#14 posted by Kinn on 2019/05/30 17:44:08

It's been well over 20 years since quake was released. We're the faithfuls, the ones that stuck around because we like quake for what it is, rather than what crazy stuff you can do with the engine, and anyway, there are so many better starting points for "crazy stuff" than the quake engine.

The people who want to turn quake into a "Disney's Pocahontas"-themed dating-sim slash 2D-fighter with QWOP controls are long gone. They left for the Q2 engine when Q2 (sorry, Wor) came out, then left for the Q3 engine when Q3 came out, and so on. Now they just cock about in the latest Unreal or Unity oh shit the engine just updated for the twelfth time this week and broke all my boob animation physics and nipple-erection blend shapes.

All of the designers here have made tons of posts in the past where they state what things they would like in the engine. There has NOT been a lack of communication in this regard. The programmers only do the bits that they personally like the sound of, and feel are worth doing, and I guess they don't really want to spend their time just working through the bullet-points of some random mapper's wish list.

From a selfish-mapper point of view, I wouldn't be coming at this from a "how do I make a programmer enjoy working with the quake engine" angle, but more of a "how do I make a programmer implement this very specific set of features that I want to build a mod around (a mod that may or may not ever be finished and released)". You can see why they probably wouldn't be queuing up at my door, unless the exchange of money is involved.


 

Posted by "8657"....


#15 posted by 8657 on 2019/05/30 19:16:34

If you want programmers to help you with what you want, you have to let them do what THEY want so they come to enjoy the engine. There's plenty of shit that is out there that's just ignored because it doesn't fit with the vision certain people think the engine should be. That's never been how this shit works.

Programmers come in because they like a certain something about what they see. It's like a hook for them to put their skills to the test. It's a challenge that they're looking for. That does benefit everyone else if it can be used in the future/built off of. Things get easier if you don't have to build from scratch.

When people say there's a lack of communication, they mean there's a lack of LISTENING. If you just talk at each other, nothing will get done. Rejecting everything out of hand also leads to alienation. That helps nobody.

I know I harp on this when this topic comes up but look at the Doom modding community and all the things they've done. It's vibrant and alive partly because there's a strong programming base. They do their own thing which helps make things better for mappers in the long run.

This isn't a business; things are going to have to evolve naturally to open new roads. Otherwise, there won't be a community worth calling it in 20 years more than likely.

 

That's Not How You Improve Things

Yeah, I know it's not. My point is that programmers just "doing whatever" might not exactly yield anything that modders and mappers are interested in doing anything with.

I am really struggling to think of ways to get new programmers excited about "just doing stuff" in the quake engine. By "new programmers" I mean programmers who aren't already doing stuff in the quake engine.

Surely, it should be about coming up with a great mod/game idea (that for some reason has to be based off the quake engine) and then trying to get coder(s) excited about working with that game/mod idea?

Someone has to come up with an inspirational concept first and then try to get a team assembled, where artists and designers and programmers are generally all on the same page and working to realise a particular vision.

This is incredibly difficult and rarely happens in the quake community, but I think that is the most noble thing to aim for, with the best chance of producing something good at the end of it.

I know I am being incredibly idealistic and naive, but I believe this approach has yielded results before - e.g. Nehahra, RMQ (even though unreleased, we got some tech from it), AD I think too... 

It seems like engine coders are lone wolves but take TrenchBroom. ericw and Sleep are doing some amazing things as a team. If egos can be put aside and focus maintained, we could see some great things.

But what features? I dunno what everyone wants. We have some great improvements in ericw's tools. A survey and a wishlist from the ppl who create content.

For my part, I am interested in ease of use for new users, future proofing (with things like Vulkan) and adding map discovery and downloads - the sort of non-graphical things only Spike and Baker focused on recently. I think those could help keep the game relevant. 

for a while, and which I just started working on, is to have a version of the engine (both software and GL) with all its caps entirely removed - that is, all MAX_***** limitations, and the whole idea of a static memory pool, gone from it.
I already talked about it a few days ago. These limitations were very useful 22 years ago, but they simply make no sense whatsoever with modern computers and devices.

By doing this my hope is to inspire map tool creators / mappers to realize the old ways are no longer something they should be subjecting themselves to, and to start bringing big, really really BIG things into the scene.

(I already removed the need for MAX_MODELS server-side in the engine - going for client-side, and whatever comes next). 

I would love to see an uncapped software version of the engine in action. 

Things like this are what's been needed for years, but AD's highlighted that point in particular by being close to if not the limit of what can be done on the mapping end with current limits. What Q1 modding needs now are gameplay and technological improvements to work in the current sandbox. Q15, while divisive, is a step in the right direction in that regard.

What could also be done is to build on the TCs and other id2 games that have been made with the Quake engine. Like I said before, it's easier to build on things that already exist rather than try to build everything from scratch. What id2 offers that the other retro engines don't is the ability to make viable TCs that won't be erased by an update. To make this point, I want to post this:

https://www.youtube.com/watch?v=8sMYBYy3BQo

This is Brutal Quake. Now you might be thinking: why have I never heard of this? Well it's not on the Quake engine. It is, in fact, on the Doom engine. Yes the engine that is a generation behind this one has gone so far ahead that it can simulate the id2 almost perfectly.

