Digital Amusement Game Development
  
|
I'm going to be MIA for the next two week or so (maybe less). I know many of you are salilvating over a beta 2, which I'll be working on via pencil and paper during my off time. Actually a lot of the design work has been away from the computer. Creating the brushes and such is the easy part right now. I'll see you all when I get back. GG!
And since the shader now works, I did some testing falling down the arches. HOLY FUCK. That's all I have to say. I'm bringing a whole new strategy with this map. Yes, that is my characters' boots. No, he is not about to lemming. Yes, he's moving very, very fast. No, there is no synthetic push assisting him. 
The arches have a nodamage shader applied to them, and as you see is suffering from the "overbright bug". I'm trying to fix that. However, with the shader applied, falling now takes on a new strategy...
I've spent the past couple days mapping, but haven't touched a computer to do it. Pencil, paper, rulers, marker board, research, etc. Wow, I almost sound like a real mapper ya think?
Of course, don't be expected to be wowed just yet. There is much more to do, and this beta signifies that the map CAN be played on both sides. Tis a milestone. The front stairs are going away because I just don't like them, and for some reason at the last minute the water shader wasn't behaving. If you load the map now, please note the following important items: stand on the midpoint of the stairs and jump; climb either side of the building; climb up the elevator and jump; fall into the water while kneeling. I'll have improved the map by next release, both play and visuals. Enjoy for now!
After some experimentation, I started wondering how tall I could build a tower. What I came out with was a 70 story tall monstrosity that takes a half hour to compile! That's why I set up my other machine to do compiling. While this is going on, I checked out ThorN's new teh_tower for design inspiration. I'm going to tell you straight out, this is one of the best looking UrT levels out there. As I'm writing this, ThorN has generously given me permission to use his skybox for my map. :) Ok, here's the goods:
I'm going to do some more design to the map to make it playable, then let the masses have fun with it. If all goes well, I'll have a pk3 up within a week!
My new map is large, and takes a lot of time to compile. Thus, I have used my networks' power to harness my spare machines' cpu. Not much more to talk about right now. Should be an update tomorrow though.
First, good news. I just discovered that there is such a thing as an overbright bug. Here's the proof and solution. About damn time I found proof that this is a bug. Now the next thing that I'm trying to prove, replicate, and fix. After I cut and pasted a shader with a new blend function from q3ase into my shader, my texture went to hell and hasn't been recoverable. Now my water texture renders as notex. However, if I dupe the shader, dupe the texture, the new shader/texture works fine. The only thing I possibly can remember breaking, and I'm pretty sure of this, is when I pasted the shader I might not have deleted the original thus having two copies of the shader. I compiled the map at that point and that was the end. I've spent quite a few hours now trying different methods to recreate the problem... nothing works so far. Nor does that texture work in any map now. I'm suspicious of gtk somehow corrupting some config file somehow. Parsing through the map file in wordpad, going through my registry, searching files for any occurance of my texture, etc, STILL no answers. I could take the shortcut, rename the texture and use that, but that doesn't fix the problem. What if that's a wall texture applied to 100 brushes, and the wall texture breaks like my texture did here? I'm not going to retex all my brushes!! Screw that. I'm going to figure this problem out. Any ideas, let me know.
The trigger_push entity will have the key/pair required for target_position and vice versa when you create and assign these two entities. You will know that a jumppad will work when the 3d window in gtk has a green line attached to both entities. If there is no line, or a grey line, then look closely at the entity view (use the N key) and be sure that the key/pairs match what commands are available for that entity. Doing a little testing with jumppads, I found that trigger_push doesn't push downward. In other words you can't spike people into the ground. Too bad, I was looking forward to that. You can however splat people on a wall. I tried midair combos of different jumppads plus the addition of the surfaceparm nodamage shader property to my floor texture. In the end, I'm leaping through the air, and falling gracefully to the ground. Lots of fun. :)
I've got terrain down now. I tried adding phong shading, but haven't been successful. I wasn't even aware that phong shading was an option with Q3. Want to see it in action? Check the detail in rommel, red flag side, in the building front right of the flag with two platforms. You'll see the floor "reflecting" the light above. Awesome detail. With terrain, it's just a matter of cutting the ground into triangle meshwork, then moving verticies. Simple, eh? FSK405 Resident has been working on terrain recently as well. He's got an idea for a map that I think would be a kick ass ctf map, has to do with an old fort. Props to another n00b mapper scaling the learning curve. :)  I also cleaned up the pond a little bit, moving old files into stuff/ and adding a file filter. Sign the pad on the left so I know ya love the pond!
There is one aspect of real life that I'm trying to capture in my map. At night, I do sit-ups, push-ups, etc in my living room, in the dark of the night. However, there is one streetlight outside that casts a really awesome orange glow through my stained glass window, creating this very interesting light casting on my wall. Honestly, I think it's beautiful. Strange eh? As I said, I'm going to attempt duplicating that. I haven't figured out the time of day that my map will occur in yet, so it's still set at night. I broke down and shoveled a bunch of questions into #urbanmappers. Big props to all the people there, always very patient and helpful. I tried to convince |NV|S to release two versions of upcoming ut_streets2, but he let me know about the challenges of mapping with guidelines, per SID's standards. It makes me think. I've seen creative limits all around in gaming. Usually it's due to hardware/technology requirements. But now that I face the situation myself... yuk. Hey, it's a challenge, and isn't the challenge part of the fun? (I hope so! :) Finally for today, I've shown off my map to a few friends. They love it. Then they go tell their friends "ya gotta see this thing Dan's making!" and I eventually show them. My friend that lives upstairs (also my land lord), has had the best quote of them all so far. "My house is a video game!" LOL. It's the astetic value of being able to frag at home, at home. What a great map to start with. I wonder where I'd be now if I had been able to get DeHackEd to build a map of my old house back in the day like I wanted?
I was testing out the best way to lay an image over a wall, like a poster. Laying a brush face (the poster brush we'll call it) up against a wall brush and adding cull none in the shader is not the answer. I thought maybe bring the brush back a little with polygonOffset would work. But, alas, it did not. It would be really cool if things actually worked as advertised! I'm probably screwing something up though. This is my rant where I say "dammit would shit just frickin work already???". :) I spent a bunch of time last night playing with brush placement, lighting, and shadering. Besides trying to figure out these shaders, lighting is not working the way I want it to either. I have so many questions, tutorials dodge around the answers, and it's just impolite to get on #urbanmappers to burn up the high-end mappers time with relentless n00b questions. As you can see, this situation is frustrating. I know why it is frustrating. I'm a programmer who likes to build from scratch, because if something high-level breaks and I can't figure it out, I can go to the low level, figure it out, and add code to make sure it doesn't happen again. No such luck here. Ok, before I can start explaining how to build these shaders I have to get them working myself. I have some work ahead indeed.
Using Q3ASE, a q3 shader design program, I've been learning different ways to make nice effects. Waves (vertex deformations), gl texture light manipulations, and interesting ways to change a texture while totally eating up fps :) have been a part of a well balanced meal. Mmm... meal. Really though, after seeing what I could do for myself with shaders, I give the Id designers a pat on the back for making simple shaders create complex textures. I spent a decent amount of time watching the torch animation in awe of how simple it really is. gj Id. Now as for Doom3... I'd like to know what new shader technology they have in store... that shader on the pipes and metal looks really good. Also, thanks to a tip from planetquake.com, Gray Matter studios is hiring a entry-level designer. Hoo hoo! I'm game. We'll see what happens with that. Tomorrow I'll talk more about what kind of shaders I'm using to create what effects, and how to do them for yourself. Next page |
