Hans Tammen

I've used a quiz with my students which requires to use frame or frame_true with jit.movie. This semester it has not worked for many, as both frame and frame_true jump only to the beginning of the file regardless of what arguments we're using.

It works fine using jit.movie~ instead of jit.movie, so there was a workaround available.

I can't narrow down the technical specs: it happened using different file formats, on Windows and Mac computers, on student's and lab computers and on mine, with no discernable pattern. It may work on the student's computer but not on mine, or the other way around, and behaved differently when presented on a lab computer. But it was consistent, if it did not work on one computer, it did not work on it the next day, too. Certainly all machines used recent OS, from the last few years.

I've used a quiz with my students which requires to use frame or frame_true with jit.movie. This semester it has not worked for many, as both frame and frame_true jump only to the beginning of the file regardless of what arguments we're using. 

I can't narrow down the technical specs: it happened using different file formats, on Windows and Mac computers, on student's and lab computers and on mine, with no discernable pattern. It may work on the student's computer but not on mine, or the other way around, and behaved differently when presented on a lab computer. But it was consistent, if it did not work on one computer, it did not work on it the next day, too. Certainly all machines used recent OS, from the last few years. 

Hi,
I've used a quiz with my students which requires to use frame or frame_true with jit.movie. This semester it has not worked for many, as

frame-frame_true-issue-in-jit-movie

hi Hans, the only thing I can think of without seeing the patch is if you are using AVF engine (the default on Mac) and not taking into account the asynchronous seeking behavior, which is described in the jit.movie reference page Discussion section (linked to in each of the reference entries for frame, jump, etc).

Hi Rob, makes sense, as it would explain why it worked on half of the machines but not on the other half. That said, I identified a student's project that does not work on my machine, and changing the video engine does not make a difference. It works with our beloved chickens, though, and it works with jit.movie~.

Is it worth looking into it for C74? I can send you the project (but it includes a 10MB movie). Maybe a link to your email address? Thanks, Hans

Not so complicated to put it together... ;-)

Is there a way to get jit.movie to work ?

I am on the latest Max version and Intel Mac.

Same problem here.
Is there a way to get jit.movie to work ?
I am on the latest Max version and Intel Mac.


Thanks for the example patch and movie file Hans. I can see jit.movie behaving unexpectedly with this file when using avf engine (though other files seem to be fine). I will have to dig in and see what's going on here.

In the meantime you can use the viddll engine which is the same engine that jit.movie~ uses (

Thanks for the example patch and movie file Hans. I can see jit.movie behaving unexpectedly with this file when using avf engine (though other files seem to be fine). I will have to dig in and see what's going on here. 

In the meantime you can use the viddll engine which is the same engine that jit.movie~ uses (jit.movie @engine viddll, or change your video engine Max preference).

Any suggestions for getting the avf engine to work ?

I am using the HAP codec and can not switch to the viddll engine because the poor playback of HAP movies. Long loading times, less precise playback and higer CPU load.

Any suggestions for getting the avf engine to work ?
I am using the HAP codec and can not switch to the viddll engine because the poor playback of HAP movies. Long loading times, less precise playback and higer CPU load.

There is no literally no documentation about the `seeknotify` message on the jit.movie Reference page. How are we supposed to use `frame_true` in tandem with `seeknotify` ?

it's documented in both the Discussion section of the jit.movie reference page and in the help patch. Does this patch clarify?

Is there any use case where we dont want to see the frame after sending the frame message? 

From my point of view the seeknotify could be implemented into the jit.movie object.

Or does it take such a long time to load the frame?

Is there any use case where we dont want to see the frame after sending the frame message? 
From my point of view the seeknotify could be implemented into the jit.movie object.
Or does it take such a long time to load the frame?

Thank you Rob this is very clear. I think it's phrased rather awkwardly in the documentation for jit.movie … I couldn't make sense of it.

frame & frame_true Issue in jit.movie