Stressed Gamer

New Site and New Game

2007-02-28T07:21:00.000-08:00

Hello All.

I have not posted in a while because I have been busy doing two things. The first is that I have a new site: www.stressedgamer.com and blog. The second is that I have been working on a new project for the last two weeks and it is now done. Head over to the new blog to find out more.

-tom

Level 1 Continued...

2007-01-26T21:33:00.000-08:00

Well I did not get everything done I wanted too on this level. It is much nicer however:

Still need to add lighting to the level, several particle effects are missing or placeholder, all decoration still needs to be done, the enemy weapons need to be modeled, and the portal still looks like a blue sphere, not to mention the text… Hopefully I can get all that done this weekend so I can get started on the intro and space mode.

-tom

Level 1 Mockup.

2007-01-19T20:42:00.001-08:00

Well I just finished mocking up the first action level of Warriors of the Shining Star. Right now just about everything in it is test art. And yes health pellets are attacking me near the end of the movie (I needed a small test mesh for a monster…):

This next week’s task will be to make that look nice and to get it balanced. To get an idea of the art assets I have left to make for it:

Map Tiles:

Base Hall
Base Hall Three Way
Base Room Corner
Base Room Corner Entrance1
Base Room Corner Entrance2
Base Room Floor
Base Room Wall
Exterior Wall
Landing Pad Floor
Landing Pad Corner
Landing Pad Edge
Landing Pad Inner Corner
Landing Pad Door
Landing Pad Wall
Landing Pad Wall End1
Landing Pad Wall End2
Black4x4
Moon Surface 4x4

Character Meshes:

Miner
Miner Boss
Possessed Miner
Possessed Fire Miner
Demon Portal
Little Demon
Sergeant Salter
Militia Fighter

Weapon Meshes:

Mining Laser
Flame Thrower

Effect Meshes:

Mining Bolt blue
Mining Bolt Red
Fire bolt

Decoration: (probably need more here…)

Transport Ship
Moon Buggy
Misc. Debris1
Misc. Debris2
Misc. Debris3
Misc. Debris4
Crate (yes I’m having crates, I like crates)

Hmm this could take me a while, but at least I will get some practice with the art tools…

-tom

The dancing AI.

2007-01-10T20:23:00.001-08:00

I’ve been working on AI tonight and I decided place down a bunch of AIs that hate each other, comedy ensues:

I made a couple different AI types tonight. The one seen in the video above is the strafe shooter AI. It is one tough bastard to actually hit because it deliberately doges your aim. The video makes it clear I need object to object physics in soon so they collide with one another to give me a chance. I already catch the collisions; I just don’t do anything yet.

This thing is actually starting to look kind of like a game, probably my favorite part of development.

-Tom

It's about character.

2007-01-09T19:24:00.000-08:00

Character modeling has never been a skill I really possessed. It is the part that has killed almost every personal project I have worked on. I wanted Warriors of the Shining Star, the name of the game I’m working on, to be semi realistic. Due to the aforementioned inability to create characters that is impossible.

So I’m going to go with something simpler and within my limited abilities. For anyone that knows me this will come as no surprise. I have decided to go with spheres (and other simple shapes) for all of my characters. While it is not the look I wanted it is a look that will provide a certain amount of characterand style to the game. Here is a sort of sample of what I’m talking about, the Warrior class:

Let there be light!

2007-01-06T21:29:00.001-08:00

I got a little side tracked this week. I planned to work on gameplay but instead worked on graphics.... In this case I did a first pass at a lighting modle for the game. I decided to keep it simple, no normal maps, no displacement maps, etc. I have done these effects before but find that I don't have the time to make art that uses them.

A nice side effect to doing this was that I now have full shader support in my game. This will give me much more flexability in what I can render over the BasicEffect that comes with XNA.

Here is a little sample movie to demonstrate:

-tom

Rolla Station Test.

2006-12-31T18:40:00.000-08:00

This is a quick test of the station tile set I've been working on. Also shows the player activating a conversation (ok an old style one sided conversation)...

We have collision...

2006-12-26T21:26:00.000-08:00

Well I have the base game framework up for the XNA game I'm playing with. I just put in avatar control and a basic collision system, 2D nothing fancy but it works.

I tried to put up a quick movie.... youtube did not like it, I'll try to get it up again later.

Update:

We have video.

Better tiles.

2006-12-24T05:32:00.000-08:00

The last screen shot had some crapy test tile. I made some slightly better ones to get some practice modeling again.

Game Studio Express

2006-12-23T19:32:00.000-08:00

Again it has been too long since I posted. I really need to work on this whole weekly posting thing. I’ve been playing around with Game Studio Express for a few days and am impressed. I think Microsoft will get what they are looking for, a lot of good fan games on the 360.

