News | Forum | People | FAQ | Links | Search | Register | Log in
Quake 2018: How Maps And Engines Are Better Than Ever.
Considering the following:

1. We're getting more map releases than we've been getting in nearly 20 years, and they're all of decent quality at the very least, and superb (that which are rivaling the undisputable classics in quality) at most.

And...

2. In 2008, for every 5 demos for a map:
- 2 would be Fitz 0.85,
- 1 would be aguirRe's AGLquake,
- 1 would be DarkPlaces,
- 1 would be JoeQuake or some other QW engine (???).

Each and all of those with their own protocols and idiosyncracies.

These days almost everybody uses Quakespasm - an actively maintained and cross-platform engine - as a standard.

Conclusion:

After a slump in the early 2010's, Quake is finally doing better than ever, with the player and mapper base growing and the game itself slowly creeping back into mainstream attention.

Discuss... or not!!
First | Previous | Next | Last
 
Theres been over 200 single player maps made for Quake in the last 2 years. I think Quakes doing just fine without all this stuff you're asking for that's already implemented in other engines no one wants to use.

Indeed. This is the stark reality. I'd like to think there's all these incredible maps and mods that are just sitting idle, waiting for Quakespasm to implement md3 before they unleash themselves into the world, but I think in reality if quake content creators actually want to use other model formats, we'd have seen some content by now. 
Give Tim Willits A Call... 
Seriously who are you going to retain to perform the labour behind developing a new Quake engine that will support fbx format, static mesh with collision, and 'ray tracing'. Do you have a 7 figure budget?

Honestly just call up id and ask to license idtech 6. This is absurd. 
Lol I Know Tim Actually, Great Guy 
kinn, i noticed that you were asking when qs would support q3 map format in another thread... in this thread you said nobody wants this newfangled crap!

the reality is people do want these things, including you, but every time it gets brought up on here there's a huge backlash against whoever asked for X new thing. quake doesn't need this shit blablalba. i think people have just given up mentioning stuff like this because it might get into a flame war for no reason. but this thread was created specifically to talk about it so here we are!

anyways, lots of great information in here so far, and in case anyone wants to consider it we have lots of ideas for improvements to make maps and engines better than ever.

@Redfield, i don't think its absurd, we already have mesh with collision or you would fall through the floor of the brushes you stand upon. none of what i mentioned requires a 7 figure budget. 
 
kinn, i noticed that you were asking when qs would support q3 map format in another thread... in this thread you said nobody wants this newfangled crap!

this weird obligation people have to pull up old opinions(3 years old in this case) of people to use against their new, current, opinion is pretty absurd. 
Echos == Otp Confirmed? 
I'm taking about the actual reality of how and why things get implemented.

Of course I think it would be cool to have md3, q3bsp etc etc. Doesn't mean I can't drop a few sodding truth bombs every now and then does it?

I'm not saying we shouldn't have these things, I'm saying this is why we don't have these things, and this is why the programmers don't think it's worth their time supporting these things. 
One More Thing. 
i keep hearing stuff like "if people wanted it we'd see X by now", here and in other threads.

i don't think it's that easy to dismiss. for example why didn't we see many q3 format maps?

because darkplaces didn't support the full implementation of the q3 shader system. this basically killed it for me and probably everyone else. and FTE uses and ungodly complex glsl shader, i would have to convert all my shaders over to that after learning glsl.

as well, model/anim implementations vary so wildly and are partial at that. if it won't support what you want to do because it's partial, it's as good as no support at all.

if you don't agree it's fine, but this is my take on things. 
And It's Pretty Dumb 
considering how obvious it is that otp just pm'd you (or whatever) with that incredibly controversial info.

Oh my god I once said q3bsp is cool.

Oh shit I still think q3bsp is cool what now. 
@kinn 
TREASON? 
It Gets Better 
spike called qs crippleware in the qss readme. that's a solid burn. what's the verdict on that?

i won't dig up any more quotes to throw at you guys for now, it's making you paranoid hahaha. 
 
Spike also released FTE with literally no documentation making any desire to design for it null from the start. 
 
The best model format is a binary format that you can fread into a VBO. Plain text is fine for intermediate formats, but for the final thing that you're going to use in-game, load times are pretty damn important.

Spike described NQ as crippleware. Not specifically QS, any kind of NQ. But Spike has a QuakeWorld background and judges things by different criteria. 
My 2ยข 
If you want to make a quake 3 map why not just, you know, map for Q3 instead of making a Q3 map then turning around and trying to run it on a Q1 engine hacked up to play Q3 maps? If the lack of quake 1 enemies is a problem just download a quake 1 enemies mod then base your map on that.

