Am I Asking Too Much of the Prop-1?

[JesusFreak1959]

I've designed & partially built a prototype of a VR ride, where simple motion, sound and video must all sync.  I thought the Prop-1 would get the job done, but I read in another post where J.Mac said that maintaining sync to a 1 -3 minute soundtrack with complex motion was asking too much of the card. 

Here is the gist of the design.  I have a 3 minute video w/sound that will be played back from a video server to an LCD screen and surround sound system.  Motion is generated by a single pneumatic cylinder with multiple sensors/valves to control speed.  1 minute into the 3 minute video is a trigger to initiate motion from a vertical to horizontal orientation that lasts about 20 seconds.  The video/sound and motion must be in sync.  About 1 minute later, another trigger initiates motion from horizontal to vertical that lasts about 2 seconds (where patrons will soil their pants).  Again, the video/sound must be in sync to the motion. 

Is the Prop-1 a good choice for this application?  (Thanks in advance for design feedback!)

[JonnyMac]

One of our big professional customers uses the EZ-8 for this.  The trigger is fed into the EZ-8 which provides the start pulse to a video player; the other outputs move and control elements of the prop.

You *may* be able to pull this off with a Prop-1 but that will depend on the length of your show and the resolution you desire -- both interplay.  The EZ-8 has a 32K EEPROM and is running at 50MHz so this is not a problem.

If you can't make the Prop-1 work I would suggest that you go with the HC-8+.  I know it's not listed as a standard controller but it can be used as one.  It uses the multi-core Propeller processor which give it huge horsepower.  We also have a microSD card add-on coming that would allow you to create show data as a text file and read it right off of there; this means you could change shows without changing your program. 

I will be doing a demo of running show data from a uSD card in the next month or so.  For my demo I will be using servos, but VSA can be setup for digital outputs as well.

[JesusFreak1959]

Well... I'm not a big professional customer; I'm a 1-man operation working for a nonprofit org on a shoestring budget.  I'm leery of investing time/money in the Prop-1 if there's only a chance it *may* work. 

You mention the interplay of resolution and length of show.  My project requires motion to media sync (resolution?) to a quarter of a second throughout the duration of the show to be effective.  The show length is 3 minutes.  Is the Prop-1 capable of handling this?   

The remainder of the reply left me slightly confused.  It appears as though the EZ-8 fits these requirements based on another client's implementation.  But then, your recommendation was for the HC-8+.  $40 for a Prop-1, or $100 for an EZ-8, or $130 for an HC-8+ may not matter to some entrepreneurs, but it does to me. 

Do the requirements of this project fit the Prop-1 or should I go with a higher end E-T product?   Thanks!

[JonnyMac]

My project requires motion to media sync (resolution?) to a quarter of a second throughout the duration of the show to be effective.  The show length is 3 minutes.  Is the Prop-1 capable of handling this? 

Nope.  Note even close.

The Prop-1 has 256 bytes (that's bytes, not kilo- or mega-) of memory (for both program and data).  At 250ms resolution you need four bytes per second and and at 3 minutes you need 720 bytes.  Sorry, no deal.

I'm recommending the HC-8+ because it's out best control product and provides infinitely more horsepower and flexibly than the EZ-8 (which will be phased out for a new product later).  I brought the EZ-8 into the discussion only to point out that the Prop-1 is not the kind of product that is viable for what you want to do; you need more memory and horsepower.  You get that with the HC-8+ and with some materials I'm working on will be able to do what you want with multi-show capability.  Nothing is more boring than having a prop do the same thing over and over and over again.

Yes, the HC-8+ costs a little more.  But I challenge you to find a similarly priced product from any vendor that will offer you the support that we do.  It's a one-time purchase.  If you don't mistreat our boards they will last a long time and you'll be able to re-use them in future projects.  This is something to consider.  There are very big companies who can buy from anyone, but they buy from EFX-TEK because of our product quality and the support we provide (at no charge, mind you).  A lot of one-person operations make the same decision for the same reasons.

