Y positive up or down? / Left or right-handed?

Garry Wiegand

unread,

Oct 30, 1986, 9:29:37 AM10/30/86

to

Yo: another vote for Y positive up! For graphics, that's what seems
to be human/natural.

A related question: in 3-D, do people like left-handed coordinates,
right-handed, or programmer-choosable ?

I was raised on left-handed, but just yesterday I had someone being
very confused while twirling the camera around -- he was thinking
right-handed! So, fine, let it be selectable - but then think of
a programmer who picks one way who calls a library routine which
was written using the other way. Muchos confusing. Better to make
everybody agree ?

I heard a rumor once that most solid geometry books are right-handed...

garry wiegand (garry%cadi...@cu-arpa.cs.cornell.edu)

(The opinions expressed above are indeed those of my company.)

Vulture of Light

unread,

Oct 31, 1986, 5:25:01 AM10/31/86

to

+ Right, right, right, right--the physics doesn't work in left, left, left!!!

+ Perhaps much of this right/left/up/down confusion is due to graphics books
that sloppily switch all around.

+ The hardware argument has no bearing.

Douglas

ARPA: tra...@locus.ucla.edu
UUCP: ..!{sch-loki,silogic,randvax,ihnp4,sdcrdcf,trwspp,ucbvax}!ucla-cs!trainor
SNAKE: tra...@eric.sidewinder.snake

Jon Leech

unread,

Oct 31, 1986, 12:27:28 PM10/31/86

to

In article <13...@batcomputer.TN.CORNELL.EDU> garry%cadi...@cu-arpa.cs.cornell.edu writes:
>A related question: in 3-D, do people like left-handed coordinates,
>right-handed, or programmer-choosable ?

>...

>I heard a rumor once that most solid geometry books are right-handed...
>
>garry wiegand (garry%cadi...@cu-arpa.cs.cornell.edu)

This is another of the innumerable religious debates in computer
graphics. An amusing story along these lines: Jim Blinn has done a great
deal of computer generated material for the PBS series ``The Mechanical
Universe''. His SIGGRAPH film 2 years ago contained portions of this
material with his own narration. At one point, while talking about
tops or some such, he said 'of course, vectors obey the right-hand rule',
which received applause from the knowedgable members of the audience.
Jim's rendering software uses LHS internally, leading to more work than would
otherwise be neeeded. The point is that if you're doing modeling based
on real physics, you work in RHS - and it makes life that much easier
if your other software talks the same coordinate system.
-- Jon Leech (j...@csvax.caltech.edu || ...seismo!cit-vax!jon)
Caltech Computer Science Graphics Group
__@/

Marc Majka

unread,

Nov 2, 1986, 6:55:48 PM11/2/86

to

In article <13...@batcomputer.TN.CORNELL.EDU> garry%cadi...@cu-arpa.cs.cornell.edu writes:

>[...] So, fine, let it be selectable - but then think of

>a programmer who picks one way who calls a library routine which
>was written using the other way. Muchos confusing. Better to make
>everybody agree ?

Better to make the library smarter.

---
Marc Majka

Doug Gwyn

unread,

Nov 3, 1986, 12:04:53 PM11/3/86

to

In article <25...@curly.ucla-cs.ARPA> trainor@hera (Vulture of Light) writes:
>+ Right, right, right, right--the physics doesn't work in left, left, left!!!

Since coordinates are purely conventional, it would be amazing
if physics "didn't work" when one chose a left-handed system.
Of course, if you don't distinguish between vectors and pseudo-
vectors (best done by using skew tensors, aka forms), you're
making a mistake that shows up when coordinate conventions are
switched on you; but it isn't the physics that's broken.