Q-SYS Corner

[Elia Orselli]

Also, one more. None of New cores can't di 96k processing?

I've found the point in the recording about that:
immagine.png

https://www.youtube.com/watch?v=Xk2Vr82tcyE

[Marin Zorica]

I just hope they do not do "easy" way of 96k.....adding 96k processing to new cores, by cutting channel count on half, compared when on 48k!

[Jay Wyatt]

I just hope they do not do "easy" way of 96k.....adding 96k processing to new cores, by cutting channel count on half, compared when on 48k!

I mean it does double one's bitrate so I wouldn't necessarily consider halving the processing count cheating. Theres always bigger cores to size up to, so maybe a studio running ~64channels that'd typically rely on a Nano or core110 would size up to the new 24f. One thing to keep in mind with 96kHz and mega-channel counts is network saturation. QoS becomes more important as more multicast traffic is dumped onto the switch. 384channels (48 x 8ch streams) is ~108% of a 1Gb port's utilization.

[Leo Enticknap]

If 96KHz audio streams are going to strain the LAN and switching infrastructure, then adding networked video to the mix is really going to strain it...

image.png

So at its highest resolution, frame, and sampling rates, a single stream of 2160p HDMI will consume almost a gigabit of bandwidth. If you have dozens of endpoints in use, the worst case scenario could be multiple high end switches, each costing in the low to mid four figures, linked by 10 GBPS fiber LAGs. Even at lesser extremes, Q-LAN bandwidth budgeting becomes a far more restrictive issue when video is involved, compared to a system that just consists of audio and control.

[Steve Guttag]

For sure, the 1G port and typical switch bandwidth is going to become an increasing factor as people (probably not as fast in cinema) start using high bandwidth devices like 96KHz and video. Video is definitely one where you have to pay attention, already. One reason for the popularity of the Netgear A/V switches is that they have a very high bandwidth as compared to typical networks switches. It is one thing to keep track of how much bandwidth you are using on any one device...it is another to keep track of how much bandwidth you are using in your entire infrastructure and that falls on the network switch.

Moving to fiber can address endpoint bandwidth but you need it plugged into a switch that can deal with many devices running and moving that data about.

Whereas the 26f has double the DSP processing of the 110f, I guess we'll wait an see what impact 96KHz has, if it happens during the product's lifetime.

[Harold Hallikainen]

Is 96 kHz audio worth the effort? As I recall, only one movie has been released with 96 kHz audio (from Sony, I think). USL processors did all the DSP at 96 kHz, but the output was typically analog audio. When they did have a digital output (either AES2 or BluLink), the sample rate on the output was 96 kHz. Is processing and distributing the audio at 96 kHz worth it?

[Steve Guttag]

Harold, Q-SYS doesn't factor cinema into their planning anymore (or if they do...it is WAY down on the list). The A/V people have been complaining about the lack of 96KHz for years now. They are interfacing with 96KHz equipment all of the time and now Q-SYS is the bottleneck.

I'd be more than willing to do a shootout of 48KHz versus 96KHz in a theatre or most any other space to see if ANYONE can not only tell the difference but can accurately identify the 96KHz one at at rate of 75% of the time or better. That aside, I've definitely heard the complaints from a LOT of people in the A/V world. If Q-SYS doesn't offer a solution, they might find themselves excluded from those systems.

It's kind of like the Dolby CP950 not having multi-channel analog inputs. For many/most systems, that is a non-issue. But, if you need to support multi-channel analog devices (a film sound processor, for instance), that just eliminates the product from consideration as it does not meet the minimum feature requirement.

[Leo Enticknap]

One reason for the popularity of the Netgear A/V switches is that they have a very high bandwidth as compared to typical networks switches.

