ETC Lustr 2 LED over Conventionals

[EdSavoie]

To be fair, LED tape shouldn't really be controlled over straight DMX... Aside that most lighting consoles aren't really entirely designed to control pixel surfaces, the amount of universes required to control anything with a reasonably decent resolution is crazy.

Using DMX to trigger a controller with prerecorded scenes is another matter though.

 

[soundofsparks]

If by "point channels" you mean channels that nest under a main channel, that how fixture attributes work.

You don't need to make channels 1-3 as Red, Green, Blue respectively. You make channel 1 a device with RGB attributes. That's how you can use the color picker, etc.

Would that give you more than 2048 outputs? Not alone.

Also--- saying that the Ion has 2048 channels is disingenuous. You mean it has a max of 2048 outputs or addresses. If you had a fixture with 2048 parameters you could put them all as channel 1, but then you're maxed out. In terms of software capacity, Ion can have up to 16000 channels.

Also also --- you can upgrade the console for more outputs. Ion can be upgraded to 6144. Eos Ti can go up to 32768. The reality is that output limitation is more related to a company's available money than the limitations of the console/software.

Until the ion adds point channels (or whatever ETC comes up with to do something similar) I would not suggest using the pixel modes on the spiiders, you're just asking for a mess. I do think that console manufacturers need to rethink how they license consoles, when $150 of led tape from china can max out a $10000+ console, something needs to change...

 

[danTt]

If by "point channels" you mean channels that nest under a main channel, that how fixture attributes work.

You don't need to make channels 1-3 as Red, Green, Blue respectively. You make channel 1 a device with RGB attributes. That's how you can use the color picker, etc.

Would that give you more than 2048 outputs? Not alone.

When i talk about point channels, I'm not talking specificly about attributes. I'm talking about multiple intensity fixtures. My response was to a person talking about using all of the modes on the spiiders. Two of those modes target individual pixel control That means your fixture has 19 red attributes, 19 green attributes, 19 blue attributes, and 19 intensity attributes. (Oh, and 19 Amber attributes if you use mode 10). Even that's misleading, because you also have a separate "master" red, green, blue, and intensity attribute, as well as a parameter in the profile that allows you to choose how the values on the master channel and pixel channel interact.

You cannot use the color picker successfully on a multiple intensity fixture on etc software. The workaround is to split the profile up, making a "master" channel that has most atttributes, and then a number of separate channels that are generic RGB/RGBA, and either targetting them directly or as a group to adjust the color. This is fairly gross, takes up a lot of screen real estate, and doesn't work nicely when the profile puts the pixel channels in the middle of a profile.

The way MA software has tried to address this is through the use of "point channels". Fixture/Channel 21 would be your master channel, and then fixture/channel 21.1-21.?? would be your pixel channels. This way it all lives inside the profile, and is clearer to understand what selecting a given number might do. It still isn't ideal, the screen reall estate is managable on the MA but would not work so well with etc's current channel views. With the way the industry is going ETC needs to come up with a solution to this, it's just a question of when and how.

Also--- saying that the Ion has 2048 channels is disingenuous. You mean it has a max of 2048 outputs or addresses. If you had a fixture with 2048 parameters you could put them all as channel 1, but then you're maxed out. In terms of software capacity, Ion can have up to 16000 channels.

Also also --- you can upgrade the console for more outputs. Ion can be upgraded to 6144. Eos Ti can go up to 32768. The reality is that output limitation is more related to a company's available money than the limitations of the console/software.

This is true, to a point. Yes, a 2048 output ion allows outputting 2048 DMX values across any number of universes, and the use of channel was unclear (though more than a few consoles refer to "DMX Channels" to mean addresses... so I'm not sure whether @OscarMelectrified was speaking of addresses or console channels.

And yes, you can upgrade your consoles. However, my point still stands. WS2812 tape comes in at approximately 180 addresses/meter. A 2048 output console can run all of ten meters of tape, which can be purchased for ~$70 on amazon. Console upgrades are roughly $1/output (very roughly, depends on all sorts of factors.) This means for $5,000 of LED tape, I would need to spend $30,000 on console outputs. (And a decent amount on alcohol to make me forget all of the LED tape hell I'd be putting myself in). It's similar when you look at other fixtures. Consoles licensing makes sense when you look at a number of small parameter fixtures, but when your devices are eating 90+ parameters without trying, the console licensing approach does not scale well to handle it.

