257. Chris Tsanjoures Of Rational Acoustics

This week, the hosts get nerdy with Chris Tsanjoures, one of the Rational Acoustics folks behind the Smaart measurement platform, a veteran podcast guest and an avid enthusiast of squiggly lines and measurement geekery. He answers all of Sean and Andy’s questions about Rational’s new Smaart Data Modeler in addition to addressing some of their other Smaart-related questions. This episode is sponsored by Allen & Heath and RCF.

Smaart Data Modeler is a new DSP modeling software tool designed to allow you to perform virtual audio system alignment and optimization through the use of level, equalization, polarity, and timing adjustments and to view the effects of those adjustments on Smaart measurement data.

Episode Links:
Smaart Data Modeler
Smaart Data Modeler 10-day Demo
Smaart Data Modeler Support Page (Sample Projects & Data)
CheatSheet (MacOS)
Episode 257 Transcript

Be sure to check out the Signal To Noise Facebook Group and Discord Server. Both are spaces for listeners to create to generate conversations around the people and topics covered in the podcast — we want your questions and comments!

Also please check out and support The Roadie Clinic, Their mission is simple. “We exist to empower & heal roadies and their families by providing resources & services tailored to the struggles of the touring lifestyle.”

The Signal To Noise Podcast on ProSoundWeb is co-hosted by pro audio veterans Andy Leviss and Sean Walker.

Want to be a part of the show? If you have a quick tip to share, or a question for the hosts, past or future guests, or listeners at home, we’d love to include it in a future episode. You can send it to us one of two ways:
1) If you want to send it in as text and have us read it, or record your own short audio file, send it to signal2noise@prosoundweb.com with the subject “Tips” or “Questions”
2) If you want a quick easy way to do a short (90s or less) audio recording, go to https://www.speakpipe.com/S2N and leave us a voicemail there

Transcript

Signal To Noise Podcast, Episode 257: Chris Tsanjoures Of Rational Acoustics


Note: This is an automatically generated transcript, so there might be mistakes--if you have any notes or feedback on it, please send them to us at signal2noise@prosoundweb.com so we can improve the transcripts for those who use them!

Voiceover: You’re listening to Signal to Noise, part of the ProSoundWeb podcast network, proudly brought to you this week by the following sponsors:

Allen & Heath, introducing their new CQ series, a trio of compact digital mixers for musicians, bands, audio engineers, home producers, small venues, and installers that puts ease of use and speed of setup at the heart of the user experience.

RCF, who has just unveiled their new TT+ Audio brand, including the high performance GTX series line arrays and the GTS29 subwoofer. Be sure to check it out at rcf-usa.com. That's rcf-usa.com.

Music: “Break Free” by Mike Green


Andy Leviss: Hey everybody, welcome to Signal to Noise. I'm your host, Andy Levis, and with me, as per usual, is Sean Walker, the SS to my L. What's up, Sean? 

Sean Walker: Wow, dude. You went there, huh? 

Andy Leviss: I went there. I, you know, it's, it's, if I'm going to do it every episode, sometimes I'm just going to look around like ADHD on the desk and see what hits me. 

Sean Walker: Oh my God. I love it. Nice. Good choice though. Good choice. I like 

Andy Leviss: yeah, you know. I mean, I'm coming to you live through my SSL 2 so, 

Sean Walker: Nice. All right. I got you. I got 

Andy Leviss: yeah, it was right there in front of me, literally. 

Sean Walker: Oh, that's awesome. I will say this is way better timing. The early one was brutal, dude. This is way, way better after some sleep. 

Andy Leviss: Yeah, we just, uh, everybody, uh, last week's episode, y'all heard, uh, we, we were recording that at, uh, 6. 30am, right after Sean's 3. 30, so, uh, he's, he's gotten a little rest and we're back, we're back at it. 

Sean Walker: Yeah, it's, it's way, I'm way more of a coherent human being at this point. I w I was barely a person at that point. 

Andy Leviss: I mean on a, on a, grading on a curve of coherent anyway, like at least as far, as far as Sean is coherent any given day. 

Sean Walker: Obvious, obvious. Come on. Yeah, totally. 

Andy Leviss: Cool. 

Sean Walker: with you, bud? What are you, what are you doing? 

Andy Leviss: uh, just hanging out, getting ready to take a week off and go up to the beach in Massachusetts for a week, 

Sean Walker: Nice. That'll be fun. 

Andy Leviss: yeah, gonna try and, try and get some, some podcast stuff for next week, uh, set, and by the time you're listening to this, I'll have batched into the little thing, so it just pews out on its own. 

Not really on its own, the, the PSW overlords will take care of that for us. We love you guys. 

Sean Walker: They're awesome. Behind the scenes, people are super awesome. 

Andy Leviss: Yep, so Keith, Sam, Julie, we love you all, we appreciate you, and uh, thanks for making sure this one gets, this one gets out so everybody can listen to it right now while I am off on vacation. 

Sean Walker: Totally. I'm also surprisingly enough trying to head off on vacation tomorrow to go spend some time with the family, which is. Which is nice. And, uh, we are in the middle of our busiest week of the year, as I was kind of telling before you hit the go button. So right about the time where we're doing a super complicated bunch of moving parts show, I'm going to just peace out for vacation. 

Let the boys deal with it. 

Andy Leviss: That's a, that's some bad boss energy right there. Yeah, I'm, I'm heading for what may or may not end up being my, my last vacation before, uh, before we fully engage dad mode in the fall. So we'll see, maybe we sneak in like one other trip before then, but, uh, 

Sean Walker: Oh, dude, do it if you can. Before I became a dad, I was on tour and I was hoping to come back just in time to like, not be late for the kid's birth. I was like, I'm getting divorced if I don't make this, it's over right now, you know? 

Andy Leviss: It's like, honey, hold on, hold on, hold on. 

Sean Walker: Yeah, right? Don't push! Don't! I haven't landed yet, I'm coming! 

Andy Leviss: Oh no. 

Sean Walker: Oh, 

Andy Leviss: Well, that said, we've all landed here. We're all still alive. As is. As is this week's guest, Chris Sancho of Rational Acoustics. 

Chris Tsanjoures: What's up? 

