Using DP live

[Shooshie]

I've had a lot of experience using Macs live in concert with DP. Unfortunately, I have not used VI's in concert. That part of my career came to a close before VI's were reliable enough and easy enough to use, but toward the end I was making plans to try VI's mixed in with live instruments and see if I could get away with it. But I never really got to try that with an audience.

Here was my thinking, though: for live performance, computers are scary. The tiniest touch of a mouse or key can make everything go in a different direction, and before you can fix it, it's way too late to avoid interrupting the song. So, you want to simplify. Plan your sets, reduce your CPU usage down to recorded stuff where possible. It can be MIDI recordings if the instruments are reliable enough, or audio if MIDI would lead to complex situations that can go south in a hurry. Rather than use heavy plugins, bounce all that to final two-track form so that you are playing straight audio.

While redundancy is essential for live setups (unless you just like to gamble and want to feel what it's like when you eventually lose), even redundancy does not guarantee a seamless patch over a glitch. Most likely by the time you realize you have a problem, it will take time to switch to the redundant system and sync it to the place you were at when the glitch occurred, so it's best to have a contingency plan: prepare your patter with the audience as part of your show, with jokes about technology, or about the fictitious character "backstage" who is running around trying to keep things together while in reality your are flipping some switches and clicking a mouse. "Joe, the Sound Man, keeps tripping over his beard back there. I've been telling him to trim it. 'Joe,' I says; 'that ZZ Top beard is gonna get you in trouble, and when it does, I'm docking your pay.' Well, that didn't work, so now I'm gonna dock his Jack Daniels…" You get the idea.

Preparation, practice, redundancy, simplification, and make lists of what you must do before every show, so you don't forget under pressure with 20 people shouting your name constantly before curtain. The music director running a show with a computer is the focal point of everything that happens in the show behind the scenes, so things get pretty hectic after a day of loading in, sound checks, final preps, crew calls, curtain, and so forth. It takes a cool head and quick thinking. Anything you can do to make your job easier and more reliable, by all means do it!

Shooshie

[mjmoody]

Thanks!

I'm particularly interested in finding ways to repeat a section on the fly and to cue up a section - but to have it come in on the downbeat of a measure once you decide you want to go there. In other words it would finish the current measure and/or section before jumping to the selected marker, chunk, or section.

Is there a way to do that?

John

[Shooshie]

Thanks!

I'm particularly interested in finding ways to repeat a section on the fly and to cue up a section - but to have it come in on the downbeat of a measure once you decide you want to go there. In other words it would finish the current measure and/or section before jumping to the selected marker, chunk, or section.

Is there a way to do that?

John

I'm not sure. It depends on exactly what you expect it to do, and what you think you can get away with.

For example, it's possible to queue up Chunks and use Chunk Chaining for a set list, and it would be possible to re-order those Chunks in real time, but there is always a small pause between chunks. It's not seamless.

You can use Markers to jump to different sections, but to my knowledge there is not a "jump to marker after current measure is complete" command. It would be possible to set up a repeat loop (Memory loop) so that at the end of a section or bar, you would jump back to a previous section. It used to be possible to jump ahead that way, but that trick did not survive the programming transition from Classic Mac to OSX to Cocoa. But setting up repeats in real time while playing live in front of an audience? Not me! I wouldn't even attempt it unless there are some helpful commands in the COMMANDS Window of which I'm unaware.

Another way to do it might be to use multiple rigs. You can be cueing one rig while playing on another, and then at the appropriate moment -- they're time synced, of course -- you just switch audio from one to the other. This would take a special switcher. Mike McKnight mentions that he had one custom made with 72 inputs and 24 outputs -- enough for three full 24-channel feeds to be switched in or out in their entireties without a hiccup. See the interview with Mike here:
http://www.motu.com/newsitems/atnewsite ... =Interview" onclick="window.open(this.href);return false;

You might find some of the other interviews interesting. Type "Interview" into the search field on the MOTU page, and you'll get a list of interviews with various artists who use DP. Well… here it is:
http://www.motu.com/search?SearchableText=Interview" onclick="window.open(this.href);return false;

It all depends on your motivation, your budget, and what is acceptable to you. There may be other ways, but I'm for keeping things simple. If I were going to try it, though, I'd be looking for some way -- maybe via a Custom Console (an often neglected DP feature, though it doesn't always do exactly what you need) -- to incorporate a foot-pedal unit to send something like patch changes or other controller information that would set the location and cue the event, awaiting your final tap on the enter key, or something like that. Something where a few options have already been programmed for easy access through a pedal, so that you're not fumbling around with transport controls and repeat markers on stage.

