split colour fade

[Jason Brameld]

So I had an interetsting request for a Fringe show yesterday that I couldn't figure immediately so can I pick your brains to see if there's a better way than the one I subsequently came up with?

The requirement was for a split colour fade on an RGB fixture as follows:

1. <GO> Start with general FOH wash (tungsten) 
2. <GO> Add in red from side LED fixtures on a 5 minute fade up time (so far so good....)
3. <GO> Start a 5 minute fade out of the reds (still easy.....) 
4. <AUTO 1 min after start of Q3>, start a 5 minute fade in of blues on the same fixtures as the reds (accepting that this will pass through various lovely shades of purple on the way (now I'm stuck.....red fade would then finish 1 min before the blues did).

Eventually we compromised on a 6 minute colour fade between red and blue and that worked fine.

The only workaround I came up with in my head afterwards was:

1. <GO> Start with general FOH wash (tungsten)
2. <GO> Add in red from side LED fixtures on a 5 minute fade up time
3. <GO> Start a 1 minute colour fade of the red leds to 80% of full value (lose 1/5 of value in 1/5 of the fade time)
4. <AUTO after end of Q3>Start a 4 minute colour crossfade to take reds out and bring blues to 80% of full value (bring in 4/5 of blues in 4/5 of the time - take out remaining 4/5 of reds)
5. <AUTO after end of Q4) Start a 1 minute colour fade of blues from 80% to 100% of full value

Leaving fixture intensity at fixed level through Q3-5.

Is there any smarter way of doing that?

Thanks,

Jason.

 

 

 

 

[kgallen]

10 minutes ago, Jason Brameld said:

4. <AUTO 1 min after start of Q3>, start a 5 minute fade in of blues on the same fixtures as the reds (accepting that this will pass through various lovely shades of purple on the way (now I'm stuck.....red fade would then finish 1 min before the blues did).

Q: So on the last part your timeline mismatch by 1 minute as you say. In what way were you wanting to make them align?

If you wanted to keep the red and blue independent, then on your red fade out you would only tag the red colour and for the blue fade up only tag the blue colour. I believe this would require the desk to be set "keep parameters separate" for colour (that's probably the default). In both cues the intensity is retained at 100% so the fading of the colours is via the colour channel not the intensity channel.

1. I've probably misunderstood your requirements!

2. Yes I'd struggle to do this too!

3. Edward will give us the proper recipe in a minute!

[Jason Brameld]

54 minutes ago, kgallen said:

Q: So on the last part your timeline mismatch by 1 minute as you say. In what way were you wanting to make them align?

That's part of the problem - they don't time align - it's a 5 minute fade out of the reds, a 5 minute fade in of the blues, and a minute gap between the starts of the 5-minute fades. 

so:

At time T+0: Start 5 minute fade down of reds

@T+1 min - start 5 minute fade up of blues

@T+5 mins - red fade out complete

@T+6 mins - Blue fade in complete

This was the subject of much confusion as it was being asked for by the designer (who wasn't very good at explaining things) as well!!

Jason.

 

[kgallen]

What did the designer want the output to look like? Surely they realise the blue and red fades won't finish together if they are of equal duration but a minute skewed! So they either have 1 minute of blue fade alone, or the blue fade is reduced to 4 minutes or...

I sympathise with you! But you can't programme a requirement when the requester isn't even clear on what they want!

I have similar experiences. Some directors are good at visualising what they want to achieve, others aren't. Others don't have a mental image of what they want a scene to look like.

[Jason Brameld]

He knew what he wanted, which was non-time-aligned fades that didn't finish together - hence it was tricky - but he was rubbish at explaining it.  I had programmed it completely wrong first time based on my initial understanding of his confused explanation!

Anyway - hopefully Ed will be along with some sanity soon.....

Jason.

[Edward Z88]

Hi Jason and Kevin,

2 hours ago, Jason Brameld said:

This was the subject of much confusion as it was being asked for by the designer (who wasn't very good at explaining things) as well!!

Ultimately, it sounds like the designer was trying to explain what he wanted as if there were red PARs and blue PARs, not taking into account that for you, a red PAR and blue PAR are the same fixture when they're LED!

3 hours ago, Jason Brameld said:

<AUTO 1 min after start of Q3>, start a 5 minute fade in of blues on the same fixtures as the reds (accepting that this will pass through various lovely shades of purple on the way (now I'm stuck.....red fade would then finish 1 min before the blues did).

Maybe I have misunderstood this, but it sounds to me this should look like this...

Your reds are up in cue 1.

Your reds start to fade down in cue 2.

After a minute, cue 3 fades your lights back up again, taking 5 minutes to get to their intensity, whilst they crossfade from red to blue, also over 5 minutes. Remember as far as the console is concerned, there are no lights fading down in cue 3. In terms of your output window, you will see them dip to 80% over a minute, and then raise to full over 5 minutes.

If after a minute of the reds fading down, you needed the fixtures to halt at 80% for 4 minutes, then raise to 100% over a minute, it could have looked like this...

Your reds are up in cue 1.

Reds fade down to 80% in cue 2 over a minute.