Andy Leviss: what's up Chris? So, uh, yeah, like, I mean, Chris has been, has been an friend of mine as a new friend of Sean's as of about four minutes ago. , um, has been a rational for long enough that we won't acknowledge exactly how long on the air. 

'cause he can't, we can't possibly be that old 

Chris Tsanjoures: I know, I know. 

Andy Leviss: Um. But, uh, we wanted to hang and talk because, uh, I mean, we all know Smart is like the big, you know, flagship product that, uh, Rational has, but y'all have a new, uh, a new kind of, is it, is it fair to call it a spin off of Smart? 

Chris Tsanjoures: Uh, yeah, like another tool in the box? Yeah, sure. 

Andy Leviss: uh, yeah, so Smart Data Modeler. So, uh, we were like, you know, Chris, what's, what's the deal? Tell us about it. So, uh, yeah, 

Chris Tsanjoures: Sure, uh, 

Andy Leviss: Tell us about it. 

Chris Tsanjoures: What's the deal with the data? Uh, yeah, so, you know, something, Doing equalization or virtual equalization, I'm just having that sort of modeling Uh, tuning capability in smart has been something that we've wanted for a long time. Um, and then when, uh, when we launched version nine, uh, two years ago now, um, you know, that was like one of the, the, the biggest sort of like, is this the version that has virtually EQ, you know? 

And, uh, and it's like, no, you know, cause that's, it's something that we've, you know, obviously it's been on our to do list, but, um, You know, we tend to overthink a lot of stuff and I'm not saying that's a bad thing. I think it's a really good thing, but, uh, you know, the ability to just like put an equalizer on a measurement in smart was always something that we could have done, um, but it wasn't that interesting for us, like in terms of like creating a use case, like what do you, okay, so now you have an EQ on a trace, how does that help you tune a system? 

It helps you equalize this measurement, but like, what is, you know. Is that actually any less, any more beneficial than just, you know, changing the EQ on your DSP and refreshing the measurement, you know, so, um, with Data Modeler is sort of like take a holistic approach to, uh, system alignment and optimization and create an interface that you can utilize to tune it. 

Multiple outputs of a system. So not just one channel of DSP, but like as many as you have. So, you know, if you're doing a theme park or a stadium or you know, commissioning of anything from a, you know, office to a, I don't know, whatever, you know, you can take the data and smart, throw it into data modeler and model each of the DSP outputs. 

So like when we say like DSP modeling, like, you know, that's one of the things that data model is really cool for, um. You know, modeler has a connotation, right? So when you say modeling, you're thinking like venue modeling or prediction or, you know, we're modeling the response of the system based on this CAD file or something. 

We're not doing that. We're, we're modeling the DSP configuration of the system. And, you know, we're staying in our lane where, you know, people use smart to tune systems. You know, the system may already be, it's already been installed. It's, you know, you're doing it after you've done your setup or whatever, you know, whether it's before a show or you're using a system that has all the manufacturer's prediction stuff. 

And so you theoretically know exactly what it's going to do in all these positions that you've modeled in this prediction software. What, you know, pick your favorite manufacturer, they all have some sort of tool for this. But, you know, once you get into reality, and you actually have that system in a room, you know, the, there's, there's variables that that software didn't account for. 

And, and you do, you know, you do some amount of engineering on that system so you can gather your measurement data and, um, you know, turn off, you know, The noise and put that data into smart as DSP channels and I say smart into data modeler as DSP channels and you can go through and model the whole, you know, the whole thing. 

Um, you know, whether it's EQ delay level, uh, you can model the summation between different systems at positions that you capture. Um, and then, you know, if you have the ability to based on whether or not we have, uh, you know, reached that level of cooperation with various manufacturers, you know, you can, uh, then push those settings to, uh, to the actual hardware software that you're, you know, that you're modeling. 

Andy Leviss: Right. So it, I mean, I know going back like old school, like even back to the smart five days, like a lot of us, and I know when like the request for this would always come up on like the Facebook group or whatever, like those, those of us who were like, you know, cranky old smart users would always be like, just, just take a while, straight out of the queue, invert the trace and kind of do it. 

So is it basically like, Or at least to a certain degree, is it basically just kind of doing that virtually instead? 

Chris Tsanjoures: Yeah. I mean, we actually, we have a mode in Modeler. It's called System EQ Result Mode, which is, um, something that, uh, Sim users will be familiar with. If you've used Miresound Sim, where, you know, you can see the EQ response and then invert it and then view that against the system's response. So like the, the post EQ, Equalized response of a system. 

And then we can look at the result, which is the equalized response of a system. So we can do that all virtually. Um, so, um, yeah, uh, sorry, I don't know if that was a tangent or if that answered your question. Uh, 

Andy Leviss: Yeah, I, I think so. Um, and yeah, I mean, I guess, uh, yeah, like, what else is it taking the account past, like, what I would get out of, aside from the fact that I don't need to have a tap out of the device, which certainly I'm like, uh, when the DSPs in the amp gets a lot trickier, is there, like, I, I imagine there's a little more going on under the hood than just 

Chris Tsanjoures: Yeah, well, um, you know, one of the things that makes Data Modeler kind of, you know, cool or unique or maybe, I don't know if it's unique or not, um, I have the, you know, for better or for worse, I tend not to play with other analyzers very much, so, um, You know, when we do, you know, when we're in our development, we're thinking about smart and how we can make the smart platform better. 

It's not necessarily, um, based on what else exists. Um, but one of the things that we did with modelers, we partnered with a company called Eclipse Audio from Australia and they have the ability to model specific branded IIR filters. So with data model, you can actually select from a list of DSP processors and then you will be modeling. 

The exact, um, you know, bandwidth or queue settings and the different filter topologies that that manufacturer is using for their, um, coefficients. Um, you know, some of the under the hood stuff is already accounted for, like, you know, averaging exists in smart, you know, we can average data and smart, we can average data and data modeler. 

The thing that took the most amount of R& D and, you know, frankly, uh, sweating on our part was making sure that summation modeling was correct. Um. You know, there's some, some specific, like, you know, where things are picked off and how the phase, you know, response interacts with the summation modeling and, and getting it right. 