I'm not saying you'll succeed in finding such a method; only that this is the approach I'd take, and I can't say there isn't any such method, because I have not tried it. But always, go for simplicity. Things happen fast and in a blur on stage, and too much fiddling with controls will eventually lead to chaos of the sort that makes you look like you don't have a clue what you're doing.

Shooshie

[SixStringGeek]

For example, it's possible to queue up Chunks and use Chunk Chaining for a set list, and it would be possible to re-order those Chunks in real time, but there is always a small pause between chunks. It's not seamless.

This is THE feature in Vision I miss the most. Every chunk or sequence had a single letter identifier and you could just type some letters while stuff was playing and those sequences got added to the end of the current play queue. Playback continued seamlessly. I used to use this to do two or three different song transitions and then play with organizing them into different orders while building the song structure.

That would be a nifty feature to add to DP in the next release.

[Umbrella]

Still, there are some basics that it would be nice to have some tutorials on. Here are some situations that are common in live situations that I don't know how to do in DP:

1. Be able to vamp a section at will - then be able to jump to a new section at a LOGICAL point - for instance the ending of a section, or measure. I have found ways to instantaneously jump to new sections in DP - but what good is that live?? The millisecond I press the command to go to section B it goes to section B - no matter where I am in the form or measure.

2. The ability to loop more than one section of playback. This would include a glitch-free method to disable the loop (memory-cycle) function whenever it has been decided to go on to the next section.

Being able to line up all the chunks is great. A complete set list is right there - but I can't see how you can do anything creative on the fly except to just play back the sequence. I know that DP has been used to run countless live shows, but the interviews I have seen involve two Macs and a person dedicated to syncing with whatever is going on on-stage at the moment. Wonder if you are just a keyboard player, or drummer, trying to do something spontaneous (like a vamp, or maybe one more chorus for a jazz solo) and you are controlling the sequence with a foot controller or something - while keeping track of the form and keeping playing something? I can't see how to do it. There is nothing in the manual that I can find, and beyond testimonials of how wonderful DP is in a live performance setting - I can't find any answers on how to really do anything in a live context.

So, I'm with Prime Mover on this - it would be helpful to have someone who uses DP in a live context to give a few brief tutorials - maybe even as part of this thread. . .

John

One method that might accomplish what you're describing in #1 if you wish to trigger-switch from one main looped section to another is to use the "chain chunks" command (pg. 47) to "jump", as you put it, from one section (or sequence, technically speaking) to another at a specified end point. Since DP is not optimized for real-time seamless chunk chaining, busy or CPU taxing sequences might incur unworkable gaps between them. Actually I haven't used this feature in a long time, but I seem to remember that in the old days on ancient Mac Quadras and Performas, DP could handle MIDI chunk chaining no problem; as I remember it, having audio in a sequence would throw off the timing of a transition. One would think that this is a non-issue for today's computers.

For #2, there are a couple of options within DP that can get you close... For MIDI looping control, check out the Echo and Arpeggiator MIDI plugins - pg. 816 - the "As Played" and "Hold Notes" arpeggiator settings pg. 819 should give you something to work with. As such I think you'd need to set up a custom console to handle the toggling between one triggered phrase/loop and another - there are multiple ways to approach this. Also check out "Cycle through device group assignments" pg. 818 for more ideas... For looped audio, Polar can almost do what you're asking with the "Toggle Group Play Enable" command on pg. 263. Unfortunately Polar doesn't offer an easy way to "chain" (as it were) its passes, afaik, but you can certainly trigger them (mute or unmute) via MIDI controller.

It would be nice if DP had some functionality similar to Logic's "touch-tracks" environment object.

One thing I discovered just now in looking through the manual is that DP's MIDI plugins are in fact an open format and that MOTU offers an SDK for interested developers!@ (pg. 535)

[Shooshie]

For example, it's possible to queue up Chunks and use Chunk Chaining for a set list, and it would be possible to re-order those Chunks in real time, but there is always a small pause between chunks. It's not seamless.

This is THE feature in Vision I miss the most. Every chunk or sequence had a single letter identifier and you could just type some letters while stuff was playing and those sequences got added to the end of the current play queue. Playback continued seamlessly. I used to use this to do two or three different song transitions and then play with organizing them into different orders while building the song structure.

That would be a nifty feature to add to DP in the next release.

No wonder everyone liked Vision so much. That indeed would have been handy. Like the Song window, only live.

