im trying to place a 3d object relative to existing mesh. however, i couldn't seem to get the rotation right
var quaternion = new THREE.Quaternion().setFromUnitVectors( (new THREE.Vector3( x1, y1, z1 )), (new THREE.Vector3( x2, y2, z2 ))); quaternion.normalize();
i'm using quaternion based on the two vectors, but it seems to be giving me a perpendicular angle to the one that i'm expecting
maybe i should give you the context of what i'm trying to do - user draws 2d floor plan, the program then construct a room with walls based on the plan - so from 2d vectors, i constructed walls using extrude, so no, i don't really have the rotation of the walls) - use chooses a wall, add windows to it on the 2d plane - for the windows in 3d, i just use box geometry, size it accordingly
var quaternion = new THREE.Quaternion().setFromUnitVectors(new THREE.Vector3(1, 0, 0), (new THREE.Vector3(x2, y2, z2) - new THREE.Vector3(x1, y1, z1)).normalise());
try that
you might have to play with the first vector. it'll be 1 or -1 in one of the 3 directions. depends on how the two walls are oriented.
err wait
there, that edit should do it
and I'm not sure you can directly subtract vectors with three.js, lol, but there should be a similar method
var right = new THREE.Vector3(1, 0, 0);
var wallDirection = (new THREE.Vector3(x2, y2, z2) - new THREE.Vector3(x1, y1, z1)).normalize();
var quaternion = new THREE.Quaternion().setFromUnitVectors(right, wallDirection);
what setFromUnitVectors() does is finds the rotation required to get from right to wallDirection
what you had before was interpreting the beginning and end of your wall as directions and trying to figure out the rotation to get from one to the other, which doesn't make much sense at all
you need a rotation matrix (or quaternion) an an origin. then for each vertex **v** you do... vec3 newPos = (v - origin) * rotMat + origin
or if each wall has its own position then you can replace v with that position and move the whole wall at once
aaand depending on whether your matrix is row or column major, you might have to do "rotMat * (v - origin) + origin" instead
that explanation probably doesn't do you a lot of good unless you're familiar with vector math... I really should write a blog post on the subject to link to people.
I am...kind of. But only part time, and one of my school projects is going to kill me for the rest of the semester so I probably shouldn't pick up a new job yet anyway. Next semester might be better if my capstone doesn't kill me in the same way.
longer than you'd think. getting accustomed to a new project enough to actually work with it is painful. and we'd have to work out payment and all that. really not worth it for either of us. sorry.
if I was picking up random freelance projects routinely, I would totally take you up on the offer.
I might start doing exactly that next semester, but not yet
no problem. I hang out here a lot, and there are a few other people here who are great with math like that, so feel free to ask more questions whenever.
I made my own bare bones game framework and started on a math library just because I didn't have any good ideas for a real game and I wanted to learn how it all works. I probably wouldn't have been able to help if I didn't do that.
even gamedev degrees focus too much on "how to program" and not much on "how to solve gamedev related problems"
which is necessary, but sometimes sad
not sure if that changes for postgrad degrees. haven't gotten there yet.
well, my point is that "how to program" is more fundamental than the abstract "subtracting a vector from another vector just changes the origin of the first vector and this will be very useful very often"
most people enter college not even knowing how to program, so those bigger picture, abstract things are pretty hard to teach
People take game classes because games are fun. Who doesn't like games? What they don't realize is that they're not giving people who actually want experience in that industry a chance and hogging up class resources so that they can have some downtime (they don't know how hard game dev is until they do it) or have fun.
it's not at my uni. class sizes are fairly small and the profs are usually more than happy to answer questions after class or during their office hours.
I haven't ever felt like anyone is holding me back there
except for the requirement to get a degree in the first place... so many companies want it to even consider hiring someone. I think that's kind of stupid.
nothing against the uni or any of the students though
As for UnrealC#, there are currently limitations: Our initial release supports Mac and Windows, since these were the easiest to develop against and ship.
Internally, we have proof of concepts running on Android and iOS, and based on demand, we might complete those efforts.
it seems, these days anyone can make a GPU particle simulation
it's so damn simple, too
rendering forward and back between two copies of position/velocity data, pixel shader being the simulation code, collision checking being done by texture sampling
also, what if you want to implement some custom particle behavior, that requires more data, compared to simple velocity/position package? for example, it needs something like 10 floats per particle?
@snake5 I meant, not every device supports geometry shaders, or even render to texture
@Ali.S why would you need geometry shaders? and who cares if some device doesn't support anything - nobody actually said this sample supports all devices
I'm at an impasse: I'm too lazy to beat Mono into submission, but I'm too lazy to finish up my recursive descent parser.
Do I just throw Lua in there, or do I just forget about that and hack the gameplay together in C++ too
I mean I hate Lua so much, I might as well use C++ because it's more of a "quick and easy" thing for me than Lua scripting
Last night I was struggling with my laziness bad, I evaluated Unity, Unreal, and OGRE again just to be sure I wouldn't rather suffer through getting one of those working
I'm honestly getting to the point where I might just hire goons
Give them a giant UML diagram and all that fun shit