So that, you know, it's, it's accurate to what would be expected and not doing, you know, weird things that, you know, causing comb filters and there shouldn't be a comb filter and you're getting addition when you should be getting addition. And, you know, that was, that was novel for us, you know, so that took the most amount of sort of like, let's make sure we get this right, or else this isn't a viable product. 

Sean Walker: What filters are in there? Like, what's in that list of things you can pick from from different manufacturers? 

Chris Tsanjoures: So there are, I believe, like 13 or 14, maybe something like that. So the current list and I'll, I'll sort of rattle them off for you. Um, as soon as data modeler launches, um, there's a generic filter, uh, which is, um, uh, based on some sort of standardized topography and, um, It's the default setting, but beyond that, it's in alphabetical order. 

Um, BiAmp, Blaze Audio, which is also Pascal amplifiers, uh, BSS, uh, Lake Mesa. So, um, if you think about a Lake processor, there's the, there's the sort of output. Processor mode, the loudspeaker processing, what I think they call the contour mode. And then there's the Mesa mode, which is like the, what you would use to tune a system. 

Um, so we, we can model the Mesa mode, um, which is actually, uh, this is not something I knew before getting into this stuff, but the, it's a, it's sort of a specialized thing. They, you know, their filters are actually like I, uh, FIR based and, but we're doing IIR emulation. So we're doing a conversion all things made possible by Eclipse Audio to work out the math for that. 

Um, Linear Research, Mirani, Pascal Audio, again, because Blaze is Pascal, um, Powersoft, Q SYS, Ram Audio, RMS Acoustics and Symmetrix. Uh, this is the current list, and that list is actually twice as long as it was when we started development of this product. So, um, you know, we keep, you know, after Infocomm this year, which was last week, as of the recording of this podcast, um, we met with, uh, quite a few different live sound, you know, speaker manufacturers who had lots of interest in. 

Getting their company in the list too. So I expect it to be something that is going to sort of grow organically as our relationships develop with these other manufacturers. And as, um, you know, things become available. 

Andy Leviss: Awesome. And I, am I imagining that it's the sort of thing that, like, it, if, If or listeners like want a certain manufacturer in there, it's worth our time to reach out to them and say, Hey, are you guys going to work with Rational? 

Chris Tsanjoures: that's how it's done, man. Um, you know, I've already taken a bunch of screenshots from our user group and sent them off to the manufacturers that the people are requesting. You know, that's, it's, it's, that's what it's about. I mean, we've, you know, we've always been sort of Switzerland here too. So, um, You know, we, we, we recognize that our software, you know, smart is used by people that use all sorts of different, you know, manufacturers, loudspeakers, and, um, and you know, we'd like to help them have a better day. 

And, you know, you would, you would think and hope that the manufacturer also wants them to have a better day. So, you know, whatever we can do to all work together, I think is the right move. So that's sort of the, the path where we're going down as a cooperation. 

Andy Leviss: I mean, do you want to talk, cause I know there's like, you've got a few different modes in there. Do you want to like talk a little bit about exactly what those different modes are? Cause I know like, I haven't had a chance to fully poke at the software yet, but watching like the intro videos and stuff, like I sort of get it, but sort of, um, like what exactly does Swapping between those different modeler modes do. 

Chris Tsanjoures: Yeah. So, um, when we were in initial development, we just had this one interface, which we, which we call modeler data modeler. So that's the default interface. It's, it's a, you know, a smart screen that you're very familiar looking at. It's going to phase response over a magnitude response. And there's data on the left and control on the right, except that the data on the left is actually DSP channels that you're modeling. 

And then on the right you have DSP control instead of measurement control. And so modelers kind of It's a free run, you know, like a sandbox for doing stuff. You can add measurements in and equalize them and adjust their level, adjust their polarity, their delay. You can, you can pick multiple measurements and sum them together. 

We actually don't limit the amount of measurements you can sum together right now. So you could, you know, uh, theoretically turn, you know, go through measure three delay lines and then sum them all together and do the timing for all three You know, and one sort of move, um, which is kind of neat, um, you know, a product of let's do this and see how people use it. 

Um, and so then, yeah, ModelerView is the default, then SER, System EQ Result Mode is, is designed for creating what, you know, what we call complementary equalization. So, you can take your EQ response and then invert the, uh, amplitude. So when you on the screen, move that filter up, it actually cuts the EQ. And so, um, you can take your measurement and normalize it in such a way where you, you know, you're just going to be doing cuts on that EQ. 

And then just sort of trace out the, uh, the response to create, uh, an equalized response based on the response of the system. Um, I definitely, you know, definitely check out our videos on, on doing that cause it's, it may not be intuitive if you've never heard of doing that before, but once you see it happening, you're like, Oh, okay. 

So you're just like literally taking the EQ and tracing out a measurement, you know, with some amount of smoothing or, you know, and, um, And it creates an equalized response. 

Andy Leviss: Yeah, that's, that's what we were like talking about a little bit earlier with the like that like old school, like even back in 5 there was that invert, you know, source and measurement button. I remember learning that trick from Jamie way back in like the EAW days. 

Chris Tsanjoures: Back when I was, uh, teaching smart classes before the pandemic and before children, uh, you know, that was something that in every class I did, that was, that was like one of the, you know, okay, let's hammer this home. This is what an EQ does, you know, and here's a, here's some really cool stuff we can do in the analyzer to show it. 

Andy Leviss: Yeah. 

Sean Walker: Oh my god, will you please teach me what an EQ does? Someday I want to be good at this job. Please. 

Chris Tsanjoures: by hook or by crook, we will all become EQ masters. 

Sean Walker: Yes! Oh, I knew I was in the right place. 

Chris Tsanjoures: So, uh, the third mode is, uh, summation. So it's a, it's a, we, you know, we have the measurement delay for all the measurements that we take. You know, when you, when you take a measurement smart, you set the reference delay and then capture the measurement. 

And so in summation mode, what we're doing is reading the. TraceInfo for the contributors to the sum and then reading that measurement delay time. And I'm, you know, I could go into like acronym city right now because my head's in the weeds with this stuff. But, um, so we added a, we added a bunch of stuff with data modeler. 

That's, you know, all sort of getting slowly ported over to the mainline smart. But one of the main ones is something we call PTR or plot time reference or phase plot time reference. Bear with me. And um, what we can do is we can change the phase response of capture data. So it seems like, yeah, of course you can. 

