Digital Workshop

It took me about three hours to re-write this simulator and now I've spent 2 days with this silly little issue. I've checked, checked, and double checked, even re-rendered the clip in question, and I am now convinced this has to be an Opus bug. No matter what I do the fourth clip from the left shows the last frame instead of the first until other clips have been functioned, when it will suddenly revert to the first frame and work just fine.

I'll be specific, it that will help.

If I go in order and function the first lever on the left, then the second, and then the fourth (normal operating order for this machine). the video is on the first frame and the audio plays, but the video does not and diplays the LAST frame of the video.

If I then simply hover over either lever 1 or 2 (lever 3 does not have a "go to specific frame" command) and return it is on the first frame and works fine.

Now, if I start fresh, then start from the right and move down the center (lever 6, 5, 4) when I get there it shows the FIRST frame, plays sound, and does not move. Then, if I move over lever 5, then back to 4, it is on the FIRST frame and works fine.

Spooky. Frustrating. Expensive.

Dave

Maybe you've tried this before, or, maybe it's a dumb answer to your problem, but have you tried just putting one video object on the stage and setting it to a variable and then use the buttons to "load" the video object with clips based on variable names?

I don't know about anyone else, but I've always found multiframes to be VERY quirky, and it seems like your setup could open yourself up to those quirks.

Maybe I'm not understanding your needs enough, but just thought it'd be worth mentioning.

NOW you tell me...

I get so used to Opus working reliably that I assume it must be me when something does not behave as expected.

The clips here are locked off shots of a machine operated by the levers below. The issue is to have the levers work as realistically as possible. That involves having clips come to the front rapidly as as close registered as possible. I tried this a variety of ways and the multiframe appeared to be the best option. I did not know it was not reliable.

As the clips have to be AVI due to the reverse function, they are large and need to be in place.

Unless I am not understanding your approach, I cannot see how it could be accomplished with mouse over triggers.

Since I have your attention, perhaps you have some thoughts on how reverse video might be accomplished with proportional speed control?

Also, I am curious about one oddity of the reverse video routine you provided last year that has worked so well for me. When set to 1 ms back as first posted, the reverse was twice as fast as forward. When I looked at the input for min:sec:ms (which really should be min:sec:frames since this is video!), it appeared that there was only one position available for ms. Being adventurous, I entered :05 anyway. It appeared to take, and the reverse speed was then perfect. However, when I return to look at it, it is set to :0...though it still works fine.

Just wondering how that worked. Only thing I can think of is that it is not set to :05 at all but to default, 30fps.

Dave

Don't quote my on multiframes being quirky, that's just been MY experience.


The only other thing I can think of off the top of my head that might make things easier is to be convert the AVIs to Flash and then use JavaScript to control them from within an HTML page. I think you could do everything you want that way, but not 100% sure. You probably don't want to go down a different path at this point though.

As for the reverse action, I can't even remember that far back so I don't think I'd be of any further help to you on that. Perhaps one of the scripting gods might have a scripting solution for you.

Sorry, not much help, just passing through today and thought I might be able to quickly help folks.

OK, so quirkyness in multiframes is just yours and my experience. How many before we can call it a BUG?

You are correct. I need clean, Opus executables. I am no fan of flash...way too limiting and overhead intensive.

Browsers? We don't need no stinkin' browsers!

I don't give up until I have conclusive proof it can't be made to work.

Dave

PS - Because this issue is reliably repeatable, I believe it is fixable.

I've been working with an advertising agency on their credentials - they have over 200Mb images, text, brandscape videos, radio clips delivered via multiframes so they can get at them from all sorts of directions. No quirkiness in any of them!

We can't test the attached IMP properly because it doesn't have the video you refer to. Let us have the resource and we can look into it.

In the meantime I have a trick which I find often works in these circumstances even if I don't know why. Copy the offending page into a new publication and delete the original. Save and close both publications, then reopne them and paste the buggy page back into the original publication and reconnect any actions which were lost. Probably doesn't always work but saved me on several occasions.

The other thing to try is to rebuild the offedning object or page again in a new clean publication. If that works copy it into the working publication.

Hope this helps.

Paul Harris

Well, I'll have a go at uploading the whole thing via zip. Since it is 105mb and the limit is 5, I do not expect it to go.

It didn't.

If you will provide me with an ftp address via my dave.mallette@hpidc account I will gladly upload it. I really need this fixed. Once going, I'll clean up a bit as the video needs to be "registered" so movement from action to action is more realistic, but the basic engine is pretty much what we need. The existing one I did a year ago is much less sophisticated but quite highly regarded by the instructors and students.

One thing this demonstrates is the usefulness of reverse video. While this routine works pretty well, I still would really like proportional speed control as this is required to make many simulations more realistic.

Regards,
Dave