Cue 3 runs once the fixtures are at 80%, and the fixtures then start to change to blue over 5 minutes, however remain at 80% for 4 minutes, due to cue 3's delay time. They then raise back to full over a minute.

3 hours ago, kgallen said:

If you wanted to keep the red and blue independent, then on your red fade out you would only tag the red colour and for the blue fade up only tag the blue colour. I believe this would require the desk to be set "keep parameters separate" for colour (that's probably the default). In both cues the intensity is retained at 100% so the fading of the colours is via the colour channel not the intensity channel.

Unless I have misunderstood, I therefore don't believe this is down to tagging. Worth mentioning, "Keep Parameters Separate" is set to "No" for colour, to ensure that with SmartTag disabled, if you move a colour encoder, all colour parameters will be tagged, to ensure all parameters are stored, equating to your desired colour.

I hope this helps and makes sense, if you have any questions let me know.

Edward

 

[Jason Brameld]

Hi Ed,

Your summary of the problem about separate PARS but in the same fixture is spot on!

This is what we want to achieve:

I think your first example is closer to the mark. - I may give this a try as an exercise

So in your cue 2 - that is the fixture intensity at zero and the red colour remaining at 100% (G and B at 0)

In your cue 3, fixture intensity 100%, R and G at 0, B at 100%

So the question in my head on this one is would the colour crossafade in Q3 effectively slow down the red fade out in Q2, as the red fade down should be competed 5 minutes after the GO on cue 2, and not at the end of the 5 minute colour crossfade started when Q3 starts (If that makes sense?!)

Also if you have the intensity dip and the colour crossfade do you end up with the brightness dropping more in the middle doing it this way?

I'll have to go and play tonight now to satisfy my curiousity!

Jason.

[Edward Z88]

Hi Jason,

39 minutes ago, Jason Brameld said:

This is what we want to achieve:

Thanks for the graph, that is great! Bet you wish your designer could have shown you that at the start!

In that case, we'll be looking at this...

In cue 1, your red lights are up.

In cue 2, your red lights are fading down to 0% over 5 minutes. No colour information included, therefore no colour times are shown. [Minute 0 in your graph]

In cue 3, your lights then begin the 5 minute cross fade from red to blue, which begins 1 minute after the reds started fading. There is no intensity information included, therefore no fade up/down times shown. [Minute 1 in your graph].

In cue 4, your lights then begin to fade back up to full over 3 minutes, 2 minutes after they started to change to blue. There is no colour information included, therefore no colour fades are shown. [Minute 3 in your graph].

Result once cue 4 has completed - red to blue fade ends at the same time as blue hitting 100%. [Minute 6 in your graph].

39 minutes ago, Jason Brameld said:

Also if you have the intensity dip and the colour crossfade do you end up with the brightness dropping more in the middle doing it this way?

In theory the intensity of the colour will remain the same (assuming the red and blue LEDs are equally intense, which is unlikely), as you will go from one LED colour at full, to both LED colours at half (equaling the same intensity as one LED colour at full), to the other LED colour at full.

In your graph though of course, you are dipping to 40% of the overall intensity mid fade.

Hope this helps,

Edward

[Jason Brameld]

Ah - but the red fade out should finish at minute 5, and not minute 6.....leaving the blues to continue from 80% to 100% in the last minute.

The actual time they spend cross-fading is only 4 minutes - in the first minute at the beginning only the red is coming down and in the last minute at the end only the blue is going up.

Good brain teaser huh?

[Edward Z88]

Hi Jason,

32 minutes ago, Jason Brameld said:

Ah - but the red fade out should finish at minute 5, and not minute 6.....leaving the blues to continue from 80% to 100% in the last minute.

Good brain teaser huh?

The intensity of red, as shown in your graph, will have completed its fade to 0 at minute 5. This is true of cue 2 in my image above, due to the 5 minute fade down. Red fading down between minute 3 and 5 is theoretical in this instance, as the intensity will have begun to rise again at minute 3 (cue 4 starts).

However when red is at "0", there is still a minute of colour mixing left to complete, due to the 5 minute colour fade in my cue 2. If the colour needs to be full blue at minute 5, then simply change the colour fade in my cue 3 from 5 minutes, to 4 minutes. As cue 4 is set to "Auto With", not "Auto After", changing the timings of cue 3 will not affect cue 4.

This would then give in my example:

• In cue 1, your red lights are up.
• In cue 2, your red lights are fading down to 0% over 5 minutes. No colour information included, therefore no colour times are shown. [Minute 0 in your graph]
• In cue 3, your lights then begin the 4 minute cross fade from red to blue, which begins 1 minute after the reds started fading. There is no intensity information included, therefore no fade up/down times shown. [Minute 1 in your graph].
• In cue 4, your lights then begin to fade back up to full over 3 minutes, 2 minutes after they started to change to blue. This starts when they are magenta - 50% red and 50% blue, mid colour cross fade. There is no colour information included, therefore no colour fades are shown. [Minute 3 in your graph].
• Result once cue 4 has completed - a minute after completing the fade to blue, the fixtures hits 100%. [Minute 6 in your graph].