Um, but until data modeler, we couldn't. So, um, what we're doing with summation mode is say you have two measurements. They're captured from the same location, but one of them's a delay system, one of them's a main system or whatever. So they have different delay times, you know, say the main system arrives at 50 milliseconds, but the delay system is closer and arrives at 10 milliseconds, right? 

So there's a 40 millisecond difference. What we do in summation mode is automatically calculate that value, that delta value of 40 milliseconds, and then add it as positive time offset. to the delay system. So now, you know, when you look at the data without anything applied, they say they have the same phase response and you're good. 

You're timed, but you're not timed. One system's actually arriving 40 milliseconds earlier. So by adding 40 milliseconds of positive time offset, that's the exact same value you need to add to the DSP output to make them aligned. So we just, we just figured that number out for you. And then, um, and then you are allowed to enter it into your DSP delay or adjust the DSP delay manually, um, you know, which becomes somewhat more fun if you're using loudspeakers that don't have matching phase responses where you have to make a decision where, you know, that time might work great for the top octave, but be totally out of polarity for the bottom. 

So, 

Sean Walker: On the, on the manufacturers that you have integration with right now, or you have the filters for right now, is there any plan to be able to Import or sync that data to their processor. So if you had a Galaxy or an HM or a whatever, you know what I mean? That you could do that or your buy in or your BSS. 

Could you take, you know, take all these, do all this, and then just fire that off to your DSP and let her rip 

Chris Tsanjoures: Yeah, actually, um, as of now, we are able to export to the Lake LM and l uh, is it LPX? The new, the new Lake processor? Um, 

Sean Walker: the new, yeah, yeah, I got you. The new, the new Behringer, I mean, Lake processor. 

Chris Tsanjoures: Yeah. Yeah. Um, so they, they contacted us like three weeks before InfoComm and we're like, Hey, let's make this happen. And we were like, great, let's make it happen. Cool. So, um, so I, you know, as of, as of last week or two weeks ago from today, in fact, um, export directly to Lake processors as possible. And then, you know, uh, Yeah, it's something that we want for everything. 

I mean, that's the goal, right? Any, any filter mode that we can model, we want to be able to push settings directly to that processor. And it's just going to be a sort of ongoing and long term, you know. Um, exercise of, you know, getting involved with those manufacturers and getting to the point where we can. 

Um, it gets a little bit more interesting when you talk about like the open architecture based, um, processors. So like a, you know, a Q SYS or even the BSS, like a London or something, there isn't necessarily a physical, you know, chain associated with an output. It's, you know, it might be a bunch of different things in different places. 

So it's 

Sean Walker: But a lake or a galaxy is more straightforward so you can kind of do that 

Chris Tsanjoures: Yeah, yeah, yeah, that's, you know, Galaxy is a perfect example of something that, you know, we would love to be able to push right 

Andy Leviss: Yeah. And in theory, it's like, yeah, it's all, it speaks OSC for like everything. So it should be. I say in that way that like, as somebody who used to do software support, I know is the way that everybody in software hates hearing, but like, 

Sean Walker: All you gotta do is Yeah. All you gotta do is You son of a bitch. 

Chris Tsanjoures: Yeah, it's just typing. Yeah. 

Andy Leviss: Like I mean, I don't know why you guys don't have this already. It's, it's so, it's right there. Yeah. No. 

Chris Tsanjoures: No, that's, I mean, yeah, like that, like totally dude. And that's the, um, you know, I, I would love to be, you know, to have the same conversation, you know, catch up to in a year and be able to say, oh yeah, we can directly push to like, uh, these seven processors or something, you know, and it's, and then ideally those are things that like our users are actively using, you know, And not some, um, well, I'll stop myself there, but not, you know, not the, not the less common processors, I should say. 

Andy Leviss: like I'm trying to think how that would even like work on the back end with Qt, like I almost, it almost would have to be like a plug in or something I think 

Chris Tsanjoures: Yeah. Yeah. I don't know, you know, whether they can maybe address certain things by IP address. I don't know. I don't know. You know, that's going to be something that we're going to have to figure out with them as interests. comes because I mean right now, you know, like they're not going to be interested in doing any of this work with us until they have users that are like, we really want to use this tool. 

And 

Sean Walker: I really want to use this tool. Can I just say that? 

Andy Leviss: and like, 

Sean Walker: want to use this tool. I want to push that to my galaxy, please. 

Chris Tsanjoures: Yeah. 

Andy Leviss: yeah, and like, like Q SYS, I feel like it'll always be that race between like, what, what Q SYS will do themselves or like, what somebody like, forward thinking might do, 

Chris Tsanjoures: Yeah. Yeah. 

Andy Leviss: plugin that I think, I don't know if it's updated for Smart 9 yet, but I know for Smart 8 they've got their, their like, interface for that that integrates pretty nicely. 

Chris Tsanjoures: I talked to them a few months back. I think it should be, I mean, they definitely have what they need for it to be. Um, but yeah, with, with the version nine update. It's no longer just one smart program. Now we have three additions. So API adopters do have to, do you have to tag which edition to look for on the network to get stuff in? 

So that's a little bit of a change. 

Andy Leviss: gotcha. Yeah, so for like, for those who don't know, like I know we've talked about Q SYS DSPs before, and so Forward Thinking Design is a company that does a lot of, uh, AV programming for like theme parks and that sort of thing and they also like they're kick ass for it like I've used them to subcontract on projects to build like custom Q SYS plugins and one of the ones they do it like ties into smart and basically both pulls in the smart trace data right into your Q SYS UCI and then also gives you the EQ controls right there so similar to like how like the like the lake integration would be or something like that you can actually like pull the trace up and the EQ right on top of each other live you In, in the Q 

Chris Tsanjoures: it's like a whole processing chain. Like doesn't the, their plugin do delay and polarity and everything else 

Andy Leviss: Yeah, it's got all the, yeah, it's, they've basically built all the, like, EQ blocks and stuff that you need in there. So it's, it's, it's almost like you're dragging and dropping, like, a galaxy with an interfe Or I guess at that point then, it's probably more accurate to compare it to a lake, because it's got that. 

