What are you working on?

[Julius]

@toneddu2000: great, was wondering what came out of this after the quake expo. Please put in on github soon

Also: will it have an Linux, Android and WebGL version as those are supported by FTEQW?

Edit: Does it include the CSQC realtime light editor and support for Trenchbroom for easy level editing?

[toneddu2000]

Thanks Julius! I'll sure put a download link when finished (I never used GIT before so I should look at some documentation), but probably it will be just a link to a Sourceforge zipped file. The whole source is quite big (almost 5 GB) so I need to understand if I can upload all that stuff on GitHub (imagine that every texture source is PSD with mininum resolution of 2048x2048 pixels, so..).
It should run on Linux without problems(ODE lib is a mistery anyway), but forget about Android: too many drawcalls. For web, ehr..I could try (probably with a little Spike's help) but I can't be sure it will run above 15/20 fps in WebGL(probably ditching all rtlights).

Edit: Does it include the CSQC realtime light editor and support for Trenchbroom for easy level editing?

Realtime lights are very few and placed in game via code (99% of lighting is made via q3map2). Editor used is NetRadiant, so, sorry about Trenchbroom. But, are you sure Trenchbroom can handle q3bsp maps? Last I saw it, I understood it could handle only q1bsp ones.

[Julius]

Well, you could just put all the QC code into Github so that others can easily contribute fixes and it would also give you a nice issue tracker that many people use. The artistic source files and all the rest can stay as a separate download on sourceforge or such.

Is FTEQW somewhat inefficient with real-time lights? I would have imagined that when starting a new game from scratch it would have been a good time to get away from the legacy tech. With only real-time lights you could have also used a more simple map-format that is supported by trenchbroom (currently Q1 & Q2 bsp, but Q3 will come sooner or later).

Don't get me wrong, I really applaud you for this cool one-man project, and it is nice to finally see a game that uses FTEQW... but it feels a bit redundant at this point (a bit like a poor man's copy of Xonotic). What I would really like to see (and maybe it is not too late for that) is a quake based game that ditches all the legacy stuff and focuses on being really easy and fun to edit. So basically a small fun base game, that's fully open-source and comes with easy to edit source files for everything, as well as a modern level-editor that comes ready setup with the download etc. I think the way QC works, makes it actually a really nice modding platform (as the original intend and as evident by the still existing community after all these years), but the rest of the quake 'ecosystem', especially mapping and and ready to go modern base-game without overly complicated stuff is kind of missing these days.

[toneddu2000]

Well, you could just put all the QC code into Github so that others can easily contribute fixes and it would also give you a nice issue tracker that many people use. The artistic source files and all the rest can stay as a separate download on sourceforge or such.

Yeah, that could be a good idea

Is FTEQW somewhat inefficient with real-time lights? I would have imagined that when starting a new game from scratch it would have been a good time to get away from the legacy tech. With only real-time lights you could have also used a more simple map-format that is supported by trenchbroom (currently Q1 & Q2 bsp, but Q3 will come sooner or later).

Well, when I started projectUnknown I imagined to trash all the lightmaps and go for direct lights only. After 2/3 weeks of tests I changed my mind: unfortunately FTEQW is VERY inefficient with dynamic lights (just to make an example, Tesseract, runs on my pc at 60 fps stable with dynamics lights, global illumination, anti aliasing at maximum, etc. FTE at 1920 x 1080 with 4 dynamic lights slow down to 19/25 fps, ugh..). Another point is global illumination: without lightmaps, you have only direct lighting in game. This is an aspect I didn't take in account went I went to "full dynamic" approach, but, after played A LOT with q3map2 bouncing feature, I noticed how beautiful could be high resolution lightmaps with colors blending each others. Spike said several times (I ask him every month or so ) that he's too paranoid to take in consideration to add gi to real time rendering system, because of performance issues, so...
A good thing would be porting Tesseract rendering system to FTE (or adding fteqcc support in Tesseratct, I don't which is the craziest thing! ) but it's not a task that I can handle with my poor programming skills. But I',m pretty sure that, with this addition, a stable, functional physics system (ODE, no thanks), and a decent site with wiki, lot of documentation ,etc. FTE could be considered, with no problems, a real alternative to Godot engine, for making games with open source tools.

