. I think some shots from the devs of upcoming/in progress advances would be cool, but they know best when/if to do that sort of thing.

    The game has definitely surpassed 1.69 at this point. It does have noticeable and interesting visible improvements. The lighting and post-processing improvements make it feel more scenic, and meshes pretty well with D&D in general. The problem arises when you try to open up NWNEE to a broader audience. Those critic reviews smash on the graphics because they don't care about (or ignorantly miss) all of the technical enhancements Beamdog made for multiplayer, and the ludicrously impressive mod support. Most prospective new players will miss that too. They don't really care. The fickle gamer of 2020 especially wants to see a remaster without any ambiguity. Most toons look ugly. That may be sufficient to warrant a bad review :( .

    I'll try to stay optimistic. Admittedly it's naïve of me to suggest that there's no good to come with OpenGL. I'm not a game developer nor am I a graphics programmer, and there's no hard evidence of that. So the premise of my argument is admittedly flawed. I guess the issue is more that this feature was dropped after so many tweets, posts, and live streams. It kind of dims the hopes for some massive graphical makeover that draws in a new flock. But we shall see.
    Post edited by bladeshriek on
  • JFKJFK Member Posts: 214
    edited June 2020
    I understand where you're coming from, @bladeshriek . Many people don't want to or even know how to delve into what has been done, let alone what IS being done. But take heart a little. There actually are graphical improvements from what I can see that haven't yet made their way through to retail. Have you checked out the nwvault Discord?
    Again, I'm not trying to be a Pollyanna about the situation, but realistically, it does look a lot better with new advancements that are happening visually, without an actual 'new renderer'.

    FreshLemonBun has a point too: CC creators have gotten a lot of new abilities to exploit visually, including textures, swapping, shaders, scaling, translation (moving), way more bones for animations, more animation slots, etc., etc. The Toolset has been improved, and continues to improve. So, it is now (and really has been all along) somewhat in the hands of the custom content makers to actually put those new things to use. Beamdog has done another thing that to me is huge: they have accepted and integrated what was once custom content (GREAT custom content) into the base game.

    If you haven't yet, I encourage you to check out the nwvault discord. Look back through it, or ask around about upcoming improvements.

    Regards,

    JFK

    edit: Here's the invite link:

    https://discord.gg/pWVqMRX
    bladeshriek
  • AncarionAncarion Member Posts: 155
    I agree that EE does look objectively better than 1.69, and with a few more tweaks (eg like better control over vertex normals) it would be more than adequate overall. And I'm surely in the minority, but I personally don't care for the current normal mapping - I think the graphics would be better served to keep a more stylized and "artistic" feeling, rather than trying for too much realism. I just think it suits the overall spirit of the game better.

    As for the need for new assets, these things take a lot of work. Most custom content was made back in the day when even a new crate model would get a thousand downloads from the Vault. Very few are going to volunteer the effort needed now, when their work will get little if any attention. Take a look at the current Vault; the lauded, near-mythic Bloodwar tileset doesn't even have a hundred downloads. Who's selfish enough to demand anyone make tons of new assets (for free!) with the game in its current state?

    A larger player base is the number one priority, and we should be working towards that, not arguing with each other for its own sake.

    Personally, I think NWN would be better served by new, engaging modules that might bring more people in to play the game. I regret to say that BD's most recent offering probably convinced more people to stay away...

    We all enjoy the simple, hack and slashers, and the PnP conversions, and all the earnest amateur attempts, but we're not the people who need to be won over. I really think the game needs a professional-level, voice-acted module (or series) with a strong story and traditional DnD characters (not all the weird stuff developers seem to be pushing these days, that are often hit-or-miss). And it needs to be widely advertised.

    For such a small audience as us to demand very much is a bit unreasonable. But neither do we have the means to widen the player base as much as BD could have. It's all a shame, really, but I don't think it's too late yet. The cultural climate is ripe for nostalgia now, and despite what reviewers say, I believe new and former players could be brought in even with less-than-perfect graphics. A good game experience, meaning a fun, well-done module, is more important for that.
    bladeshriekJFKGrymlorde
  • FreshLemonBunFreshLemonBun Member Posts: 909
    Obviously there are improvements with a different graphics API future ready for AAA graphics of 2030 but for those who don't spend their time just examining a technical specification it's a pointless waste of time. Even putting the suggestion that the community must develop it in their free time aside.

    "NWN graphics" or at least what is possible with the old graphics API are on the right and graphics for a newer API are on the left. You decide if it's worth another half year or year of development with no visible improvement to the actual game since the art assets still wont use the new features.



    All I'm saying is don't get carried away and prioritize the best most impactful things. It's a shame the community hasn't taken full advantage of it all but that can change any time someone decides to do some modeling tutorials over the summer. If you just wait and see then it might not happen though, it's a lot of work for hobbyists, keep that in mind.
    bladeshriek
  • bladeshriekbladeshriek Member Posts: 16
    edited June 2020
    Obviously there are improvements with a different graphics API future ready for AAA graphics of 2030 but for those who don't spend their time just examining a technical specification it's a pointless waste of time. Even putting the suggestion that the community must develop it in their free time aside.