Hope this helps,

Edward

[Jason Brameld]

OK - Thanks - I'll have a play.  Given that we seem to need 3 cues to do this by your way or mine, would my way described at the top of the thread also work?

Thanks!

Jason.

[Edward Z88]

Hi Jason,

29 minutes ago, Jason Brameld said:

Given that we seem to need 3 cues to do this by your way or mine,

Yes 3 cues is easiest here, as we turn lights off, change their colour, then turn them on again. You could probably get rid of “Cue 3” in my latest example, and also have the colour change from red to blue in cue 2, with colour having a 4 minute fade, and a 1 minute delay. My cue 4, would then need an Auto With wait time of 3 minutes. In many ways though, it is easier separating it out into separate cues for each part. 

6 hours ago, Jason Brameld said:

1. <GO> Start with general FOH wash (tungsten)
2. <GO> Add in red from side LED fixtures on a 5 minute fade up time
3. <GO> Start a 1 minute colour fade of the red leds to 80% of full value (lose 1/5 of value in 1/5 of the fade time)
4. <AUTO after end of Q3>Start a 4 minute colour crossfade to take reds out and bring blues to 80% of full value (bring in 4/5 of blues in 4/5 of the time - take out remaining 4/5 of reds)
5. <AUTO after end of Q4) Start a 1 minute colour fade of blues from 80% to 100% of full value

Leaving fixture intensity at fixed level through Q3-5.

This works, however the fixtures will be halted at 80% for the duration of the colour fade, rather than dipping in intensity, however this may be what you’re after.

Edward

[Jason Brameld]

This has been a great thought experiment for me.  Thanks for all your input Ed.  I was after the straight line performance in the graph without a significant intensity dip.  Here are some more graphs:

If the intensiity values are run on an HTP basis with a 5 minute down at minute 0 and a 3-minute up at minute 3, then I think you get what's shown on the intensity graph below

If you then add in a 4-minute colour crossfade, and multiply the colour values by the intensity values to arrive at the fixture output for each colour then you get (with a slight approximation on the resultant intensity values) the output in the second graph below. The actual output values are not very linear and there is quite a dip in overall brightness at the 3 minute point.  The eye doesn't really see the non-linerity, but the brightness dip is noticeable.

Your last post above however did cause me to and figure out how to add a delay to a fade on a cue which i hadn't done before - very handy to know.

Lastly, in programming your cue sequence I couldn't use Smart Tag for all the cues as cue 2 has the fixtures 'off' and cue 3 has no brightness information in so to program the color fade I had to use 'selected fixtures' to program that cue otherwise it didn't take any information into it - would that be right?

Thanks again for the brain exercise!

Jason

 

[Edward Z88]

Hi Jason,

Here's another thought - we could just look at doing this with colour fades, and no intensity changes. This would mean we'd need to be careful with tagging as Kevin mentioned earlier.

The cue stack would look like this...

In cue 1, my fixtures are at full, in red.

In cue 2, my fixtures are at full, in black. You can create black quickly by using the slider in the colour picker window to pull all colour mixing down. When cue 2 is triggered, I therefore have a 5 minute fade out of red.

In cue 3, my fixtures fade up to full blue, a minute after cue 2 was triggered. The key however, is ensuring that only blue is tagged in cue 3. You can do this by creating blue, but then tapping on the "Red" and "Green" parameter displays so they have a dark blue background rather than light blue, which will display "Untag [Red]" and "Untag [Green]" in the command line. If red is tagged, it will not continue being controlled by cue 2, and so its fade out would slow, so that red would be out at the same time you reach full blue.

11 hours ago, Jason Brameld said:

Lastly, in programming your cue sequence I couldn't use Smart Tag for all the cues as cue 2 has the fixtures 'off' and cue 3 has no brightness information in

If SmartTag is enabled, the console decides what gets stored. When SmartTag is disabled, you decide. Therefore for all of this, I have had SmartTag disabled, as I need to be specific about what does and what doesn't get included.

I hope this helps, let me know your thoughts.

Edward

 

[Jason Brameld]

Hi Ed,

That all makes sense, and using colour fades only was my method as described in my original post - I was just using changes to colour values (no change to fixture intensity throughout) so my sequence was:

1. Full reds to start (red at 100%, intensity at 100%)
2. Fade red colour values 100% to 80% over 1 minute (minute 1)
3. Crossfade red 80%-0 and blue 0-80% over 4 minutes (start at end of Q2) (minutes 2-5)
4. Fade blue color values 80%-100% over 1 minute (starting at the end of Q3) (minute 6)

I could do all of this with Smart Tag on and no tagging/untagging of different colours, but I had to do the maths (admittedly not difficult!) to split the fades up, and your method can do it with one fewer cues.

There's always more than one way to skin a cat!

Thanks again for the excellent contribution to the debate.

[Edward Z88]

Hi Jason,

1 hour ago, Jason Brameld said:

There's always more than one way to skin a cat!

Thanks again for the excellent contribution to the debate.

Yes there certainly is! No problem at all!

Edward