 

[soundofsparks]

When i talk about point channels, I'm not talking specificly about attributes. I'm talking about multiple intensity fixtures. My response was to a person talking about using all of the modes on the spiiders. Two of those modes target individual pixel control That means your fixture has 19 red attributes, 19 green attributes, 19 blue attributes, and 19 intensity attributes. (Oh, and 19 Amber attributes if you use mode 10). Even that's misleading, because you also have a separate "master" red, green, blue, and intensity attribute, as well as a parameter in the profile that allows you to choose how the values on the master channel and pixel channel interact.

You cannot use the color picker successfully on a multiple intensity fixture on etc software. The workaround is to split the profile up, making a "master" channel that has most atttributes, and then a number of separate channels that are generic RGB/RGBA, and either targetting them directly or as a group to adjust the color. This is fairly gross, takes up a lot of screen real estate, and doesn't work nicely when the profile puts the pixel channels in the middle of a profile.

The way MA software has tried to address this is through the use of "point channels". Fixture/Channel 21 would be your master channel, and then fixture/channel 21.1-21.?? would be your pixel channels. This way it all lives inside the profile, and is clearer to understand what selecting a given number might do. It still isn't ideal, the screen reall estate is managable on the MA but would not work so well with etc's current channel views. With the way the industry is going ETC needs to come up with a solution to this, it's just a question of when and how.



This is true, to a point. Yes, a 2048 output ion allows outputting 2048 DMX values across any number of universes, and the use of channel was unclear (though more than a few consoles refer to "DMX Channels" to mean addresses... so I'm not sure whether @OscarMelectrified was speaking of addresses or console channels.

And yes, you can upgrade your consoles. However, my point still stands. WS2812 tape comes in at approximately 180 addresses/meter. A 2048 output console can run all of ten meters of tape, which can be purchased for ~$70 on amazon. Console upgrades are roughly $1/output (very roughly, depends on all sorts of factors.) This means for $5,000 of LED tape, I would need to spend $30,000 on console outputs. (And a decent amount on alcohol to make me forget all of the LED tape hell I'd be putting myself in). It's similar when you look at other fixtures. Consoles licensing makes sense when you look at a number of small parameter fixtures, but when your devices are eating 90+ parameters without trying, the console licensing approach does not scale well to handle it.

It really comes down to what you're trying to do:

If you want to control a Spiider from an Eos and have full discrete control over each pixel, you can build a custom profile based off the default M10 profile that breaks each pixel into unique labeled attributes. Then you'd have full manual control over all 123 parameters built within one channel heading. You could type: Chan 1 Red 1 @ 50, etc.

If you wanted to use the color picker or the pixel mapper on the individual pixels, you'd have to make two custom profiles. First would be all the master control, dimming, master color, pan/tilt, etc... Second would be color control for one pixel. Granted, yes, this wouldn't nest nicely. You now would have to dedicated channels 1-20 as your first Spiider. But, at that point, I would make a magic sheet that has my main channel as a tombstone and all the other channels laid out to show the face of the fixture. Now you could click on a "pixel" to select it and then click on the color picker to change the color and see the change on your screen. This patch set-up would allow you to use the built-in pixel mapping to map animations across the face of the fixture(s) with relative ease.

But my question would be, do you really need all that control? Sure, I can imagine wanting it, but then wouldn't I also have abandoned DMX all together? Functionally, the difficulty is not a problem with ETC's console, but a problem with the limitations of the DMX Protocol.


In terms of cost:

The max MA session can have only 59,850 parameters. By my quick count, a full Spiider is 113 parameters over 123 addresses (correct that, please). The max Eos can be is 32768. I think a maxed out Eos Ti should run about $50,000. The GrandMA2 Full runs about $55,000 with 8192 parameters. To max out the session, you'd need 14 MA NPUs at $4,000 each or another $56,000. That's $111,000.