[JackMan]

JesusFreak1959,
   I may be looking at this from a totally different direction but from your description of what your prop needs to do I think a Prop-1 will work. It would take a little time and some tweaking to get the program exactly where you want it but I think it's very doable. You would need the Prop-1 to initiate the video player and then it's just a matter of nailing down the timing for the 2 events that you need. I have several Prop-1's set up with crucial sync points to sound and while it does take a little patience to get the program dead on, they work flawlessly as long as your video/audio is consistant.

[JonnyMac]

Jack,

You may be right, but I can't tell what the exact IO requirements needed here.  Given ambiguous requirements I'm forced to lean toward more powerful products to guarantee success. 

[JackMan]

Yep, I agree. It is a little unclear as to exactly what the Prop-1 needs to accomplish here. He did mention simple motion and it appears that there is just a 20s duration of possible multi-valve activation followed by a 2s activation 1m later. Again, the timing of the code would take a little trial and error. I'm guessing it's possible but I could be completely wrong depending on the actual requirements.

[JesusFreak1959]

Thanks for the feedback & help JM1 & JM2.  I really don't want to look anywhere else for product for the reasons you mention.  In this day and age, it is very hard to find good product support, and God knows I need it.  I suppose I can go ahead and purchase the Prop-1 and see if it will work.  It won't be hard to put it to good use in a different scenario if it fails to cut the muster. 

JackMan - it sounds like Prop-1 code execution timing is pretty reliable for simple tasks.  Can you give me more details about what it is your prop does and how long your show is?   

Just curious... how do the big guys implement media/motion?  I'd assume media playback drives the motion rather than vice versa.  Do they use show control software (like Medialon, Venue Magic, QLab) to act as master clock to all subordinate routines, such as pneumatic movement and video/audio playback?

[JackMan]

I have 17 animated props and characters in my show along with a lot of lights, sounds, and music tracks. The show is 14 minutes long. With the exception of one musical segment and 2 of the animated characters being run from a laptop (which might change soon to an HC-8+), everything is controlled with Prop-1's, AP-16+'s, some hacked MP3 players, and my Audio Servo Controllers. Just as an example, one of my props is a full size wooden coffin that is programmed to open at a precise time in the show. The coffin is laying flat and the lid slowly opens with a loud creaking sound effect, fog spills out, an animatronic skeleton hand slowly reaches out, and a skull (with a cigarette in mouth and animatronic jaw) raises as if peeking out at onlookers. When the lid is open about 12" it stops and "Skully" says, "Hey, can you keep it down out there, I'm tryin' to get some sleep!" Then the lid slowly closes (with sound EFX) and the hand retracts into the coffin. The whole sequence is controlled with a single Prop-1 and an MP3 player. The jaw is servo operated and is precisely timed to the audio by the Prop-1 with PULSOUT commands. Like I said, it takes a little time and effort to tweak the program code but it can be done with perfect results as long as you don't need more eeprom space than the Prop-1 has to work with. There are other ways (more complex and expensive) to sync video/audio triggered events, one is with DTMF controllers (dual tone multi frequency). This system uses imbedded frequencies in the video/audio to send signals to a DTMF relay controller. Like I said earlier, by your description of what you need to do, I think you'll be alright with a Prop-1 but it is hard to tell without knowing exactly what the requirements are.  8)

[JonnyMac]

JackMan - it sounds like Prop-1 code execution timing is pretty reliable for simple tasks.  Can you give me more details about what it is your prop does and how long your show is?

As Jack pointed out if your outputs -- which you haven't defined very clearly -- are very simple, the show length is not a real issue.   There is a PAUSE command in the Prop-1 that can create timing delays up to 65 seconds (due to the 16-bit math limitation of the controller).  If you need a two minute delay it takes two command that consume just a few bytes of memory:

  PAUSE 60000
  PAUSE 60000

I answer your original question thinking you had multiple outputs that you could be turning on and off at 1/4 second intervals.  Perhaps if you provided a more detailed explanation of your control inputs and outputs and a timeline of their behavior, we could make a determination from the jump.  What you're running into is common: the hardest part of building a prop with support from others is explaining to others exactly how it's supposed to work.