It is very easy to pick up and just use. The content pipeline is especially easy, which is probably my favorite part. I usually spend a few weeks setting up content pipelines on hobby projects only to work on the actual project for just few days. Frequently the content problem has been enough to make me not work on a side project. With Game Studio Express I just dropped in my content and it just worked.

Extending it has been a breeze so far. I wanted to add camera frustum culling, so I went about setting up the math only to discover they had all the frustum math code already made. I just needed to set it up in my camera class and call it. Most features I wanted to add have been this way, the base code was already written and I just needed to put it into my framework.

The biggest hurdle has been learning C# which actually is not that big a deal. The IDE is very helpful in this regard, almost keeping me from doing anything illegal. Finding how to do certain tasks could have been made easier, but I should probably buy a book for that anyway.

So I have quickly set up a 3D tile map editor:

I think I am going to do a simple dungeon running game with it. Not 100% sure at this point. I need to see if there is a C# physics package floating around to hook up to this.

I also need to figure out how to do my characters for this. I’m not really capable of modeling them myself so I may need to do something 2D.

The Challenge

2006-11-03T20:24:00.000-08:00

Anthony had challenged himself to write a game in one page of code, 80x60 characters. He finished with a pretty neat game called sandworm.

I thought the idea was pretty cool, so I tried to do it myself. I decided to see if I could do mine in 3D. After playing around a bit I figured out I could get the 3D and windows going in about 11 lines leaving me with plenty of space for my game. Then as I started to play with some ideas I got a simple action RPG game going, very Zelda like. I figured if I was careful I could get about 4 dungeons connected by an over world together.

This is when things started to go wrong. I kept adding little details; new rooms, tile types, monster types, weapons… Pretty soon I was up to a page and change and only had my first dungeon done. At this point I realized I was not going to get it under a page and have everything I wanted. I was about ready to give up when Anthony (to get me to finish) added the two page challenge to his web site. He did it just to make me finish it.

So now I am thinking to myself, “well the framework is already written I should be able to get it all in the next 40 lines.” Yeah right. Being me I kept adding extra stuff.

Well so much for the two page RPG… Let’s make it three…

So at this point I’ve pretty much blown the 120 line budget. Now the game weighs in at 165 lines. Only 45 lines over budget, maybe no one will notice. I could probably remove about 10 lines from the game before I have to start removing actual working code to get it down. But at this point I like the game how it is so I don’t want to remove anything.

So I have decided to call it done just so people can play it.

Here is the game.
Here is the code.

The goal of the game is to gather the Quad Force together. I would have needed four pages to add the princess to rescue. Your character is the little blue sphere in the middle. Movement is done with the wasd keys. Heath are the bubbles in the upper left hand corner. Pieces of the Quad force are in the lower right corner as you find them. There are items you will pick up along the way that will give you new abilities. Three of these are tied to the 3 mouse buttons and will be displayed in the upper right corner as you get them. Play around with them to figure out what they do.

Edges

2006-10-14T19:19:00.000-07:00

So I finally got a chance to sit down and work on the render system I outlined. Overall I’m pretty happy. Still needs tweaking but the basic idea is sound. You can see the crayon outline effect from these screen shots:

However you can also see I need a better way to set up the UV coordinates on my models. The scale of the UV cords is so large that the art tool has trouble with it. I may switch to some kind of method to generate the UV coordinates from a set of base coordinates in the mesh. The real issue is that I want to get orientation data of the texture from the UV coordinates but not the scale information. I need to also preserve edge boundaries of the between polygons so I don’t get seams when I change the UVs. I think I read about a patching algorithm along these lines at some point… have to hunt that down. If anyone has a good idea, feel free to let me know…

Far too long...

2006-10-11T17:30:00.000-07:00

It has been far too long since my last post. I’ve been crunching on finishing up a project at work, the PC port to TopSpin2. I have not had a chance to work on anything for Alexander’s Kingdom in a while. Right now the project is sort of in limbo because of art problems. Basically I’m having trouble actually sitting down and finishing any of the art.

The big hurdle has been learning how to rig and animate as skinned mesh. I’ve figured it out for the most part but don’t seem to have the time to invest in it. Hopefully I’ll get a chance soon.

On the graphics side I’ve been thinking about a solution for the crayon line edge problem. I want to render out a depth/normal (Nx,Ny,D) buffer of the scene at the same time I do the base color pass. As a full screen post pass I take this buffer and look for large changes in either depth or normal values between texels. Where I see a large change I render a black pixel modulated with a paper texture. This should give me edge lines. I’m hoping I can remove the jagged edges that are standard with this solution by over sampling the depth/normal texture to blur out the test, we will have to see.