That is reflected in their price: roughly between a grand and $4K, depending on the model. If you want 10 GBPS SFP slots, the PoE variant needed to run the more power hungry Q-Sys video endpoints, and/or switching throughput capacity north of 60 GBPS per switch (about 12 streams of Q-Sys encoded HDMI at the maximum resolution, bandwidth, frame rate, and chroma sampling), you're looking at the upper end of that range. If you're used to buying switches for a system that just handles Q-LAN audio and control costing in the low to mid three figures, you're into a new level of budgeting process once networked video enters the equation. Even a significant number of 96KHz audio channels without any video could, as Jay notes, make the switches that have traditionally been used in Q-Sys installations no longer up to the job.

[Marcel Birgelen]

I'd be more than willing to do a shootout of 48KHz versus 96KHz in a theatre or most any other space to see if ANYONE can not only tell the difference but can accurately identify the 96KHz one at at rate of 75% of the time or better. That aside, I've definitely heard the complaints from a LOT of people in the A/V world. If Q-SYS doesn't offer a solution, they might find themselves excluded from those systems.

Nobody can. Period. This has been proven so often, it's actually ridiculous we're still having this discussion in 2025.

In a recording studio environment, those high-bitrate audio streams make perfect sense, as they give you the resolution to "manipulate time" without any noticeable loss in quality. In a "playback only" situation, any money spent on this is as wasted as using it as fuel for your bbq.

As for 10G networking: We're coming to a point where 10G at the edge becomes economically feasible for many solutions. Prices of switches have gone down significantly over the past years. One of the final issues is the extra costs for the optics/SFPs. Those can be pretty expensive, especially if you solely rely on manufacturer branded "known to be good" optics.

As an alternative, 2.5G is also picking up. The problem with 2.5G is equipment support, but it works well on most existing UTP/STP wiring. Doing 10G over existing copper/aluminium wiring has been a hit and miss for me, so much that I generally want to avoid it.

In the age of SSD-powered integrated servers, having 10G between machines makes sense, even more so than for just audio transport.

[Steve Guttag]

As Leo notes...when I do a "typical" Q-SYS system that is just audio and control, my switches are in the low-mid hundreds of dollars. I've been using TP-Link stuff with success. However, at a site where I'm doing video for an Entertainment Center type infrastructure, the switch that is transporting that content was about $2,800 and a Netgear.

One isn't rewarded for having multiple layers of switches in their topology, when doing video. For a typical cinema where I have one Core running multiple screens, i'll typically have a pair of 24-port switches in a "Central Audio Rack" with 8-port switches in each theatre to handle the amps, DCIO-H and touchscreen. If video was involved, that would change things because you are no longer merely moving a 1G bandwidth (max, and really, a lot less) between the two. You'd start to want 1-2 10G fibres between the two, depending on what is going on in each theatre, video wise. Even if you are using multiple switches in a central location to get your port count up, you are going to want to keep the bandwidth between each switch high with, again, 2-4 10G fibres. Things grow very fast and the cost grows geometrically. In the Entertainment Center I mentioned, I opted to get a larger Netgear switch and used a pair of TP-Links on the audio-only and control stuff to preserve the expensive ports for those needing the bandwidth. So, touchscreens and the amplifiers, instead of consuming 10 or so NICs on the high-dollar switch are sitting on relatively low-cost switches so more NICs were available for the video endpoints. It actually worked quit well. If I had to support everything on Netgear's A/V switches, it would have significantly increased the cost for switching and wouldn't have improved the user experience.

[Harold Hallikainen]

Nobody can. Period. This has been proven so often, it's actually ridiculous we're still having this discussion in 2025.

Thanks! The Nyquist frequency with a 48 kHz sample rate would be 24 kHz. So, the highest frequency that could be reproduced would be somewhat under this (depending on the low pass filters on the record and playback sides). ST 202 lists 16 kHz as the highest frequency in the X - curve (16 kHz is down 11 dB). I think of the X-cuve as a pre-emphais / de-emphasis system (similar to phonograph records, magnetic tape, FM broadcast) where the de-emphasis is supplied by the perforated screen. Pre-emphasis is provided on the record side by adjusting the equalization so the audio sounds good through the perforated screen on the dub stage. Do most theaters meet the X-curve specification at 16 kHz (+3, -6 dB)? Does content include 16 kHz? A 48 kHz sample rate should be sufficient to transmit 16 kHz.