My rough calculations would show the Max MA set-up could do 579 Spiiders for $191/fixture and the Max Eos set-up could do 266 Spiiders for $187/fixture.



I feel that the Eos Line can do more than folks here want to admit, so I think it's important to really challenge the notion that the console can't do something just because it doesn't do it in the same way one's used to, you know?

Also --- It's worth noting that if you're maxing out your Ion 2K with Spiiders. That's 16+ Spiiders at $5,000/each. That person has the money to expand their console. And if you're pixel mapping LED Tape on any real scale, you aren't using DMX.

 

[danTt]

It really comes down to what you're trying to do:

If you want to control a Spiider from an Eos and have full discrete control over each pixel, you can build a custom profile based off the default M10 profile that breaks each pixel into unique labeled attributes. Then you'd have full manual control over all 123 parameters built within one channel heading. You could type: Chan 1 Red 1 @ 50, etc.

If you wanted to use the color picker or the pixel mapper on the individual pixels, you'd have to make two custom profiles. First would be all the master control, dimming, master color, pan/tilt, etc... Second would be color control for one pixel. Granted, yes, this wouldn't nest nicely. You now would have to dedicated channels 1-20 as your first Spiider. But, at that point, I would make a magic sheet that has my main channel as a tombstone and all the other channels laid out to show the face of the fixture. Now you could click on a "pixel" to select it and then click on the color picker to change the color and see the change on your screen. This patch set-up would allow you to use the built-in pixel mapping to map animations across the face of the fixture(s) with relative ease.

I am in no way saying that it's impossible to use the eos to control spiiders--in fact I do it farily frequently. What I am saying is that the interface is very clunky if we talking about the fancier modes.

Of course I could setup a profile with Red1..Red19, Blue1..Blue19, Green1..Green19, Intens1..Intens19. I could then spend an absurd amount of time mashing my color button on the encoder tabs to reach the appropriate page to control the color for a given pixel, adjust it, hit the color button again, adjust it, etc. That doesn't sound like a reasonable approach does it?

And of course I could patch the fixture as a master channel and a bunch of generic RGB channels. However, this would lose any color calibration that the Eos software may provide for a given fixture. At this point I do not believe the spiiders are calibrated, but what about something like that is? Are we going to end up with ETC shipping "master" and "sub" profiles that are calibrated? That seems like not an ideal solution.

But my question would be, do you really need all that control? Sure, I can imagine wanting it, but then wouldn't I also have abandoned DMX all together? Functionally, the difficulty is not a problem with ETC's console, but a problem with the limitations of the DMX Protocol.

I agree with the idea that a lot of this is limitations of the DMX protocol, and I hope that we see innovations in this area soon, because fixtures are only getting more complicated. sACN/Artnet do not seem to fix the underlying problem, they just hide the limiations of DMX behind a layer of ethernet abstraction. Hopefully we see a move native ethernet based protocol come along, although that does present limiations of requiring star topology/active switching compared to the "just works" passive chaining of DMX. Or maybe we'll see wireless take off in ways we haven't yet comprehended and all of this DMX discussion will be a thing of the past...

In terms of cost:

The max MA session can have only 59,850 parameters. By my quick count, a full Spiider is 113 parameters over 123 addresses (correct that, please). The max Eos can be is 32768. I think a maxed out Eos Ti should run about $50,000. The GrandMA2 Full runs about $55,000 with 8192 parameters. To max out the session, you'd need 14 MA NPUs at $4,000 each or another $56,000. That's $111,000.

My rough calculations would show the Max MA set-up could do 579 Spiiders for $191/fixture and the Max Eos set-up could do 266 Spiiders for $187/fixture.



I feel that the Eos Line can do more than folks here want to admit, so I think it's important to really challenge the notion that the console can't do something just because it doesn't do it in the same way one's used to, you know?

Also --- It's worth noting that if you're maxing out your Ion 2K with Spiiders. That's 16+ Spiiders at $5,000/each. That person has the money to expand their console. And if you're pixel mapping LED Tape on any real scale, you aren't using DMX.