After that I was thinking of generating a line color buffer to decide what color to draw the edge lines. This would be another render target like the depth/Normal buffer (or possible an indexed palled using the last channel in the Depth/Normal buffer. When an edge is detected it renders the edge color of the sample point nearest the camera. No idea if it will work but it should be better than black lines.

I’ve also wanted to code a distortion buffer for a long time but I don’t think it will match the crayon rendering style of this project. Ahhh well, maybe on the next project.

I'm Back

2006-07-23T17:32:00.000-07:00

I have been crunching a little at work and that is why I have not been posting. To get back to work on this project I figured I'd start with somthing small but rewarding to implement. I hooked up game pad control to my avatar so I can move him arround in a more game like fashion.

My son decided he wanted to play and watching him play the game made me really want to get back into it. I was begining to loose intrest by not working on it.

Basic Animation

2006-06-26T17:38:00.000-07:00

I now have basic animation in and working, both skinned and ridged body will work in the system. I used the D3DX animation system. It is has all the features I need, but the way it works is kind of warped. I have implemented three different skinned animation systems now and this is by far the least efficient I have seen. It also requires the most work to use, probably not a coincidence. But it is much faster the implementing an animation system myself from scratch.

I’m going to have to hack my export of the .x files to add in more animations. By default the exporter only exports a single animation even though the file format support multiple. Seems like it will be easy enough to write an app to do the combining, it will just be a hassle to use every time I want to export something.

Now I just need to animate my test character and get character driven animations in.

Work Continues

2006-06-23T08:04:00.000-07:00

Work has been fairly slow as I have been teaching myself how to skin characters in Maya. This knowledge will be used to generate animated charachters which hopefully will be comming soon.

I also tried a colored lighing experiment which created one of the ugliest scenes I have ever seen. I'm going for a game drawn in crayon by a 6 year old. With the lighting on it looked more like a 2 year old with random collers appearing everywhere. I do not belive I will be able to come up with a convincing way to do colored lighting. Still I'll keep thinking about it.

Ohh yeah and I've been playing Kingdom Hearts which has not been helping me get work done...

Improved shading model

2006-06-14T20:50:00.000-07:00

I have not worked on this a whole lot due to a book. My wife finished The Da Vinci Code and I figured I see if I liked it by reading the first few pages….

Anyway I did at least get something done, I added improved shading to the crayon shader. It can now handle 2 different crayon colors and map them to different light intensities. This gives you the look of layered passes of different colored crayons to generate the shadows. If my sample texture were a little less black and white you could probably see the effect better, but it is still much improved.

I also simplified the texture format to only need two textures to do the eight intensity channels for the crayon. Each channel stores a different intensity now so I can get 4 per texture.

ODE is cool.

2006-06-07T05:43:00.000-07:00

I now have physics. ODE is a very cool little piece of code, I’m going to need to tweak the settings for a while but initial integration took about 3 hours. Most of that was taking my data and getting into a useable format for ODE.

Thank you Mr. Russel Smith. You saved me a lot of time.

Portal Render

2006-06-04T20:32:00.000-07:00

This went a lot faster than I anticipated. After a couple of hours I now have a portal render capable of handling the scene graph for my game. Well….at least the skeleton or the portal system anyway. It only handle portal clipping between rooms. It does not clip the contents of the next room other than clipping out further portals in the next room.

Here is a screen shot of looking through a portal into a hallway through another portal into another room.

Crayon Renderer Test

2006-06-03T10:30:00.000-07:00

Here is my current pet project. I wanted to see if I could make a game that looked like it was drawn by a 6 year old, essentially a crayon renderer. I think my test case turned out pretty well.

The challeng is to keep the crayon lines the same size on screen regardless of the distance from the object. At first I though I could trick the mip mapping hardware to do this for me using a mip map that kept the stroke size the same at all mip levels ( a normal mip map would shrink the image). But it turns out doing this generates artifacts on places where the polygons are at steep angles to the camera and causes severe tiling in the distance.. The step angle bug makes the line width of the crayons huge where the polygons are end on with the viewer.

I think I came up with a fairly good solution that was even more memory efficient, although it took two extra texture look ups. I loaded my textures without mip maps and did the mip mapping myself based on only distance. My next mip map down was the same texture as the first instead of using a mip chain, only the texture coordinates needed to be updated to make the texture look twice as large.

Also supported is lighting although it is kind of hard to tell in my test shots.

First Post

2006-06-03T10:02:00.000-07:00

Hello all.

This will be my first attempt a running a blog. My goal is to keep the blog updated with projects that I work on in my spare time. Hopefully I can update it at least once a week, we shale see…