[NetBehaviour] success at last, just for a moment
Alan Sondheim
sondheim at panix.com
Fri Jan 11 17:37:22 CET 2008
success at last, just for a moment
it has been a difficult few days in the virtual environments laboratory.
we moved most of the motion capture equipment to 109 to use, but couldn't
locate the bvh computer (which receives signals from an old 486 used with
proprietary hardware). the explanations written by the interns were miss-
ing as well. we knew the bvh machine had to be around. we found an old
dell and gary put windows 2000 on that, getting the key from another mach-
ine. we put bvh software on the old dell, but the signals from the 486,
which, it turned out, had to go through a router since we had no crossover
connectors, were crashing it. so we were back to square one. i figured if
i could locate a photo of the old bvh computer we'd used years ago, we
might be able to use it to match with one of the bunches of similar units
in 302c. i looked through archive after archive, until i located it; the
image was slightly blurred, but the layout of the masking tape label (un-
readable in the image) on top of the machine allowed us to narrow poss- ,
ible candidates to two. we found we couldn't get into either without pass-
words. i discovered a scrap of paper in a corner of the room, 302c, with a
password on it; it worked. so now we set them both up and one of them,
which had been used as a projector connection to a virtual wall, was it.
we brought the machine down to 109 and it almost ran; at least the soft-
ware didn't crash. i found changing the antenna parameters for the motion
capture and moving from ascii to binary and from tcp to udp did the trick.
gary set up the 5100 port and addresses and we produced slomo.bvh as he
moved the sensors slowly around; i wanted to duplicate the speed of the
5-7 frames per second output of the 486, instead of working high-speed as
i usually do. we brought the bvh file to poser on a compaq laptop which
for some reason is now permanently misconfigured but usable; the result is
http://www.alansondheim.org/slowmo.mov which i quite quite like; if you
ignore the hand fluttering, the movement is closer to what we need to work
with at the moment. as far as the hand-fluttering is concerned, we're
going to internally turn off the hand sensors by modifying the capture
program, written in c, but with clear documentation; even a non-programmer
like myself can get into it a bit. now to slow things down further so the
inconceivable happens at the speed of amoeba-learning; in addition, to
transform avatar into deity-emanent, following, introjecting yamantaka or
one or another manifestation, moving mind and sheave-body into other
realms far beyond 486, 2000, 109, 302c, 5-7, until what ever may appear in
the future brought quickly back to earth.
http://www.alansondheim.org/slowmo.mov
More information about the NetBehaviour
mailing list