Almost. A crap ton of money goes into military research and so they sometimes intentionally create things useful for other industries, and sometimes they fail at something but it is good elsewhere
Revolutionary inventions come from pushing for innovation without caring about the cost. So the fact is we spend an insane amount of money on military, without any strings attached... But it could come from anywhere else. And if we were to spend that money directly on research, we'd probably have flying cars and energy matter converters by now...
1969 is when the academic research was done, but it took many more years to apply it to personal computers. We had to wait until PC's were fast enough/had enough memory to use it. In this case it was about 23 years.
Great video, I have been thinking through various ways of presenting database information in the most easily digestible way to any user, and binary partitioning seems like something to keep in mind, assigning values to data which make them more or less prominent to the user based on certain criteria. Thanks!
@@PaulSebastianM The splitting is precomputed because it takes a very long time. This is why it's not used anymore, you can't really change the scene without computing it again. Also the GPUs are so fast now that it gives no advantage to use BSPs
@@dude2542Unfortunately, even now there is no way without BSP, it’s just that no one uses it anymore, now everyone uses an alternative to this, BVH, the newest of them is CWBVH, which also split all the primitives in the scene. Nowadays, the scene has increasingly begun to be split not into right and left, but into 8 parts (If you're interested, i can explain why). Also, such trees can be built very quickly, allowing you to rebuild them every frame
BSP trees were used well before Carmack's work on Doom. In the Softimage ray tracer as of 1988, Mike Sweeney used them very effectively to dramatically speed up the ray-triangle intersection algorithm. In fact, many of the technologies that would become mainstream had their start at Softimage in Montreal ; for instance, inverse kinematics using jacobian matrix math, the brainchild of Dominique Boisvert from his days at ETH Zurich.
It's just so fucking incomprehensible to me that people were able to come up with this stuff. My intelligence is just so far beneath them it's not even funny lmao.
I remember reading the doom source code like 20 years ago and being amazed with how this I game I grew up with actually worked. Also seeing it rendering scenes slowed down thousands of times helped a lot understanding how it works. and the it is ported everywhere is so cool!
