some i7 if ECC memory sounds an oddity to you). If the model is big you'll need a decent Kepler Quadro as well ... say a K4200 (I hate game cards including Titan).
http://www.tsplines.com/
Alternatively use a top dog subdivision app (Modo eats them all for breakfast) but that works with DX and that brings us back to game cards.…
the algorithm is unstable and not optimal. Unlikely to be able to somehow use it or learn something on it)
https://dl.dropbox.com/s/2x15zic85gsbfwb/%D0%BE%D0%B1%D0%BE%D0%BB%D0%BE%D1%87%D0%BA%D0%B0%20100.3dm?dl=1
https://dl.dropbox.com/s/a2avwghhdmb5tv1/%D1%83%D0%BD%D0%B8%D0%B2%D0%B5%D1%80%D1%81%D0%B0%D0%BB%D1%8C%D0%BD%D0%B0%D1%8F%20%D0%BE%D0%B1%D0%BE%D0%BB%D0%BE%D1%87%D0%BA%D0%B0%20100.gh?dl=1…
gone with the wind topic: since this is utterly Academic the main issue here is to oversimplify LBS (in real life: a collection of columns/beams/slabs/X members + tube frame rigid members (shafts/elevators/cats/dogs)). Reason is that if we use the real "solids" (turned into meshes) as the "node" pool for the hinges required ... only HAL 9000 could solve it in "real-time" (for instance an E5 Xeon 1630 v3 takes ... several minutes). And this is ... er ... challenging I must say. This is a typical case where "simplifying" means "stupidity" almost instantly.
Spam on:
where's my collection of "bend-a-truss-that-looks-like-a-tower" K1 demo defs? Is in this workstation or in another? (blame Alzheimer).
Spam off.
More soon.…