Intermittent problem with timecode triggering

  • I am running a set of sequences with isadora using a timecode actor i got from this forum, i also installed the patches that were suggested. It worked for 3 weeks at home and now it occasionally doesn't do something. Last nights 1st show the dowser didn't open on the projector and i had to leave the stage and manually hit the actor to send the serial data. I'm in the position where i push go and then i'm on stage performing as Isadora handles a few music cues and triggers several devices with serial at exact moments in the music via the timecode trigger. if any of them don't get triggered i'm screwed since this is a magic show. I'm using the timecode actor to trigger the "send serial" actor to trigger a device(s) via wireless transmission through an xbee network. for example the dowser has an arduino with an xbee receiving the code, from the mac book pro via an xbee usb explorer, to either open or close the dowser which is a piece of thin metal attached to a servo that moves in and out of the light path. It's been flawless at home in my studio but I set it up at the magic castle this week 1/21-1/27 and so far just the dowser issue occurred once, i'm dreading each time as to what might happen. Now I do realize that I'm assuming it's the Timecode actor because when i triggered the send serial it opened the dowser. but maybe it's the xbee? again i have no idea, it worked all month at home. what changed all of a sudden? same laptop, same file.
    macbook pro 13" 2009 4gb ram standard video.

  • What distance is there between the two Xbee modules?
    Are you using the regular (1mw) or pro (60mw) version of the xbee?
    Is the xbee in the dowser enclosed in a plastic box?
    I have had signal drop outs with xbee's before.  Try upgrading to the pro version and/or use one of the versions with an external antenna.

  • I'm doing a show at the moment with xbee modules and the signal response is quite different in the venue compared to my studio - the performers bodies can block the signal. If you are suspicious of the timecode not working an alternative to try would be to use the position output of the movie player with an inside range actor set to the position of the trigger point in the movie as the low setting and +0.5 for the high. I prefer this method to using comparators.

  • Izzy Guru

    I also first tried to use the position output and the inside range actor to send a serial command to an arduino for a museum installation, but the result was not very accurate, sometimes it was perfectly sync and sometimes not. The problem of the inside range actor is, that if your trigger has to be exact on one frame it can't really be, because it can be anywhere inside of that range. I endet up using trigger delays (see attached image), the trigger was started together with the movie and this worked really great.


  • Difficult to say without more information.

    How different is the positioning of things to how it was in your studio.
    Is there anything between the radios that wasn't present in the studio (walls, curtains, people, lights) all of these could interfere with transmission, lots of people with mobile phones won't help either. For a critical action that must be fired it can be a good idea to send a reply from the receiver to acknowledge receiving the signal - if acknowledge hasnt come back within a certain time, send again or take other action.
    Other possibility - are you testing for an exact time, or whether you have passed a certain time to trigger - if you are testing for an exact match of time, if the patch is doing lots of other things it may not perform the test at the actual time, and once it has passed will never be fired.

  • The setup is with the laptop stage right and there's curtains and drywall between the dowser and the laptop. The arduino is in a plastic box with external antenna. the distance is the same as in my studio, only it was in an enclosed metal building and this is in a converted concrete parking garage in hollywood. Last night i had several more issues. I will look into everyone's suggestions. I appreciate the fast response. I do feel that it's a timing thing with the output.
    I have several timecode triggers that trigger, serial data, sound files, video files. 1st trigger is audio, 2nd is for dowser to open, 3rd is for video playback and a second or two later the trigger for a curtain to be released into place. Last night the curtain didn't release so i ran off stage, then the dowser didn't close and i had projector black for the rest of the act which sucked for blackouts.
    I appreciate the fast response of everyone and will look into your suggestions.

  • Ok, so it seems to be a transmission problem. had a disasterous show where everything worked except the dowser didn't open. even after i ran off stage to manually hit the send serial button. so it seems to be some sort of lack of communication. Does anyone know a good way to output serial data from a mac book pro via a wire. i could use an arduino but want to know what others have been using.

  • Izzy Guru

    I always use the Keyspan USB serial adapter.


  • Are your send serial actors inside user actors? It appears that in my patch at the moment if the send serial actors are inside a user actor they aren't triggering unless I open up the user actor to trigger them.

  • Do you have xbee devices in multiple places - if thats the case then using wired serial instead is more difficult as you would need a serial adapter for each device, or add code to pass messages on from one device through another on an additional software serial port on the arduino. This would be TTL serial (at 5 volts), wont manage long distances in noisy electrical enviroments well. If its a bog standard uno then a long usb cable to it will give you a serial connection to the board, but usb can be problematic over long distance as well. Each board could plug in, you would need to configure multiple serial ports, and send the relevent messages to the correct serial port.

    Protocols like midi and dmx are basically ways of transmitting serial data between devices with more electrical protection/robustness. Dmx includes support for multiple devices on a cable run, midi can do it if devices have passthrough on them.
    Probably difficult to implement in time for this run, but dmx would be the route I would be inclined to look at for wiring them up in the long term.
  • Izzy Guru

    Yes my send serial actors are inside a user actor and it works. You can see an example HERE.


  • I'm sorry to hear about the dowser problem. I must say that I am personally always dubious about using RF for anything unless absolutely necessary. As you experienced, the physical layout of the space -- and even the number of people in the audience -- can negatively affect the situation. I could list my own personal disasters in this regard here, but perhaps it suffices to say they happened to me in the past. "Once burned, twice shy" as we say in English.

