So this is a very interesting coincidence. For IBC we received some users feedbacks that asked many similar things. I even proposed a potential option for achieving some of this stuff.
But after thinking about it… it would be INTERESTING.. to make Houdini NLEngine.
Maybe a new commercial product line for the high end editorial, DI, 4K, Color and Comp crowd.
Autodesk Flame and Smoke have the richest 3d for a comp app sure
but what if a NLE engine for houdini could be applied as a NODE in something like MISTIKA?
that way you can add 3d to a video clip or COPS to a video clip and CHOPS to a audio clip on a EDIT TIMELINE.
this would be used by me on a daily basis. It would be worth its weight in gold for all the people i deal with.
Being able to select several audio timeline clips and apply a global CHOPs to it would be aweomse.
Then taking that CHOP and linking it to a 3D Houdini FX on the video tracks would be just as cool!
Making Music videos would be a JOY all the time, i can tell you that much.
But making a NLE engine could be used in lots of apps like
Nuke , Fusion, Mistika, Mamba, Resolve, Smoke, Flame, Scratch, Mist, Rain, Ocean, Piranha or just about anything like that.
My vote for all of the above comes with a bit of specifics tho
The choice to make Houdini a PLUGIN LOADER for other apps is very interesting.
Rendering to the timeline or video storage could be done at both global timeline render OR individual cllip render.
WHY NOT JUST DO THIS SEPARATELY? well we knot how that works already…. but what if we can get a TRUE MASTER CGX suite. Im coining the new acronym CGX as a combo of CG and VFX so we know what the difference is.
We could manage an entire production from Houdini NLE Timeline. version management of shots could be controlled and then distributed rendered via SAN clusters… o my would that be HEAVEN for middle sizes studios.
The 3d tracker tho… its a MUST HAVE to make that all happen.
My features for a 3dtracker are specific so you can do image modeling right in houdini with both POINT CLOUD MOLDING and POLY VERTICE TIME MODELING
If your not familiar with the latter… you basically take a shot, 3d camera solve it, then draw a polygon shape over an object on frame 1… then on end frame you move the polygon points to match again.. then in the middle do the same.
at that point your camera solve can be resolved thanks to the geometry, AND you get a 3d model that you can project the footage on and bake it for camera replacement or 3d stabilization AND OR set extension painting.
YOU GOT MY VOTES ON ALL
i only wish i could help work on it.
thanks Z3
cMaynard