Why I Quit Using Aux Fed Subs (And Do This Instead)

►► Get my audio math survival spreadsheet found in my audio toolkit:
www.producedbymkc.com/audioto...
I’m going to come right out and say it. I think aux fed subs are the wrong way to manage your low end. I’m not saying you should forego control of your subs, but an aux or a bus is the wrong tool for the job.
Today I’ve got four reasons why you shouldn’t use aux fed subs, then I’ll walk you through how I set up my console and system processor to handle my low end.
My name's Michael I love helping you level up your audio and business skills so you can build a career you love.
We’ll cover:
- Why the mono bus (as well as auxes/buses) set your low end up for failure
- What bad habits we’re borrowing from film world
- How your brain has to think twice as hard about mixing low end with an aux-driven setup
00:00 - Intro
01:20 - Clarification, Defining Aux Fed Subs
02:29 - Reason #1
07:01 - Reason #2
07:58 - Reason #3
09:24 - Reason #4
11:39 - How I Route and Control My Subwoofers
14:56 - Conclusion

[@MichaelCurtisAudio]

My updated take on the topic: kzbin.info/www/bejne/raK7h4ijidtnn7s

[@BarnabY07]

Hey Michael, I saw your update. But I didn't understand how we can apply the philosophy of Dave Rat (sub on separate send) without using an aux. Maybe using a group feeding a matrix dedicate to the sub. That seems very similar. Thanks for sharing your knowledge.

[@jeremynothman9512]

Thanks for this article. I have been constantly wondering about this since working in live sound and tend to be of the same opinion. I've done it each way, but it really does seem to me (coming from the studio, and thinking from a translation perspective - be it for streaming/broadcast or for a feed to another room) that it's better dealing with the PA as if it were a single full-range playback system. I can understand the business with the transfer function, but even working in a 1000 person capacity room where visiting engineers often use SMAART, I really haven't seen them monitoring it during the show. They're just using their ears then and working the mix, or EQ'ing the master GEQ a bit to compensate for what they're hearing.

[@lfisher8154]

Great thoughts. Related to #4 I would also add that when designing systems for permanent installs like churches, schools, community theater and similar venues where you will have a wide range of mix engineers, including volunteers, using the matrix approach and a single fader will be much more user friendly and keep them from getting lost. You set the system up properly and turn over the system with confidence.

[@MichaelCurtisAudio]

Great point! Yes, I do feel like it's much more intuitive than having to worry about all of your send levels.

[@steve01010]

It's like a lot of things in audio: it's one way of doing it. I've found over the years that it's better to stay away from holding an opinion that states someone should never do something. Some of the reasons I love using aux fed subs is that I don't have to HPF nearly everything. I don't want acoustic going to the subs and I don't want a HPF on it. I also love having the ability to give the subs more level at lower volume to compensate or giving them a bit of a boost to emphasize some moments. Yes, the acoustic crossover changes, but while system design is linear and analytical, running sound is not. There is room for some of the artistic opinions and if no one gets hurt, then go for it.

[@MichaelCurtisAudio]

Thanks for weighing in here, Steve. You're totally right. It's one way of doing it and you're more than welcome to run your shows how you see fit. And you probably get great results for how you put together a mix.

[@steve01010]

@@MichaelCurtisAudio Great content by the way, even if I still like aux fed subs lol. Keep up the good work!

[@MichaelCurtisAudio]

@@steve01010 Thank you very much! I appreciate you being able to hold the tension of "sure, we don't agree on everything", but still appreciate what else I have to say. Glad it's been helpful to you.

[@__Nata_]

Can you not just get a sub with a hpf?

[@FathomSoundProAudio]

@@__Nata_he’s saying he doesn’t want a HPF on the acoustic instrument (and many other inputs), which usually cuts off most content going to the subs (though the HPF on most digital consoles is adjustable).

[@inkrebel1136]

Sick theory. That comparison to surround sound was brilliant. Thanks for sharing!

[@MichaelCurtisAudio]

Thanks!

[@aaronduesterhoeft4866]

I try to maintain an open mind. I do typically run my subs aux style. However, I do like the concept of this approach and the benefits that could come out of it with regard to tonality, TF translation and stream balance. I might have to try it out at my venue! Can't knock it till ya try it! If it doesn't seem to work, it's easy enough to switch back. Keeps ya fresh!

[@MichaelCurtisAudio]

Give it a shot and let me know how it goes!

[@chacano1]

Great video! I would suggest to handle the proximity effect of the vocals in a more "elegant" way, to use a dynamic EQ or multiband compressor on your vocal group. This way you only trim down those muddy frequencies when proximity effect comes in play. (I do understand this tool is not always available in every mixer)

[@MichaelCurtisAudio]

Yes, great input here. Dynamic EQ is a more elegant solution, for sure, especially if the vocalist is inconsistent with their mic placement.

[@jen3800]

this makes perfect sense to me. never stop challenging norms . it's how we grow. there will always be people who say outside ideas are invalid. even f they eventually are, trying new ideas is how we grow. i love this channel. one of the most eloquent educators out there. i wish i had your savvy at your age but i'm a late bloomer. by the time you are my age you will be top notch, no doubt.

[@MichaelCurtisAudio]

Thank you for the encouragement!

[@centralaudiofl]

Hi Michael, First of all, thanks for your awesome videos. They share a wealth of information and straightforward approaches. Wished I found your channel way before!. I pretty much agree on your aux for subs philosophy. Especially when you have system processors that make all the division and parameters from within by just simply feeding a LR signal (like old 2-5 ways analog x-overs). Just wanted to share that I use a Yamaha TF5 which has a Sub fader. It conveniently fits to the system processor which has L+R+ Sub inputs. Depending on the gig, I use the mixer Sub fader unprocessed or LPF it around 90Hz. In addition, when needed, I can bring it up or down to get an instant tonal balance at the FOH if any sudden environmental/unforeseen issues might happen. Streaming or recording are via Stereo Matrix fed from the L+R (all low end instruments like kick, bass, an others are always routed to LR main). Thanks! Abdel

[@MichaelCurtisAudio]

Thanks for the input here, Abdel. Glad you've found my videos helpful! Yes, having your subs on a Matrix or its use Master bus is super helpful, especially it's a new rig or your environment might change. The TF5 is a neat little desk.

[@timschad5810]

Glad to hear you're keeping the vid up. You'll never please everyone all the time. Do you !!

[@MichaelCurtisAudio]

Thanks, Tim!

[@bendadams]

Question here. I've been learning about aux subs and watched this to combat what I've been learning. Mixing at a smaller church. 50x80 room with overflow rooms open to main room (zone 2) I have a external PA processor with only 2 inputs. How do I control different zones levels and subs on the console but still use the EV DC-one processor. You use 1&2. Left/Right. 3. front fills (in my case overflow zone 2) 4. subs. with matrices, but i only have 2 inputs on my processor. Would I bypass the processor to do this and just go straight to amps? Or the processing on the matrices would output the stereo bus to the processor post fader with the sub fader affecting the stereo output? I have new gear on the way and i want to understand this stuff before i route the new console.

[@MrRoberacer]

Thanks for your thoughts. I see one issue with sending a copy of your entire mix to the subs and that is that in the case of vocals you are right in that a recording doesn't have the big bottom on the vocals that many people seem to leave in them in live situations but I think we need to explore why that is a little more. Yes, some folks have not quite understood why we commonly don't use much bottom end in vocals on recordings. That is true but the thing that some people don't understand is that the best way to achieve any specific sound especially in a recording is at the source. In the best recordings the vocalists are a good distance from the mic thereby creating far less proximity effect than with mics on a stage. Heck more often than not even lectern mics are closer to the face of an orator than a common studio vocal set up. Also something to note is that lectern mics are commonly hyper or super cardioid designs because they need the off axis rejection but with that the trade off is that they suffer from excessive proximity effect. Some consoles offer a -24db/oct cut but with that if you look at the curve you will note that above the cut frequency they also ten to boost information which is most commonly not a desirable effect either. While using the crossover as a low cut is not going to fix your recording it can at least make the house mix more usable when using substandard quality mics (we commonly record with mics costing in excess of $5K. Stage vocal mics are commonly some of the cheapest mics in the box. While the cost of a given technology is not a direct correlation to sonic quality it does in fact factor in.) An alternate to correct for the record feed which I mostly do from stereo matrixes also is to run the vocal through a buss that is doing some of the global corrections for the vocals/ for example commonly engineers spec the same mic for all of the vocalists who are performing together. There are some good reasons for that for those who don't understand but that is another topic altogether. Suffice it to say that undoing the "58 curve" is something you can do as a global setting on a subgroup while using your channel eq to fix specific issues for the various singers as well as some consoles allow you to low cut in both sections or at least "low shelf" the bottom. Also inserted eq can be your friend there as well.