DP has the ability to switch chunks on the fly using nothing more than a letter typed in. At least I assume that's still there. I haven't looked at it in quite some time. There is a preference for it somewhere, but I've even forgotten where THAT is. If the preference is selected, it will automatically assign a letter to every new chunk that is created. But then if you touch that letter on your keyboard, it'll switch chunks! I left that off except for about a year of a particular live performance when I found it useful. Anyway, it's not like Vision's. It doesn't give a seamless swap, nor does typing the letter queue up the next chunk, but just switches to it. Or does it? It's been 15 years since I last used that. Seems like if you type the letter while it's playing, nothing happens, but if you type it while stopped, it switches chunks.

This is something that would be worth writing MOTU about. It would be great if they could create a feature that allowed seamless chunk-switching at your choice of, say, "at the next marker," or at the end of the current chunk (when you have the manual end time set), or "on cue from source xyz" where xyz is a pre-determined note, controller, patch change, etc. And with V-Racks keeping all the instruments "live," why wouldn't that be possible? It should work without a hitch, with only a little programming to look ahead and eliminate the pause between chunks.

Shooshie

[mjmoody]

Another way to do it might be to use multiple rigs. You can be cueing one rig while playing on another, and then at the appropriate moment -- they're time synced, of course -- you just switch audio from one to the other. This would take a special switcher. Mike McKnight mentions that he had one custom made with 72 inputs and 24 outputs -- enough for three full 24-channel feeds to be switched in or out in their entireties without a hiccup. See the interview with Mike here:
http://www.motu.com/newsitems/atnewsite ... =Interview" onclick="window.open(this.href);return false;

You might find some of the other interviews interesting. Type "Interview" into the search field on the MOTU page, and you'll get a list of interviews with various artists who use DP. Well… here it is:
http://www.motu.com/search?SearchableText=Interview" onclick="window.open(this.href);return false;

Shooshie

I know it's been three years since anyone posted to this thread - but I still find it interesting! I've been using DP to provide backing tracks to a Show Choir. I would just make the chunks line up for the show - it has been great. Up until a few days ago I would just hook up the stereo outputs of my Ultralite audio interface into two open channels of the mixer we were using. As of two days ago, however, we just got a digital mixer (Presonus), where I can actually use DP for the singers' voices. I am pretty excited about this. I am hoping to get into some live digital effects, etc. using DP.

So, I started by adding vocal tracks to all my backing tracks we were using. This is fine, except there is a noticeable sound gap between each chunk, where DP is changing chunks. This also occurs when I accompany the group on my piano - which is just a MIDI controller, controlling various VIs. I've learned to stop "noodling" around on the piano between chunks, but part of the "show" features patter by the members of the show choir between songs. Having the vocal parts hi-cup between each chunk would mess up any chance of patter, it seems.

I reread the interview with Mike McKnight, and I was struck that he said he was running DP 45 minutes straight. So, that's several songs worth of time! It almost sounds like rather than lining up different sequences as "chunks", he just has 6 or 7 chunks that covers the whole show - but these are LONG stretches of time. Here is a quote from the article:

"What happens when you start the show?

I send timecode to lights and video, and once we start it runs for 45 minutes straight, non-stop. Then there are about seven or eight individual songs, then finally another stretch of 30 minutes or so to close the show.

I see Digital Performer on the screen there. Is that your show file?

Yes, let me show you. As you can see, there are tons of tracks. In addition to the backing tracks, I've got everybody's part for the entire set. In case somebody in the band gets sick, I can just fly in their part. This file is now very consolidated. For a while, the show was just enormous. All the stuff I had to have in here – I had to keep paring it down."

It seems to me that in order to get DP to trigger any special vocal effects, the actual vocal parts need to be in the sequence being monitored - then the effects can be triggered via automation. This would be opposed to, for instance, keeping the vocal tracks live via Aux tracks. If they were always "on" via aux tracks, I don't think the vocal effects would work - but I don't know.

I guess I've always been assuming these big shows rely on cuing up various DP Chunks, but I wonder if, instead, they are just a few chunks that represent each "set". If they DO use the different chunks, how do people keep the vocal and instrument parts sounding live - without little "gaps" in between each chunk?

John

[Shooshie]

No time for a detailed response, but my guess is that live effects are done FOH. McKnight probably outputs to a big bus backstage which goes FOH to the sound guy, who does the actual balancing of the show. McKnight is under the stage, or on stage, either of which is impossible for doing sound. He just provides the basic tracks. The sound guy does the… well… the sound!

