But as long as it�s made optional with a .lit file, I don�t have a problem.

indeed.

i'm veering slightly towards willem's point of view (on coloured lighting in quake), although i don't mind it as a novelty every once in a while. certainly not a necessity though

i think skyboxes & fog are fantastic however, so i'm far from a puritan in that respect 8) 

lol engine holywars

for a record: only fitz and DP have overbright lighting. in all the other gl engines lighting looks rather flat

http://speeds.quaddicted.com/quake_light.jpg 

"realworld" example
http://www.speeds.quaddicted.com/q1light.jpg 

If you want to get on #terrafusion occasionally, but, like me, couldn't be arsed finding and installing a decent IRC client, try this:

http://www.mibbit.com/

It's a web based thing that seems to work pretty well. All you have to do is choose quakenet as your server and enter #terrafusion as the channel (and enter your nickname, of course).

There was no obvious downloading or installing of stuff, or any other shenanigans... it just worked, and it was pretty good. 

hey, that's kinda cool. 

roar 

I guess I deserve this thread for having the gall to state a preference. Sorry all! 

etc 

is a really good engine!
I mean for a balance between compatibility and sheer eye candy.

And AguirRe's engine is immortal. 

"I see little reason for colored lighting, interpolation, or any other fancy stuff in Quake. Quake is Quake. Accept it for the greatness that it is or map for another game."

is not being a purist. It is being a puritan.

That's an excellent way of putting it, I like the analogy. There's no need to religiously stick to the original engines when there are engines out there that do the same things better. Fitzquake is pretty much just GLQuake minus the bugs, and all the other changes bring it closer to the original feel of software quake.

I don't mind new features as long as they're tastefully done, basically. Coloured lighting is a gray area, it isn't distasteful of itself, but it's difficult to do it well. 

roar - but louder. 

for me, interpolation only highlights all the limitations in the Quake models - mainly the jelly effect due to the fact that each vertex coordinate only has 8-bit accuracy; and the fairly simplistic 10FPS animation that was never designed to look good when interpolated.

Similarly, anything other than white light seems to turn Quake's 256 shades of brown into homogeneous flattened dog turd. 

Seeing as the current topic is engines - how feasible would it be to eliminate wateralpha from the cfg and make it a by texture number, configured mapside.

It'd allow a fair few tricks - animated fog being the first idea, as well as proper teleports alongside water that can be more or less transparent depending on how dirty it's supposed to be. 

16-32 bit rendering would mean that the shades of colour added to the original colours of the pallette wouldnt look crappy because of the higher bit-depth. Because a high bit depth allows for bla bla bla no no, no.

It wasn't a technical statement. It was an artistic one. Quake's color pallete is much more muted, and there's a lot of subtlety in the differences between shades and hues, and in the way id's texture artists combined them. Richly colored lighting has a way of overriding and cancelling out texture colors (see: red light on blue object), and Quake's palette is particularly unhappy with that kind of abuse.

Check out some of the copper textures, the ones with the nice green oxidation on them. The brown and green are roughly the same luminance, and even just a tinted light can ruin the hues that separate them, turning them into, as Kinn stated, visually undifferentiated poo. 

but rickyt your colored lighting was good, since it was suble and not obvious and made different feeling to inside and outside parts of the map.

I can understand why some people want to play without colored lighting.

I can NOT understand why people want to play with vanilla glquake crappy lighting, no fullbrights, gl_flashblend yellow spheres... It's much worse than software for chrissakes, those are bugs that were in the first glquake that have since been fixed in fitz, fuh etc... 

Well, there might be various ways to associate an alpha value or any other attribute with a texture in quake, using either a texture config file or some sort of shader system. There isn't really a standard way that engines do it (unless you ask the engine coder himself, and he'll say his way is standard -- as soon as others adopt it) but that's how i would see doing it.

However, in any engine with per-entity alpha, you could simply make your objects func_walls and give them the alpha you wanted. I'm not sure how other engines do it, but in the version of fitzquake I'm working on, I made entity alpha completely override wateralpha, (rather than scale it or something) so that you could have your faded ground fog, and opaque teleport texture, just by manually setting alphas on the entities. (though, i think you might need to set the teleporter alpha to 0.999 if 1.0 is considered "default" -- maybe I should go examine that.) 

But turning brushes to entities means they can't be used as water - so to have, say, two different transparencies for water brushes isn't possible?

It'll also stop the water 'animation' or warping effect, AFAIK.

Just thinking aloud here.

Looking forward to the next version of FitzQuake. 