Two way communication with smart in it, but like that's a quick way like if if if that project you'd been doing the other week we were talking about Sean was on on Qsys you potentially could have just like licensed that plugin dropped it in and there you've got your like you know 16 channels of EQ talking to smart and doing all the things 

Sean Walker: Shit, yeah, dude. 

Chris Tsanjoures: I mean, for what it's worth, uh, the first time I used Data Modeler on, uh, commissioning, you know, sometime last fall. Um, but you know, it was, it was in alpha, but it was, it was, it was in a working state and, um, the system was driven by Q SYS. And so I chose Q SYS as a filter reference mode and went through and captured all the data and, um, did the thing in Data Modeler. 

And then, you know, at that point, you know, now we're manually entering those settings into Q SYS, but It was, like, time saving. It was, it was pretty interesting and it was really satisfying to just be like, all right, let's check and make sure that this stuff is right. And then you go and check your measurements and everything's exactly what you modeled. 

It was just like, you know, high fives all around. This is really cool. Um, you know, it's, um, 

Andy Leviss: yeah like I'm pulling up the example like pdf report You know, you guys have on the website for it to show how it comes out and yeah, it's good. It's like every, yeah, it's like second by like main left and main right, you know, and then broken down for like each speaker shows, yeah, it's showing me polarity, level delay, what all, what my EQ frequencies and cuts or boosts and bandwidth are. 

Chris Tsanjoures: Yeah, I, uh, I selfishly made that PDF, you know, because one of the things that, uh, that always was sort of a housekeeping thing. But whenever, whenever I've done commissionings, I'm like, you know, half of the, half of the time is spent in Excel writing down all the changes that I'm doing. So that, cause I want to be able to deliver, you know, that client or whatever the, um, you know, here's all the things that we did. 

You know, and so if something goes wrong and you need to replace that hardware thing or whatever, you have all the recall settings. And so I've always wanted just a quick way to like create that. And so I'm happy to have the PDF export as like, um, here's, here's, here's this thing. So here's the things that we did to your system. 

Andy Leviss: There, there's a question then, um, because I know like that's one of the, one of the things that like when you start like getting down that EQ nerdery is the whole. Uh, despite the fact that theoretically there's an equation for Q versus bandwidth that doesn't actually translate necessarily from manufacturer to manufacturer, which is part of the reason y'all are emulating the different filter types. 

Is there a, is there a way in Data Modeler or, or, or would there be a way in the future to like, if I do it all with like, you know, like Q SYS filters and then later on I've gotta switch it to like, you know, linear research or something to switch it over and make whatever adjustments might be needed to make that translate? 

Chris Tsanjoures: Yeah, so, um, that's one of the key limitations between our licensing agreement with Eclipse Audio is that they retained the filter conversion. So um, if you were to use FIR Designer, which is more tailored towards individual loudspeaker outputs like, um, you know, at a manufacturer's level perhaps, you know, or you're, you know, building a loudspeaker and you want to, you know, make the sort of default settings for that loudspeaker. 

Um, you can switch between different things cause you know, as a manufacturer, you might, you know, I want to make my settings for this device and then, but the, um, you know, end user may be running it from whatever they're running it from. So you can create different files for each of those. Um, but yeah, for, for data modeler, once you've, once you've started adding EQ blocks, that ability to change the filter reference mode is, is then locked down. 

Andy Leviss: Gotcha. So, but, like, would you be able to, like, if, could you do a, like, could you, could you then add, like, another DSP channel that's a different EQ mode, or it's all, it's global for, like, 

Chris Tsanjoures: Yeah, so each project is, each project is in a DSP filter mode, so, you know, so the, the, you know, the sort of the, the overall thought with data modeler is that, you know, what you're targeting when you do the, when you do the, the thing, so like, you know, I'm tuning this system, this system's installed, it is running a Q SYS system, I'm going to use the Q SYS, and like, that's not, you know, that's the idea, it's more, you know, project based in a sort of the chain of when things happen. 

We're talking about like the, you know, the sort of right before you actually use the system chain, like in the chain, you know? 

Andy Leviss: Gotcha. Um, I mean, I don't know, if, if Sean doesn't have any other questions he wants to jump in front of this with, I was going to ask if, like, do you want to kind of talk us through, like, you're doing a commissioning, like, using this and SmartTogether, like, give us the quick and dirty, like, where this fits into the workflow and how you would use it? 

Chris Tsanjoures: Yeah, yeah, sure. Um, so with smart, and this can be any version of, uh, of smart that's made by Rational Acoustics. All the data that we've made since version seven is compatible with data modeler. Um, the next update of smart version nine, which would be L E R T and sweet. I'll have. hooks into Data Modeler directly. 

So, um, if you're using a version nine application, you capture the measurement positions that you would need to tune your system. So for example, um, you have, uh, left and right front fills and subwoofers. I don't know, just something simple, just, just to, for brevity. But, um, You know, capture the positions that you need to equalize the system. 

So maybe left and right, uh, you know, three or four or five positions in their coverage area. Um, if they sound the same, then we're good. Don't need, uh, you know, to capture necessarily both sides. Um, and so you capture your data for that and then once you have the equalization positions, then you would move the microphone to your timing location. 

So, you know, maybe go to, uh, where you want the front fills to be generally in time with the mains and understandably that this is something that you're probably going to listen to later and then tweak, but at least you have a starting point. Um, especially for equalization, you can capture a few positions for the front fills. 

Now, as we capture this data, you're capturing it into a folder. So you have your session folder and then you have a folder for your mains, you know, a folder for your front fills, a folder for your subwoofers, whatever. Um, So we've got, say, I don't know, five traces for the main system, three positions for equalization, a position for your subwoofer timing, and a position for your front fill timing. 

You've got all those traces in one folder called main, right? And then you have your front fills, and you take maybe, again, three positions for the equalization of the front fills, and then a position where the timing between the front fills and the main system will be. And then your subwoofer systems as well. 

So, I mean, there's a whole can of worms we can get into about like the system design and where, you know, what, where to do this stuff. But for the sake of, uh, you know, the alignment, you, you know, you, you're eventually, you're gonna have to, you're gonna have to pick a location, right? So you have a location for the subwoofer timing and level and, you know, where you want that to be matching the mains. 