So, how about the audio processing at the theater? Is processing at 96 kHz audibly better than processing at 48 kHz? As mentioned before, USL processors had sample rate converters that converted whatever the incoming sample rate was (typically 48 kHz) to 96 kHz for processing. The output was either analog (most common) or 96 kHz AES3 or BluLink. Even if the DSP sample rate were 48 kHz, we would have used sample rate converters on the inputs to comensate for the slight variation in sample rate between the 48 kHz coming from the server and the 48 kHz in the sound processors. We did not want to have to sync the sound processors DSP to the incoming AES3 clock. The sample rate converters were from AKM and, I believe, are no longer available due to the AKM fire. They were great chips since they were a DIR/SRC that decoded the AES3 and then did the sample rate conversion, all in one chip.

Anyway, does 96 kHz processing (typically DSP filters) sound better than 48 kHz?

[Marcel Birgelen]

If with Audio Processing you mean A/D and D/A conversion, then common wisdom has shown that "higher is better", because your filters simply aren't perfect. So, upsampling a 48 kHz signal to 96 kHz, because your filters simply aren't perfect and might otherwise cut into the audible sound spectrum, sounds like an acceptable solution to an "imperfect analog world problem". Still, this avoids the unnecessary bulk of carrying the entire 96 kHz through the digital circuits.

It also depends on whether you're designing a system for humans or bats. The number of people that can hear anything beyond 20 kHz is... well, minimal. So if we keep 20 kHz as a "safe limit", any filter would still have roughly 4 kHz of headroom if processing at 48 kHz.

There are people that claim that humans can "register" frequencies up to 50 kHz via bone conduction... That's nice and great, but I doubt any recordings are created with those aspects in mind... Maybe some dance and rock records are produced with "ground conducting bass effects" in mind...

As Leo notes...when I do a "typical" Q-SYS system that is just audio and control, my switches are in the low-mid hundreds of dollars. I've been using TP-Link stuff with success. However, at a site where I'm doing video for an Entertainment Center type infrastructure, the switch that is transporting that content was about $2,800 and a Netgear.

Yeah, a 24-port 10GE switch will still cost anywhere between $2000 for some no-name brand and $3500 for some more common brand as long as it's not Cisco, which still go at premium rates, even with 80% discount on list-prices.. But it's way down from a few years back when a switch with similar port density would cost you upwards of $10K~15K without optics and without any fancy options like redundant PSUs.

But I've seen no-name brand manageable switches with 8 10GE ports for under $150 recently. At those prices, 10GE starts to become attainable for small businesses and home lab usage. I would be reluctant to deploy such switches into anything that requires some serious uptime though.

[Steve Guttag]

Q-SYS For Cinema
Blog-15, QDS–Part-11, Sample 7.1 Part-10, Logic and Control Part-4, Booth Monitor

Current QDS Versions: 9.13.0 and 9.4.8 LTS.
Sample 7.1 design version: 4.3.0.0

Introduction

This blog will pick up right where I left off on blog 13 (Format Selection) and 14 (linear versus dB fader) and continue through the Sample 7.1 Design, starting with the booth monitor.

Disclaimer

If any of the content in this blog happens to show up in a Q-SYS exam, it is not my intention to provide an answer-sheet beyond the discussion of good practice. I have not seen any form of the cinema final exam (my Level-1 was before there was a cinema version).

Disclosure

I do not, in any way, work for QSC/Q-SYS. These thoughts are my own based on my own interactions with the product(s) and implementing Q-SYS within actual cinema environments. I do work for a dealer that has sold QSC products since the 1980s, including Q-SYS and its predecessors. For the purposes of this blog, I represent only myself and not my employer(s) or any other company.