[@MichaelCurtisAudio]

Great thoughts here! Thank you for sharing. That all makes a lot of sense and I totally agree that the proximity effect becomes much more of a bear in live sound. Yes, I a great "compromise" approach is to have a Band subgroup that passes through to the LR unaffected then a Vocals subgroup that has any needed low end shaping so you don't have too much rumble. I can see how that's especially effective if you're mixing monitors from FOH and the lead vocalist doesn't want to hear their voice too kneecapped on the low end if you don't want any excess rumble in the PA.

[@08_crown_vic]

Use the mix minus method, to control only sending lows to the subs using aux pre and post auxiliaries

[@artmathias9725]

@@MichaelCurtisAudioBringing up the subject of fold-back is a whooooooooole different topic, LOL.

[@Sound.Church]

I really don't think your tone was bad at all... thank you for the info. I really agree and enjoy your content, even when I disagree. Keep it up!

[@MichaelCurtisAudio]

Thank you for the support!

[@mirarzaamplification902]

This is what I'm thingking too. If a reference music can go to a PA without mudding the vocals that means we can do it in live also, its only the matter on how to use your eq on your channel.

[@MichaelCurtisAudio]

You got it!

[@janorolin8622]

Thanks for the video but how to propertly fed subs when I don't have any matrix possiblity right on the mixing console? Is there any other way than aux fed subs or using external crosover processor?

[@levifoe7967]

Helpful and interesting ideas here. Thanks man!

[@nathanlively]

Makes sense to me. I think most people arrive at the same conclusion considering these questions. Any reason I can find to make things easier.

[@montelcarlos143]

Thanks to both of you gentleman for guidance🙏

[@66fitton]

I still just use a crossover lol. I will explore this method for sure. I've tried aux fed subs and just went back to an external crossover because I was constantly "balancing" the system. This would eliminate that👍 Oh, and I always MUTE the walk-in music channels right after I fade them out now. I had the same thing happen lol Great video! Cheers

[@MichaelCurtisAudio]

Thanks a ton, Steve!

[@artmathias9725]

If you have a multi-speaker system, you will still need to use a crossover, whether it is passive or active. Now-a-days, it is mostly programmable amps that take care of crossover functions. If you are still running analog "dumb" amplifiers, you will need an active crossover to run your speakers correctly. You said "I've tried aux fed subs and just went back to an external crossover because I was constantly "balancing" the system." This is kinda confusing. Without the external crossover, subs on aux doesn't work in the first place. Unless you are saying that you do something like low-passing the buhjesus out of your sub bus, and then send that signal straight to your amps. That is the only other way that I can see this happening, and that is just silly talk, lol. Shout back if I confused you, that ability seems to come naturally to me, lol.

[@HAZZASIB]

For me I run aux fed subs. I set the subs up so they are roughly the same volume as my tops and so I can have as much of an even frequency response across my PA. Once i've done that I only send things to the sub at unity and then forget about it (and also link the PA master LR with aux/mono. I use the aux feed as more of an on / off switch rather than a 'how loud I want the bass' switch. I like this as it acts as an immediate high pass and saves more energy for things that are routed to the subs. I do agree you can get caught up in thinking about routing more than actually mixing the channel but for me its a 'set and forget'.

[@MichaelCurtisAudio]

Glad that approach is working for you!

[@billschnake6378]

Michael, I have done this both ways. On very large shows I have done the Left/Right to 4 matrixes, Left/Right/Sub/Fill. We always do it in that order. We do get the request to set it up Left/Right/Sub/Fill for many of the artist we support where we are doing a Aux for sub and Matrix for fill. I myself prefer the way that you suggest and as you said if we need to do a live feed for recording, we can just simply pick another Matrix and send it to the live record feed. We did this at a corp show last week. As I said, I prefer to do my feeds as you suggest, however I have clients who prefer the aux method and when they are paying the bills we do as they wish. Nice video.

[@MichaelCurtisAudio]

Thanks for sharing, Bill! Yes, totally agree that your system setup must be flexible enough to accommodate other engineers who run their mixes with an aux send as well as fills.

[@gracenotes5379]

Good points, well made. But how do you respond to those who would say that a typical mixer channel strip HPF doesn't have sufficient rejection to keep microphone handling noise and other unmusical LF content out of your subs? (I've heard it argued that you would need an ultra-steep HPF slope to properly eliminate the unwanted microphone LF, but even if you had such a sharp HPF available, you would risk excessive group delay distortion near the passband edge.)

[@robroilen4441]

I'm the engineer at a relatively well known venue and the only touring engineers who show up and fiddle with some alternative to aux fed subs are younger than 40 and haven't been doing it for very long. On the flip side, every single 40+ engineer uses aux fed subs. The only real difference at the end of the day is that the younger engineers are all stressed out and having a bad time because they've inevitably overcomplicated things in an attempt to... not do it the way that's worked for decades? All the routing tomfoolery in the world won't make up for a fundamentally bad mix, and typically these engineers end up wasting valuable time when I hand them Left + Right + Sub drive lines that just go straight to the amp rack and it doesn't jive with their Proprietary Mix Magic. Of course, to each their own and there are myriad ways to run sound, but the older guys aren't doing it because they're luddites, they're doing it because it's simple and effective.

[@MichaelCurtisAudio]

I'm 1000% with you that this pervasively overcomplicated mix trend, especially with young engineers, is debilitating and unhelpful. I've seen too many talented young folks get so bogged down in miles long Waves channel strips on the hi hat that they forget to ride the lead guitar for the solo. Rant over : ) A good mix is a good mix, full stop. I'm not trying to stir the pot and say that these veteran engineers can't get great results out of an aux fed rig, I'm just saying with my workflow I prefer using matrices and here's why. That's all. If I'm system engineering for a FOH mixer who wants to send LR+Sub, they're more than welcome to.

[@fearofnone2]

I am running in the next 18 for local shows since it does not have any matrixes. How would you suggest that I control the lower end not using aux fed subs?

[@roberthaley1]

I have tried this (matrix to sub) and aux or mix. Correct me if I’m wrong, sub send on a matrix; you are sending all content after L&R to the matrix. Are you creating a crossover? I mainly work in the corporate AV world.

[@MichaelCurtisAudio]

Hey, Robert. Great question! Yes, all channels you want going to your full PA go to your LR mix, then that LR mix goes to your mains, subs, and fill systems matrices. I usually run matrix 1/2 as mains, 3 as front fills, and 4 as subs, then use more if i've got side hangs, delays, etc. You would apply the EQ, crossover, and time alignment processing on the matrices, so that when they acoustically sum in the space you get the results you want.

[@toddmoore9138]

When I first heard of the subs on aux idea, I had the same thoughts as you are expressing here. Thanks for confirming what I've been thinking about this for some time. Good video.

[@artmathias9725]

I have been a concert engineer for over 20 years now, so I live by the motto of not telling another engineer how he should do things, just like I wouldn't tell a musician how to play HIS instrument. That said, I feel that some of the things you mentioned were made for the wrong reasons. I am neither an electrical engineer nor a comp programmer, so I don't know the exact physics of what is going on under the hood. But, there is a stereo summation from Main LR, that gets decorrelated when moving to Matrics. When I was first shown the difference, I had to go back to my shop and experiment for myself. With a 2 tops over 2 subs a side, stacks, I wired 1 top to 1 sub on each side with one amp system, while hooking up the second speaker set to a like amp system. Being able to A/B the two set-ups with one being fed main bus and the other from matrics. There is a noticeable fidelity change from one output to the other. As for the lack of full range on your main L/R bus, when using subs on an aux, your left/right outputs are still seeing full range, and anything that you matrix off of your L/R is still full range. Your sub is only affected after your system crosses over at its designed frequency, With that being said, your sub send will also send a full range signal to the sub channel of your amp, where at that point, your system will cross over, filtering all other frequencies from that full bandwidth signal. With proper "system" gain structure, there should be no audible change from a full range system to a 'subs on aux' system with channel aux send at unity. You shouldn't have to treat your channel equing any differently than mixing in full range. It isn't about having more steps to do. It is about having less to do! In most "Band" set-ups and sound, it is easier, and quicker to dial that sub aux when you only have a few instruments that you want to go to your subs vs. having to run a high-pass on 20 other channels to get them out of the sub bandwidth. This also has added benefit to your amps, being that they have to work less when you don't have all the summated noise floor that every channel ads to your outputs. The sub drivers will also become tighter if freed up from having to try to reproduce sounds from the entire mix, just like the example of the pre-fader aux send that is missed on your source music channel, you will still be able to recognize a song due to the whole stereo mix of the source channel. "if you send it there, it will try to produce it", best option is to only send what you want to exploit that frequency band, i.e. kick drum, floor tom/low tom, Bass guit, keys, and other instruments that live in that bandwidth, your overheads, for example, should be as far away from your sub bandwidth as possible, even if high-passed. I also do that to front-fills when we are playing a smaller venue. It is only the channels that get lost over the stage volume, i.e. vocals, acoustic instruments, etc, that I feed into front fills. This only a cursory explanation of why I use subs on aux and not matrix or mono. I am happy to get more in-depth on this, if interested, but this comment is already turning into a book.