    "NWN graphics" or at least what is possible with the old graphics API are on the right and graphics for a newer API are on the left. You decide if it's worth another half year or year of development with no visible improvement to the actual game since the art assets still wont use the new features.



    All I'm saying is don't get carried away and prioritize the best most impactful things. It's a shame the community hasn't taken full advantage of it all but that can change any time someone decides to do some modeling tutorials over the summer. If you just wait and see then it might not happen though, it's a lot of work for hobbyists, keep that in mind.

    Yea, this video shows that modern DirectX versions wouldn't be a panacea for Beamdog. The Steam comments and Reddit posts on this game's DX11 support are a little more optimistic, but they also say the built-in enhancements and upscaling are dependent on GPU models and may tank your FPS. At this point, I tend to agree with you that effort towards new models and assets would probably have been a lot more beneficial than a dead-end project on DX11 or Vulkan. Most vanilla toons are just so ugly, it's a shame this is what most players see as well when they start up the game, and especially on console. I'm going to look into the modeling aspect to see what I can do but I don't have any experience with it at all. All my experience modding has been with scripting and programming. But at the very least maybe I start by making a new face for my toons in single-player :D .
  • TarotRedhandTarotRedhand Member Posts: 1,481
    Before you go to all that work it may pay you to take a look at what has already been done over on the vault. One thing to bear in mind is that in order to get new faces you actually need to create new heads. So combining those two things here is a list of over 10 pages of search results for heads on the vault (may include some NwN 2 content too).

    TR
    bladeshriek
  • bladeshriekbladeshriek Member Posts: 16
    edited June 2020
    Before you go to all that work it may pay you to take a look at what has already been done over on the vault. One thing to bear in mind is that in order to get new faces you actually need to create new heads. So combining those two things here is a list of over 10 pages of search results for heads on the vault (may include some NwN 2 content too).

    TR

    That's a ton of head/face packs. I was just looking at the CEP's offerings. I'll go through this list but part of what I'm interested in testing is: what is the highest poly and best textured model that I can load into the game for a single character, before I notice FPS drops and jitter. My very limited understanding is that OpenGL and early versions of DirectX aren't nearly as optimized as DX11+ and Vulkan, and tend to perform poorly with many high poly models on screen. But I'm curious to see how much I can push the game for a single character. I'm definitely going to go through this list though to see if someone has already done this lol (chances are given the extent of this list it's already been done).
  • FreshLemonBunFreshLemonBun Member Posts: 909
    If you decide to learn modeling there is a tutorial series called "Do It In Blender" I believe which is also on the vault and quite simple to follow to get the fundamentals of 3D modelling for NWN. It starts with placeables and tilesets however and it's also a little out of date here and there. Additionally it's worth looking at CrazyBump and looking on Youtube for tutorials for it, that should help in getting an idea of making advanced textures for NWNEE.

    The ceiling for how good even old graphics API can look is pretty high but there is still a lot Beamdog can do to improve the general look. The work on their new lighting model might give some good results too. I think NWN currently works in RGB color space for lighting, but a lot of newer games, movies, and 3D artists transform it to an expanded color space which helps preserve color information. The result is a more photo realistic look if I understood the theory correctly.

    Anyway we'll have to see what they come up with in the next few updates.
    bladeshriek
  • FaydarkFaydark Member Posts: 279
    My very limited understanding is that OpenGL and early versions of DirectX aren't nearly as optimized as DX11+ and Vulkan, and tend to perform poorly with many high poly models on screen.

    You're partially correct. It didn't really have anything to do with the number of polygons directly, but the number of separate objects drawn each frame aka "number of draw calls". Setting up rendering costs CPU time, and modern APIs have changed how the CPU passes information to the GPU to reduce these costs.

    In regards to OpenGL vs Vulkan vs DirectX 11/12, I think the important thing to note here is that OpenGL is not a fixed API. It has undergone many optimisations and improvements over time, and many of the rendering hardware vendors actively support and work on the API. On top of that, OpenGL has many extensions, often provided by the hardware vendors in order to improve optimisation or add additional features.

    I confess that I haven't been keeping up with rendering benchmarks between the various rendering API's a lot lately, as my work has mainly been focused on OpenGL and OpenGL-like APIs. So I am not sure what the benchmarks are like between OpenGL 4.X with extensions vs Vulkan vs DirectX 12. But even if Vulkan is wiping the floor with OpenGL, it really only applies to PCs - it's not widely & robustly supported on other platforms yet.

    Additionally, changing a game using rendering methodologies from back when NWN was made originally, to use the rendering methodologies required by DirectX 12 etc is a much much much larger task than it seems at first glance, and I suspect this is what Beamdog ran into. Additionally, many of these changes would equally apply to modern OpenGL anyway - changing to another rendering API probably isn't what NWN needs, as much as changing the engine to do the rendering in a modern way as the first step to improvement.
    TarotRedhandDrewbert_ahoybladeshriek
Sign In or Register to comment.