Malfunctioning Random Group Effect

OK, I know I'm crazy but I really want to recreate an Effect that I have used for years in the past.

Thirty consecutive channels, in random order, popping on in a zero count, dwelling a short time, then fading in a time about four times as long as the step.

This should be a fairly straight forward task.  I ask any of you with five minutes of time to try this and watch what happens;

New Effect #1

Steps 1 through 30, channels 1 through 30.

Step time, 1.  In time 0.  Dwell time 1,  Decay time 4.

Select Forward, Positive, Random Group.

Run this effect Live while holding the Data Key down.  If you see what I see, the effect will seem relatively Random at first, but over the course of a mintue or less, you will see the steps become less and less correct.  At this point, each step will activate four consecutive channels in what seems to be a snapshot of the non random effect.  Also, the decays are not fading in four seconds.

It really quite fascinating to watch a computer progress from randomness to another state, but its not what I want or need.

If I get some people to agree with me that this is a problem, then I can ask for ideas about workarounds.  We have discovered that you can't place and Entry time on an Effect without a token dummy Exit time of at least .01 secs.  We have discovered that Capture doesn't work correctly if the  newly rereleased RFR is attached to your console.  We know that you can't use the Fader Page Key in a Macro, and I have recently discovered the same for the Stop Effect Key.  I can live with these issues and whatever else we find as long as they are recognized as problems, and I am assured that they are being worked on.

B

  • Brent,

    You are absolutely right, I have tried the effect you describe and am seeing a very similar thing. I get a grouping of five consectutive channels coming on together, and as you say, the fade doesn't seem to last 4 seconds. When I have created random effects before it has been with fewer channels, so the problem hasn't really occurred. It does seem as though there may be some sort of problem with the Random Step Generator thingy wotsit. (thats a technical term!!)

    Sorry I can't provide a solution, but at least you know it's not just you! 

  • I see it here too, thanks for the step by step reproducer. I think the defect only occurs if the steps overlap, meaning the step time is less than the other three times combined. I'll be investigating it further this morning and will let you know what I find.
  • Hi Brent

    Tried your example and got the same results as everyone else.  

    Had a play to try and figure out a work-around and the best I could come up with is to use a Linear Effect.  Once you have created the linear effect, edit the graph so instead of being a sine wave, it is a ramp down.  It also works best if you keep the line you draw in the +'ve on the X axis (things that are +'ve on the X axis add on top of the background value.  things that are -'ve on the X axis subtract from the background value).

    Set your entry/exit times as required.  Attribute field should be set only to forward (don't set random group here).  It probably will work best if you set entry to "cascade / fade by size".  Leave Grouping set to "spread" and trail to "even". Once you're happy go into live.  For the purpose of this example, let's say the lights you want to apply the effect to are Ch's 1 > 30.  

    On the command line, enter [1 > 30 {offset} {random}  effect # enter].  You can set the ch's to a base level and they will pulse above the base level and then ramp down back to their base lvl.  If you want them to be pulsing down to zero, leave the ch's at 0 and apply the effect over the top.  if you want them to be brighter, increase the size.  Obviously, if you want them to go faster or slower, adjust the rate.  If you find you want them to overlap more (or less) adjust the trail setting in the effect.  If the trail is set to "even" over 30 chs, then it will be running at 100/30 = 3.3%  For more overlapping, try 2%, less overlapping 4 or 5 or greater%.

    Another thought... If it is an effect you will be wanting to apply randomly a fair bit, you may want to make a group with the chans stored in the group randomly.  This will save you having to enter in [chs 1 >30 {offset random}....] each time.

    This should generate something close to what you were after with the step effect

    Hope that makes sense and helps you out of your bind.  Have fun!

    Cheers 

     


  • Dude,

    I can't tell you how much I appreciate your free time and enthusiasm.  I can't wait to try that out tomorrow.

    B

  • Happy to help anytime - Helps me get my head around the desk better too! yay!

    Hope it all works out for you.  Just read your post using absolute effect - i'll have to try that one out tonight when I get into the theatre

    Cheers 

Related