And that goes into a folder that you call sub. So then in smart, you just right click. These folders, you'd multi select these folders, so you have, you know, the three folders selected, right click. And then create DSP from selected and then a thing will pop up and it'll say do you want to create a new project or do you want to add these channels to an existing project? 

And so right from there you can say create new project, call it, you know, whatever the show is, whatever the gig is, whatever the job is. Um, from there you can directly choose your filter reference mode. Um, if you have something that you can choose or just stay on generic and then you press export and then your data modeler just launches with those folders, uh, automatically with a DSP channels associated for each folder. 

So you'd have, um, you know, three channels of DSP. If you, um, if you do need to have the right side as a DSP channel as well, you can, you know, you can create an additional channel of DSP and then just copy the settings from the main into that, into that, uh, that fourth channel. And then you can link them as well if you want, you know, so your settings will track between the two. 

Um, and then just you sort of just go through the process of tuning the system. So you equalize the mains position, the main system based on the three positions. And then, um, that's, that becomes your target EQ for your front fills, you know, and then you EQ your front fills to that. And then, um, You'll have the ability to change the level of the, of the front fills based on the timing position that you chose between the front fills and the mains. 

Um, you know, choose the polarity that's the best match for the both. And then, you know, when you go to sum, sum them, and then summation mode, you'll see which one's arriving first, or second, or First and then, uh, delay the first rival to the later rival. And it's just, it's going through the motions. I mean, it's something, if you've ever taken a smart class and learned the sort of relapse process that we talk about, you know, response level, polarity, timing, it's just doing that in rinse and repeat on. 

On the DSP channels, uh, that being said, you know, one of the, one of the, you know, other primary things with making data model is having an application that we could use in smart classes to talk about this, the concepts that we're doing, you know, in, in, in a classroom environment where not everyone is connected to a DSP, you know, now you, you effectively are, you know, effectively you have a, have a, the ability to do DSP work with smart data, you know, without being on a system. 

Andy Leviss: Yeah, it's like another tool in the box too to, to kinda help push us towards that goal of like getting people to stop blasting 20 straight minutes of pink noise every time they tune a system. 

Chris Tsanjoures: Yeah. You know, it's, uh, I, I, when I, when I was a little younger and would, would, uh, feel more inclined to be on my soapbox and talk about how you don't have to do that, but now it's just like, what people are going to do what they're going to do. So, you know, we added, we actually added a feature in smart that stopped the signal generator after capture. 

Andy Leviss: Love it. It's one of my favorite things. 

Chris Tsanjoures: Yeah. 

Sean Walker: Dude, that and ramp up are killer, where it like, 

Chris Tsanjoures: Yep. Yeah. The fade in. Yep. 

Sean Walker: you're like, Oh, I'm such less of an asshole to these people on site right now, like, holy 

Andy Leviss: we'll engage, don't be a douchebag mode. 

Sean Walker: dude, totally. 

Chris Tsanjoures: Yeah, the other thing that is, you know, a sort of common misconception is that you need to like measure very loud and you really don't like, uh, just turn the system up until like, you know, it's the coherence stops getting better cause it's not going to get any better once, once you've reached that. 

And that could be, you know, a relatively tame, like 80 dB SPL or something, you know, it doesn't have to be, uh, you know, show level. Unless you really want to, you know, clear the room and then crank that baby up, everyone will scatter like roaches. 

Andy Leviss: Um, yeah, I mean, that was actually, that was going to be my next question was asked, like, while we've sort of gotten onto that tangent and back into smart is like, if, if you've got any, like, quick tips or tricks for that, for like, both working faster in it and working in a way that just doesn't annoy the crap out of everybody else in the room or, or burn your own ears 

Chris Tsanjoures: Yeah. I mean, the thing that's been the biggest sort of game changer for me is just wireless microphones. You know, the, I mean, depending on what sort of situation you're in, you know, um, I, you know, I, I'm not a touring engineer. I'm not doing shows in big venues. You know, the stuff that I do is, is like outdoor. 

Farm, you know, fair festivals, agricultural fairs and stuff for like small capacity, you know, a thousand people or less theaters out here in the woods because like we're out in Woodstock, Connecticut. I mean, there's just not, not too much going on out here in terms of like Wi Fi congestion. That said, I've really fallen in love with the, the X5. 

Plug on transmitter. It's like a really inexpensive, it's got like six presets or something and it works in the 2. 4 gigahertz range. But, um, for basically anything that I do, the fact that I can just not have a cable and walk around with my microphone and I'm never going to be, you know, further than like 75 feet away from front of house or whatever the, you know, the measurement stuff is. 

And that's been my biggest like workflow improvement, uh, thing is just putting one of those in my kit and using it. 

Andy Leviss: Yeah, I mean, I went, I mean, I, you probably remember, I think I bought, at the point, like, at the height of the, uh, that godforsaken phrase, supply chain issues, uh, I think I got the last four of the MIPRO, like, 5. 8 gig transmitters y'all had, right? And then I literally bought them without a receiver for them, just the, I'm gonna buy them all I can, and I will find a receiver somewhere, and I was trying to eBay it everywhere, and found like one vendor in France that had a pile of them on the shelf, and unfortunately, in confirming that they had it actually in stock, I clued them into the fact that they were the only people who had it in stock, so I got it for the price they had it listed at, and they almost immediately jacked the price up by like a hundred euros right 

Chris Tsanjoures: Wow. Yeah, I actually, we, um, we had a meeting with the, um, the MyPro distributor for us, uh, last week in InfoComm and, uh, it seemed to have very positive news that there will be MyPro, um, hitting our shelves in the next, uh, few months, which is, which is great. Cause yeah, I mean, once you get into, you like real wireless where you need, uh, you know, to be able to choose a frequency and you need some reasonable range and stuff. 

Uh, it gets expensive really quick. So the fact that my pro offers something that does that and is not, um, you know, 2, 000 a channel is, uh, is great. 