[@jonathanjabezlopez6868]

Thank you for the video. 😊 But i have a question on routing it. How do i route it on our x32 in our church. Because LR is on 15 and 16 i used to use subs on mono bus in out 14. Where is the best out to put it in?(matrices) Thank you. 😊

[@MichaelCurtisAudio]

Hey, Jonathan. You can make your matrix outputs any output you need. You can replace outputs 15 and 16 on the stagebox with Matrices 1 & 2. Then use Matrix 3 as your sub matrix for output 14. You're able to adjust what source is "filling" those outputs on the routing tab, then go over to outputs. Then route outputs 1-16 to physical outputs 1-16 on your X32.

[@claybob7729]

Great points. This video made me subscribe to your channel. Excellent and clear explanations.

[@MichaelCurtisAudio]

Thank you so much, Clay!

[@midtowndjs2889]

Can you recommend an ideal setup for a DJ? I only have my mixing console with XLR outs to the amp. A lot of DJs are adding external mixers for more inputs and better control of the mics. They say it sounds better too. I'm wondering if thats really needed for better sub control?

[@MichaelCurtisAudio]

I'd say you can get appropriate control from just a single console.

[@QuicksilverSG]

Aux fed subs are appropriate for hiphop and DJ-centric venues. In those situations you often ride the gain on the subs to suit each particular track or performer, as well as punch up the energy in the room. For live bands, however, it's better to use FOH speaker management via well-tuned external crossovers. That way your master fader controls everything as it should, and there's no need to complicate things with a layer of matrix outputs.

[@MichaelCurtisAudio]

Thanks for sharing.

[@philipjaltenburg8214]

Some of the old crossover/processors in clubs were right in front of the DJ for easy access. Now it seams the "hands off" approach is more common.

[@QuicksilverSG]

@Bogdan Chepurny - For a parallel livestream mix, I run a separate stereo submix that is routed post-fader, and set all faders in the active channels in the livestream at 0db. That way, I can choose which FOH channels are also routed to the livestream, and the livestream channels track the FOH channel faders. I don't use compression on FOH, but have found that livestreams need extra compression to maintain a more consistent volume level.

[@artmathias9725]

Try a multi-band comp (Like C6) or a Dynamic EQ (F6). I do this for every show with pre-recorded source tracks as the main source of music. Most of the time, the source material is mixed over emphasizing the lows and highs. If no multi-comp or dynamic eq is available to you, you can use an analog compressor with a sweepable key frequency. If you insert it onto you sub bus, it should give you the control that you are trying to achieve.

[@davidjunius6869]

Hey, long time aux fed subs guy here, however greatly considering a different method. Recently I’ve been trying to record shows and noticed the mix sounds great but is lacking some low end in the recording, plus I just bought a measurement rig, so I can see how going forward I will get poor results in the measurement. Also, if I use parallel compression in groups, that won’t get to subs on an aux! So now, if I go to a matrix, are you saying to make the sub matrix mix from L/R or from subgroups, and why? Also, what is the benefit to running the L/R from a matrix instead of the L/R outputs? I haven’t used fill speakers yet, but might consider that for some shows in the future and completely understand why those should be on a matrix.

[@MichaelCurtisAudio]

Hey, David. Thanks for considering a new approach even with having your current workflow serving you well for a long time! Yes, signal flow would be subgroups --> L/R --> Matrices for speaker zones. And those zones would be a stereo PA matrix for left and right (assuming you're running left and right on a gig), a mono matrix for subs, then other matrices as you see fit for front fills, delays, lobby/overflow rooms, etc. I like running it this way so you can make sure every zone in your gig gets the same mix. You're not having to worry about mixing differently for each place your mix might end up. I like having the L/R mix always flow through a matrix for a few reasons: first is that I still have a quick way to turn down the ENTIRE rig with one fader, assuming that every speaker zone is fed by a matrix that is downstream of my LR. I can also apply global EQ to the LR bus for mix shaping purposes, then reserve the EQ's on the matrices themselves to shape the speakers. It helps separate out the "am I doing this to make my MIX sound better, or make the SYSTEM sound better?"

[@samo071112002]

Very good video I'm using aux fed to fed my sub, but my livestreams sound weak but the room sound very good. Now where should I plug my subs, strait from the main speaker or else where.

[@JamesEatonMedia]

I’ve run subs in an aux for so long. I’m used to some of the “oddities” like walk in music and things that are hitting those subs differently. BUT. I have been watching your stuff and I am trying the LR to Matrix mixing for some clients to see how it works. I’ve got a room to test soon and I’m rebuilding their X32 for matrix feed and I will try the subs this way and A/B it to see how it works. I also see the benefit from a mixing standpoint. You just mix faders at the end of the day. All those tweaks should represent a whole system that sounds good. Plus it cleans up your LR mix for reference and review. When I think about training new people to mix, I can totally see this being a great choice. It’s more wholistic!

[@MichaelCurtisAudio]

Thanks for considering this new approach! Let me know how it goes.

[@keisenb]

Thanks for sharing. This was really helpful!

[@MichaelCurtisAudio]

You got it!

[@Mayoman7]

There are a lot of considerations when it comes to routing signal through your console. It not only depends on the capabilities of the console (some can't route groups to auxes and some can't route inputs direct to a matrix, or some don't have matrices), it also depends on what the PA is, how it is setup physically, and what system processing is in place. My first preference is to use group-fed subs so I don't have to worry about getting pre-post incorrect. The channel processing is still full range so my LR buss is still getting all the same low-end information which is getting recorded. It's up to my external system processor to handle the crossover between subs and mains. In some cases, if you have flown mains and ground subs, you may not wish to high-pass the mains anyway. In that case, the subs are definitely just icing on the cake and need to be treated more alike an LFE. If I really need the subs to carry the whole PA, then maybe the approach changes. I strongly agree that variable sends to subs is a straight up terrible idea. I'd still use an aux in a pinch but it would only ever be unity sends and let the system tech do his thing (I'm also primarily a system tech, so there's that). On an M7CL, you can't have groups, only "fixed auxes" which is the same thing anyway. I sort of don't care about the transfer function reasoning - I think you should mix with your ears, not your eyes. I've never mixed live with a constant TF and don't feel the need to. I still argue that with group-fed subs, the TF won't be completely wonky, it'll just appear to have a low-shelf boost because, again, your LR buss still has a copy of the same information as the subwoofer group. Lastly, I strongly believe that high-pass alone is insufficient for low-end management. The HPF frequency is the 3dB down-point and the roll off is still a curve. A 4th order LR filter at 100 Hz is only 6dB down at around 80Hz. For every non-high-passed mic on stage, you're gaining 3dB. If every mic on stage is only high-passed at 100Hz, you've now added enough low-end energy back into the system to fight your crossover or high-pass frequency set points. It's still totally possible to get an 80Hz or 100Hz ring in this setup. You got the point about a studio track through a PA incorrect. Once a 2-track is printed, the high-pass filters used on each channel become printed and there are no open mics. Microphones feed back, not speakers. Everything in the track will go to your subs, yes, but that's only if the information is in the track and that's why we use a system processor to low-pass the subs.

[@MichaelCurtisAudio]

All great points here. Like you've mentioned, the routing and processing available to you on each gig will dictate how flexible the system can be. I've personally found the high pass filter alone on my live music gigs to be sufficient, but on this most recent corporate gig of mine when I ran my LR Bus as the source for all speaker destinations I had to do more LF ring out on my lav group than I usually do since the PA tilt had a significant LF tilt in the PA because there was also a live band. So there the HPF alone didn't work, which proves your point. All things considered, a variable aux alone I don't think is a good idea, but I'm starting to change my tune a bit about strictly sending the LR mix into the processor and having the processor ONLY manage the LF. I'm actually in the middle of rebuilding some of my corporate console file templates and will likely be sticking with LRSF and taking my lavs, handhelds, and podium groups out of the subs. I appreciate your feedback.