As for models I think FBX would be a great idea too although there should be an FBX > MDL converter made so people can convert their model for use in more vanilla-ish engines that don't/won't support more modern advanced formats. I would also be kind of concerned about performance concerns of really high poly meshes (at least really high poly in quake terms anyway)on lower spec pcs, the way I see it quake is a great game for lower end pcs and laptops- and why wouldn't it be considering it is over 20 years old at this point- and the mdl format helps with that because it gives strict well defined limits on what you can do and those limits also happen to help keep the models lightweight and able to help maps run better, my lower spec laptop already apparently can't run quakespasm without -noglsl which makes no sense since the gpu SHOULD support that, it isn't particularly happy at the thought of quake having compatibility with a model format with unlimited vertex/face count texture resolution etc, surprisingly enough. Potato lives matter you know ;~; You know there will always be that one guy that spams free high poly models that they downloaded from sites like Model Resource in their maps just like how in Doom mapping you have some people that just decide it would be a great idea to make their map in the most advanced format just so they can copy and paste a bunch of random custom shit in it from Realm 667/. Then they will post it for people to play and shockingly to no one, it runs like shit even if it's vised unless you have a super epic high spec system which like I said makes no freaking sense for a 20+ year old game but oh well that's just how it is.

I get that I'm obviously kind of biased in that regard but still I think that's something that should be taken into consideration. 
 
i'm really just speaking as a modeller, animator, and mapper here. i can code a bit and know a few technical details. for example i know vbo is a vertex buffer object.

my suggestion is you could have the engine parse the collada at first use and save it out in its own native format in the userdir, that's what many games are doing.

blendshape anims should be allowed to deform along with the skeletal for doing facial anims on characters.

this would give the ideal solution for endusers, if they want it or not is another question. how many people wanted this and got ran off by an angry mob years ago, or how many want it now and are afraid to say so? lol

it shouldn't matter who wants what, if the engine is going to support models and animations it should do it in a smart way. coding the engine features by taking votes from an angry mob isn't a good way to go in my opinion!

so says the black sheep of the quake commune, baaa! 
 
just out of curiosity, whats the last Q1SP youve played, echos? 
 
@muk: it was one from sock, i really like his mapping style.

@therektafire you made some good points, i won't cover all of those but i will say most of your concerns are because of newbie mappers over-using triangles. there's nothing you can do about it, just don't play their map.

vis won't help unless it can seperate the map into smaller pvs sets. if a pvs is using too many triangles then you will have a slideshow. a good mapper knows there is overdraw on top of that, and performance takes a hit when moving between pvs's since more is visible from that spot.

i have tested an iqm mesh with 50K tris in fte and it worked fine for me with no slowdown, and that's using a midrange older gen graphics card.

newer gen cards would allow higher res meshes or more of them to be used in the same area. 
Echos 
Good luck with your project. Sounds like you have it all figured out. 
 
thanks dumptruck, my project is more about the models and animations than anything. i wanted to put some monsters into quake which are more modern looking.

since everyone is going to attack me again saying i'm the only one who wants to do this... no. i'm not the only one.

see for yourself on youtube:
https://www.youtube.com/watch?v=X9Dtdq_wBBU
https://www.youtube.com/watch?v=8-x91pvL97A
https://www.youtube.com/watch?v=qi7iLEolsn8

it may be better for me to just take all my models and put them into another game entirely if everyone here really hates newer looking models. i would have no audience to appreciate them. at the moment a project like mine can only run on fte or dp engine, which you guys say nobody wants to use any longer, this is also a problem. for now i am just making my models, and i don't know where they will end up at. i'm only going to do 3 maps at the most to put these models into, maybe only 1 map.

i made a map a while ago but it relies upon lots of q3 patchmesh, if i could convert these patchmeshes into mdl models maybe it would be playable in qs as a bsp2 map. i can convert the brushes easily but not patches. i would probably have to write my own app to convert patch data into obj files and then convert those obj into mdl.

i'm only a hack at coding so the ai for my models would be utter shit, but since it's quake the ai was always shit so purists can rejoice lol. 
 
at the moment a project like mine can only run on fte or dp engine, which you guys say nobody wants to use any longer

People don't want to use FTE or DP to run content which already runs perfectly under QS.

If you make something that's targeted specifically at DP or FTE, then that's a completely different story and (shock, horror) you might actually have an audience.

