I’m trying to composite text with video by sending two matrices into a pwindow. The video is no problem, but the user entered text is another thing.
I’m using jit.gl.text3d to great effect. I can have the UI accept text and then manipulate it with all the wonderful gl messages such as scale, position, color, etc. This all comes up quite nicely in it’s own pwindow. I’ve also created a handy video mixer.
The problem is mixing the gl context with the simple video matrix. I thought I could use the Matrix (out) from the gl.text pwindow, but I don’t think its a standard 4 char matrix. As such, I can’t just route the output matrix into a composite pwindow with the video.
Some of the forums suggested getting info from the GPU, because it’s writing to the screen as a 4 char matrix, but I didn’t quite follow how to get that info into a new pwindow. It just seems that there’s simple way to write whatever shows up in a pwindow into a standard 4 char matrix…perhaps? I’m open to suggestions.
I found the message draw_pixels that inputs to jit.gl.render and, in theory, outputs a 4 char matrix that can be shown in a window, pwindow, etc. But where is the jit.matrix routed from: the gl.render pwindow, the rendering context or the jit.gl.render itself…?
Basically I think you should draw to a matrix object with a name say "Fritzli".
Then the jit. gl . render object
and the jit.3d.text object
they all should receive a message called "drawto Fritzli"
Now you should receive all in the "Fritzli" Matrix.
Be sure to activate Blend enable everywhere you can.
It shure works if either there is a jit.window named Fritzli,
or a patcher window called Fritzli.
With a Matrix called Fritzli it might work and actually probably should, although I run into problems there two trying to draw to videoplanes in the same Matrix.
but try – it should work.