Thye 45 minute chunk is probably one chunk. It may be several chunks in a Song, but not likely. Chances are, there's enough live music going on that he can chain chunks and nobody will hear any gap between songs. Plus, the applause covers a lot.

Shoosh

[mjmoody]

Thanks! What's FOH?

[James Steele]

Thanks! What's FOH?

Front Of House.

[mjmoody]

Thye 45 minute chunk is probably one chunk. It may be several chunks in a Song, but not likely. Chances are, there's enough live music going on that he can chain chunks and nobody will hear any gap between songs. Plus, the applause covers a lot.

Shoosh

One chunk vs. chaining several chunks implies very different things. I'm thinking it might be better to have several different tunes - a set, maybe - in one sequence. Markers could be set for different songs, tempos could be adjusted with the conductor track, and patch changes could be specified as needed.

Perhaps a quick way to do it would be to create a "song", that would basically be the set - then merge the song into a sequence. As long as markers still indicate where different songs are in the sequence, some flexibility could still be had. The sequence could "pause" at various points - for instance, after a song - then one could jump to the marker of the next song, which could be just a count-off for the upcoming song. None of this would disengage the mics. like jumping to a new chunk, however. I'll experiment with it tomorrow.

John

[Shooshie]

Years of performing live with DP pretty much cemented a style of working for me. When the music was continuous, I used one chunk. When it stopped, I used separate chunks for sending light cues to set up the stage for patter with the audience, moving around, and so forth. Then the light cue for the next song would be the first thing in the next musical chunk. Those separate chunks gave me flexibility with the time. If the audience went long with applause, I merely waited before the next light cue. If the star wanted more time to talk, no problem. We both knew the words that would trigger my next move. I had all those verbal cues written into the comments field beside the chunks.

Of course, all lighting within the music was cued from the musical sequence.

The chunks were chained, but only to cue up the next chunk and stop, waiting for me to hit play. Woe unto him who hits Play before its time… the lights start changing, and you'll be lucky to get it stopped before the music comes in. Then you have to go back to the previous lighting chunk and resend the cue for the current patter or monologue segment. It only takes once to learn to keep your hands far away from the keyboard until it's actually time to begin.

When you add effects and automation, you make your file more complex. I'm not saying not to do that, but do it sparingly; only what they can't do front-of-house. Keep it simple. A simple file will never, never misfire or freeze up or anything. In 20 years of using DP on stage, it never failed. Not once. It rarely crashed even when working, composing and arranging music, adding light cues with SysEx, or any of the many other ways of doing cues. At about the 15 year marker I actually looked back and counted the crashes. There were less than 5. (I guess that means 3 or 4, and at least one of those was caused by a lightning storm that cause power fluctuations, and another was caused by a faulty IDE cable.)

The box that McKnight uses to switch rigs at the instant a signal stops sounding from his main rig can now be bought, or at least one similar to it. I think he had to have his made custom, but they can be had now. They monitor a tone piped through a bus in DP, and the instant that tone stops, they switch dozens of inputs and outputs; the whole rig, really, and suddenly you're playing on the backup computer, which has been synced to the main one up to that point.

I wish I had such a rig when I was on stage. I had a backup rig, but I had to perform the switch manually. It only took a few seconds, but that's an interruption. Fortunately, in all those years I never had to use it. It just didn't fail.

Shooshie

[Sergievsky]

I'm a little disappointed that Motu hasn't incorporated the "jump to marker after current measure is complete" command that was referred to earlier in this thread. This is what makes Ableton Live so invaluable. You can kinda make DP do this when Live is ReWired to DP, but it still skips. I messed around with the buffers and such and managed to minimize the interruption, but then it got unstable. Hopefully they can manage to squeeze this in...someday

[mjmoody]

Years of performing live with DP pretty much cemented a style of working for me. When the music was continuous, I used one chunk. When it stopped, I used separate chunks for sending light cues to set up the stage for patter with the audience, moving around, and so forth. Then the light cue for the next song would be the first thing in the next musical chunk. Those separate chunks gave me flexibility with the time. If the audience went long with applause, I merely waited before the next light cue. If the star wanted more time to talk, no problem. We both knew the words that would trigger my next move. I had all those verbal cues written into the comments field beside the chunks.
Shooshie

Shooshie, thank you so much for these words of wisdom! Motunation continually amazes me with all the talent and experience continually exhibited at this site!!

