JF make 322
JF Nav
JF Nav
Creation Date: 2004-02-02
My server is up, but my ftp is still down. So this news is not exactly up to date. But I want to get it down for the record. It's a bit exciting. The hard parts of Demo 1 are finished. I have 2 easy technologies to add and the rest is creating the models. That means that I am just 2 days behind schedule if I can do the easy stuff tomorrow. 2 days behind is nothing. In fact, the models I am doing tomorrow will be part of not just demo 1, but demo 2. So demo 2 will likely take less than a week. Even though I plan to spend time at 2600 on Friday, I can still make the Saturday night deadline for demo 2. That would be really amazing to me. There are times when I throw myself around and verbally abuse myself (half are untrue, but definately unproductive). Missing deadlines can be a horrible experience. It often compounds upon itself; it makes easy deadlines into more missed deadlines. If a missed deadline can destroy a person's confidence, I say that person is better off ignoring the deadline entirely. The confidence will motivate the person to do the work better -- in the end, faster.

My mistakes were terrible today. They were in physics and logic, my two strong points. The first physics mistake was made Friday. I got the cross product wrong. Remember:
c = a X b
c.x = a.y * b.z - a.z * b.y
c.y = a.x * b.z - a.z * b.x
c.z = a.y * b.x - a.x * b.y
Why did I make the mistake? I was working fast and I thought I knew this by heart. I did not. The second physics mistake was done Friday also. I slowed down the object, but did not take that into account with my optimized collision time prediction. Okay, I may be talking greek to most of you, but I can put it into simple words:
A maglev train (A) is going from NY to LA (5000 km) at 250 km/h. It makes no stops between. How long will it take to get there?
5000 km * (1 hour / 250 km) = 20 hours.
A second train (B) is going from NY to LA (5000 km) at 250 km/h. It slows down to 200 km/h between the great divide and LA (1000 km). How long will it take to get there?
First  Leg: 4000 km * (1 hour / 250 km) = 16 hours.
Second Leg: 1000 km * (1 hour / 200 km) =  5 hours.
Total: 21 hours.
So if Alice on train A wants to meet Bob on train B in LA, Alice should get an hour long massage in LA train station.

The third mistake was a logic error where I rotated the acceleration to be perpendicular to the normal of the ramp. I had a brilliant plan:
1) check to see if the acceleration is perpendicular to the normal using the dot product.
2) rotate it.
The problem was sillyness. I totally forgot that I always accelerate for a short amount of time and then stop. So when I would rotate the first acceleration, it would take effect, but the stopping overwrote the rotated acceleration. A very strange thing. The solution was to simply add the rotation to the path. That says that whenever I accelerate, make certain that it is perpendicular to the normal. And it works.

The final problem I won't even explain. I forgot to initialize a matrix. What a maroon type of thing, huh? The solution? mRotMatrix.setIdentity(); It was a doozy because it seemed like it was caused by working with the mesh. That's what you call common misdirection. Every fraud uses it, even when the fraud is a few lines of code.

A scanner is coming on a UPS truck. It was a great deal. I have to thank my friend Colin for that. I just hope YMMV does not mean Murphy's Law. Why do I need a scanner? The real decision I made right before I visited my parents. I had 40 character sketches to make for Hack Mars. I had them in my sketchbook. I needed a way to import them. So I imported them using my parents' scanner. Well, that's cool that things worked out like that. But what happens two weeks from now when I want another sketch? I buy my own, so I did. Expect to see the beauty of JF and HM increase as I draw more and more. But don't expect some miracle. Tools and such don't make the artist, the artist makes the tools work for him/her.

More about the deadlines if you wish to read about it. It took me almost 10 hours today to fix those bugs I talked about above. I spent half of Friday and half of Saturday working on them also. Why did I only spend half of the two days? Because I spent the other half miserating over the deadline and slacking off. If I succeed and Hack Mars makes a profit, I will forget the misery it took me to create it and I'll make a game even better than it immediately afterward.
Home Characters Making Of Technical Mail News Links |< First < Prev Next > Latest >|  bandwidth version Goto Scene