Jim Lefevre
4
COW Coins
Forum Replies Created
-
Yes, so far as I’ve experimented annoyingly bringing on the .ma data gives you a wonderfully tracked camera with nulls for the tracking points BUT not the 2D tracking point data that I need (which you rightly state is within the txt data.
Your suggestion looks promising though…
Paraphrased from another post the After Effects position keyframe data looks like this…
Adobe After Effects 8.0 Keyframe DataUnits Per Second 25
Source Width 1920
Source Height 1080
Source Pixel Aspect Ratio 1
Comp Pixel Aspect Ratio 1Transform Position
Frame X pixels Y pixels Z pixels
0 400 300 0
10 400 600 0End of Keyframe Data
but the Matchmover tracking data is laid out like this…
TrackName Track_01
Frame X Y Correlation
1.00 817.748 100.708 1.000
2.00 813.603 100.676 1.000
3.00 806.605 100.605 1.000
4.00 797.781 102.009 1.000
5.00 789.841 103.557 1.000So I’m trying almost every thing under the sun to try and re-format although I’m not entirely sure what are the things that I need to reformat to make it After Effects valid.
Obviously there’s the bookends (all the bumpf before and after the numbers) and then I also have to take out the ‘correlation’ column – easy with find/replace) but it’s still not working. I’m assuming that maybe the fact that I have numbers after the decimal points although can’t find an easy way to get rid of those (have 7 tracking points of 500 frames so it’ll be a massive manual slog!)
I’ve tried using Xcel but it doesn’t interpret the figures correctly into columns….
Ho hum.. will post my progress if I get any!!!
x
j -
Brilliant!
Thanks Declan and as always, thanks Dan!
That’s incredibly stripped down and tight and the second time I’ve come across that ‘linear’ thing. I’m going to go back to the last time I used it and try and extrapolate a bit more meaning from it and see if I can hold it in my brain!!!
jim
-
Declan,
ahhhh… yes… ok, so that’s the thought-process-kick-in-cerebellum I needed.
It’s not working yet but that feels entirely due to user error on this side.
I think the main two things I might not have got clear are…
i) When you say ‘clamp the rotation to your scale’ it’s probably really obvious but you mean the scale of the values here not the ‘scale’ of the layer. Actually even in writing that I know that it is. Anyhow, consider that baby ‘clamped’
ii) When you say ‘rotationCurrentValue=360; // This needs to be linked to your actual rotation value’ … am I replacing the ‘rotationCurrentValue’ with a pick whip to the rotation value of the SUN or the bit that says ‘360’? It feels fairly obvious that it’s latter as that’s the only option that gives a variable driven by the SUN.
I guess I only ask as the SLANT range (i.e. the returned results after the expression) that I’m getting when I do the latter is between 100 and 300 which is promising as it’s definitely giving me a 200 value range.
I’ve put in
(newValue-200)*-1
at the end and that’s pretty much cracked it but I was wondering if I’d done something wrong before that point…
Brilliant though, really brilliant, and also so very useful to get the idea of interpolating one range into another too..
Oh and love the Llangollen Narrow boat film. Funnily enough was going to do the saw stretch last year with my Parents in Law but ended up doing one south of the Chippendale museum, mostly as we have a 2 year old daughter and my Parents in Law had awful visions of small children and high aqueducts!
-
Ahhh… a stacked if – else -if
That’s the puppy.
In actual fact I managed a bit of a work around in the meantime which is far less elegant which pre-works-out a value before it heads into the main if statement
if (SLIDER_value < 0) then {VALUE_onoff = 0} else {VALUE_onoff = 100);
VALUE_var = SLIDER_value;if (SLIDER_value > 0) then {VALUE_var} else {VALUE_onoff)
but I’ll go back in and try that other suggestion for the sakes of personal education.
Thank you thank you thank you
x
jim -
Darby,
Thank you so much!
Yes, I’d actually forgotten about that pointing away from camera thing although actually it wasn’t what I needed as I ended up needing layers to be visible for only a quarter of a revolution – 90 degrees – and not the 180 that that gives.
More importantly what you’ve suggested is far more valuable as an introduction to the && and || operators ( although it’s taken me a fair few minutes to find the ‘|’ key on my keyboard (upper case ” if anyone is interested)!!
I shall start to experiment and see how stackable they are too.
Many many thanks Darby and apologies for the late reply – sudden deadline swamped me.
-
Thank you! Yes, I’m there! And what’s more I think I understand!!
As always, many many thanks Dan.
-
Aaaaahhhh… ok… after some more digging and help from a colleague here at work I’ve stumbled up this!…
maskOp = thisComp.layer(“MOON_REVOLVE”).transform.orientation[1];
if (maskOp > 90 && maskOp <270){
[0];
}else{ [100] ; }so there you have it!
&&
works a treat!
-
Ahhh… of course, I seem to remember this thing from a while back about the value of seconds and frames in the decimal returns.
Oh my goodness… it works a treat!
I know you hear it a lot but you are an absolute star Dan, thank you so much!!
I’m going to study the expression and try and properly understand it!
-
ahhhhhh… a bit of thinking and I’m taking my own grown up steps forward…
realising I eventually need
[x,y,z]
I’ve got Dan’s expression onto a ‘slider’ in each of the layers and am referring it back to that so with the slider with Dan’s expression but having the range as 0 – 100 …
x=100-(effect(“Slider Control”)(“Slider”)*80);
y=(effect(“Slider Control”)(“Slider”));
z=transform.scale[2];
[x,y,z]I’m getting there!!!
I’m sure this is a bit clunky but I feel a tiny bit proud of myself!!!
jim
-
I popped Dan’s expression in to start with as I suddenly needed to get a render out and it works spectacularly.
I’m assuming that the ease(x,x,x,x) expression works in a similar way to the linear (x,x,x,x) and I’m going to start playing with Darby’s as that looks a little simpler for me to understand the linear expression.
One thing I’m now trying to do (and this is as much about the basics of expressions) is to apply a similar set-up but to scale which has three values it needs returning (it’s in a 3D space). How would that set-up that’s applied to a single value (the x rotation) be applied to one that has needs three values returning but apply it only to the x-scale?
I’m aware that AE needs to end up with an array of something like [x,y,z].
I’ve tried to drop the code in there
x= linear(my_rot, -500, 0, 0, 181)
but obviously there’s something wrong here…
head aching but getting closer!!!