    That being said, I wanted to add one note to Michel's previous post. First this is what wikipedia has to say about RS-232 cable length:
    _The standard does not define a maximum cable length but instead defines the maximum capacitance that a compliant drive circuit must tolerate. A widely used rule of thumb indicates that cables more than 50 feet (15 m) long will have too much capacitance, unless special cables are used. By using low-capacitance cables, full speed communication can be maintained over larger distances up to about 1,000 feet (300 m).[9] For longer distances, other signal standards are better suited to maintain high speed. (See
    RS-485 serial (where you have a balanced/twisted pair, like an XLR mic cable) can go much further. At 9600 baud, you can apparently have a cable length of over 2000m. (See formula here: The higher the frequency, the short the maximum length. (They say on the same page that at 3 mega-baud (3000 kbaud) you can still have a 100m cable.
    So if RS-485 is an option, one should go in that direction.
    Best Wishes,

  • dmx uses rs-485

    rs232 is the full old school serial spec with 12v signalling (can be done with microcontrollers but needs translation chip - if you are doing that you may as well go to rs-485) serial at 5v is likely to get lower ranges than that

  • Unfortunately I don't have the time to reconfigure the devices to be wired this week, but for the future this will happen. I will probably go with a RS-485 option. I did located some extended range 2.4ghz antenna and implemented that yesterday, also moved the transmitter higher up, and pointed all the antenna towards the transmitter. it all worked. But i was still uneasy about it. I know what i'll be working on next week! Also I'll be getting some aluminum to wrap my head with.
    Thanks for all the comments!

  • Well the week wrapped up well. We had 3 flawless shows on sunday night. everything worked as it was supposed to. I have no idea. But i will modify the system so there's a physical backup system in place that can control the stuff.
    Thank you for all the feedback.

  • Flawless shows... that's music to my ears.


  • So an update on this whole thing. I had to get a new computer, old one had logic board problems and was probably part of the issue. It works great and stable. Ran it for 2 weeks all day long at home. I took it to the magic castle this week and it worked for rehearsal and then went nuts during the shows. Now the time code triggers all go off randomly. When I hit play everything goes off. All the sound, music and video files triggered over and over again. Effects that I had remotely controlled all went off as well. I don't know what happened between tuesday night and wednesday afternoon. It sort of worked on tuesday, i was able to get through 2 shows, but wednesday afternoon i went in early to test things and trouble shoot and ended up with a completely useless program. I uninstalled it and reinstalled it half a dozen times. I ended up, after a long, long process, exporting the video out and making a sound track with video for the one main scene I needed. So now I hit play in Quicktime player, it plays one music track and then the video. I am manually activating a few things that I was able to make manually controlled. I have a scrim that comes in on stage, and then travels off stage. used to work with electronics, now there's a string you pull off stage to release the drop, then i have worked my choreography to pull the scrim open rather than having it open on it's own. the other stuff i am just doing without. So at this point I have no idea what is causing all the time code triggers to trigger. there's two sound files that play one after the other. there's timecode triggers connected to each sound player. why would timecode triggers go off when the sound file they are attached to isn't playing? also just now i opened the file and ran the mouse across the menu items selecting "Output" and it triggered some sound effects? (can't replicate it now) what is going on? all this happened after a great 2 weeks of hard core testing and leaving it on for hours on end. I had NO problems. it was all 100% then i go take it to the magic castle on monday and it all goes to hell. At this point I can't use the program at all. I even tried making a new file, which would take hours, but i started it and tried the first few items and it still went nuts. it's like there's something inside the computer and it moved in on monday and for some reason it's not letting me do stuff that i want to do, only what it wants to do.

    2013 macbook air 11" with mavericks(could be problem?) thinking about putting mountain lion on it, but it was working great till show time!
    Out putting serial via xbee to other xbee's connected to arduino's and a projector connected with a vga to rca converter box.

    I read someone partitioned a mountain lion os on their comp. maybe that's an option, also how do i properly uninstall and reinstall Isadora. Are there prefs that maybe are screwed up? or some files that need to be removed manually? again I did nothing to this except take it to the venue and plug it into power and it all went nuts over the course of 3 days. worse an worse each day to the point now that the program doesn't work.

  • Hello @magicdominic
    Which version of Isadora are you using?

    Here is a knowledge base link to installing more than one version of Isadora.
    Also to let you know there is a new pre-release version available publicly since yesterday, with some Maverick fixes.

    It would be good at this point perhaps, if you could send a Bug Report (select topic 'Bug report'), so we can start a ticket to investigate.

  • @magicdominik - When you say "I took it to the magic castle this week and it worked for rehearsal and then went nuts during the shows," I am inclined to think it was not an Isadora issue, i.e., that something changed in the setup from the rehearsal to the show. But let's make sure. @primaldivine can you please follow up with @magicdominik directly to get a more complete report and to find out what went wrong? Thanks.
    Best Wishes,

  • I dont know how many of your problems were for events that were related to xbee communicated parameters but for those that were xbees are not really suitable for high density rf environments. A room full of people with smartphones and other wireless devices can easily ruin your xbee communication. 

    Onyx Ashanti talks about it here (as well as his switch to RNXV over xbee - these instructions have a few mistakes though).
    This also has its problems, but can be easier to deal with in combating the hipster force field than xbees. Apart from cell phones other devices (large motors for moving stage elements) can also give off enough interference to kill xbee communication. They are not really venue safe.