Instead of this weird hostile attitude of somehow assuming no-one here wants to play your stuff before you've even made it...you could just try, you know, making something for DP and FTE? If it's good, people will play it. 
 
I don't think you're being attacked.

What's happening here is that there are - for the purpose of this discussion - two directions for the Quake engine.

One of these directions is a platform to create lots of varied and semi-generic content on.

The other is a platform to play Quake on.

So, this site and this particular part of the community has a preference for the second direction, which is why an engine like QS is the way that it is. New formats typically don't exist and when/if they are created, they are in response to very specific hard limits in the old formats.

That's why BSP2 exists but MDL2 doesn't. Stock BSP has limits of 65536 or 32767 in several of it's content lumps, mappers started hitting those limits, so a response was required. Stock MDL may look crap but there's nothing about it actually preventing people from doing stuff.

That this part of the community prefers the second direction doesn't invalidate the first. By all means use DarkPlaces, which is very much tooled towards the first. Few people in this part of the community use it, but it certainly is widely used elsewhere. 
 
I would say that people in this community would use DarkPlaces, if there was good DP-specific content to play.

There isn't really yet though.

This community is driven by good content. So far, that's almost exclusively been stuff that you can just run under QS.

I think there was that one time when Tronyn did a big release that only worked in FTE. Guess what? We all downloaded FTE and played it. Holy shit. Oh and also guess what - future versions of QS then added support for it. Mind blown.

Admittedly, that was more about limits being broken, rather than radical changes to file formats, but I'm using it to illustrate my point that this community is content-driven, and changes happen because we want our engines of choice to support good existing content (that might otherwise only work under less-popular engines). 
 
kinn: Instead of this weird hostile attitude of somehow assuming no-one here wants to play your stuff before you've even made it...you could just try, you know, making something for DP and FTE? If it's good, people will play it.

also kinn: If you want anyone to play your map, target Quakespasm, otherwise GLWT.

where did i get a hostile attitude and assumption? it was from what you told me when i first came here.

now you are blaming me for holding hostile assumptions but they are just the "facts" that you all gave to me. i only know what you told me, and i have been away from quake for years.

i'm not tronyn so if i release something i highly doubt anyone would be making a special case out of it to code support for my mod into all the engines. nobody here knows me. qs doesn't even support nehara though either, so... popular or even a retro style mod doesn't always mean supported. 
 
There's no contradiction really. Yes by having something that only runs on FTE and DP you are initially limiting your audience, but then you can mitigate this by having this thing you've made just for FTE / DP be really good and worth playing.

It's a pretty easy concept - if you make something average and unremarkable, that only runs in DP/FTE, then I doubt many people will bother to play it, because no-one can be bothered to go out of their way to play something completely unremarkable in an engine they almost never use.

If you make something totally awesome that only runs in DP/FTE, then you have a very good chance of finding an audience.

So again, good luck with that, and I mean that in a sincere way. 
 
If you make something totally awesome that only runs in DP/FTE, then you have a very good chance of finding an audience.

True. Also you said that you want to make a TC. If so, you just bundle FTE with your mod and done. 
 
a completely faithful re-creation of the original quake monsters would be possible with a new mdl2 format. or new monsters which fit a retro theme. higher res and smoothly animated, not a godawful eye-rape like you guys think i am going for lol.

i understand the desire for a retro look and i support it, that's my preference too. i don't think high-res is mutually exclusive to a retro look though.

as it stands, if i tried to re-do original monsters to look better but still identically the same as they ever did, i hit a hard limit. also as mentioned earlier you can't do things like a terrain with mdl very well. original quake did have some outdoor areas, i think it would be interesting to do some as mesh.

nobody really expects me to create and release monsters that don't work in any engine before limits are increased do they? i know for a fact some mapper didn't invent the bsp2 format himself, then make maps in it, then upload these and say hey here it is, please implement my new format in your engines. he would of also had to make the .lit file spec and the compilers and everything else himself too.

that's basically what you guys are telling me to do with my models here, there isn't really the ideal model format existing yet. iqm is good but not completely ideal since i don't think it supports vertex anims, just skeletal.

i don't expect anyone to code support for this for me specifically, but i'm just putting it out there to illustrate a problem for modellers and animators in general.

you could ask on polycount what they would want for more opinions from actual modellers and animators, since this is mostly just a mapping forum. 
 
Show me your portfolio, or at least examples of models you did in the past. 
First | Previous | Next | Last
You must be logged in to post in this thread.
Website copyright © 2002-2024 John Fitzgibbons. All posts are copyright their respective authors.