Cygwin. I would go insane in sort order if I had to use a windows machine for something other then games with out cygwin and all the the lovely stuff it brings on to a windows box.
While I was a student I worked for a lab on campus that did a lot off
stuff with 3D graphics. I remember how much a pain in the butt it was
try to move things between different packages ( Autocad, ProE, Catia,
Solidworks, 3D Studio Max, Wavefront Advanced Visualizer, Alias
Studio, Softimage, Maya, you name it. )
While a universal format sounds like a great idea I am a bit skeptical
about this working out. Most of the packages I have run into play
with a neutral format called IGES, I use the term play, because in my experience moving any useful amount of data around in it
is hit or miss at best.
With 3D data you are messing with so many variables a task
specific representation is almost required. Audio isn't really that
complicated, compressions techniques aside, most are just based on a
pulse code modulation. 2D images start to get interesting, sure there
is the good old raster images, but once you go vector there are lots
of different formats. The brute force raster like approach to 3D is
voxel based images, which are huge, and hardly generally useful in
proportion to their size. Maybe we could model the solids, no use
using storage space representing empty space. We could just store
solid primitives and logic operations on then. This of course can get
really big for complicated surfaces. Maybe the surface is all we need,
then how do we go about it, polygon approximations? NURBS?
Part of the reason for the different formats are differences just from
reinventing the wheel, but some of them are there because the
representation they are using is better suited to the application at
hand.
Drop that wave in an editing app and speed it up from 8kHz to 44.1kHz, and I swear it sounds exactly like a sound effect from your typical scifi flick.
The most recent kernel version that I see there is 2.4.13 though. IBM is claimed to have released the code with in 18 months of when this whole stuff started. I don't have a release date for 2.4.13, but the changelog for it on ftp.kernel.org has a date stamp of 10/24/2001. Could it be that the kernel that SCO is distributing doesn't have the questioned code? Wouldn't that shot a hole in the "they released it under the GPL" argument?
It sounds in concept like condor being developed by the CS department at UW-Wisconsin Madison, unfortunately the condor folks aren't letting their source code out except for special special cases, and then you have to beg them for it.
Re:Keep in mind that most of these "irrationally..
on
Irrational Exuberance
·
· Score: 1
Slashdot Mirror
Cygwin. I would go insane in sort order if I had to use a windows machine for something other then games with out cygwin and all the the lovely stuff it brings on to a windows box.
While I was a student I worked for a lab on campus that did a lot off stuff with 3D graphics. I remember how much a pain in the butt it was try to move things between different packages ( Autocad, ProE, Catia, Solidworks, 3D Studio Max, Wavefront Advanced Visualizer, Alias Studio, Softimage, Maya, you name it. )
While a universal format sounds like a great idea I am a bit skeptical about this working out. Most of the packages I have run into play with a neutral format called IGES, I use the term play, because in my experience moving any useful amount of data around in it is hit or miss at best.
With 3D data you are messing with so many variables a task specific representation is almost required. Audio isn't really that complicated, compressions techniques aside, most are just based on a pulse code modulation. 2D images start to get interesting, sure there is the good old raster images, but once you go vector there are lots of different formats. The brute force raster like approach to 3D is voxel based images, which are huge, and hardly generally useful in proportion to their size. Maybe we could model the solids, no use using storage space representing empty space. We could just store solid primitives and logic operations on then. This of course can get really big for complicated surfaces. Maybe the surface is all we need, then how do we go about it, polygon approximations? NURBS?
Part of the reason for the different formats are differences just from reinventing the wheel, but some of them are there because the representation they are using is better suited to the application at hand.
Drop that wave in an editing app and speed it up from 8kHz to 44.1kHz, and I swear it sounds exactly like a sound effect from your typical scifi flick.
The most recent kernel version that I see there is 2.4.13 though. IBM is claimed to have released the code with in 18 months of when this whole stuff started. I don't have a release date for 2.4.13, but the changelog for it on ftp.kernel.org has a date stamp of 10/24/2001. Could it be that the kernel that SCO is distributing doesn't have the questioned code? Wouldn't that shot a hole in the "they released it under the GPL" argument?
It sounds in concept like condor being developed by the CS department at UW-Wisconsin Madison, unfortunately the condor folks aren't letting their source code out except for special special cases, and then you have to beg them for it.
Actually according to CBO that last time we had a surplus was 1960. http://www.cbo.gov/show doc.cfm?index=1820&sequence=2&from=5