Monitor Level and Control
Blog15Image1.png

So, what does the “Monitor Level” do? Do we need/want to change it from a dB control to a linear one?

In reverse order, no there is no need to change the fader type. In fact, there is no need for a unit/number to the level at all. It is a booth monitor. There is no reference for it. Just louder/quieter. It is my philosophy that, when it comes to UCIs and other user interfaces, minimize the amount of information so that what remains can be of higher importance.

But first, let’s see just what the fader controls. How are we going to do that? Our friend <CTL-F>. Select the fader and hit <CTL-F>

Blog15Image2.png

Let’s jump to the source of the control.

Blog15Image3.png

As we discovered from a previous blog, it is the SPKR level control within the DCIO-H. Anyone remember why I think that is a poor choice? The DCIO-H is a Left Side Pane (LSP) device so it doesn’t copy/paste between designs. If we duplicate this design, we have to go through the process of replicating it each time.

Let’s improve this. First, expose the “Amp Gain” pin (it is SPKR internally and Amp externally…cool, huh?). Also, expose the “Amp Mute” pin. Then put a suitable control on that.

Blog15Image4.png

Since this is a booth monitor, the “Custom Control” I used was a “Level fader w/taper (dB)” as well as a Mute button.

Blog15Image5.png

My preference, instead of controlling the DCIO-H, is to add another Gain control. Then use the Level fader w/taper (dB) to control that.

I’m thinking, something like this:

Blog15Image6.png

This gets us out of controlling the DCIO-H (with respect to the monitor). Using a Gain component also opens up options of using ramping controls, which would be handier for a monitor than fixed adjustment amount controls (String Buttons). I also like the Level fader w/taper because it allows the fader’s range to be more useful than a straight up dB fader. As discussed in Blog 14, most of the useful range of a fader used for volume is towards the top of the range so putting the detail there is beneficial. The taper (audio taper) version helps with this.

I’m going to drag that fader out of the custom control and put it into the design:

Blog15Image7.png

Alright, we have fader, of sorts. But just what are we listening to? Well, up and to the right (in the schematic), is the Booth Monitor Router.

Blog15Image8.png

There is nothing fancy there. It is just a router so we can only select one thing at a time. Fortunately, if you recall from a previous blog, they supplied an L-C-R mix but still, you can only choose what has been provided. I added the Mic into the monitor system because I injected the Mics after the Format selection so you could have both Mics and other sound active at the same time (think Power Point presentations).

So, how does this show up on the UCI?

[Blog-15, Page 1 of 3]

[Steve Guttag]

image.png

I sure hope that fixing this UCI’s presentation is part of the test. As such, I don’t want to go into too much but this is sort of a rough UCI, in my opinion. The “Input” buttons are just the buttons from the router copied out and placed/sized on the UCI. The “Loudspeaker” button is button-10. It is placed with the “Listen” buttons in the Loudspeaker section.

I suppose I should bring my “Mic” button out to be consistent. And while I’m being consistent, I won’t space it uniformly either.

Let’s get rid of the text box on the level (what do we or, more importantly, the user, care what dB it is and what is it in reference to?). I’ll bring my custom control fader out to the UCI so the user can quickly drag the level. Note, being able to operate the fader with your finger may seem like a cool idea, however, often, you don’t have the precision on a UCI to do that with any accuracy. But, on a booth monitor, there is no need for accuracy. The user is just going to set the level to their preference.

Based on experience, I’ll bring the “Clip” LED out since the DCIO has a rather low-power amplifier. If they are judging distortion, you don’t want the monitor clipping to be causing the user to think there is a problem in the theatre. To size/shape the Clip LED, change its presentation to a “button” (which, basically, means, present it as a rectangle; it doesn’t mean that it is a pressable button). Then, I merely added “Clip” text to it, for clarity.

Blog15Image10.png

So, without really changing the UCI, that is where I’d leave this one. If you were to want a person to actually use it, probably some grouping, button arrangement, color coordination should be considered.