[@henryvanschalkwyk2133]

I like to run aux fed subs because, i use 2x dual 18 subs in the middle of the stage, then i have 2x single 15/18 subs on the side with my Array tops ontop for my LR mix. Then i send the outside mic (for low-end only) of the kickdrum to the dual subs(aux fed) and the inside of kickdrum mic with a low cut of 100hz so i get mostly attack to the LR system. then bass guitar and the rest of the mix also to the LR speakers. Is this an effective way to get separation of kick and bass guitar, low-end spread and a cleaner sound? what issues may occur?

[@MichaelCurtisAudio]

Although I like your creativity, I think this overcomplicates things a little bit. Your technique borrows some elements from mixing on a front-field immersive rig. Although immersive mixing let's you physically separate out where audio sources are coming from, every major manufacturer I'm aware of with an immersive platform still recommends all low frequency heavy sources come from the same center sub cluster. You've probably heard of Dave Rat double micing sources and panning them out left/right to get separation. I know you're not double micing EVERY source, but it is common to have an In and Out mic on the kick like you're doing. What I think is worth investigating is how much actual correlation is occurring between those two mics. You can only mitigate comb filtering when two sources arriving at the same point are either decorrelated OR greater than 10dB in level from each other. Given that, I'd run some tests to see if the kick In and kick Out mic are actually all that "different". Hopefully that makes sense!

[@henryvanschalkwyk2133]

@@MichaelCurtisAudio Thank you Michael. your content is epic. I still have some questions, Is it not preferable to separate the kick and bass guitar's with frequency like this, kick drum focus is 50hz to 100hz Bass guitar low cut till 90hz with focus on 90hz to 400hz. This way you don't need the bass guitar going to the centre subs because you have made space for it in the LR mix just above the kickdrum frequency, even better separation? Thank you for your feedback

[@MichaelCurtisAudio]

That's definitely a school of thought when it comes to "carving out space" in your mix. Although this makes intuitive sense, I as a bass player wouldn't want the bottom octave of my instrument to be missing : ) I think there are other ways to make sure both the kick and bass can live in the same space without clouding the other. That largely comes down to the arrangement and players first, though.

[@aaronl7669]

Thanks for this video! I think I stop using subs on aux for awhile and see how it goes. I was always afraid of bad stuff getting into the sub, but I'll try your EQ strategy for vocals with the low cut, and two high shelves. Do you have some other recommendations for how to EQ guitars, bass and drums? I need to learn more about using EQ for better sound. Just looking around your channel now.

[@MichaelCurtisAudio]

Thanks a lot! I have a free PDF in my Audio Toolkit called the 9 EQ Pivot Points that I think would be helpful for you when it comes to EQ. Although there are specific techniques and frames of mind for each instrument when using EQ, having these "macro" ways of thinking about each part of the frequency spectrum applies to all sources.

[@aaronl7669]

@@MichaelCurtisAudio I've been playing around with not using the aux-fed subs some. One thing I noticed is there is definitely *some* vocal content that is going through the sub even if I use a 160hz hpf. i guess it's not possible to completely eliminate the vocals, etc. from the sub with this method, just reduce them? that may not be a big deal of course. If I'm applying the hpf and maybe some low shelves to eliminate more of certain channels from the sub, will live performers be upset that their monitors make their guitars or vocals sound too thin? I'm wondering if it really makes much of a difference for my small setup whether I use aux-fed subs or not. I usually just have 1 sub and 2 tops at gigs (crossover is 100 Hz b/w sub and tops.) Is the phase issue really that big of a deal if I have the sub adjusted independently of the mains in that case? I never do a live stream, so I could see that being an issue with the aux-fed method, as you indicate. Is there a difference between increasing or decreasing the volume of the sub through the aux and doing it through turning the volume control on the sub up or down?

[@jourellbacani9793]

i do a full range setup and depending on the venue too if there are lots of low energy then i adjust my hpf either from 120 to 200 just depends on the venue, but im just doing small events not big events

[@philwood6739]

Hi Michael, I’d love to know if you still feel the same for corporate work. I usually do exactly this for music work, but for corporate I often do buses of different inputs (lecturn, lavs etc) straight to my various matrices to allow me to slightly shade down the lecturn in front fills for example to give me a tiny bit more headroom for the rest of the space. Thoughts?

[@08_crown_vic]

Can I use mix minus Theory with aux post fade into a channal and send that channal to aux pre to the sub with only lows to the sub instead of sending highs, mids and lows to the sub?

[@MichaelCurtisAudio]

You can, but I think that's way overcomplicating things. The subs will take care of filtering out high frequency information.

[@RogueFisher80920]

hey Michael, love the video.... Could you maybe provide some details (or a video perhaps) around how you would set up what you are discussing here on an x32... ? It would be greatly appreciated. THANKS!

[@MichaelCurtisAudio]

Thanks a ton, Marty! Glad it was helpful for you. That's a great idea and I'll definitely throw that in the video pipeline. Is there any specific part of the routing scheme on an X32 when it comes to Matrices that's confusing to you?

[@RogueFisher80920]

@@MichaelCurtisAudio I am bit confused on how to set the Martices up for the subs. I am still learning the geography of this board, and alas, I am "one of those" who is using an Aux channel for my Subs. I am also using the boards abilites to manage cross-overs managing a 6 speaker setup, (Subs, Mids, Highgs) vs using hardware cross-overs. So anything you have and can clarify would be awesome. AN ADD TO THE ABOVE, If I use Matrices for In-ears, will that mess up using these for SUBS, and would I want to use them for mains as well? THANKS!

[@MichaelCurtisAudio]

@@RogueFisher80920 I would still use auxes for your in-ear mixes, not matrices! Using the console to also split up the feeds for lows/mids/highs in a three-way speaker definitely makes things more complicated as well. It's ultimately up to the specs of the speaker and the power amps you have attached to each channel. The simplest solution would be to send your LR mix at unity to all six matrices, then apply any filtering across each matrix to take care of any crossover/frequency divider duties. Each matrix output would then feed into an input channel on a power amp.

[@Billsoundmaster]

I tend to think of aux or group fed subs as a hold over from analog desks. To process the subs separately from the RL bus. And some analog desks had limited matrices. Great video.. oh I like the idea of audio nerding over thanksgiving.. as heard on the Signal to Noise pod.

[@MichaelCurtisAudio]

Thanks a ton, Bill! And yes, that's a great point on the analog hold over. Makes a ton of sense.

[@ernietalks]

I use the main L/R output for my subs. I can take R for the tops and L for the subs after coming out of crossover with the high and low. I use the PAN knob on the channel strip to assign each things at channel level to the subs. For example, for kickdrum I PAN hard left and also compliment with the EQ low knob on the channel strip. For vocals , I PAN right and also use the EQ on the Channel strip to achieve a balance for the low ends and high properties of vocals . It works perfectly for me in house of worship and all applications

[@MichaelCurtisAudio]

Thanks for sharing your workflow!

[@WahalaboyPromotions]

This is what i do too

[@Ali-my5xo]

Hahah ıts crazy

[@zamirbetancourt6493]

Hello, how can I make some bleachers in a Manuel way with my mapp3d app for places like fields, stadiums, etc.

[@MichaelCurtisAudio]

Check this video out - kzbin.info/www/bejne/e2q5fYKQm96JjZI

[@nathankabat1276]

What are your thoughts on aux fed subs for a DJ situation, where it’s mostly recorded music and a few speeches, would it be beneficial to be able to turn subs down for speeches and background music and then turn them up when it’s time for dancing? (Thinking weddings!) Love your channel, I’m learning a ton about pro audio!

[@MichaelCurtisAudio]

Great question. You still can still turn up and turn down your subs with a matrix. If you're just using a DJ mixer that doesn't have a matrix, then yes next thing would be an aux if you need that type of control.

[@mattnewcomb_official]

I'd stick with a 160Hz High-Pass filter on the "speech" mic channel, and drastically cut anything below that on the channel's EQ as extra insurance. Speeches at weddings are usually not made by men (the crying in public thing), so unless there's a Betty Baritone giving a speech (or a belching contest), you shouldn't have to worry about anyone hitting 80Hz. I'd also make a steep sloping cut to the highs starting at about 8K to weed out the high screechies of excited women (it's a wedding) and sibilance when the party-goers start slurring their Ss. Put it at - infinity by the 12K band. It'll sound a bit boxey, like a bad radio, but speeches aren't entertaining, so they don't require the same clarity. I'd also notch down the immediate center 900-1.3K on this channel, since this is where someone with a nasal speaking voice (Fran Drescher) will cook your tweeters. You want to keep speakers audible but tolerable. If there is a belching contest, though, you want to kill the HPF so the full guttural richness comes shining through. And some Lysol or a replacement mic ball and windscreen.

