31 replies to this topic

[TheNut]

Not a lot of interesting talk lately so I figure I would start-up up a social thread to see what everyone's working on. Advertise your stuff if you want, or just kick back and discuss what you're doing.

I have two exciting projects I'm working on right now. The first one is a game in Silverlight that's based off a classic remake of Scorched Earth in space. I don't remember what the name of that game was, but it was highly addictive. My friends and I played it day and night back in the old days (mid 90s?). I'm taking a unique twist to developing it though. I'm not producing any art, sounds, or music for it. Instead, the whole game is procedurally created. Levels, graphics such as planets, nebulas, and starfields, music, are all built on the fly (and cached if I need to reuse it). Most of the tech is in place, so I'm focusing more on the game content now and preparing data and new algorithms to generate the tunes, sounds, and tweak the graphics for performance vs quality. I make use of as many cores as I can, but even on a single core the load times are quite good. This leads into my second project.

When I was creating tunes for the game, I got really pumped about the whole thing. So I started up a new synthesizer project (also in Silverlight), which is also a great way to test my sound engine. It already has adjustable oscillators and a working piano roll. I'm gearing up to add instrument support, saving and loading projects, and exporting to WAV. I'm hoping to wrap up the initial version by the end of the weekend so I can start using it to build the tunes I'll be using for the game. Nothing orchestral of course, but it's not like I'm capable of doing that even with expensive software ;)

Here's a screen grab of the current version.

[starstutter]

Im up to homework man... lots and lots of homework *~*

One thing I am working on latley is a universal instancing system. Using DirectX9 (meanong no geometry shader of course), I'm setting up a system that allows for automatic instancing of similar objects on screen.

The objects are first stored by material and mesh. During rendering, it is sensed if an object is followed by an identical one next up on the list. If so, instead of using the object types regular mesh, it draws from a different mesh which has object indicies as part of the verticies (stored in normal.w). These instancing meshes have to be created during startup but its not too difficult to predict which meshes will be drawn the most often.

Just wanted to share :)

Nice work so far though.

[Mattias Gustavsson]

I'm working on several games (yes, games and not just tech :happy: ), which I'm jumping between as I please, taking advantage of the lazy and laid back life of being a student :yes:

Right now, I've just switched back to my top-down RPG:
Escape from the Dungeon

Three adventurers have become separated while exploring a dungeon, and you control all three, trying to reunite them and find the exit.

Download wip version here
Read more about it here



I've also just started working on a first person RPG:
Untitled "Dungeon Master"-style game

Where I'm hoping to create a very immersive and interactive dungeon crawl experience in the style of the old classics.

Read more about it here


I also have this other RPG (notice a theme? :blush:)
Thieves and Warriors

This game is inspired by old pen&paper RPG's and games like Pirates! or the Phantasie games, and I'm going for complex and comprehensive simulation here, with little in terms of animation.

Read more about it here



And of course, I also have my side-scrolling RPG:
The Jade Figurines

Which I made over a weekend for a 48-hour competition, and which I fully intend to keep working on - but I need a break from it right now to regroup and rethink some things about it.

Download compo version here
Read more about it here
Watch video here


So yeah, that's the stuff keeping me busy at the moment :-)

[TheNut]

You guys are lucky to be students (well, until exams come around, mwhaha). I have to put in my work hours before I get to indulge myself. Even now I'm pretty antsy to work on my projects, but I have to control myself ;)

starstutter,
I remember you posting a lot of interesting discussions over the past while about shadows and other advanced topics. How did that turn out? Are you using this towards any goal or just learning for fun?

Mattias,
You're an animal with Poser. I couldn't quite get lighting to work right in that app, but your screenshots are very good. Of all those projects, I would say the Dungeon Master style game is the most interesting. I pretty much grew up on those games and they were really fun. I still remember games like Ultima and EOB. Great classics. I've also grown very fond of those style of games, including Myst like games where you solve puzzles and explore beautiful worlds. It's a good break from the typical run 'n gun style games.

[karligula]

I've been doing a cinematic engine for Blender files, and I'm currently working on some trapezoidal shadow mapping but my trapezoid transform isn't working yet... WAAAAAAH...

[roel]

This vacation I was working on a game:




... but I got bored of it: I couldn't get myself creating decent art, because I got a cooler idea: a voxel based terrain engine, very early stuff:

http://www.youtube.com/watch?v=DktCwxgOR6s

and experiments to simplify the poly mesh:

http://www.youtube.com/watch?v=BB89sdO3cAQ

[fireside]

I'm working on a Flash adventure game that will use a nodal system with 4 rotating views. Early stages, artwork done in Blender.