Monitor Channel Selection
As with the other blogs on the Sample 7.1 Design, I have my preferences on this one too. I’m not a big fan of using a Router for monitor selection. This is the DSP equivalent of using a rotary knob to choose which channel to listen to. What I’d rather be able to do is to mix any/all channels of interest at once. The key word there was “mix.”

Adding the mixer isn’t a trick. There are 9 inputs and 1 output. We can copy the inputs over:

Blog15Image11.png

You can’t copy the output (if you want to leave the original router) because they have to be unique. So, if you are going to switch over to a mixer, then you can cut/paste the Signal Name over too. You may have also noticed we went from 11 inputs down to 9. We don’t need the L-C-R mixer since we can do that in the monitor itself. We also don’t want the amplifier output to be part of our mix but we will have to contend with that too.

So, how do we implement a mixer-based booth monitor? With the router, we had those handy buttons to copy/paste over. What do we have in the mixer that we could use (and don’t forget to label your inputs and outputs as well as set your gains to 0dB)?

Blog15Image12.png

It’s the Mute buttons.

So, lets copy the monitor UCI page to a new page. Just select everything and <CTL-C> and then paste it to the new page <CTL-V>. If you press and hold the <Shift> key before you click, it will snap everything in the same place as it was on the original monitor page. The HOME and METERS buttons are “Navigation Buttons” and have a trick to overcome before they can be duplicated but since that may be part of the exam, I’ll leave that for you to discover.

We’ll want to get rid of the original Router buttons and set aside the “Listen” buttons. Then copy the Mute buttons from our new mixer:

Blog15Image13.png

The problem with using the mute buttons, as they are normally used, is that they are not intuitive for a monitor selector. They light when they are not part of the mix. So, if we select them (all) and go over to properties, change the “Reverse Action” to “Yes” and then they’ll function as one would think (lit when part of the mix).

Blog15Image14.png
Blog15Image15.png

If the client has a color theme to their company or facility, those colors should be incorporated into the colors of the UCI. I would probably start to rearrange things a bit and group them more in logical order.

[Blog-15, Page 2 of 3]

[Steve Guttag]

Remember your intended user. For instance, Left, Right, and Center are how the channels enter the audio path but they are not how a user would think of them. They are Left, Center, and Right.

Maybe, something like this:

Blog15Image16.png

So, what to do with the amplifier outputs? How about, add a router that selects between processor and amplifier? Then, use the router to not only choose which source to listen to (processor or amplifier) but also switch layers on the UCI. It’s just a thought as there are many ways to solve the problem.

How about this:

Blog15Image17.png
Blog15Image18.png

The schematic would look something like this:

Blog15Image19.png

They are pretty minor changes, really. Just a couple of “clicks” here and there but I think it greatly improves the user experience. Given the amount of space, I’d probably also enlarge the buttons. On a touch panel, things can get small and we have plenty of space, once we switched to using layers for each type of monitoring.

The UCI Layer Controller I added turns the visibility on/off of the layers of interest and the same router that is choosing the audio is choosing the layer. So, the booth monitor has three total layers. The background layer (always visible) that is used for both types of monitoring and the two monitor layers. I also added the text to let the user know that when monitoring amplifier outputs, they can only select one channel, unlike when monitoring the processor. Using different colored buttons also helps the user know that they are monitoring different types of signals.

Blog15Image20.png

Conclusions
There is a lot more leeway on how to implement a booth monitor. I think it will greatly depend on what environment the system is in. It may be sufficient, for instance, to have a mere on/off switch with a basic L/C/R mix to just know that sound should be playing. Most of the time, in a typical cinema, there is nobody to listen to the monitor. In a screening room, you may need to provide quite a bit of control/detail so the operator can adequately hear/diagnose the sound. In our theatres that still support film, we will often have more robust monitoring system, including well-amplified monitor speakers (by the view ports) so that the projectionist can hear the soundtrack when making changeovers.

©2025 by Steve Guttag

[Blog-15, Page 3 of 3, End of Blog]