[@vicseven6456]

In your example, Matrix 4 feeds the sub. Im assuming you are simply selecting LR main fader, choosing sends and raising Matrix 4 fader to unity. How is X32 summing that to mono from LR feed?

[@MichaelCurtisAudio]

That's exactly it! You can also do it from the "sends" tab when you select your LR fader, then use the encoder on the bottom of the screen to send that to Matrix 4. The X32 automatically sums down a stereo source to mono when sending it to a mono destination.

[@Bobby_Uterus]

Mr. Curtis, you have been doing an awesome job with your videos. I’ve learned a lot. Do you have a video for how you process IEM feeds? Like EQ and an FX. What you generally do and any rules you have?

[@MichaelCurtisAudio]

Thanks a lot, Bobby! Here's a few things I keep in mind for IEM feeds: 1. Always go stereo if you have the outputs. That opens up the stereo field so you can pan things around and create space. The "separation" really helps create clarity. 2. I don't usually end up applying an EQ across the whole IEM mix. Only if the performer tells me their overall mix is bright or dark. 3. Less is more. I usually start with the instrument the performer's playing at unity, their support instruments at -5dB, then everything else at -10dB or -15dB. 4. Use ambient/crowd mics if you got the time and inputs. Helps the performer feel less isolated. 5. Run FX returns back to performers post fader, for sure.

[@Bobby_Uterus]

@@MichaelCurtisAudio Thanks for always taking the time to respond. Doing most of the things you mentioned here. Makes me feel more confident knowing you have confirmed I’m doing the right things. Thanks again for the reply’s and videos man. Much appreciated, keep it up.

[@MichaelCurtisAudio]

@@Bobby_Uterus You're very welcome! Happy to help.

[@danbei-yv6qm]

Reason #4 is a very good point that I sometimes forget, well said. I sometimes have to use an analog console, with fixed slope and frequency for HPF. In this case I would argue in favor of using an aux. Nevertheless, thank you for your well argued insights!

[@MichaelCurtisAudio]

Yes, for that specific analog use case with a limited toolset an aux can definitely be helpful!

[@kobek5272]

How would you run the subs on a xair18? That’s the mixer I’m currently using. I usually put it in bus 6 and go from there. Is there a different way you would recommend to run the subs

[@MichaelCurtisAudio]

I don't think Matrices are available on the xair18, unfortunately. So, I would just use an Aux with everything post-fader to it, then have all the sends to that Aux at unity.

[@kobek5272]

I appreciate the reply. Thanks 💯

[@jakegodman]

Referring to how you talked about using a matrix for your fills. What would be the benefit of using a matrix over using an aux in post? Wouldn’t an aux send in post have the same effect in terms of levels following the LR and getting the same mix?

[@MichaelCurtisAudio]

Great question. With an aux I would not be able to send me Main LR to it as an input source. The LR on most desks can only flow "downstream" to a matrix. Furthermore, if I apply any master bus processing to main mix and the fills were on an aux, then my fill speakers wouldn't have my master bus processing.

[@etienometiabasiudo799]

How can one apply this an analogue mixer with two matrixes. I use the Yamaha MGP 24x

[@MichaelCurtisAudio]

It depends how much you care about "stereo". You could send your main to Matrix 1, then another copy of it to Matrix 2. Then run Matrix 1 to your left and right mains, then matrix 2 to subs.

[@theflyingalamo]

Great material Michael. My primary mixer is an Allen & Heath SQ-5. I only have access to 3 Matrices. What would you recommend as a workaround for my limitation?

[@MichaelCurtisAudio]

It's up to how many speaker zones you have, but I'm pretty sure those can each be run as stereo. So, you could have Mtx 1 feed your main LR speakers, Mtx 2 feed your subs, then Mtx 3 can still feed any delays or front fills. And this is all assuming you don't have a system processor to handle those duties for you.

[@theflyingalamo]

@@MichaelCurtisAudio yessir. Not sure why that the MTXs are stereo wasn't clicking. Thanks. No processor as of now. Smaller events for me for the time being.

[@MichaelCurtisAudio]

@@theflyingalamo No worries! Glad this was helpful.

[@victorbesedinsky1935]

Hello Michael. Could you please explain how to connect subs using matrix ? But I don’t understand what if I don’t need some instruments sounding in my subs? My mixer is Allen heath sq 5. Thanks for your answers! Happy New Year!!!

[@MichaelCurtisAudio]

Hey, Victor. The whole point of running it on a matrix is for you stop thinking about, "What instruments do I want in my subs?" and trust the tonality of your system working as a whole from the get go. I'm sure you don't want a flute going to your subs, but there's not frequency content down there anyway. And you can use a high pass filter on an instrument that may have sub content, but you can then shape to fit into whatever tonal space you'd like. On the SQ5 you have four stereo matrices. You'd route your LR bus to Matrix 1 to feed your mains, then route the LR again to Matrix 2 to feed your subs.

[@victorbesedinsky1935]

Thank you so much!!!

[@sensibleandrational6682]

Grab your sub matrix input from the sub aux send and you’re good to go. It’s the FOH engineer’s choice to either use speaker management or dedicated sub feed, but understanding why you’re doing it one way or the other is equally important.

[@MichaelCurtisAudio]

Like you said, lots of ways to manage LF and that's definitely a valid approach.

[@marzbanirani1412]

also wont changing sub woofer level AKA aux fed subs change the acoustic crossover point?

[@MichaelCurtisAudio]

Same thing with Matrix fed subs, but yes that will change. Not a dealbreaker in my opinion, though.

[@nickwarren131]

Would love to see a video on the use cases for a FOH reference mic. Generally, I understand it helps you get a sense for how your direct mix translates to the room, but are there specific things you are looking for in the signal? What is your preferred mic, and how do you decide where to place it?

[@MichaelCurtisAudio]

When I use a mic at FOH it's to gauge SPL and also have a look at a few other graphs, mainly spectrogram and RTA. I use the dual-channel functionality of my analyzer during tuning, but single channel during show.

[@nickwarren131]

@@MichaelCurtisAudio Thanks for the response! This made it clear I need to develop a more robust metering system. During tuning, are you typically looking to achieve a "flat" response from your system? Or are there cases in which you may want the output transfer function to be nonlinear? I have usually taken this on a case-by-case basis (also primarily going by ear). I am wondering if I should get in the habit of staying flatter, especially in the balance between subs and mains. As you alluded to in this video, these types of nonlinearities can force you to think as both a system tech and mix engineer simultaneously.

[@MichaelCurtisAudio]

@@nickwarren131 The "target curve" I end up using and staying with most often is from Michael Lawrence, a world class systems engineer. He uses it for the large scale systems he's tuning for musical performing acts. There's a -2dB drop off in high frequencies from 1kHz up to 16kHz. Think of it as a very broad shelf. Then for low end, starting at 300Hz, there's a fairly steep slope that goes +15dB up at 50Hz. So, the system is "flat-ish" from 300Hz on up, but then has a generous low end up to get some meat.

[@user-lh6tj4tr4c]

Hi Michael, could you indicate a link or simulate two diferent size speakers SPL map (6" and 20", for example) at same frequency and intensity? This is for academical purposes. It would be very interesting to see how speaker size affects SPL using same conditions.

[@artmathias9725]

This may sound simplistic and insulting, but it's not meant to be. Sound is the movement of air. How much air you can move determines the amplitude (perceived volume) of your waveform. You're 20" speaker has to much mass to push and pull at 2000 cycles per second. Likewise, a 6" driver just doesn't have enough surface area to produce the same amplitude of wave at low frequencies. I hope that helped.

[@martinvera918]

I tried to do this setup but I always run into issues when the band is asking for more subs. But I do like the fact that the setup mentioned in your video is better for live streaming and recording. Do you offer any tutoring or classes?

[@MichaelCurtisAudio]

I do! I have a course called Making Sense of Sound: How Sound Actually Works In Live PA Systems": www.producedbymkc.com/makingsenseofsound I'm currently working on my followup courses that will focus on sound system design and tuning. Will drop this year.

[@timogletree8754]

I think there's a general misunderstanding as to the intention of aux fed subs. It was never intended to use as a volume control for the sub freqs of an input. All the sends & the aux master should be set to nominal. It was intended to simply keep the subs as clean as possible by not assigning channels that you don't want in the subs which also reduces ambient low end on stage from leaking back through those mics into the subs. With digital consoles these days you have a lot more at your disposal to control the low end on an input as needed. That was not always the case. It's arguable as to which method get's you there quicker. One thing for sure, there's a lot of misunderstanding & therefore bad use. Anytime you adjust these faders off nominal you're changing the crossover point...

