Fades

Message Topic Search Topic Options Post Reply Create New Topic Printable Version

   Just fired our show up for the first dry run.  I was on spectrum last year, and it seems like the fading is much less subtle than it was with Spectrum.

Like where there's a long fade out from 100 to 0, the channel seems to turn completely off well before the end of the interval.

Anyone else seen this?  Also it seems like short quick fades that worked ok in spectrum now act erratically in Aurora.  More of a quick flash than a quick fade.

Any suggestions on how to handle this better?

Author	Message Topic Search Topic Options Post Reply Create New Topic Printable Version
beavis Members Profile Send Private Message Find Members Posts Add to Buddy List Groupie Joined: 01 Sep 2008 Online Status: Offline Posts: 15	Quote Reply Topic: Fades Posted: 25 Nov 2008 at 11:27pm
	Just fired our show up for the first dry run. I was on spectrum last year, and it seems like the fading is much less subtle than it was with Spectrum. Like where there's a long fade out from 100 to 0, the channel seems to turn completely off well before the end of the interval. Anyone else seen this? Also it seems like short quick fades that worked ok in spectrum now act erratically in Aurora. More of a quick flash than a quick fade. Any suggestions on how to handle this better?

LightsOnLogan Members Profile Send Private Message Find Members Posts Add to Buddy List Admin Group Joined: 11 Oct 2007 Online Status: Offline Posts: 3187	Quote Reply Posted: 25 Nov 2008 at 11:33pm
	First question... what firmware is the controller running? D-Light has been working on fixes for ramp timing so the ramps may vary depending on the firmware version .


beavis Members Profile Send Private Message Find Members Posts Add to Buddy List Groupie Joined: 01 Sep 2008 Online Status: Offline Posts: 15	Quote Reply Posted: 25 Nov 2008 at 11:40pm
	My first controller from last year is running 1.13. The new one I bought this year is 1.14, which I think only adds wireless support. I haven't updated the 1.13 controller since I used it last year though, and it's driving the same fades it was last year.

Pony_God Members Profile Send Private Message Find Members Posts Add to Buddy List Senior Member Joined: 01 Sep 2008 Location: Naples, FL Online Status: Offline Posts: 551	Quote Reply Posted: 26 Nov 2008 at 8:25am
	How short/long are the fades?
	Fine. You're so smart you rig up the lights. D-Light users Unite!

Jonathan Members Profile Send Private Message Find Members Posts Visit Members Homepage Add to Buddy List Beta Testers Joined: 07 Sep 2008 Location: SoCal Online Status: Offline Posts: 1237	Quote Reply Posted: 26 Nov 2008 at 12:51pm
	If you are using LED's, you'll see this. That's because they only need a fraction of the power incandescents use. So while an incandescent uses 0 - 100 in 240 steps, the LED's will do the same thing in far fewer steps.
	~Jonathan

LightsOnLogan Members Profile Send Private Message Find Members Posts Add to Buddy List Admin Group Joined: 11 Oct 2007 Online Status: Offline Posts: 3187	Quote Reply Posted: 26 Nov 2008 at 1:28pm
	Jonathan is partially correct (LEDs behave differently and do so in fewer steps). The exact reason isn't the power requirements though. In the most basic circuit design with a way oversimplified explaination, it has to do with the fact that the "perceived resistance" of the LED is not fixed as it is with an incandescent. As the current changes, so does the resistance that must be used in ohms law calculations. The controllers are designed to alter the overall power going into a fixed resistance, not a varying one. The result is a non-linear gamma curve. Add to that the fact that some LED vendors are now using voltage doubler circuits and other capacitive devices and the exact shape of the gamma curve becomes even more complex. The only good way to get a linear fade out of an LED is through current regulation on a DC circuit. Firefly is a good example of this. Now back to the topic at hand... we don't even know if the OP has LEDs... I do know D-Light has been working on numerous fixes to the ramp timings, so this might be a firmware thing. The ramp calculations in Aurora are the correct ones. I've noticed the same effect on Firefly though when helping in the Red to black through yellow thread. It might be possible to add an option in playback preferences to "pad" the ramps some, but I suspect the option would look really bad on LOR or hybrid LOR/D-Light networks. I will look into this. Edited by LightsOnLogan - 26 Nov 2008 at 1:29pm


beavis Members Profile Send Private Message Find Members Posts Add to Buddy List Groupie Joined: 01 Sep 2008 Online Status: Offline Posts: 15	Quote Reply Posted: 26 Nov 2008 at 10:25pm
	These circuits aren't LED's. It's wierd because it looks good in the visualizer, but the ramps are definitely happening much faster than the interval. I sequenced a short test sequence, and same thing. Fades out are happening way faster in lights than they do in design (and as shown in the visualizer) It's really bad in one sequence that's fade heavy. I think I'll hold it out of the show on Friday until I figure out how to fix it.

beavis Members Profile Send Private Message Find Members Posts Add to Buddy List Groupie Joined: 01 Sep 2008 Online Status: Offline Posts: 15	Quote Reply Posted: 26 Nov 2008 at 10:56pm
	Changed some of the fades to bottom out at 15%. Seems like they race to 15%, and then stay at that same brightness for a noticable time before switching off at the end of the interval. The fading definitely doesn't seem linear. I also copied the same test sequence over to some channels on my 1.14 controller so I could watch them together, and it performs exactly the same as the 1.13 controller.

LightsOnLogan Members Profile Send Private Message Find Members Posts Add to Buddy List Admin Group Joined: 11 Oct 2007 Online Status: Offline Posts: 3187	Quote Reply Posted: 28 Nov 2008 at 10:49am
	~~If they seem to hang at 15% and/or non-linear then this is probably a firmware issue.~~ Edit: Misunderstood the above but I get it now... I will look into this and involve the beta testers early next week, but as I mentioned earlier this might be a firmware issue and not Aurora itself. Edited by LightsOnLogan - 28 Nov 2008 at 11:22am