Just curious... how do the big guys implement media/motion?  I'd assume media playback drives the motion rather than vice versa.  Do they use show control software (like Medialon, Venue Magic, QLab) to act as master clock to all subordinate routines, such as pneumatic movement and video/audio playback?

Which big guys to you mean?  I work with a large, mouse-friendly amusement park from time-to-time and their shows play SMPTE timecode through a control track. The props "listen" to this track for their own trigger (time) and run behaviors from there.   I'm working on a SMPTE interface for the HC-8+.  In the HC-8+ thread I showed how to use DMX as a trigger.  DMX is easy to broadcast from boards and show-control software like Venue Magic.

[bsnut]

JesusFreak1959,

Here's another idea for you and that is, to try Vixen with the Prop-1 sequencer to handle the timing.  All this can be downloaded to the Prop-1 after you have gotten timing right. 

[freakyland]

A few years ago I put together our entire haunted house with prop-1s using Vixen to create the timings.  At that time I had NO programming experience, but with the amazing support from the EFX-Tek guys, it was very easy to get things going.  Our most complex prop was our elevator simulators that pushed the prop-1 to its limit.  We had 2 valves controlling airbags, a sound board trigger, 2 valves for pneumatic doors, a push button start button.  To get any sort of interesting show (45 seconds) I was constantly fighting with the memory limit on the card and some creative tweaking of the basic program.  The programs synced up to the sound effects excellently, but the problem was in the memory as John discussed.   At the time, the owner was extremely cost conscious and didn't even want to move up to the Prop-2s which would have solved that problem (the HC-8+ didn't exist at that time).

Four years later, I am now writing code for the HC-8+ and can't begin to describe how much time is saved by not having to worry about memory.  Rather than hunting through the sequence to find spots to eliminate ride features and compact the code to fit the board, I can now write and write and write and not hit the memory limit.  Yes, it MAY be doable with the Prop-1, but having more than enough board will save you many many headaches in the long run!

[JesusFreak1959]

We finally finished the design & ordered the parts for the pneumatic system.  Here is the gist of what will be required of the card:

VERTICAL to HORIZONTAL MOTION
* sequence initiated on Prop1 & kicks off a video w/sound
* Prop1 sends msg to main valve to retract cylinder (engage/open)
* Prop1 receives msg from cylinder sensor#1 to initiate next action
* Prop1 sends msg to 2ndary valve#1 to speed up cylinder retraction (engage/open)
* Prop1 receives msg from from cylinder sensor#2 to initiate next action
* Prop1 sends msg to 2ndary valve#1 to slow down cylinder retraction (disengage/close)
* Prop1 receives msg from cylinder sensor#3 to initiate next action
* Prop1 sends msg to main valve to stop retracting cylinder (disengage/close)

HORIZONTAL to VERTICAL MOTION
* cued by the end of a Prop1 pause, and while the video continues to play,
* Prop1 sends msg to main valve to extend cylinder (engage/open)
* Prop1 receives msg from cylinder sensor#2 to initiate next action
* Prop1 sends msg to 2ndary valve#2 to speed up cylinder extension (engage/open)
* Prop1 receives msg from from cylinder sensor#1 to initiate next action
* Prop1 sends msg to 2ndary valve#2 to slow down cylinder extension (disengage/close)
* Prop1 receives msg from cylinder sensor#0 to initiate next action
* Prop1 sends msg to main valve to stop extending cylinder (disengage/close)

Asking too much?  From what I've seen the card do at another haunt, and from what JackMan said, I'm thinking it can pass muster.  Thoughts JM?

[JackMan]

Sounds like you plan to have a lot of end switches connected to the pneumatics? It may take a little extra time to get the timing nailed down but the wiring and programming would be much easier without having to deal with the end switches.

[JonnyMac]

I started to put together a program for you but -- like Jack -- I don't think you have enough IO points on a Prop-1 to do what you suggest.