My discussions regarding cost were not designed to try and compare/contrast the MA vs the eos. I think that all console lines are going to need to evaluate their per-address licensing as LED and granularity takes off. Even at a basic level, look at a new auditorium. Say 250 static lights, maybe a few movers. 250 lights. A few years ago we'd be talking about convetional source 4's, using one address a piece. Now we're talking colorsource (5 addresses each) or lustr (10-15 addresses each), and when you compare the cost of dimming + conventionals+A/C distribution to LED + relay+data distribution the numbers come out fairly close cost wise these days. However, we're using five or ten times as many addresses to use our basic system. a 1024 output ion starts looking like a thing of the past fairly quickly, doesn't it?

 

[soundofsparks]

I am in no way saying that it's impossible to use the eos to control spiiders--in fact I do it farily frequently. What I am saying is that the interface is very clunky if we talking about the fancier modes.

Of course I could setup a profile with Red1..Red19, Blue1..Blue19, Green1..Green19, Intens1..Intens19. I could then spend an absurd amount of time mashing my color button on the encoder tabs to reach the appropriate page to control the color for a given pixel, adjust it, hit the color button again, adjust it, etc. That doesn't sound like a reasonable approach does it?

I wouldn't use the encoder tabs like that. In fact I find that for most parameters, it's quicker to key in values:

[Chan][1]{Red 1}[@][50]

or I'd set up a magic sheet layout that gave me the buttons I needed upfront or I'd use the ML controls. If something you're going to use is buried, you put it somewhere else.

And of course I could patch the fixture as a master channel and a bunch of generic RGB channels. However, this would lose any color calibration that the Eos software may provide for a given fixture. At this point I do not believe the spiiders are calibrated, but what about something like that is? Are we going to end up with ETC shipping "master" and "sub" profiles that are calibrated? That seems like not an ideal solution.

So you're saying the generic profile won't match another unit? Probably. How often does unit A match unit B anyway? Does a VL3500 and a Lustr+ inhereantly look the same when you tell them to be "R08"? You use the quick buttons to get to the ballpark, then you tweek with the color functions (+Red, +Brightness, etc.) then save as a color palette so that the look you want is reusable from show to show.

I agree with the idea that a lot of this is limitations of the DMX protocol, and I hope that we see innovations in this area soon, because fixtures are only getting more complicated. sACN/Artnet do not seem to fix the underlying problem, they just hide the limiations of DMX behind a layer of ethernet abstraction. Hopefully we see a move native ethernet based protocol come along, although that does present limiations of requiring star topology/active switching compared to the "just works" passive chaining of DMX. Or maybe we'll see wireless take off in ways we haven't yet comprehended and all of this DMX discussion will be a thing of the past...

Yes. This.

My discussions regarding cost were not designed to try and compare/contrast the MA vs the eos. I think that all console lines are going to need to evaluate their per-address licensing as LED and granularity takes off. Even at a basic level, look at a new auditorium. Say 250 static lights, maybe a few movers. 250 lights. A few years ago we'd be talking about convetional source 4's, using one address a piece. Now we're talking colorsource (5 addresses each) or lustr (10-15 addresses each), and when you compare the cost of dimming + conventionals+A/C distribution to LED + relay+data distribution the numbers come out fairly close cost wise these days. However, we're using five or ten times as many addresses to use our basic system. a 1024 output ion starts looking like a thing of the past fairly quickly, doesn't it?

The point that I'm getting at here is that functionality is related to cost. An Eos Family rig can accommodate the work you want to do if you want to pay to expand it and if you need to pay to expand it, that is because you've already bought expensive equipment to use on it and you probably can afford to expand the rig. You can't expand your fixture inventory without simultaneously expanding your control and back end infrastructure.

If I have a ULD with 6 Source Fours as my rig and then I replace them with 6 VL4Ks, my console needs and back end needs grow in tandem.

The thing I think is really great here is the ease of scalability. You can get up on your feet on a ETC Client with a Gadget for pretty cheap. That can run your ULD, but when you upgrade the fixtures, you got to upgrade the console. But you get to upgrade to a console that you basically already know!

