samlittlewood

There was a NES and later SNES prototype (By Pete Warnes and Carl Graham) that was IIRC the focus of the 3D chip discussions. X was cool, and a good selling point (and predated the NES work), but there was (again IIRC) no interest on going further with 3D on GB. Seeing custom hardware on carts (Pilot wings, Mario Kart?) and having dealt with the custom hardware on the Konix, gave Jez a good avenue to explore with Nintendo.

So tempted to try this myself in another language - just so I can call in ‘tungsten’.

My goto for this is ‘strace’

Cool! I saw this link and your books came to mind as another example of how material like this is vital for seeing (amongst other things) how creative endeavours cope with ‘the other 90 percent’ of projects.

Please no! People will start hanging around trying to do clever programmer stuff, and then we will be in all sorts of trouble.

Quite a few games (eg: Starfox) in the 90's would use BSP in a slightly different way - purely within each object. The tree would be generated offline (mechanically or by hand), and would often be some sort of branching bytecode (test this, draw that etc.)

It was very useful to be able to use one plane test to discard a whole bunch of geometry, eg: a surface plus its detail, or everything visible through a window.

This still left the problem of sorting between objects - mostly a depth sort was just fine. Special cases like weapons on hardpoints, or vehicles in hangars could be handled by having some sort of 'call submodel' for spaces within the parent model. Beyond that, just dial up the hostile firepower until players stop complaining about rendering artefacts.

It might be interesting to explore with the students why this would not be quite right for a general 3D view, and how it could be fixed. Conic sections, etc.

Long time! Maybe we should wait until Dan back over, get the old gang together, and see if we can finally agree on a restaurant.