So, I'm sorry for being so dense, but there are some logistics I'm just not figuring out (and my show choir has a big gig tomorrow). Now that we have the Digital Mixer, and I can actually use it more as an audio interface with DP, I'm trying to figure out what to do with the live vocals. Do I make them activated tracks within DP, or do I turn off the Digital "link" with them, and just run the vocals through the mixer (and have the mixer act like a mixer and not a digital audio interface)? I have been thinking I would like to add some effects that DP can provide directly to a live show - however, when the chunks change, the vocal sound is interrupted for a moment.

In these big shows where DP is involved in lighting and such, are the vocals actually going through DP too, or are they just going directly in a mixer, and DP is running backing tracks and click tracks, and the like? If the vocal tracks are going through DP, then how do people deal with the momentary interruption in sound as a new chunk is cued up?

I experimented with making the vocal tracks Aux Tracks, but then, I couldn't figure out how to do any "track specific" things with them. So, I tried having Aux vocal tracks AND vocal tracks activated within the chunks - but, then each voice sounded like two voices. In today's practice for the gig, I just made one long track with markers for the different songs - so, it is taking up one chunk, and everything is going through DP, including vocals. I just think there's probably some other way of doing it that would also incorporate using the different chunks for different songs.

Thanks!

John

[Shooshie]

Years of performing live with DP pretty much cemented a style of working for me. When the music was continuous, I used one chunk. When it stopped, I used separate chunks for sending light cues to set up the stage for patter with the audience, moving around, and so forth. Then the light cue for the next song would be the first thing in the next musical chunk. Those separate chunks gave me flexibility with the time. If the audience went long with applause, I merely waited before the next light cue. If the star wanted more time to talk, no problem. We both knew the words that would trigger my next move. I had all those verbal cues written into the comments field beside the chunks.
Shooshie

Shooshie, thank you so much for these words of wisdom! Motunation continually amazes me with all the talent and experience continually exhibited at this site!!

So, I'm sorry for being so dense, but there are some logistics I'm just not figuring out (and my show choir has a big gig tomorrow). Now that we have the Digital Mixer, and I can actually use it more as an audio interface with DP, I'm trying to figure out what to do with the live vocals. Do I make them activated tracks within DP, or do I turn off the Digital "link" with them, and just run the vocals through the mixer (and have the mixer act like a mixer and not a digital audio interface)? I have been thinking I would like to add some effects that DP can provide directly to a live show - however, when the chunks change, the vocal sound is interrupted for a moment.

In these big shows where DP is involved in lighting and such, are the vocals actually going through DP too, or are they just going directly in a mixer, and DP is running backing tracks and click tracks, and the like? If the vocal tracks are going through DP, then how do people deal with the momentary interruption in sound as a new chunk is cued up?

I experimented with making the vocal tracks Aux Tracks, but then, I couldn't figure out how to do any "track specific" things with them. So, I tried having Aux vocal tracks AND vocal tracks activated within the chunks - but, then each voice sounded like two voices. In today's practice for the gig, I just made one long track with markers for the different songs - so, it is taking up one chunk, and everything is going through DP, including vocals. I just think there's probably some other way of doing it that would also incorporate using the different chunks for different songs.

Thanks!

John

Hi John,

Personally, I'd run the choir through a mixer and not try to use DP as a live mixer. It's there to provide what the live musicians can't. A front-of-house mixer is the best place to do effects for the choir. DP should also be running through that FOH mixer. The sound man will adjust all sound -- the chorus, the soloists, and anything coming out of DP -- for the best blend and/or emphasis.

In a pinch, I guess you could use DP in that regard, but I think it's taking a risk. If you do choose to use DP for live vocal effects, I suggest using MOTU's built-in plugins, such as the graphic EQs. Avoid look-ahead plugins, because they'll cause too much latency. On a modern Mac Pro, you'll not want a buffer to exceed 128 for live sound. Hopefully, the hall will provide acoustic reverb so you won't have to worry about that.

Still, I'd go with FOH effects rather than routing live music through DP. But if that's not possible, and if you're happy with how it's working out, disregard me and just do it!

By the way, if you have spots in which the chorus sings without DP, and you need to keep them "on," set up cues for your conducting, and use the SHIFT command (Command-L) to move them forward or back until they make rhythmic sense for you. For some reason, perfectly timed cues don't work for conducting. I found myself pushing them back as much as a tenth of a second. Your mileage may vary. Be sure when shifting to use "preserve original performance." that way the timing of the cue itself will not change if you push it back through a drastic tempo change.

Shooshie