[@MichaelCurtisAudio]

Great point here, Tim. Yes, before I switched to either feeding just my LR mix or running subs on a matrix I ran my aux fed subs that same way. Everything sent at unity. If you do end up running into feedback issues with certain sources in your subs, that is certainly helpful.

[@Cbamptronious]

@Michael Curtis So on the WING would you still use matrices or would you use the 4 mains?

[@MichaelCurtisAudio]

@@Cbamptronious Definitely the eight matrices.

[@joelmathejojo]

Your really shaking the table here haha, But thank you for giving us some food for thought. I kinda agree with you, it doesn't make sense to tune the system, making it sound great using a reference track only to ignore and bypass the whole process by using auxes. Mixing without aux fed subs and doing it well is a challenge and a test of someone's mixing capabilities, If a reference track can sound good, you should be able as well.... I like it 😁😁 Personally what I've been doing is a separate system tuning for the aux subs and mains separately, because doing it as a whole never translated when mixing, I think it makes sense now. Will give it a go and see.

[@MichaelCurtisAudio]

It was definitely a big transition for me as well when I stopped using aux fed subs. Give it a go on your next gig and let me know if it was helpful!

[@leifthune9071]

Really valuable information to consider about system and mixing workflow. I think it may be great advice for beginner engineers/techs. I just have a few questions that I need clarification on. 1. Can’t you utilize bus processing separately and mix levels on DCAs for instruments? I do this a lot. 2. Using a hpf on channels and lpf on the system how much different is this than using a crossover? If you are streaming with subs on aux, it always makes sense to reach for eq as well for tonal changes unless you’re compensating for serious abnormalities live. 3. Transfer function point is valid, but if you keep the system close to full range and use the subs as low end enhancement. Would this allow your full range PA mix to be sent to the transfer function? It seems viable to feel out the low end needs based on the event/crowd engagement, etc. Thoughts?

[@MichaelCurtisAudio]

Hey, Leif. Great points here. 1. I'm not quite sure the routing you're trying to get across here. Buses on most desks can't be routed to another bus, so the aux doesn't get the processing? 2. A hpf on a channel is very different in that it is much farther upstream. All monitor sends, PGM feed sends, main LR sends, etc. would be affected by that. And I'm not saying it's impossible to have the aux make a sensible balance of low end vs the stream, but I do think it introduces another unnecessary variable. 3. If I'm on a system where the tops barely get below 80Hz, then the subs are more than just low end enhancement. It's a good idea in theory to use the subs more like an LFE channel, but on the shows I'm on they're treated like an extension of the whole PA, not for effect.

[@leifthune9071]

1. Ah, right. I got turned around in my thoughts about parallel processing vs inline bus processing. It wouldn’t make sense to send the bus with HF content to the matrix either. The engineers I’ve worked with process the kick and toms individually and only send bus processing to the tops (paying attention to phase coherence). Then mix level on the DCAs, but this does require extra brainwork in routing and mix decisions. 2. You’re absolutely right in saying the methods of system design/routing will be more work to account for in your design/mixing. 3. I understand clearly now, not sure if I missed this in your video. It is important that your subs feel cohesive to the entire system with similar processing. I think your information can be quite valuable in setup especially with a similar system and managing the entire show smoothly.

[@Sg435]

At my regular gig, I use an m32 and the gain on my PA never changes, so I run everything to main L/R then use 3 matrix, one for L, one for R and a mono sub through out 14, 15 and 16, and 9 times out of 10, my sub matrix will be at -20 and rarely ever moves

[@djgazaaka5000]

How about if the mixer have zone /sub output?

[@davidwellington4186]

Re X32 Buss to Aux (8:32): Yes you can route a Buss to an Aux, or any other channel, (and therefore also route a Bus to another Bus, via a Channel) I agree with the Subs (indeed all the PA) on Matrix after the LR, for simplicity of control and trouble shooting. After all, having spent the time (and money if you have bought Smart) to get a good even frequency response in the venue, would you mess with the PA balance (subs level) during the show rather than ‘tuning’ the band sound with Ch or Group Eq. Thanks for all your brilliant tutorials, especially on cardiod Subs, going to be experimenting with this next time I can get into the venue on a non show day.

[@MichaelCurtisAudio]

Ah yes, totally forgot I can put a bus on a channel and do routing that way. Thanks for the reminder there. I'm glad you've found my videos helpful! Let me know how the cardioid setup goes for you.

[@TimmyP1955]

When I encounter a system that's set up with aux-fed subs, I leave it that way, but I set the levels "at unity", so the system is still flat, I'm just taking advantage of the "tighter" high pass filtration. (Boosting the sub level on a channel seldom gives me the result I want, so I don't bother if I am configuring the system.)

[@scivirus3563]

people seem to miss the point of aux fed subs ..the point is you send only sound sources that has sub frequencies you want to the aux .so like plufs on vocals is not a problem .don't need subs for snares , rides ,crashes ,hi toms splash Banjos and ,Vocals ..there mics will only pick up wind and handling noises hi pass filter do not stop every thing

[@jeremynothman9512]

I might be wrong, but I don't think @MichaelCurtisAudio misunderstood the idea, he just said that modern consoles offer variable HPF and low shelf/peak cuts on every channel, and therefore this point is moot. IMO, if you are used to mixing in the studio on full-range speakers, the idea of mixing full-range on a PA just makes sense.

[@Ldawn318]

Is this also suggested for DJ applications as well?

[@reinhardleonardmarpaung1422]

I think it's only good for a big pro mixers or digital mixers, but for smaller mixer (like a Yamaha MGP16X or Beringer Xenyx XL2400) they doesn't have a matrix (a good one), so you must route subs from a Group, Aux, or mono(even Yamaha doesn't have a mono channel control)

[@Bobby_Uterus]

Mr. Curtis, this is great info. would you consider doing a layout demonstration video using the M32 edit program?

[@MichaelCurtisAudio]

Hey, Bobby. My latest video (which I'll link to below) documents this using M32 edit. Let me know if that answers your question: kzbin.info/www/bejne/sHK4iZ-MeKtkkMU

[@Bobby_Uterus]

@@MichaelCurtisAudio you are awesome! keep killing it man!

[@MichaelCurtisAudio]

@@Bobby_Uterus You got it!

[@robendj]

I think it all comes down to approach, coming from a sound engineer background a LR mix makes more sense. It's how I've always mixed at venues and it's worked fine to date. If people prefer aux fed subs, all the power to them. I don't believe there's one better way over another, if you can make it work, you made it work. The crowd don't know, they're not sitting there like "can you hear this A1 isnt using aux fed subs!!".

[@BERMANSOUND-nk2fr]

08:30 Sure you can send group to an aux on x32: 1. take 2 free channels, link them to be stereo 2. choose the input of those channels - the buses of the group 3. send the channels to an aux as usual

[@MichaelCurtisAudio]

You're totally right! Great point.

[@kevinzolitor]

Good food for thought. I was taught to subs on aux, this all makes sense. Especially for someone like me that doesn't mix a lot, it's one less thing to keep in mind during the show. But I have a question. If your main PA is on matrix, what is plugged in to the console master out? Nothing? Is the purpose of the main PA on a matrix just that you can raise and lower the main PA level without also raising and lowering the fills/lobby/etc?

[@MichaelCurtisAudio]

Great question! Yes, your LR master feed is not going directly to any output source. It is first flowing through a matrix. Being able to change the level of the main PA is one benefit. You can also apply EQ independently, etc. Any processing I'm doing to make the MIX sound better goes across the LR bus. Anything I'm doing to make a part of the PA itself sound better goes across the matrix.

[@kevinzolitor]

@@MichaelCurtisAudio I guess I never thought of doing that, I guess I learned in the analog days where matrices were limited and the EQ was done outboard. Cool though. I like the organization. but what about front fills? I also was taught that you would do these on an aux (usually post fader) so if there were things that were naturally loud on stage that didn't need to be amplified for the people in the front row, you would create a separate mix and delay it so precedence would place the image on stage and not in the FF speakers for those in the expensive seats. sounds like you don't do that either if the FFs are on a matrix.

[@MichaelCurtisAudio]

@@kevinzolitor If you want to get super granular and create a separate submix for the front fills, than totally go for it. Even if it's just on a matrix, you can still time align it. I just find it hard to keep track of mixing so many destinations differently, and if I'm at FOH I can't hear the FF mix. So why not get the PA sounding great everywhere, then trust that one mix will translate everywhere? I totally understand that if a band has live drums, open guitar amps, etc., that the front row will definitely feel different, though. But even left to right across the front of the stage you'll be hearing different amps at different times, will get monitor mix bleed differently, etc. I think that approach could be handy if you're working the same band night after night, but not as practical if it's a one-off.

