<FreeFull>
AFAIK quaternions do also work for 4D rotations
<zid`>
and then telling that you multiply on one side to do a rotation around the global axes and on the other to rotate around the one you made from your last rotation
<zid`>
instead of this ijk nonsense
<nikolar>
zid`: the ijk nonsense is how you mulitply though
<zid`>
no it isn't, that matrix is :P
<zid`>
I live in real life
<zid`>
not in mathland
<nikolar>
(unless you convert them to matrices)
<nikolar>
and that matrix is derived from the ijk nonsense as you put it
<zid`>
and then one more thing about (r,x,y,z) being the opposite of (r,-x,-y,-z)
<zid`>
r,x,y,z >>> ijk1 nonsense
<nikolar>
conjugate :P
<FreeFull>
A quaternion is just 4 numbers, a full rotation matrix is 9
<nikolar>
r,x,y,z stands for r + x*y + y*j + z*k
<zid`>
no it doesn't
<nikolar>
like it's still the ijk nonsense :P
<nikolar>
it does
<zid`>
100% does not stand for that
<nikolar>
it literally does
<nikolar>
that's why you have minuses only in front of x,y,z and not r
<nikolar>
for the "opposite"
<zid`>
It doesn't stand for anything
<zid`>
it's a vec4
<FreeFull>
Euler angles aren't too great
<zid`>
with certain properties under multiplication etc, which you can *describe* with the ijk shit
<nikolar>
which you use to represent quaternions
<zid`>
but it doesn't 'stand for' anything
<FreeFull>
Unfortunately by the time quaternions were invented, Euler was already dead for 60 years
<zid`>
It's like saying 0xFF0000 stands for 700nm. Despite nothing on a computer giving a flying fuck about nanometers or wavelengths
<zid`>
it means red
<nikolar>
sure, nothing on the computer cares, but if you say 0xff0000 represents 700nm in your program, then it's 700nm
<zid`>
And yet, it doesn't
<zid`>
because I don't
<zid`>
r,x,y,z is not me foolishly using a stand-in for the proper ijk mess, that's abstract math that means nothing. rxyz is a concrete implementation derived from that.
<FreeFull>
You can interpolate quaternions
<FreeFull>
how would you interpolate a rotation matrix?
<zid`>
Like how a SAT solver is not 'a stand in for algebra'
<FreeFull>
As in, take two rotation matrices, and interpolate the rotation between them
simjnd has joined #osdev
simjnd has quit [Ping timeout: 248 seconds]
simjnd has joined #osdev
vdamewood has joined #osdev
simjnd has quit [Ping timeout: 252 seconds]
chiselfuse has quit [Ping timeout: 264 seconds]
Lucretia has quit [Remote host closed the connection]
simjnd has joined #osdev
simjnd has quit [Ping timeout: 248 seconds]
Turn_Left has quit [Read error: Connection reset by peer]
jcoding2 has joined #osdev
<jcoding2>
I'd like to know - what's your go-to editor for OSDev?
jcoding2_alt has joined #osdev
<jcoding2_alt>
Making sure that everything's OK on my thelounge setup - bye