What's New in S3...

[bob]

Bob, the sample sequence that you linked to is not downloading for me. However, I think I might at least partially understand what you're asking now. Please correct me if I'm wrong:

You are hoping for something like the Chase tool, except with the ability to chase several consecutive source channels/RGB channels over the destination channels/RGB channels, instead of only a single source channel/RGB channel over the destination channels/RGB channels. Is that correct?

[rstately]

DevMike wrote:

Mark Showalter wrote:

Still trying to get my brain around everything new in S3. Do you know if the wizards will help create a Weber tree in the visualizer?

I'm not sure exactly what a 'Weber' tree is. However, the initial release (again, as far as I know and it is subject to change) only does 'normal' trees. The Demo @00:57 shows all the options that are currently available for trees.

But again, you or someone else, could draw one by hand and then share it with the 'export/import' functions. With as many sequences as you guys share, I fully expect a large library of shared Fixtures/Props to quickly become available.

Mark's question just clicked with me. If you were to make a ccb mega tree, it could be programed to be a Weber, mega or anything, with the strings run vertically. I'm sure the superstar software makes that easy...

[Mark Showalter]

rstately wrote:

DevMike wrote:

Mark Showalter wrote:

Still trying to get my brain around everything new in S3. Do you know if the wizards will help create a Weber tree in the visualizer?

I'm not sure exactly what a 'Weber' tree is. However, the initial release (again, as far as I know and it is subject to change) only does 'normal' trees. The Demo @00:57 shows all the options that are currently available for trees.

But again, you or someone else, could draw one by hand and then share it with the 'export/import' functions. With as many sequences as you guys share, I fully expect a large library of shared Fixtures/Props to quickly become available.

Mark's question just clicked with me. If you were to make a ccb mega tree, it could be programed to be a Weber, mega or anything, with the strings run vertically. I'm sure the superstar software makes that easy...

And that would be true with a number of common lighting elements used. Think about any number of straight poles (fire sticks, shoots, towers whatever they're called) put together side by side. Then think about the pixels needed to craft animation/video on a TV screen. If the poles were vertical CCB strings, all close together, with each bulb acting like an independent pixel you could create some really cool and stunning effect .

[ItsMeBobO]

bob wrote:

Bob, the sample sequence that you linked to is not downloading for me.  However, I think I might at least partially understand what you're asking now.  Please correct me if I'm wrong:

You are hoping for something like the Chase tool, except with the ability to chase several consecutive source channels/RGB channels over the destination channels/RGB channels, instead of only a single source channel/RGB channel over the destination channels/RGB channels.  Is that correct?

Not sure what you mean now. Arg. I will email the example and here is another try.

http://itsmebob.com/Christmas/2011/MatrixChaseExample.las

Trying to put this in other terms. A group can have a logical purpose. They are separate channels, sometimes you want them all to come on together as if they were all one channel. The on/off & intensity do that from the unit level. But the video shows the chase has different behavior. It does not set all component channels to the same segment of the chase. It steps through each channel as if the group was not there.

Seems inconsistent that the chase would work differently. I hope for an option for the number of steps in a unit chase to not include the component channels.

Attached files MatrixChaseExample.las

[bob]

I mean, for example, that I think you want a way for chases on channel groups to behave the same way that chases on RGB channels do:

An RGB channel is actually a group of three channels. But doing a chase across five RGB channels is not the same thing as doing a chase across fifteen channels:

If it were a chase across fifteen channels, it would take the first RGB channel's red channel, and chase it through the other fourteen channels. Notably, the first RGB channel's green and blue channels would be overwritten.

But that's not what happens. Instead, it takes the first three channels, and chases them as a set through the next four sets of three channels.

I think that you want a way for chases on channel groups to do the same: For example, if you do a chase across five channel groups, each of which has fifty channels, you want it to chase the set of the first fifty channels, as a set, through the next four sets of fifty channels, rather than chasing the first channel through the next 249 channels.

Is that correct?

[bob]

Here are a couple videos to demonstrate what I mean I think you mean:

I believe that rather than (or in addition to) being able to do this:

http://www.lightorama.com/bobpublic/BobOChase1.swf

You would like to be able to do this:

http://www.lightorama.com/bobpublic/BobOChase2.swf

Is that correct?