And then there's this as a parallel project of sorts (skip to 1:10):

https://www.youtube.com/watch?v=gAkwzB9fJdw

ShadesMan is working a project for YPOD called Eternal that will help update it to the modern day. Things like this should be encouraged as well as improvements on the base game. There are other things, but I don't want to flood this too much. 

a version of the engine (both software and GL) with all its caps entirely removed - that is, all MAX_***** limitations, and the whole idea of a static memory pool, gone from it.
I already talked about it a few days ago. These limitations were very useful 22 years ago, but they simply make no sense whatsoever with modern computers and devices.

When using modern Quakespasm, FTE etc, what limits do mappers/modders brush up against in practice?

I'd be interested to know what limits people are actually still hitting. 

Never in a million years would I have ever seen that coming and I'll bet that I'm not the only one.

What's next, someone springs a surprise Navy Seals episode? 

There are already a lot of coders doing amazing work. If you're talking strictly the Quake engine, then dumptruck is right. EricW continues to do great work and he has worked directly with mappers to implement features, such as for Orl's Ter Shib maps. Spike did cool stuff with QSS and continues to push forward with FTE. R00k is also actively developing his port last time I checked. Aside from directly modifying the engine, several people are making awesome mods and doing cool stuff with quakec like khreathor and snaut.

When it comes to people complaining about adding "crazy stuff," keep in mind that it could just be a vocal minority. The people who are the angriest about something are often the loudest. I also doubt that people who have a problem with new additions would care so much if they were off by default or easily disabled.

I'm not sure if the challenge of improving the engine is enough to attract anyone. I think that any talented developer coming in would have a love for the game. If people continue making great mods and maps, then maybe interest will continue to grow and eventually a talented newcomer will show up.

In spite of the naysayers, stuff like AD has done a lot to generate interest in the game and Wrath will surely foster even more interest. Hopefully potential newcomers have thick skin and don't take the perpetual complainers too seriously. 

Wrath trailer got me back into the scene, back in the 90's I only made maps and assets for myself.

Currently I'm working on a refactor of the 1.06 QuakeC files to make it more comprehensible for new joiners. In the end I plan to provide a fully sanitized codebase with clear and flexible API functions for each aspect available. 

Years ago the community was more receptive. Nowadays, it isn't.

Dunno. The community seems just as receptive of high quality additions that enhance the game, fit in with it's style, etc.

Some maps in some map jams are very poor, yet people don't reject maps jams because of them. Nobody says "this jam is bad because of that map", but they say "this engine is bad because of that feature". It's sad to see lots of good work being ignored because of small stuff.

That's different. A bad map in a jam pack is a 5 minute experience you never repeat, or just skip over entirely. And yes, some people don't enjoy jam packs as much because there can be some low quality "filler" in them. A bad feature in an engine, unless it's easily turned off, will be a nuisance every time you try to play in that engine.

In spite of the naysayers, stuff like AD has done a lot to generate interest in the game and Wrath will surely foster even more interest. Hopefully potential newcomers have thick skin and don't take the perpetual complainers too seriously.

You been on the booze along with mankrip, matching him shot for shot?? I dunno who these imaginary "perpetual complainers" are, but your very examples of AD and the promise of Wrath are things that have been almost universally welcomed as great enhancements / advancements on Quake - prime examples of well-themed high quality. You've disproved your own point there.

The question remains: How to make engine coders enjoy working with the Quake engine?

Trickier one. How many engine coders would enjoy working with the Quake engine to produce great, Quake-themed, harmonious end results that will be enjoyable for players, mappers and modders?? Rather than say, tinkering with the code for their own pleasure or experimenting around to see how far they can push things?? (FWIW it seems the vast majority of mappers do it for the former aim).

I guess the ideal will be when a coder's desires / ideas and what works great in the Quake style both coincide. This is pretty easy to visualise - just look at stuff like AD, or the examples I gave in an earlier post (coloured lighting, fog, skyboxes, particles, transparency, increased limits, props, high quality model replacements etc etc). To what extent it will happen in reality is less certain - maybe just ask the coders "what can you bring to this to enhance it further??" 

Mappers, modders and players alike, all have a favourite engine.

For many reasons, Quakespasm is that favourite engine for a lot of us.

Let's say I have an idea for a cool feature I'd like to use in a map - volumetric fog. Suddenly volumetric fog appears in Darkplaces. Will I suddenly start mapping for Darkplaces? No. Because it's not Quakespasm, and I have to now put up with all the stuff I don't like about Darkplaces, and I no longer get to enjoy all the stuff I do like about Quakespasm.

Let's be honest, when a mapper wants feature X, what he wants is feature X in his favourite engine. 

...and as Bal just pointed out in #tf, the mapper also wants that feature X in all the engines people are going to be playing his map in.

