PDA

View Full Version : Renard Sequencing (Why you shouldn't use 25mS)



xmus
11-17-2007, 05:05 PM
Finally getting the hang of this, after perhaps 200 hours using vixen (over the year) I just made Jingle Bells (Jim Reeves, 1min 50 seconds) in only 2 hours!

My advice (I'm NO EXPERT, but if your a newbie this might help)

1) learn about beat tracks, and make some. (tap in the beat using ctrl key, you'll find this in sequence->audio.) Make a beat track for the basic rhythm, and then make additional beat tracks for other things you want to synchronize to (lyrics, drums, etc)
2) learn how to use the "effects-chase lines-paste from clipboard" feature for chase sequences.... and when making long chase sequences, learn to build them instantly with that feature, not try to create them manually, manually they will look choppy (some fast, some slow, awkward).

The trick is to make a single channel with blocks of on/off (or ramp) that is very long (and perfectly symmetrical), and copy the ENTIRE thing, and then use the "paste from clipboard" feature, you'll be amazed. They will look perfect, and you can make them in minutes (from your beat tracks).

LAST TIP:
In my humble opinion, 25mS is unnecessary, I wish I had never used it on my first two songs. Too late now, they are done. It makes SO MUCH more work to have double the boxes to fill in, especially with the "125, 126,1267 issue, where two bytes are sent via the serial port to do a single channel rather than one, I had "too many" (in the worst case, not the nominal case) channels on a serial port, so when ramping, I had to manually remove instances of 125,126,127 and replace them with 124, and 128 (respectively?). If I had simply used 50mS I could have avoided that all together.

To prove to yourself that 25mS is unessisary try the following experiment. Take a bunch of channels, i'll give the example with min-trees (which is what I used) and have every other channel turn on at time X, and then all the other channels turn on 25mS later. You cant see it.


Now if you repeat the experiment above with 50mS resolution, you can detect the difference in timing. And honestly, after HOURS of hacking sequences, I never had a light I couldn't time perfectly with the beat at 50mS.

Again, I'm sure others will disagree with me, but from now on, I'm only using 50mS in the future.

For the record the WORST case maximum channel count for Renard @ 25mS is 43 channels, and 86 @ 50mS (Given the 125,126,127 issue).

Now, 90% of the time, you are fine with more channels than that, I am running 96 channels per serial port at 25mS and I only rarely see issues, so i need to manually go in and get rid of 125/126/127 sometimes (what a time waster!)


Just my 2 cents worth :)

For more info:
http://www.christmasinshirley.com/forum/viewtopic.php?t=921&highlight=124

NogginBoink
11-17-2007, 06:16 PM
It seems that KC (or one of us) could write a Renard plug-in that would automatically strip values 125-127 and replace then with 124, 124, 128 to avoid this whole issue.

xmus
11-17-2007, 06:30 PM
It seems that KC (or one of us) could write a Renard plug-in that would automatically strip values 125-127 and replace then with 124, 124, 128 to avoid this whole issue.

I agree, Phil even says it would be easy, but he doesn't want to step on any toes. Apparently there is some concern about having more than 1 release of the Renard dimmer released. The assumption is, most folks would want "perfect dimming", and not our "more efficient in the worst case" dimming.

NogginBoink
11-17-2007, 06:34 PM
It seems that KC (or one of us) could write a Renard plug-in that would automatically strip values 125-127 and replace then with 124, 124, 128 to avoid this whole issue.

I agree, Phil even says it would be easy, but he doesn't want to step on any toes. Apparently there is some concern about having more than 1 release of the Renard dimmer released. The assumption is, most folks would want "perfect dimming", and not our "more efficient in the worst case" dimming.

A checkbox in the plugin setup dialog, "skip dimming values 125-127" would take care of that, right?

And for the perfectionists, we're talking of steps less than 0.5%, fer cryin' out loud!

P. Short
11-17-2007, 07:29 PM
Here is an experimental version of a Renard plugin that skips over the two-char codes (replacing them with the nearest values), and has one other minor speedup. Unzip the file, and copy the resulting DLL into the vixen output plugins directory. This .dll has a different file name from the regular Renard plugin, so it will appear as a separate entry in the list of plugins (and your sequence will need to be changed).

Use at your own risk...it's only had very cursory testing.

--

Phil

Edit - I removed the Version 2 Protocol stuff, use the standard plugin if you need it.

--

Phil

xmus
11-17-2007, 07:40 PM
Phil thanks so much, that is awesome!

xmus
11-17-2007, 11:04 PM
Hi Phil,

It seems to work(I'm having some weirdness, but I don't think it is your fault)

Does it matter that there is no checkbox for "hold port open" on your modified plugin? I don't know what that is for, but..... just asking.

Thanks,
Dave

P. Short
11-18-2007, 12:18 AM
Don't know...I'm still on the last version of Vixen, so I haven't looked at what that is about.

--

Phil

xmus
11-18-2007, 12:59 AM
Don't know...I'm still on the last version of Vixen, so I haven't looked at what that is about.

--

Phil

okeydokey, I've got the same question posted on another thread, so maybe I'll get an answer over there. Your plug in seems to be working just fine.... The weirdness I was reporting earlier was due to a strobe causing some flakiness (I posted that in the Renard->General area).

xmus
11-18-2007, 02:44 AM
Well I'm not 100% sure the modified plug in is helping. It definitely is not hurting anything, but I still see some occasional stuff, and when I look at the sequence, sure enough I find a 125,6,7 there... when i clean it up, the problem goes away....

P. Short
11-18-2007, 12:01 PM
Are you sure that you are actually using the modified plugin for the sequences in question?

Anyway, I changed over to Beta 4, and I see that KC added code to help with multiple plugins and cross-fading, so I need to do some work on this.

--

Phil

xmus
11-18-2007, 01:33 PM
Are you sure that you are actually using the modified plugin for the sequences in question?

Anyway, I changed over to Beta 4, and I see that KC added code to help with multiple plugins and cross-fading, so I need to do some work on this.

--

Phil

I'm definitly using the correct plug gin. It says "Modified Renard Plug in".

P. Short
11-18-2007, 02:26 PM
Let's take this to PM for a bit...

kmc123
01-01-2008, 10:47 PM
Hi Everyone - What has become of this issue? Does the latest beta remove the problem dimmer levels? Or is the test dll still needed?

xmus
01-01-2008, 11:07 PM
Howdy,

Phil released the modified plugin that does not use the double byte characters. I think it is in this thread a little higher ....
And it is working perfectly.

It is also here
http://www.ritzfam.com/xmus/

P. Short
01-01-2008, 11:09 PM
Xmus,

I've been told that KC made a modification to that modified plugin, something to do with the comm port errors. I don't have a copy of the changed version, though.

--

Phil

xmus
01-01-2008, 11:42 PM
Thanks Phil,

I did notice that if you press stop in the middle of a sequence you get com errors and have to reboot. But if you just let the scheduler manage things, it runs perfectly. (let a sequence finish by itself don't press stop).

Thanks!
Dave