[@kevinzolitor]

@@MichaelCurtisAudio i did not think about the left-to-right variation. Once again thanks for the insight.

[@techmetal2292]

Curious what slopes you run your high pass at on your vocal mics?

[@MichaelCurtisAudio]

Usually second order butterworth. Sometimes third or fourth order.

[@mauricerogerson5825]

It never even occurred to me to run the subs out an aux channel. I just use the sub out of my FOH EQ, and adjust the LF channels at the mixer as needed. I'll give your approach a try though.

[@owenjoseph6097]

Thanks for this excellent explaination of your preference. If im mixing multiple different sources. Ie headset mics, HH mics, close instrument mic, rifle mics etc. I set the console inp-grp-mtx. So bypassing the stereo bus. (Hear me out) So if I have big loud brass section who are loud onstage i can "dip" the send from the group to the front fill mtx. It makes the setup a little more complex but gives you a chance of evening out acousticly loud instruments. The default is grp send is unity so it makes it possible to do a full consistsnt sound check. So i only use this to attenuate levels never increase them. I think im going to try your way the next time though! great vid, keep up this great content

[@MichaelCurtisAudio]

That's a cool workflow and makes a lot of sense! The feedback is much appreciated.

[@djfrenchi]

Mono to sub fan here. I like your theory as it forces you to make good EQ decisions, however it does not account for most consoles having only one HP filter slope on individual channels that might be less steep than what you’d want going to your subs.

[@MichaelCurtisAudio]

Good point here. HPF flexibility is helpful.

[@TheOtherPhillip]

Great video. I don’t disagree, but do have a counterpoint. With live PAs, we generally have an exaggerated sub to main ratio. Some level of a haystack in the bass frequencies. With aux fed subs, the mix engineer is able to prevent vocal mics from going to the subs at all. Without aux fed subs, the only tool is the high pass filter. Do you find yourself high passing at a higher frequency to be more effective in the sub region, when the subs are boosted 6-10db above the mains? If more aggressive on the HPF, that would affect other mixes for streams, etc. Would love to hear your thoughts…

[@tmc3p1]

Hi Phillip that's a good point. I had a vocal with 85hz high pass filter. But I noticed it was still sending a bit of low frequency energy to the subs even though my xover freq was 80hz. I thought it had to do with a slope of the high pass filter. Anyway, that made me almost force me to use aux fed subs. Yes, I'd too like to hear your thoughts on this.

[@MichaelCurtisAudio]

Great counterpoint here, Phillip! Glad you liked the video. As much as it feels like vocals in the subs is the issue, in my own systems (without floor wedges) I have trouble with vocal feedback in the 200-600Hz range, which sits squarely in the custody of the mains. The highest crossover frequency I use for subs for most rigs is 100Hz, which would be a low G for a bass singer (and the open G string on a bass guitar). There's of course a slope downwards in level or frequency for the subs after that, but they're usually more than 10dB down by 200Hz. I feel like the bigger issue is too big of a tilt in the LF on the MAINS. If I'm running a show that has both a big dialogue component AND a big a band component, I'll have the pink shift/LF shaping on the PA geared to sound great for the band, then use a low shelf to shape low end for my Lav/HH bus. My system target curve shoots for a flat PA from 1kHz to 10kHz, then some tapered HF dropoff after that as you move from front row to back row. For LF shaping, there's a +9dB rise from 1kHz down to 100Hz, then it's flat from 100Hz down to 31Hz. So, it's PA systems that have a much bigger LF lift down towards the subs that I think pose a bigger issue. Also, on BIG big PA systems the mains easily get to 60Hz. So, even if you did take your vocals out of the subs there, you've still got a significant amount of LF power coming from just your mains. If you still want to make sure vocals never hit the subs, I'd run a Band and Vox subgroup, then route accordingly to matrices. That just makes it harder to do any overall PA tonal shaping without duplicating processing. As far as vocal processing and EQ, these days I'm running HPF at 160Hz to start, then having a low shelf at 500Hz and a high shelf at 2kHz. These high and low shelves act as a "tilt EQ" together. If too much proximity effect, bring down the low shelf. If not enough top end to cut through, boost the high shelf. Then I have one parametric filter sitting at 400Hz to cut any mud. Hope this helps!

[@srendyhr5844]

@@MichaelCurtisAudioAlso, think of the harmonic's the subs unavoidable are creating ... there is actually more distortion in a "punchy" sub, so even if you have a 4th order crossover at 90 hz vill 120 hz and 180 hz come creeping in as byproducts.

[@srendyhr5844]

There is another point to be made about the cumulative effekt of having a wast number of inputs lowcutted at the same frequency, the ripple or bump just at cut off - is pronounced

[@MichaelCurtisAudio]

@@srendyhr5844 Yup! That can cause a ton of phase shift and weird frequency responses.

[@thecolb1398]

You probably don’t remember but several years back at KPC you and I actually had a conversation about this exact thing. I hate mixing separate to the subs for all of the reasons you give!

[@MichaelCurtisAudio]

I do!

[@nisios]

Isn’t the mono bus just another matrix in the M32 for instance, what’s the point of setting and wasting a generic matrix?

[@MichaelCurtisAudio]

The matrix sits downstream of the LR Bus and inherits its signal and processing. The mono bus is a separate signal path. So, if you applied mix bus compression on your LR mix the same processing would not apply to your Mono bus.

[@user-ms8ec8cg7i]

Hello, Michael. Big follower of you here. Quick question. Is putting the sub on a matrix essentially the same as aux feeding the sub? Since you are controlling the overall sub volume?

[@MichaelCurtisAudio]

Hey, Aaron! It's different in that I choose to send my full LR mix to the matrix (not a piecemeal collection of channels). Yes you're right in that both have a fader and you can control overall level, but the big difference for me is the contents of the signal, not that you can adjust the level of the signal.

[@user-ms8ec8cg7i]

@@MichaelCurtisAudio Awesome. Thank you, Michael!

[@sebastienrichard3270]

On #3 I still always feed the subs from a matrix its just a matter of what feeds that matrix. If I have multiple mix engineers on the same show with different viewpoints I can accommodate aux or full range mixing within the same console config.

[@MichaelCurtisAudio]

Exactly! Yes, gives you flexibility with the production desk or DSP. I had to do this same thing Friday night on a gig where I was mixing the entertainment for the night, but then there was a wedding band after that brought their own desk.

[@jordanthurt]

For me, it boils down to several things: 1. What console am I working with? 2. How many outputs do I need and how many are left? 3. What type of event am I mixing for? 4. So I have any external processing control for the PA? All of that is going to influence whether I use aux-fed subs or not.

[@broadcastedtube]

Great Sub guidance for live mixing!

[@MichaelCurtisAudio]

Glad this was helpful to you!

[@xgamesnet4160]

For me it really depends on the system and how your console brand implements auxes and matrixes. I find, specifically on digico consoles, running subs as a matrix that you can choose weather or not a given channel sends is nice. Some systems implement this differently, and I find using whatever gives me the most control over the sound is generally the best to use. If I’m on a large PA with delay subs and massive delay line arrays as fills, aux subs may not give me the even sound across the audience that’s desired. It’s just a tool that’s sometimes helpful and sometimes not :) love the content btw

[@MichaelCurtisAudio]

Thanks for sharing!

[@artmathias9725]

I would agree. If I am in an arena with multiple hangs, Each zone of PA is treated as it's own. For your example of large PA with delays and out fills, I may have the Main L/R zones with sub on aux, where my delays and out hangs are run full range. Again, like you said, different rooms and different gear builds, have to be treated with flexibility. Last month, I have had to build a show with a mono-block of subs in front of the stage, a regular point source set-up, an inline cardioid subs, and once a di-pole cardioid set-up. 3 weeks, same band, 4 different sub set-ups. Again, good job to Michael Curtis for initiating this dialog. And good job to you for bringing more information to the discussion.

[@user-uk7ts6qc6p]

Ung matrix system for FOH SUBS AND FRONFILL and send it to LR is my main standard in mixing since i used x32 mixer,.. I insert eq to all remain in flat and just invert sa polarity signal of the sub matrix mix.

[@maizesoft]

The problem I found using matrix is: even after I lowcut the vocal at 120hz, there is still some reproduced by the subwoofer which cause muddiness

[@MichaelCurtisAudio]

Then move the low cut higher?