Oh shit son... 

So, we want a solution that concilliates what I want (or bother) to do, with what the rest of humanity needs me to do, if I want my work to be known and played for that precise instance of humanity.

Man, if you find the answer for that, you’d find the solution for all of the problems humanity has right now.

For now, the only thing we can do, if we want segment X of the market to enjoy our work, is to step up of our confort zone, find out what segment X uses, learn about it, and do whatever it takes to implement our work on it and give them what they need to see it. I see no two ways about it. Sorry. 

The answer is to diversify the engines so that each engine does a certain thing others don't so it's worth playing them. Quakespasm's worth is that it simulates Quake in a better quality than the base game, but it struggles with simulating other builds that don't fit in the neat little bow.

In contrast, Darkplaces and its forks are the TC engine in the sense that it's more flexible than QS when it comes to these things. This is why Quake 1.5 (Q15) started on that engine in the first place. Epsilon is a superior version of Darkplaces that lets you play Quake at higher fidelity closer to modern games. Mark V tries to go to the real experience of the original Quake using many different parts of past engines. And that's not counting the MP engines.

QS can still be the main favored engine, but it also doesn't have to be the only one. 

Epsilon is a SEVERELY OUTDATED version of Darkplaces. Also, you may want to check the definition of fidelity. 

Epsilon is a superior version of Darkplaces that lets you play Quake at higher fidelity closer to modern games.

https://youtu.be/1wiz0UsBPac 

That is what you call fishing for responses, though it does look very shiny. -_-

Basically, the curiosity of id2 is how unexplored it is in comparison to every other retro engine. The Quake engine, despite all that has been done with it, has never truly been pushed to the full extent of what it can do as well as taking from successor engines as far as I can tell. A project that pushes the engine as far as it can go would be an interesting prospect. Something like a tech demo... 

FTE also simulates Quake like Quakespasm, has multiplayer support, and has a lot to offer to mods or total conversions just like Darkplaces (or even more)... but FTE should come with "vanilla look" as default for newcomers (i know there's some premade configs that show up when you run the engine for the first time, but it's confusing to newcomers, because they don't know what that is or how the game will look with each premade config)... Another thing that is confusing, is that some things you have to write on the console to turn it on/off (i have a txt file with a lot of console commands, so i don't forget). All the customizations should be on the menu/options (like in rain, snow, turn on/off effect.info...)

And if, for some reason, a mod/TC needs some of the effects turned on (like realtime shadows or something) just make an autoexec.cfg and pack with the mod.

And, of couse, people need tutorials and examples to learn how to mod for it... FTE could be the ultimate quake engine for all we need, but most of the times it seems too advanced for my little understanding :P 

The answer is to diversify the engines

This causes problems as well. You have something like GZDoom, which is the Emacs of Doom source ports - at this point, it's more like a separate engine that just happens to run Doom. It inadvertently pushes its own format by offering un-Doomlike features such as true room over room, slopes, and much more. A chunk of the GZDoom camp is incredulous that anyone would want to use Boom or Doom format when mapping. If you prefer PrBoom, you're out of luck if someone used features specific to GZDoom. If you have a large, detailed map i.e. Frozen Time, you're out of luck if you want to use GZDoom due to its abysmal performance compared to PrBoom.

Incompatibilities aside, you end up with a lot of talented coders single-handedly keeping their project afloat. I don't think that PrBoom+ has been updated in years, and that's the problem when the single developer gets bored and moves on. There's a case to be made for developers consolidating their efforts into fewer source ports to make it the best that it can be and providing a fallback if one decides to quit the project.

It's good that different developers who have different ideas about where they want to take the project are doing their own thing but there are still a lot of drawbacks to consider. 

I agree 100% about FTE. I literally was just tweaking a new install of FTE to my liking and even with the presets I needed to fiddle around for quite a while (and I sort of know what I am doing!) But I agree with the power under the hood there. If FTE defaulted to a Quakespasm look and didn't default configs to the Home directory I think it would be very good competition for QS for casual users.

As it is now, there are almost too many options. 

And the most impressive thing about FTE is that Spike is always fixing things and adding even more features as you can see here:

https://sourceforge.net/p/fteqw/activity/?page=0&limit=100#5cef13cef0d347678db86202 

It's better than having engines/source ports locked bytheir creators and not giving the source code for public use. And knowing how things usually go, something will replace PrBoom+ once it becomes a problem. In practice, Doom players use specific ports, but there aren't any serious hang-ups against making maps for something like Chocolate Doom for example.

Branching out doesn't hurt when it calls for it, but you're right that there doesn't need to be massive forking at the moment. 

Good point. Doom engines can either use the GPL, or the non-commercial closed source id licence. For developers who wants to do extremely innovative stuff, the closed source may be more attractive since they can finish implementing their vision and still get open beta testing feedback before releasing their code.

I wonder how many active Doom engine projects still uses the closed source licence, though. It's hard to say that the closed source option has any significant weight on Doom engine diversity.