(I want to be clear that the second video is a mock-up; it doesn't show functionality that is currently in S3).

[ItsMeBobO]

We are getting closer, but no that is not it. FYI thank you SO much Bob for taking so much time to understand.
It is a great thrill for me to have some input and I really appreciate it.

You are correct that it should work for RGB. Each RGB channel within the group should have the effect as is shown at the group level. If you were able to get my example RGB sequence, that is what is in there.

To me, one of the the great benefits of defining a group is to set all channels within it at the group level to be the same, treating it logically as one channel for this effect. But the chase is making them all different from the group level. Each component channel is stepped over a little. Defeating the tool value of making the logical group.

With individually addressable bulbs, we have the great ability to forget the concept of a string and just say 'this set of channels is my string (group)' and this set of channels next to it is another logical string.
Now I want to chase between these two logical strings. All the bulbs in one logical string should come on together. Fading into the adjacent logical string coming on all at once. Chase the entire logical string, not its component channels.

Sorry I have to go back to work so I must pick this up later. ( I have Jim W envy right now)

[jimswinder]

.Sorry I have to go back to work so I must pick this up later. ( I have Jim W envy right now)

I will trade you my job for your brain ANYTIME!!!!

[bob]

I'm sorry, I'm still not getting it. I have looked at your sequence, but I don't know what I'm supposed to see about it. Is it a "before"? If so, what's the "after"? Is it an "after"? If so, what was the "before"?

Perhaps could you create two simple example sequences - just a few channels each or whatever - one being a "before" application of the tool that you wish to exist, and the other being an "after" application of the tool that you wish to exist, and then tell me exactly what time and channel range you applied the hypothetical tool to?

Also:

To me, one of the the great benefits of defining a group is to set all channels within it at the group level to be the same, treating it logically as one channel for this effect. But the chase is making them all different from the group level. Each component channel is stepped over a little. Defeating the tool value of making the logical group.

I don't understand what you're referring to here, either. Perhaps could you tell me a time range of the video that shows what you're referring to?

[ItsMeBobO]

My vocabulary is apparently not as fast as my thoughts on this one.

Trying to rephrase. My mind is off in the 2D/3D clouds. With individually addressable bulbs, we can just forget about how they are connected with a wire and think about them in relation to each other. Groups can have a jumble of non-consecutive channels as far as the wires go. But may be consecutive in the layout so we define a group. Commonly called a matrix but the word matrix infers rows and columns which are not actually required.


At 12:40 in the video a ramp is applied at the arch group and that same ramp is applied to every component channel. There is some value to making them all the same. They will all ramp up and down at the same time as if it were one string instead of individual segments. The person sequencing decides to and intentionally sets all the same effect to treat the individual arches as if it were one string on one channel. All acting together.

What we are discussing is the level of granularity for a chase and whether the above concept (make them all the same to treat as a single channel) applies or if each channel drops its group identity and is treated like an individual channel again. The video shows that the chase always forgets the group identity and treats each as an individual.

Chase applied to 'group'. This view shows the top channel in group
There is no visual queue that group contents are same or different.
Looking at just this picture, we dont know if the four groups were filled from the ramp tool or the chase tool.
If done with the ramp tool, each component channel would be the same.


Now we see Chase was applied to the individual channels. Not the group level and its children.
The number of steps in chase is the total number of channels.



A chase which honors group identity (HGI) would look like this.
The number of steps in chase is the number of groups. Each channel of each group would be the same as when a non-chase effect is applied.
If you opened up the other arches you would see channels 2-8 all had the same ramp position as channel 1.



Sorry to repeat. One more rephrase. While 'arch' would be a common and convenient way to group, there are other reasons too. Color of led strings, adjacent location or Pattern building. A group could mean treat as a regular string where all are always treated the same when applying a tool.

[ItsMeBobO]

To answer your before or after question. All the below are after the 'matrix chase' which honors group identity is applied.

