The amazing Magnus Wootton, some stuff ive been workin on

The amazing Magnus wootton presents the stuff ive been working on the last 6 months, which has been the most gruelling 6 months yet! Im getting desperate to get this robot security dream happening, and ive got all these toys along the way!

This first one, I didnt leave in failure, I just got the urge to get into the electronics finally! Because I want the robots brain to be my own hardware, not software, but I had a webcam based 3d kinect type thing on the way, it interestingly triangulates the video and turns you into a triangle model, but its not quite 3d yet, the other cool thing is it has a key memory, and it improves its tracking ability immensly, but at the same time slowing it down.
This is a cpu mary tree implementation with 32 gigabytes of ram devoted, a pure 4 gig gpu version would kick this ones butt, although its storage ability is limited.

//youtu.be/-5MJt_jHiTo

Now this one is excellent!
I can actually take an ordinary video. (would be better if it were cube video) And I can actually through a clever billboard process let you walk around in it! Imagine using this for cheap virtual sets that actually look good for some budget movie making. (would kill hollywood finally nails in the coffin.) Makes your keyshot raytracing alot cooler too, being able to place your model in the set, and the preview process is alot more fun.
In the video, there is actually edits taken place, I can mirror the video around (in 3d!!) just like a 3d model, and its just a crummy video! this one amazed the hell out of me.

//youtu.be/tTKHE0xYY14

heres a failed attempt at bring back street fighter 2 in a markov chain, this failed, but i know how to do it now! i dont think the attention span is good enough to get out of the character select screen to the game, so it would need rejogging to get here and there.
The robot im planning to make does reconstitute its sensor in this way, but I havent gotten around to finishing this one off!!! Soon! I must finish it!!!
The red numbers on the left are patterns entering the system, you have to wait a bit before I go into simulation mode, and it fails miserably. I have a better video, but the framerate was so poor im just putting this here to mark a spot.

//youtu.be/eXbdUDKrRRc

And after that ground box program, little fun music video i made from the left overs of the code. :slight_smile: (watch out its nasty dnb :))

//youtu.be/BNV7LYAPkZ0

last but not least, some realtime(ish) environmental lighting, off simple homogeonous procedural generated skymap. (blue sky gold sun.)

//youtu.be/uGOtbgfUWCs

back with a new raytracer. inserted cook torrence, i could NOT! say i even learnt it, its more implaced before understanding. its just experimentation.
I have to thank old ken and cook for this back in the 80’s! they helped create the “analogue graphics look” that were very potent even in the early beginnings,
i say it a million times, an idea/theory is worth more than all the technology in the world, but the technology definitely helps.

can thank a vintage 80’s algorythm (cook&torrence spectacular reflection model.) for the main worth of this demo. it also runs on lambert (which is pretty much everyones bike. in complexity of an engineer coming up with it.), but trust me, its not as good, but it is still cuteish, and still shows computer like efficiency at the depth map shading task…

fully cut to a 10 minuter -

The music isnt really good enough, but at least it suits it. im not sure if a full dynamic sound

(which my ear tends to ignore too much, even tho i do still tweak for it! its one of my main tricks, just havent got the pro ear, and my alternate technique, clipping seems to push me in two opposite directions AT ONCE!!! (clipping being a “characteristic” generator.) when i make music.)

is what it even needs, i tried playing other peoples music to it, to get the full volume range, but just felt like a pirate. even tho the cook&torrence algothym i stole like a greedy selfishmin!!! I figured there was just no other option, when i saw the results.

But will say i had no part in the worth of this production, i merely set up the framework for the magic to happen.

//youtu.be/Z-8p00E_LjA

the madman himself on the mic -> (excuse the lazy brain rudeness…)

//youtu.be/LjXRDENJMSM

secrets out, the specular function is removed, and you just have diffusion with microshadowing - and it makes sheeps wool on the surface.

//youtu.be/XaNa_--jhj4