I'm completely serious. GLQuake looks and runs fine. Any new feature added either doesn't look great with the art content of Quake, or is done far better in a newer engine. That's not to say increased limits aren't nice, but I think it's strange to use more entities or tris when designers aren't even getting that much more fun out of them.

After 11 or so years, my opinion is that we haven't even explored much past the simple gameplay of the original id maps.

As it is, the Quake mapping community is still creating the same old maps, the same old gameplay, and attempting to dress it up with nicer visuals. Orges will always be above the player to lob grenades, you'll always need to get the gold key to get through a locked door, etc etc etc. At this point, you may as well use a more recent game that looks nicer since you really aren't pushing forward the gameplay standards. And if mappers aren't making all that much better stuff, why bother?



You may want to take my comments with a grain of salt, since I've come to feel that single player FPS games in general are tired and played out, an entire genre of "click on things and they die" mixed with "find random object, progress, repeat" as the fed-to-you 'storyline'. 

 

...he said 'poo'. 

reluctantly though 

you can't seriously be arguing that there hasn't been anything original. Besides, what I really get out of Quake is the atmosphere. Most new games don't have any. 

I agree almost completely with basically everything you say. There's one major flaw in the argument that I simply cannot get past though:

Standard glquake is pants.

If glquake had have been done properly (say, it was as good as FitzQuake, disregarding 'new' features like skyboxes etc) then sure. Completely agree.

However, let's face reality: glquake is riddled with flaws and bugs that detract from the experience (and is missing important features that were present or implemented better in the software renderer!)

Load up a map in glquake. Fire the lightning gun... notice the dirty grey shaft (no fullbrights) that isn't even connected to the end of the weapon. Notice the shit lighting (see speedy's pics above). Jump in the water and observe the problems that metlslime mentions.

Then change maps and crash with a texture cache mismatch :D

At the end of the day, FitzQuake is closer to proper Quake than standard glquake is. 

This is not done yet but I could use some feedback already:
http://www.quaddicted.com/feature/engine_comparison.html

First I wanted to add all known engines but luckily I quickly realised that it a) would be pointless and b) another shitload of work. So I cut them down to something I consider a good range of decent singleplayer engines.

I will add Tyrquake and Makaqu. Q2K4 will probably be excluded unless I find a way to disable the "PPC hitboxes".
Feedback on the engines list and the tested features is very welcome. Please mind the "todo".

One thing I really need is some sort of stress test demo that ideally would top the "usual" limits and bring engines with non-raised limits to their knees (crashing, buffer overflow, missing stuff or whatever). Any suggestions for maps? I thought about that crazy "rectangle filled with monsters" aguirRe and sielwolf posted about some months ago.
Or maybe several demos focusing on different limits.

Also I would like to test some recent map for bugfree-ness. Probably one of neg!ke's speedmaps would work fine, any recommendation? 

There are really two different ideas here that are getting conflated -- what is a good engine to play quake with, and what is a good platform for creativity. You seem to be saying that because quake is not a good platform for creativity, glquake is a good engine to play quake with.

Since championing glquake is semi-absurd, i'll move on to your more compelling point, which is that the overall quake mapping scene has run out of any creative relevance, and we are basically making new maps with tired old gameplay. And, if we really want to do something innovative in either the graphics, or gameplay departments, or both, we should be using a more recent game as a platform for that. This is probably true.

But, there are a number of reasons not to, some of which might be valid. Switching games also means effectively switching communities, which is difficult (to find a decent one, that is.) It also means starting over in terms of technical knowledge, which for people here that are not and never will be pro game developers, is not worth it for a side hobby. There is also the fact that after a certain amount of time, the core fans that remain with a game have become near-100% blind to its shortcomings, which means that any other game won't be as "perfect" in terms of atmosphere, gameplay, art style, etc.

For me personally, there is still value in making levels in the relatively done-to-death conceptual space of single player quake. It's a well-understood, familiar set of game mechanics, and yet I still don't feel that I've mastered the art of making levels for it. As long as it takes effort and thought to make a level, I feel like there's some further enlightenment waiting for me to puzzle out. And the familiarity of the game mechanics, technology, and authoring tools, means that I'm spending my time working on the actual craft and not working on learning new tools, tech, or gameplay.

But yeah, I probably should have moved on already, and I probably will move on eventually.

It's also possible that I will find a day job that fully satisfies my game design urges, but it hasn't happened so far -- you're always rowing to somebody else's drum, even if you're leading a team and the drum is the CEO, or the publisher, or the mass market.