In this picture from my example sequence, in the first two seconds are bars. If you assign groups to channels 1-10 11-20 21-30 31-40 41-50 you will see at the group level it looks like a chase.
But when open all channels are the same. But you must look at it from the proper track. In each track there is one grouping which looks like bars of different sizes while the others are jumbled.


The other tracks have these groups. All channels of each group are the same.



I am worn out so you probably are too. No plans to bug you again unless you ask me a question. If you want to talk on the phone it might be easier but I understand if you have other priorities.

[bob]

A chase which honors group identity (HGI) would look like this. The number of steps in chase is the number of groups. Each channel of each group would be the same as when a non-chase effect is applied. If you opened up the other arches you would see channels 2-8 all had the same ramp position as channel 1.

Bob, that is (I think) what I thought you were saying when I posted my two videos demonstrating what I thought you were saying. But then you said that that wasn't it. So I remain confused.

Did you see the second of those two videos? Here it is again:

http://www.lightorama.com/bobpublic/BobOChase2.swf

In that video (which, again, I want to explicitly point out is just a mock-up), the chase takes all eight channels in the first channel group, and chases them as a set through the next three channel groups.

How, specifically, is what I'm showing in that video different than what you are describing?

[bob]

Maybe the source of confusion is that (in my second video) the collapsed channel group shows an "on" event at time zero, but if you expand that channel group, not all of its channels have "on" events at time zero.

If so, then that doesn't really have anything to do with the behavior of chase. A collapsed channel group does not show "here are the things that are the same in all channels of this channel group". Rather, it shows "here are the contents of the first channel of this channel group".

Here's a video that shows a bit more playing around with applying effects to channel groups, and what the results look like both collapsed and expanded:

http://www.lightorama.com/bobpublic/ChannelGroupEffects.swf

[taybrynn]

To me, that S3 chase functionality on groups makes sense as it is. I mean, it is a little confusing seeing only the first channel when collapsed, but I could live with that.

I like that most singular effects (on, off, ramp-up, ramp-down) get applied to all channel in the group when applied to the collapsed group. It seems like thats what BobO wants ... and seems like it already works like that, no? Maybe the confusion with the original demo which only showed chases operating on the group. It seems like the last couple videos show the behavior desired for non-chase effects applied to the group, right?

I also think that having the chase operated within it as it does makes sense, and saves times for sure, esp. if your groups makes sense.

[ItsMeBobO]

Seems we are getting there. But I did'nt get your example2 with the staggered effect. That seemed to be a group copy. Wondering how that would work with a irregular number of rows in the groups. I do understand a collapsed group can one show the top channel. My attempt was to point out that a ramp applied changes them all to be the same as the shown row one, while a chase applied makes them all different with row one shown.


Let me try a new term. Virtual channel. Locked or handcuffed or in unison channel group.
Any word will do which means a set of channels which can be defined to be treated as if it were one channel.
When any tool is applied to that group, all do the same as if it were an old fashion string on one channel.


A new example of a roof with 100 CCB bulbs spread out. (fancy roof line for a dog house)
Even with an irregular layout, its all about patterns right? Shapes or lines.

Note the bulb numbers are mixed up but still in a line for the wire.



What I want to accomplish is an entire roof sweep by columns.
One way to do this in S2 is create tracks rearranging the channels.

I would then do a chase across the first 18 channels. 100 93 92 85 84 77 76 63 62 49 48 43 42 25 24 11 10

Then I would copy that chase in to the next set 99 94 91 86 83 78 75 64 61 50 47 44 41 26 23 12 9 and the next
But row 4 is a glitch because 2 channels need to be skipped over. A lot fussing to get a column effect by pasting rows


In S3, If I could define column group1 as 100 99 98 97 & column group2 as 93 94 95 96 etc (numbers going from top to bottom)
Then when I chase the top row. 100 93 ... 11 10 (collapsed groups), that chase would assign the same chase segment to each group channel.
The effect would then be a moving column across the roof correctly skipping over the skylight. (matrix chase)

If I want a row effect I would define 7 groups down the side. with group7 being 70 69 56 55 34 33 32 31 18 17 4 3
resulting in lines across. Then I could chase the row groups down.

[Nixay]

I had problems with loading times last year in trying to keep synced with my radio software for music between shows. I had about 80megs across two sequence files which took on average 40 secs to load.