[karligula]

Hey Mattias,

Just an idea for your Jade Figurines game... I always hate it when you defeat a character and they just flash for a bit and disappear... how about having a dog run onscreen and drag them off like they were scavenging?

[Wernaeh]

Most of my time these days is spent on preparations for my diploma thesis - nonlinear optimization using an interior point method.

Apart from that, I'm still doing some freelance programming work for the automotive industry, so there is little time left for private projects.

However, I'm currently texturing and normal mapping this geometry stuff:




so I'll finally have some proper testing environments for my deferred renderer :)

Cheers,
- Wernaeh

[JarkkoL]

Working full-time on my game engine project and have been mainly writing documentation for the past week, duh! Ported the engine to x64 recently. Evaluating various 3D modelling/animation softwares for purchase (modo 401, 3DS Max, XSI, etc.). Negotiating some deals and various other non-programming related fun tasks. Will probably continue working on the engine skeletal animation system next week (event based blend graph) and try to design robust undo/redo system for Spin-X Editor... or maybe something else if something interesting crops up (:

Also waiting for Hauppauge HD PVR to try out some HD video capturing (:

[laweya]

I've been playing around with game development for about 5 years and haven't completed a project yet (bad I know). So I'm currently working with The Zombie Engine (http://thezombieengine.sf.net) to complete my first project - an FPS. Its still in in the planning stages though.

I'm also trying to fix bugs and improve the engine where I can.

[rouncer]

I just want a virtual world to call my own, im still not even past the design stages tho... all i have for something for it is a long list of failing demos that didnt get anywhere cause they either didnt work, had too many bugs or didnt have a decent result.

but i still havent given up... i like what everyones done especially those shots from Wernaeh, they are great!

[Wernaeh]

Quote

i like what everyones done especially those shots from Wernaeh, they are great!

Thanks for the compliment :)

Quote

I just want a virtual world to call my own, im still not even past the design stages tho... all i have for something for it is a long list of failing demos that didnt get anywhere cause they either didnt work, had too many bugs or didnt have a decent result.

but i still havent given up...

That's the spirit ;) Just stick to it...

IMHO the most important three tips for motivation are:

=> Lower your expectations

You needn't have full-blown 3D graphics and lots of scripted scenes to convey atmosphere, have character development or to create your own world. Just think of old console games... You can even get away with text-only dialogue, and sparse still-frame images for cutscenes - after all, you are _not_ an A+ game company with dozens of specialized programmers and artists.

=> Use existing code and focus on the fun aspects

If you think it is fun to code a renderer, do so (I did this) - otherwise, use an existing one. You will have much more progress that way.
If you think it is fun to write a collision detection library, do so (I did _not_ do this, too much number crunching, I'm doing this professionally already!) - otherwise, use an external library.
... and so on.

=> Polish first, then integrate new features
Do not discard working code. Rather, fix any bugs, make it work properly (rewriting is the extreme case, but only sections, never start over from the ground up) - then make it work in a polished way (i.e. add proper graphics) - this way, you have something to show off to friends or whoever, and this will keep you motivated.
Perhaps maintain a blog or something where you present your weekly features, alongside a feature list that you aim at implementing for the next week.

Hope this helps.

Cheers,
- Wernaeh

[rouncer]

Ive got a few directories full of screen shots of everything I've done over the years... some of the stuff is ok looking, thats where i go if i need inspiration.

My code is cursed! If i dont touch a program for a while and come back to it after a couple of months it just wont even launch and itll say theres some memory fault.
Once Ive lost passion for a project, it'll just do that and die.

I consider being a programmer as ones profession makes you responsible to be a genius or you might as well give up and go get a shit kickers job for the rest of your life.

Im right arent I, programming is TOUGH.

But thanks for the words of wisdom. :) I listened well. Yeh why not go finish a smaller project, its a good idea.

[starstutter]

TheNut said:

starstutter,
I remember you posting a lot of interesting discussions over the past while about shadows and other advanced topics. How did that turn out? Are you using this towards any goal or just learning for fun?

quite well ^_^
Actually I've moved WAAAYY past those topics. Global illumunation, instanced grass rendering, god rays/volumetric light scattering, subsurface scattering, 5 types of new lighting models (including optimized cook-torrence/oren-nayar), cascaded shadow maps with incrimental updating (for distant shadows), and... I have to get back to homework =P

There is a lot more stuff than that, and while it is fun for me to create, I hope to use it for a 'portfolio'.