Another problem is audience. You (impersonally "you" , of course) could put a lot of effort on studiyng, learning things, creating code, assets, etc. but usually, people that would follow your work would be 3 or 4. That's it. Because 99,99% of game developers or game dev enthusiast are focusing on Unity5 or Ue4. (Even CryEngine is almost abandoned right now!! ) So, when you think to re-create everything from scratch, you ask yourself: will anybody give a *#@*% about my work?
Frustration is another thing to take in consideration

Don't get me wrong, I really applaud you for this cool one-man project, and it is nice to finally see a game that uses FTEQW... but it feels a bit redundant at this point (a bit like a poor man's copy of Xonotic). What I would really like to see (and maybe it is not too late for that) is a quake based game that ditches all the legacy stuff and focuses on being really easy and fun to edit. So basically a small fun base game, that's fully open-source and comes with easy to edit source files for everything, as well as a modern level-editor that comes ready setup with the download etc. I think the way QC works, makes it actually a really nice modding platform (as the original intend and as evident by the still existing community after all these years), but the rest of the quake 'ecosystem', especially mapping and and ready to go modern base-game without overly complicated stuff is kind of missing these days.

No, No, I don't get you wrong, it was my purpose at start to create a project from the ground up that creates a new standard (am I humble, am I?) for open source games. The problem is that I failed! Of course, if I failed, it doesn't mean that someone, when code and assets will be released one day (I'm paranoid for perfection too! ), he/she won't be able to improve it.

Another thing: projectUnknown is not a game, still, (as I described in the site). It's basically a proof of concept that I started with an idea in mind like "can I be able to make all by myself a simple arena where people can slain each others?":) It's NOT a poor man's copy of Xonotic. I'd be so happy if it was!

Regarding TrenchBroom, let's wait for q3bsp support. q2bsp sucks overall..using now TrenchBroom would seem to me like going backwards instead of afterwards

[Julius]

Ok, I see. Thanks for explaining in detail... a bit of a pity, but lets see how things develop.

Oh and there will be always some things to fix and improve, so don't worry about that and make a public release. You will probably not get too much feedback initially (so don't be too disappointed), but at least its out there and people will start taking more interest. No one benefits from it just sitting on your harddrive

[Barnes]

hdr bloom

[qbism]

hdr bloom looks good with the chroma distortion. Could there be bloom that scales- more bloom for explosions and very bright lights?

[toneddu2000]

Really cool effect Barnes, it adds a lot of realism to the scene! Compliments!
I agree with qbism, adding glsl post processing chromatic aberration would be great. I added it to my projectUnknown (converting a glsl from shadertoy to suit FTE needs) game and it really pumps the graphics up!

[Barnes]

remake bloom shape (use star shape), all work w/o hdr, more shader optimizations

[toneddu2000]

mmm... personally I prefer old one, this seems too fake, but probably there are situations where the use of this kind of effect could be more appropriate

[Barnes]

Previous versions gave too many artifacts. Strong flicker, wrong glare size on large luminous surfaces. By this, I modified the original version (star shape)

[toneddu2000]

I see. So, switching to star shape reduced artifacts. How is possible? Or is still possible to use new version with elliptical shape bloom?

[Error]

BEHOLD! My Zelda : A Link to the Past inspired WIP

[toneddu2000]

Soo cute! I wanna see it in motion, now! Video! Video!

PS: what's the glow on the left bottom flower? Mouse cursor?

[Error]

Sadly, I still haven't found a reliable way to take good videos... so I'll have to look more into that.

The light is simply that, a test light in the test map.