Position of object in Canvas does not correspond to position per Inspector

Question

Level 1

15 points

Position of object in Canvas does not correspond to position per Inspector

I am working with a 3D object in Apple Motion that I am moving from one place to another on the Canvas using Keyframes. Firstly, I note that the wire Frame for the object does not correlate to the position of the object on the screen---and the Z value for the object is always set to 0. When I use the pointer to determine the position of the object on the canvas at any time and check this against the object position per the position parameter in the Inspector, it is not even close. For example, at the termination of the object's movement, per the canvas grid, the position is approximately x = -63px, y =n -994 px; but the Properties Inspector lists the object's position as x = 3399, y = -3105, z =0. Why is there such a discrepancy?

iMac (2017 – 2020)

Posted on May 28, 2024 9:49 AM

Reply

Answer 1

fox_m

Level 6

12,595 points

May 29, 2024 2:28 PM in response to Lakewatch

Sorry for the delay...

I'm a bit confused... how do you get a 3D Object's wireframe? This is as close as I can get, but it is *not* a wireframe:

Are wireframes a feature of newer versions of Motion? (I'm stuck in 5.4.7).

As for the apparent position of any 3D object in 3D space, there are several factors at play. The actual physical location of the object; and where the camera is, how close or far it is from the object, the angle it is "looking" at the object, its Angle of View, etc...

Without seeing the screen relationships you are trying to describe (locations of object w/r/t the grid) and the parameters... and *which* parameters you are talking about, it's nearly impossible to tell what you're experiencing.

Also, you need to explain "y =n -994 px" — is this a typo? or an actual expression (if so, you're not using Apple Motion!)

Reply

Answer 2

Lakewatch Author

Level 1

15 points

May 29, 2024 3:54 PM in response to fox_m

First, let me express my appreciation and gratitude that a knowledgeable Motion user has taken the time to read and respond to my post. This is one of the great features that Apple offers with its products--the willingness of an "army" of users to share their knowledge with neophytes like me.

Second, I am using Apple Motion ver. 5.7. I am a new user (with some prior experience with FCP) and I am attempting to teach myself by reading the almost 1500 page manual--which reads in major part "like stereo instructions" (as they would say back in my day for any manual written in techno-babble for the cognoscenti).

At any rate, since it's pretty impossible to learn by simply reading the manual, I have made myself a teaching project where I am struggling to implement concepts taught in the manual by destroying my place of residence by crashing a meteor into Smith Mountain Lake. This has got me using motion tracking, particle emitters and a 3D object (the "rock" taken from the Motion library of objects) against a 2D background shot of our lake. My project is about 12 seconds long and tracks a "meteor" as it enters the picture at the upper left and crashes into the lake around the dam in the middle of the shot.

Here is a screen shot about 5 seconds into the video:

So, if you can see it in the shot, the object (the rock--which I have selected in the Inspector) which is surrounded by the flames generated by the Meteor particle emitter from the Library) appears above the red line which shows its projected path (to crash into the lake between the opening of the two mountains) and the square box (which I have referred as the wireframe) is on the path, but below and behind the rock. Incidentally, the wireframe mimics the motion of the rock itself--moving along the path and spinning on three axes.

Now, further, as you can see in the photo, the Inspector shows the position of the rock as x = -496.94, y = -877.05 and z = -170.71--but examining the grid at the boundaries of the canvas, the position of the rock (or the wireframe) would appear to be more like x = -1200 (for rock or -2050 for wireframe) and y = 300. Also, as you can see, I have chose "Front" as my 3D view so would understand that I am viewing right down the Z axis.

Now, I do see that the position displayed in the Inspector for the Rock is showing a non-zero Z value, even though I endeavored to leave a change in the Z value out of the equation by starting the rock from a "0" Z value. I see now that as I advance my video, the path of the rock does involve some movement along the Z axis--so is this what is causing the lack of correspondence that I am seeing between the position values shows by the Inspector and the grid (that the grid is showing a picture in 2D space while the Inspector tracks the rock in 3D space? Would this cause that much difference in the position values?

I have tried to "tie" the movement of the effects which I have associated with the rock (the meteor emitter and a fire plume emitter from the Library) both by motion tracking (Match Move) and simply specifying essentially the same starting and ending points of travel for both the emitter and the rock) and neither seems to achieve the result I am looking for--basically, I am trying to film the scene with a "Sweep/Zoom" camera that will bring the shot more into the path of the rock and zoom in on it around impact. But when I move my camera view out of the flat "Front" aspect, my tracking behaviors do not seem to work and the emitters become separated from the rock.

Any suggestions on how I can avoid this result?.

Have I clarified the problem for you? Again, many thanks for any assistance or suggestions that you can offer.

Reply

Answer 3

fox_m

Level 6

12,595 points

May 29, 2024 8:38 PM in response to Lakewatch

Ok... the explanation is rather complicated and does involve the relationship of "grid", objects, and the camera *view* plus things like parallax and other boring stuff like that.

From your screenshot, you are looking at your scene with the Front camera view... Switch that to Active Camera if you want to see the scene as it *really* is going to look like. The Active Camera is the only view that will export (Export Movie), you cannot choose any other camera view for export. All the other camera views (Perspective, Front, Back, Top, Bottom, etc... are only to be used as aids to aligning objects...*) Looking at Front is definitely going to make the grid look "way off". These alignment views are using the **Canvas** view for their arrangements. Active Camera is using the Camera's viewpoint to look at a scene.

This is the Active Camera View - Rock is beginning to "explode" on impact - 3D space Grid visible in the lighter regions. The Camera is rotated -15.5º(x) and -5º(y) - so it's looking at the 3D stuff at an angle. The background is a 2D Group Layer (always flat).

This is the Front View - notice the grid is flat. The Red and Green grid guides represent the intersection at 0,0 (2D). You can tell by the difference in apparent angle (which is straight on to the "front face" of the 3d scene) of the "meteor"/fire plume arrangement appear complete different in the two views.

(told you it was complicated... I know I'm probably not explaining all that well.)

Using the different camera views can be helpful, just be mindful to always reset back to the Active Camera to make sure you're getting (and seeing) the scene you want.

HTH...

Reply

Answer 4

Lakewatch Author

Level 1

15 points

May 30, 2024 5:57 AM in response to fox_m

Again, thanks so much for taking the time to analyze my problem and respond. I frankly did not understand that the active camera view was the only view that would export. Obviously, understanding this is crucial. I am now going to try to use other views like "Top" and a side view to make certain that my rock only moves in the xy plane. I will try to do this by defining beginning and end points and making sure that the z coordinate in both cases is set to "0". I then will keyframe the movement and hope that this will define a move only in the xy plane (I think that in my prior attempts I had dragged my rock to its final position--inadvertently introducing z plane movement into the equation). I will check this by examining different view perspectives. I will then try and use Motion Tracking (Match Move) to attach my meteor and fireplume effects to my rock. It is my hope that the needed close correlation between the movement of the rock and the attached effects will continue even when viewing the scene from different perspectives.

Again, thanks for taking the time to help me out.

Reply