[laweya]

Wernaeh said:

Thanks for the compliment :)



That's the spirit ;) Just stick to it...

IMHO the most important three tips for motivation are:

=> Lower your expectations

You needn't have full-blown 3D graphics and lots of scripted scenes to convey atmosphere, have character development or to create your own world. Just think of old console games... You can even get away with text-only dialogue, and sparse still-frame images for cutscenes - after all, you are _not_ an A+ game company with dozens of specialized programmers and artists.

=> Use existing code and focus on the fun aspects

If you think it is fun to code a renderer, do so (I did this) - otherwise, use an existing one. You will have much more progress that way.
If you think it is fun to write a collision detection library, do so (I did _not_ do this, too much number crunching, I'm doing this professionally already!) - otherwise, use an external library.
... and so on.

=> Polish first, then integrate new features
Do not discard working code. Rather, fix any bugs, make it work properly (rewriting is the extreme case, but only sections, never start over from the ground up) - then make it work in a polished way (i.e. add proper graphics) - this way, you have something to show off to friends or whoever, and this will keep you motivated.
Perhaps maintain a blog or something where you present your weekly features, alongside a feature list that you aim at implementing for the next week.

Hope this helps.

Cheers,
- Wernaeh

Totally agree, my biggest draw back has been trying to have something that does everything from the word go which has resulted in a lot of wasted time trying out different engines and discarding them.

It's very motivating to just have something small working and growing from it into something better and implementing features as they are needed.

[Vilem Otte]

Nowdays I have two projects on my head ... one as full time work and one as hobby one. Although there are two more projects I'm cooperating (they're child projects of my full time work one).

Iceberg lake release
I'm going to release the engine I'm working on and it is going to be this year.
This is not my hobby project (although I really enjoy it very much), right now we are mostly finishing editors and testing it (plus on last hour implementing Havok physics .... I'd say that we began SOON with this).

Some show - material editor and world editor (click to enlarge!) ... though not WYSIWYG, runtime looks better (has more effects - editing in CPU eats too much to do ray tracing for GI and other nice looking ray traced effects).





Forgotten library
This is my hobby project - though it is connected to previously named project (it uses this library).
I did some research into realtime ray tracing and this library is being built as interactive ray tracing library (it is not finished! although it can be used for achieving nice effects with hybrid rendering (case of Iceberg lake release) or even realtime ray tracing).
And as everyone interested in realtime ray tracing I'm looking forward to Larabee :P (every day checking if some new specs are out, or if it is out :D).

[Mihail121]

I'm up to tons of stuff, mainly for university. First, there is a project to convert DrawingML diagrams to SVG -- it's for a seminar I'm taking -- adapting MS Word 2007 for blind people. It's more than just hell... Then I'm writing some reports for another seminar about automated game players: you give your player a LISP-like description and it starts working... Google for "General Game Playing". I'm also researching linear shift regs for my Bachelor thesis and doing some benchmarks for a paper I'm involved in. There's no such thing as free time so I'm giving my best to like what I'm doing.

[geon]

I'm officially a game developer since the end of August, woot!

I'm working on a racing game to be released for the iPhone, using Unity.

Screenies:

[TheNut]

geon
Nice use of lighting there. It goes well with that scenery. Should be fun watching people on the subway trying to race each other. Swinging there arms around, cheering, cursing, ... :)

Vilem Otte said:

And as everyone interested in realtime ray tracing...

It's funny you mention that because not to long ago I was tossing around the idea of writing a ray tracer. I've been doing a lot of work lately around procedural texturing and there have been cases where having a ray tracer, especially on platforms without any 3D libraries or accelerated graphics cards, could benefit from it.

laweya said:

my biggest draw back has been trying to have something that does everything from the word go which has resulted in a lot of wasted time trying out different engines and discarding them

I wouldn't call it a waste of time. Even if you haven't output anything tangible, you still learned about the engines you were using and their design flaws. Knowing how other people do things I find helps you become a better developer. In my case, I didn't really think much about a signal / slot system, widely implemented in Qt. I use to do things much more "coupled", but after having learned about new ways to manage events, it really improved my productivity and code quality.