So I'm very interested in the Compressed Sequences option. Just wondering how this works and what the trade-off is (I assume there's some kind of performance or functionality trade-off for it to be an option and not the default setting?).

[JonB256]

ItsMeBobO wrote:

....
If I want a row effect I would define 7 groups down the side. with group7 being 70 69 56 55 34 33 32 31 18 17 4 3
resulting in lines across. Then I could chase the row groups down.

BobO, short of a Galaxia type thing where you use a video camera to map out a grid as the software lights each light in sequence, the way you did it is the only way I've ever seen it done.

[taybrynn]

Great illustration! That would be nice.

Last year, I was trying to write a program to parse the sequence XML and then create functional tracks from the visualizer XML information to make tracks like:

LEFT to RIGHT

RIGHT to LEFT

TOP to BOTTOM

BOTTOM to TOP

RED only

GREEN only

CLEAR only

and then combos of the above ...

it could be really only tracks ... where you could list up to 20 keywords to describe your display in a way thats meaningful to you.

Included as built into metadata for this would be color-centric capability ... which already exists in the channel color assignments in the visualizer.

ALso included in the built-in metadata might be spatial stuff like RIGHT, LEFT, CENTER, TOP, BOTTOM ... based on zones of x,y boxes you could draw onto the visualizer to custom define those to your display

the concept would be that you could enter in a list of your display centric "keywords" that you use in your channel descriptions that mean something to you ... and any keyword listed could have a spatial track created for it ... maybe that would be an incredibly useful utility .. because I think you'd likely run it on your config file sequence ... then get a list of useful tracks you can use, then apply that new config file to the sequences you want to use it with.

so say I use the keyword "treeline1" in my channel descriptions for my first tree line ... I could then have a treeline1 RIGHT-to-LEFT track created, and treeline1-RED-RIGHT-to-LEFT tracks or a treeline1-RED0-LEFT-to-RIGHT created as a functional tracks ... you'd pick the combination's you want to generate tracks for and there you go.

Anyways, it did a program to do that last year, at a basic level ... and it kind of worked ... but I abandoned that effort.

Ironically, LSP has this spatial concept down, but lacks the proper chase to apply to it, unless you make your own as a video.

I think with channel groups in S3, maybe my idea could go in that direction instead of tracks.

Basically, I'm thrilled with the addition of channel groups ... and the ability to sequence them collapsed is key, esp. chase the groups.

[jimswinder]

taybrynn wrote:

so say I use the keyword "treeline1" in my channel descriptions for my first tree line ... I could then have a treeline1 RIGHT-to-LEFT track created, and treeline1-RED-RIGHT-to-LEFT tracks or a treeline1-RED0-LEFT-to-RIGHT created as a functional tracks ... you'd pick the combination's you want to generate tracks for and there you go.

this is why I would like to be able to have a "sort" (like in Excel) where you could sort by the description (maybe even have 2 columns of "description") and also by the color assigned to the channel....and the events of the channels would stay with their assigned channel (see below).


Attached files

[jimswinder]

Bob...

Don't know if you are still monitoring this thread, but something else crossed my mind last night and it involves "Grouping Channels"...in particular the grouping of a Cosmic Color Device (and in this case a CCB)...and after watching your Power Point presentation again, I did not see it mentioned or it listed in the menu anywhere.

I think I may have an element where it would be advantageous to group different bulbs of a CCB together... such as Bulb 1, 7, 11, 34, and 49 (Group 1) and Bulb 2, 8, 12, 35 and 50 (Group 2).

Will this be possible with S3...to take individual CCB bulbs and place them into a group?

And to make it even more interesting, I would need to have a third Group that has Bulbs from Group 1 AND Group 2 IE: Bulb 1, 2, 22, 25, 50 (Group 3).

I know the first concern is having duplicate channels in multiple groups...but no group would be on at the same time.

Make sense?

If not, let me try to explain:

I am thinking about making a talking snowman out of CCB's, and to create the 6 different mouth shapes (movements) would require some of the bulbs being used in more than one mouth shape...hence the same bulb being used in a different Group.