Andy Leviss: Yeah, and they've, because they've got both like UHF frequency models and then the 5. 8, and like there's trade offs that come with that 5. 8, like you are somewhat more limited in range, but like you can get high gain antennas for it and that'll, like I've done like three, you know, three story venues with like lots of steel on them and generally had either no problem at all, or like the battery telemetry gets a little sketchy, but the audio signal still seems clean, 

Chris Tsanjoures: yeah. And I mean, you know, when we're talking about measurement, it's we're not talking about recording like we're not, you know, you're not broadcasting pink noise, you know, to an audience. So like the, the transmission, you know, can be, you know, somewhat You know, less than desirable for, for, yeah. For like, uh, an application where it needs to be audible, but for a measurement application, it's like, you know, we've got all these mechanisms built into smart for averaging and, and data protection and stuff. 

So you can, you know, you can get away with a, not just like a pretty high noise floor, but even if it's intermittent, as long as it's in more than it's mitten, then you're gonna be fine. So, 

Andy Leviss: Let's not talk about if it just gets chur, though, that's a whole 

Chris Tsanjoures: Mm-Hmm. 

Andy Leviss: Um, yeah, but I mean, it's like, and, yeah, that, like, the MyPro stuff is, like, game changing, cause, like, for the cost of, like, maybe one channel of Electrosonics, which, Electrosonics is fucking awesome, but for the cost of one channel of that, I've got four channels, that it's, like, it's Bonkers. 

And I've yet to like bring it on a gig where I'm tuning with somebody else or where I'm like, eh, you know what? I've got it here in my car anyway. Like if you want to use it, here it is. Where somebody hasn't been like, oh, I need to look for as soon as I can buy a set of these. I need a, I needed, I need the full set. 

Chris Tsanjoures: yeah, yeah. I mean, you know, there's, and there's, there's some other stuff that's kind of hitting the market lately too. Like Wizzy. com has a, has another, you know, kit that's up there in price too. But I mean the, for the measurement application, like measurement, it's just such a, it's like a side quest for a lot of these companies, you know, they're, they're really concerned with like, you know, in ear monitors and different things that, um, you know, so, um, you know, we, We're not the market driver for this kind of stuff, 

Andy Leviss: but there's so many of us, I don't understa 

Chris Tsanjoures: yeah I know, I know. 

Yeah, well three of them are here right now, so 

Sean Walker: Right. Totally. What? 

Andy Leviss: Yeah, uh, oh, I had another question on that front, 

Chris Tsanjoures: Oh, I would say, oh, so the other thing, um, just for like tips and tricks, we are like always adding hotkeys to the program and I, you know, I know that not everyone is maybe a hockey junkie as much as I am, but like, um, It's always sort of, you know, take a look at the help files and look at the keyboard shortcuts, because there's probably something there that's going to surprise you and be like, Oh yeah, this is great. 

I'll just, I'm never going to not use this again. Like just the other day, uh, I say the other day, the other day is sort of rel rel, whatever could have been. Six months. I don't know. What is a day? Um, but, uh, the ability to like press shift D and delay track all of your inputs at the same time, um, was something that, you know, it was like sort of an obvious thing, just never got in there. 

Um, and now you can, and now like, I just find myself, it's like, if I have, you know, three transfer function measurement engines, I tend to use track more than find, and you can just press shift D and everything's like synchronized basically instantly. Um, so there's a lot of stuff like that, which was kind of like, you know. 

We are in somewhat of a rapid development all the time. Like we are always putting out new stuff and always putting out new features and there's always a new beta. I mean, we launched a beta today, you know, and we launched the last beta two weeks ago. So like there's, you know, there's always sort of new stuff happening. 

And, um, if you, you know, even for, for us, it's a lot to keep track of, you know, so, um, it's good to just kind of, yeah. 

Andy Leviss: I'll throw out a tip there for the Macheads. There's a, there's a program called Cheat Sheet you can install, which is, like, I think it's included as part of Setapp subscription, or you can buy it on its own, and basically you hold down and Like the command key or like one of the basic shortcut keys and after a few seconds of holding it down, it will pop up a full screen overlay showing you every keyboard shortcut in the current program 

Chris Tsanjoures: No way. 

Andy Leviss: Yeah, it's pretty slick. It's I forget I have it and then I'll just hold down the key because I'm thinking too long and suddenly it'll pop up like, Oh, like, right. 

Chris Tsanjoures: Cheat sheet? I'm writing this down. Cheat sheet? 

Andy Leviss: Yeah. Well, and we'll link to it in the I'm going to scribble a note to to put it into the show notes too, because it's 

Chris Tsanjoures: That's awesome. 

Andy Leviss: Yeah, um, yeah, so that, that's my pro tip for that, and then I'm gonna, I'm gonna ask you, like, one more, like, one more deep question before we let you go, because I know you're running tight on time, is we, you, you mentioned the relapse, uh, like, sequence of processing again, and it's the thing that, like, I've, I know I've asked before, and I don't always, there's, At least in my mind, for like, tuning under baux, there's a little bit of a battle between the, like, EQ vs timing, cause it feels like it's a chicken and egg thing where each one of those affects the other, so which one do you do first? 

And I've heard and made arguments on like, either order of delay vs EQ first to get the best results, and like, how much doesn't matter, like, Is there a better order? Is there personal preference? 

Chris Tsanjoures: Well, I think, you know, when you're talking about big EQ decisions, like if you put a high pass filter on something and it's time to the low frequency of something else, and you made a conscious decision for that to happen, then that high pass filter is going to throw out that timing. So, you know, but if you're talking about parametric filters or something where there's sort of a minimal impact on phase, whatever, I mean, the thing to me, especially when you're talking about fill situations, uh, is, is just, do they sound appropriate? 

You know, like, I'll, I'll basically always, um, always prioritize tonality over, um, you know, just like pure perfect imaging, you know, because, you know, the imaging is great, but if it just, if it just sounds awful, that's going to probably be more distracting than the imaging. Um, and I'm saying, like, I'm not talking about, like, we're off by 140 milliseconds or something. 

I'm talking about, like, literally 180 degrees at, uh, you know, 800 hertz or something where that maybe the low frequency of a system just sounds weird. hollow. And, you know, and so it doesn't, that area won't have the same impact or just be fighting with the main system or something. So as far as the order goes, you know, it's, it's, it's, at a certain point, you kind of have, you gotta have some common sense, you know, you gotta have some, some experience on your belt that's saying, you know, this, this looks right on the screen, but it doesn't sound right to me. 

