JF make 208
JF Nav
JF Nav
Creation Date: 2002-08-27
Greetings fellow human being. I'd like to give you a pair of brief ideas. The first is motion blur. Above is my attempt to do a bit of motion blur special effects. I simply took two screenshots and superimposed the later image above the earlier image with transparency set to around 60. This is okay, but not up to my standards. For it to work for Scene 1, it will need to be absolutely spectacular. Then, I'll anti-alias it down to half a page and it'll look fair. It really isn't cool that I have to slave over something for days and days to get it perfect and the format of the comic book makes it look like I didn't try at all. But that's okay. Perhaps it'll just happen anyway -- it often does. So how does one do a correct motion blur. Well, to find that out, I set up my camera at 10 frames per second and danced a jig. I looked at it and it wasn't just two picture superimposed. It had a picture for each pixel in between. So say that the left side of my face is at pixel 32. If in the next frame, the left side of my face is over at pixel 45, I need to take 13 pictures and blend them all. Usually, it's only three or four pixels. But in this picture, it's twenty or forty. Then I'd get actual motion blur, I think. This is a property of physics. The retina or a camera pixel takes a time average of energy. That is how televisions and monitors can work with scanlines. However, remember that I'm doing a comic. Looking at my favorite comic (which is a good reference for fighting scenes), Gunnm aka Battle Angel Alita, I find the answer. Comics use motion lines. It's actually a rule or something if comic books have rules. Ever since the early days of superhero comics, they have been using lines to show movement. You actually only need to render what you want once. You have to draw a group of lines from where it was one unit of time ago until now, usually an arc. Sometimes if it's unclear where your character is, you can render twice or three times to show your reader. So that will be a challenge to myself: make motion lines that look good.


You know Lara don't you? She's a cool gal. She's sporting her skimpy training uniform from Tomb Raider 2. This is a super-low-poly model. It also has no deforming triangles. What does that mean? It means that her limbs are seperated so that the renderer only has to translate the entire triangle rather than each vertex seperately. Why would they do that? Because Tomb Raider 2 was created before most people had 3d accelerators. It used software rasterizers. Back in the day, right? JF does not require such icky things. AltSci3d uses skeletons with skinning. The program allows up to 4 matricies per triangle (2 via GeForce2 hardware and 2 via DirectX8.1 software). Another fact about this model is that it uses 24 textures for her crotch section alone. I counted. ^_^ These textures are all very small, but really lame. How lame? When I rip them from the WAD file and try to put them together, only two or three fit together. Perhaps someone was using one of those nifty 3d painter things. It's disgraceful that a professional skinner would make such ugly skins. I think there are about 120 textures for all of Lara. JF does not require textures to be broken apart, each model can use one or a dozen textures if they so choose. However, all JF models have used a single 256x256x24-bit texture. In fact, all new age games use only 1-4 textures per person. I think Half-Life uses 2-4 because they use 8-bit textures. So what am I trying to say? Think about the success of Tomb Raider on such meager technology and very low poly models. Is a good game defined by how cutting-edge its technology is? Is it defined by ease of use? It is defined by positive game play? Is it defined by large breasts? Why is the third-person survival horror genre populately largely by women? Is it a win-win situation for developers? Girls like it because they get to play a girl (Lara is a great role model ^_^). Guys on the other hand get to do the handstand (Ctrl-Shift-Up on a ledge), swimming, and wide-angle shots. It seems silly, right? But what use is a guy in third person? Girls get to awe at his muscular back? Guys don't even look in the direction of guys who aren't looking at them irl and they certainly aren't going to do it for a full twenty hours of a game. Well, I guess I'm not entirely right. GTA3 did pretty well and it was a third person view with a guy. However, it is the exception and not the rule. The lesson for today is to go with the flow and make a bunch of women models for your video game even if a real woman with dimensions 66-12-35 wouldn't last a second in a far off galaxy battling evil wearing a tank top (with wonder bra, doh!) and shorts. It's just eye candy so why should anybody care? Wait, nonono, that's not the lesson. The lesson to be learned from the Tomb Raider series is that a powerful, easily-extensible engine with ease of use for the artist is the best way to make a game that has fifteen sequels (five of which end with the hero dying).
Home Characters Making Of Technical Mail News Links |< First < Prev Next > Latest >|  bandwidth version Goto Scene