[@maizesoft]

@@MichaelCurtisAudio kzbin.info/www/bejne/nqGpfGSOhL-oi8k

[@MichaelCurtisAudio]

@@maizesoft Yes, I'm well aware of this video. Dave Rat has some good points, but I feel like it's a non-issue with most vocal mics. An SM58 already had a fairly steep LF dropoff below 160Hz and I'm usually high passing at least that high due to the proximity effect. I'm usually able to do a cardioid configuration on my subs which keeps more LF off the stage. If I do need to get some specific sources out of the subs (lav mics, podium mic) I opt to matrix out my subgroups to the appropriate destinations vs using an aux for my subs.

[@jouebien]

1A/3/4. I would best describe as make your life easy by not doubling the mental load of keeping track of where your decisions are. Similar to using a compressor - volume = volume out with the make up gain. 1B. I don't really buy the broadcast limitations. If you are using the FOH mix to feed broadcast you are already making a sub par compromise & most of the playback devices won't really produce the low end anyway. If you want to include the sub in broadcast then you can use a matrix to combine it with the LR. Ideally it should be a complete separate mix & if you have the budge a passive split (or like the Pro 2 with completely separate pres in the remote i/o). 1C. Yep shelf that low end away on everything that hasn't got bottom end you need.

[@SpaceAgeConsulting]

I came across this video because we decided to move from the Matrix sub feed and go to an Aux sub feed. I don't find that it will be any more complex than doing a stage monitor mix from the FOH console and we will be using less channels in the sub mix. I didn't understand your comments about the stream feed being affected. Are you using a matrix to feed your stream or an aux? Good video. I'll see how this works for us this weekend.

[@MichaelCurtisAudio]

Thanks! My point about the stream feed is that if you're mixing subs on an aux that's yet another variable to your low end. So, it may feel "right" in the room, but since the low end is treated differently than the mains, it's hard to match that to your stream feed if you're pulling double duty.

[@conorm2524]

Your matrix routing is slightly different than the standard in Europe (and most US venues I've toured through). Normally L, R, Subs, Fills. I agree with your views on the aux fed subs. Especially when it comes to glueing a mix together on the master buss. The subs aux won't be involved in that process. I enjoyed your S/N podcast appearance - keep up the good work!

[@MichaelCurtisAudio]

Thanks a ton, Conor! Yes, L, R, Sub, Fills is more common for sure.

[@artysanmobile]

HPF and for that matter LPF are my fundamental tools for building a mix. They often are the basis of my equalization to the point I often need far less time getting a sound on a given channel once I’ve filtered. I can’t imagine why anyone would object to this. Most of my music mixing is for broadcast or release. I know the frequencies of musical tones and use that to narrow every channel as much as is possible. In symphonic spot mics, this is done without question. Ditto an electric band perhaps even more so. If working in surround, my Sub is simply the speaker handling the lowest octave or two. I have never been happy with aux fed subs and leave them off if they are present. You are preaching to the choir here about ‘aux mixing’. A recipe for disaster and a tail-chasing exercise - I’ve never needed either.

[@rabudja48]

One thing which I would like to experiment with more is unconventional groups of channels. Next time I will run a multitrack through a rig and group bass - kick - floor, then guitars - keys - vocals. I dare say that we often suffer from such time pressure that we rarely experiment!

[@MichaelCurtisAudio]

Yes, most live events being so fast paced makes it very hard to experiment.

[@solitaire5142]

Another argument against aux fed subs is that they dont track the rest of the system if you do any LR buss riding.

[@curly_]

i work with having the Subs on a matrix linked to LR , it helps me a lot

[@passionformusic4291]

What if you mixer does not have Matrices? Behringer xAir 12

[@ryanshaughnessy828]

I may have completely missed this answer in the video... but why send LR to Matrices vs just using the LR Bus with a Matrix fed sub? Great insight in this video. Looks like I may have to rework my setup!

[@MichaelCurtisAudio]

Because a matrix fed sub still means each channel has a variable send to it. Sending the LR means whatever you're sending to your main mix also gets sent to the sub, then let your processing take care of subdividing the frequency response between the two systems.

[@ryanshaughnessy828]

@@MichaelCurtisAudio Right, got that part! I apologize, I wasn't very clear with my question. You mentioned Left Channel is sent to Matrix 1, Right Channel is sent to Matrix 2. What's the reasoning behind that?

[@MichaelCurtisAudio]

@@ryanshaughnessy828 No worries! It's so that matrices 1&2 function together as a stereo pair. This is assuming I've got a LR PA setup. Even if the coverage may not be true "stereo", I still like having left and right feeding the speakers separately.

[@timnordberg7204]

I'm no acoustical engineer, and I'm the first to admit I'm no live mixing virtuoso--but riddle me this: what's the functional difference between your preferred method (feeding a subs matrix with the LR buss signal) and feeding a subs matrix with signals from groups? Sure, you miss out on whatever processing is on the LR buss, and you may need to make a delay compensation if you use latent processing. I'm already a fan of using drum and vocal busses, and if I made an "All Instruments" or "Video World/BGM" buss, couldn't I just send each of these busses to the subwoofers at -0dB, and have nearly the same result as your method? With, I might add, the added ability to remove any of these busses from the sub feed by muting its send? I think this "subs or no subs" method splits the difference nicely between the benefits of complete sub removal on problem signals and the downsides of aux-feeding that you describe in the video. And to preempt your inevitable question of why I would want to go about this bass-ackwards way: I'd like to be able to try something like your method in a venue which has defaulted to aux-fed subs.

[@MichaelCurtisAudio]

Hey, Tim. The biggest reason for me personally is that running a matrix (or letting the system processor do everything and simply feeding it my LR mix, which is my preferred method) makes a clear divide between my console is for mixing, the processing or matrices are for wrangling the PA. When I'm choosing what's going to the subs or not, even if the send levels are at unity, I'm having to think both like a mix engineer and a systems engineer at the same time. Feeding it my LR mix saves me subgroups/buses, keeps all my processing in one place, and I can trust my board mix will sound great since I'm not having to think about the LF in my mix vs the house system, assuming my mix is great ; )

[@mrsonor1090]

So, you always run your subs in mono (unless there is a specific demand for stereo subs, I guess)? Best regards from Sweden.

[@MK1Records]

In live sound subs are generally driven mono, multi channel only in very specific use cases :)

[@bakareanthonybabatunde4653]

what is the best frequency for sub ?

[@MichaelCurtisAudio]

Do you mean crossover frequency?

[@jeremyryannoel]

Two things: first is something I discovered recently, second is something you confirmed. 1. Adding low end shelf to the sub channel isn’t necessarily a good idea, due to affecting the crossover on the parametric eq (so far this is an observation) 2. Continuing 1, eq from input channels instead of thinking it’s a global issue. That changed the way tracks sound, as well as overall mix. I use the Center channel on M32, factoring in what I do or don’t need in the mains. This works 99.9% of the time, especially since there’s a dedicated rotary knob in on the penthouse section. Even the TF series suggests a center channel via Sub out, yet I do like having the option of stereo subs depending. I’ve since then summed both sides to mono at my Sunday setup, but hasn’t caused any issues so far.

[@DrNIx123]

While the SQ series can send groups to aux, it doesn't make sense with regard to the arguments pro aux-fed subs. Because would not send the drum group to aux, but only the kick drum. However, especially on small venues, you might want to delay the drum group to align PA drum output with direct acoustic drum sound. Which would work with aux-fed subs, however at a much higher level of complexity. You'd have at least two places to visit in the console to change the delay.

[@InfamousEllis]

Unfortunately, our mixer only has 3 matrixes, but I recently started using 1 for uppers, 1 for subs, and I'm experimenting with 1 for the live stream.

[@MichaelCurtisAudio]

I'm assuming it's A&H SQ or Qu? I like those desks for the most part, but only three matrices is a beef for me : )

[@InfamousEllis]

@@MichaelCurtisAudio yep! An SQ

[@kieranniemand2939]

Dave Rat would like a word... lol jokes aside.. all valid points, especially the part about it being uncorrelated with a stereo recording/stream/ etc. However theatre world where you have a million different speakers all over the place, correlation takes a bit of back seat when you have time to rehearse and program a console for 2 weeks and then aux fed subs are way more common. Great video as always!

[@MichaelCurtisAudio]

Thanks, man! I'd love to have a conversation with Dave Rat about it. He's a brilliant dude and I think it'd b a fun talk.

[@psmail007]

​@@MichaelCurtisAudioyeah ... Dave has a video and essentially he says noise is the reason he insists on aux fed subs ... Such as the noise that comes from gusts of air from instruments - say hi hat - that aren't designed to create low freq sounds. kzbin.info/www/bejne/nqGpfGSOhL-oi8k. I don't understand enough about matrix mix to understand your approach ... But I'm up for the learning!