But you get to decide if your space needs a $5000 console or a $50000 console. If you can afford 250 Lustrs, you can afford and Eos Ti.

And I think as a nod to the OP, this is the crux of the question. If you're upgrading your space. You think about the fixtures you want to use and then buy the electrical and data infrastructure first, then a console second, and then you can buy the fixtures you want. There's no point in making a big fixture purchase if you can't run them.

I feel like you keep circling back to the fact that a Ion 1K (or 2K) can't do a big show. It's not supposed to. Yeah, if I have rig of 250 Lustrs on an Ion, that won't work so well. But that's $625,000 in lighting fixtures. If I spent that much, let's just say the cost is $700,000 and throw on an EOS Ti 32K... now I'm good up to 2,400 Lustrs. Of course the real cost is probably $800,000-$1,000,000 because I needed to replace my dimmer racks and run Cat6 everywhere. Now the cost of an EOS Ti just like a shipping surcharge.

 

[OscarMelectrified]

Whoa that's a lot of info I can use! Thanks!
Too much to read right now at work but I'll take in every little bit of advice on this...
We have an Ion 2k and has up to 2,048 addresses of output. With two 2 x 20 Fader wings. I wish to have at least 6 universes for our setup. But 4 will have to be for now.
We are purchasing a new console, will have to wait for an upgrade to the Eos Ti 4k in two years time. So for now I'll have to be VERY conservative on what modes I want to run the (12) Robe Spiiders and (10) DL7S Profiles, plus (40) D60 Selador Desires and (50) S4 Tungsten HD Lustr put together for our Ion.. Just enough to get us what we need in our space, no Pixel Mapping and no need to max out of everything just yet.. In my opinion having the DL7S on Mode 3 'Seven Colours' (which takes 59 DMX channels per unit) and for the Spiiders to be on Mode 1 '3 Zones' (which takes 49 DMX channels per unit) seems very feasible.
I certainly wish when we do get a new console, to beef up our network to NET 3 and CEM3 with new 4 port network nodes, it definitely opens up a whole new opportunity for a much broader system.

 

[sdauditorium]

We're still using conventionals for front light but all other systems (tips, tops, cyc) were replaced a few years ago with D40s, either Vivid or Lustr+. We opted for D40s for the cyclorama instead of a linear array because the PAR form factor is more versatile for use elsewhere and it is nice to have control over the focus and lensing of each cell. We can do the cheesy vertical rainbow stripes, or get 2 or 3 horizon lines for naturalistic skies depending on the needs of the show.

Recently we added some used Clay Paky Alpha Profiles. I considered getting S4 LED2 fixtures but the used Alphas were actually cheaper and far more feature rich. Lamp replacement costs are about the same as scroller replacement costs they displaced. They should get us through the next 5 years as the LED luminaire market continues to evolve.

When not using for cyc washing, what are your throw distances when used as top light and how effective/bright do you find the D40s? I'm waffling between the 40 v. 60 with 18-21 trim heights on our electrics.

 

[sk8rsdad]

We have D40s for tops independent of the D40s used by the cyc, zoned 5x3 on a 33' x 30' stage. Trim height is typically 21-25 ft lensed with what used to be called 45 degree lenses. They do their job with no complaints. Get a demo and decide.

 

[OscarMelectrified]

I found that the D60 Lustr+ we purchased for our trim set at 21'6'' was lacking in beam spread over the entire stage. I hung 8 fixtures across a 40' truss batten. They came with Narrow lenses and I must get the different type of lens to widen that spread. Otherwise you will notice the obvious gaps between the lights when looking down at the stage.

 

[SteveB]

I found that the D60 Lustr+ we purchased for our trim set at 21'6'' was lacking in beam spread over the entire stage. I hung 8 fixtures across a 40' truss batten. They came with Narrow lenses and I must get the different type of lens to widen that spread. Otherwise you will notice the obvious gaps between the lights when looking down at the stage.

Figure out the spread of whatever to used to do these washes and match with the appropriate lenses from ETC ?