And, um, we're going to do what sounds right, you know, and the measurement tools that we have. They get you to the point where you can make that decision much faster, you know, so it's not necessarily about blindly, well, I wouldn't say blindly, but, you know, just, just, uh, saying, no, this is the trace. The trace is correct. 

And so that is correct. It's like, well, you know, The measurement system's gonna get me to the point where I have the equalized response I want in way less time than if I were gonna do a by ear and just keep second guessing myself and wondering, is it, you know, is it 400 hertz or is it 500 hertz? Is it, you know, is what, you know, what is the thing I'm going after? 

How wide is that thing I'm going after? Is that, is that the room? Is it the, the, you know, is it where I'm sitting? Like what? You know, um. The measurement tools help you get there so much faster. And then after that, you really just need to listen and, and do what's right, you know, for you, do what's right for the situation. 

Andy Leviss: Yeah. And I mean, that's, that's the thing we talked about, like when we had Pat Brown on a while back too, is like the, we're going to, like, we can use the tools or whatever to get it in time perfectly, but then you're going to put, you know, something more akin to program signal into it and listen to it. 

And like, you're, if you're trying to get stuff to image to a stage, like the under balk being perfectly time aligned to the main might not actually be what gives you the result you want. Like you've always got to. 

Chris Tsanjoures: yeah, yeah. I mean, like I really listened for like transients and stuff, like especially transients that have low frequency content, like is it distracting, you know, and, and, you know, and, and, and, and, and, and, and, and, and, and. If your delay system is being asked to do a lot of work, then I will tend to, to delay that system out of the high frequency coverage of that system. 

Like I'll go right under it where the low mid stuff is because the people that are right below that delay or right in front of that delay, if they're, you know, if, if they're getting that low mid content, if you haven't high passed it out, um, you know, which could be a decision based on a number of factors, you know, but maybe you did high pass it out, but if you haven't and they're hearing it, then, you know, I'm going to time that delay to us, uh, like. 

What a snare drum would be or something, you know, like the actual hit of stuff. 

Andy Leviss: Yeah, I'll, I'll share, like, my trick lately, like, the, my two go tos for aligning, for, like, checking alignment on, on, like, bow delays or front fills or anything like that lately have been either, uh, Michael Jackson Human Nature, 

Chris Tsanjoures: Hmm. 

Andy Leviss: uh, or, uh, Not Rosanna by Toto, but often you can actually go in on, on YouTube. 

There's the isolated drum stem 

Chris Tsanjoures: and 

Andy Leviss: and I'll just take the drum stem just of that. And like, that's one of those, like, you'll hear everything start flamming if stuff isn't exactly in time. Cause it's such an intricate drum line, but so recognizable. And those are big on my too. And it's technically cheating because technically both of those are Toto songs, but that's a conversation for another time. 

Chris Tsanjoures: Uh, yeah, I really like the instrumental song by, uh, Xtreme, um, God, it's what's it called? I'm going to look it up real quick because it's going to, it's going to bother me that I can't, uh, just, I didn't, I didn't have it prepped, you know, in my, on the top of my, uh, my mind, but, um, it's just like, it starts off and it's just like a shaker and bongos and, um, and every time, you know, whenever I'm doing, um. 

You know, when I'm listening to delays, it just that, that, uh, that makes a lot of sense for me. Hmm. 

Andy Leviss: of trivia, which is about a shaker, which is, uh, if you had listened to Aerosmith's Sweet Emotion, there's like a very distinctive shaker throughout that song. And apparently the SIR in, in classic SIR fashion, they'd requested a shaker specifically for that to the studio. 

It didn't show up. So what that shaker sound is, is literally a packet of sugar that happened to be dropped in the bottom of the SIR, you know, like percussion toys box. in front of a fucking u87 and it's literally just they crank the gain out of it and and steven tyler's just sitting there with with a packet of sugar 

Sean Walker: dude. That's awesome. 

Andy Leviss: yeah and i actually the the first time i heard that like a couple folks in the like the demonstration that like was doing like a breakdown of that song and talking about it didn't believe that that could possibly be what that sound was so i actually had a slow enough day that i went and took and i didn't have a u87 but i took a 57 and cranked the gain and demoed and i was like nope nope you can make that sound with a sugar packet 

Chris Tsanjoures: Perfect. Um, the song is Midnight Express by the way. So, uh, but yeah, 

Andy Leviss: I'll go, go, go re listen to that 

Chris Tsanjoures: got like a big, it's got a big like jangly acoustic guitar part that like, it's just like, if it's not synced up, it's just like, it's just such a mess, you know? Cause Nuna Betancourt is such a percussive player. So, um, yeah, anyway. 

Midnight Express. 

Andy Leviss: Cool. Well, and on that note, I know, I know you've got a, a 5 p. m. express you need to go, 

Chris Tsanjoures: Yeah, the kids, the kids are not going to be, you know, they don't, they, they get hungry when they get hungry and it's, uh, you know, 

Andy Leviss: that's an, and, and we want to, to go with another song reference, we want to, your pop culture reference, we want the kids to be all right. 

Chris Tsanjoures: right. Well, hey, I really appreciate you guys having me on. I know it's super short notice and, um, you know, it was just cool of you to, to, you know, have me. 

Andy Leviss: No, I dig it. I always, I always love a chance to, uh, learn and pick in your brain a bit and, and, yeah, it's, it's fun to, like, you know, have this tool that, like, I'm interested in knowing what it was and to get all my stupid questions about it out in front of an audience of thousands. 

Chris Tsanjoures: problem. 

Sean Walker: Oh, now you know what it's like to be me, Andy. Just get all your stupid questions out in front of an audience of thousands. See, there it is. 

Andy Leviss: Oh, I've had my moments of complete dumbassery since we started this show, like, there's no denying that, I just bury them in amongst occasional things that sound smart, so I create the illusion. Cool, well, thanks again Chris for joining us, thanks everybody else for listening, uh. 

Sean Walker: RCF and Allen and Heath. And, uh, that's the pod y'all. See you next week. 

Chris Tsanjoures: Everyone be good. Take care.

Music: “Break Free” by Mike Green