8. Comparing Preamp Noise on Interfaces, Mixers and Recorders
Check out the Clean Cut Audio Patreon page and consider becoming a support or a student! Tons of bonus content and coaching/consulting bonuses at a SERIOUSLY reduced rate!
When it comes to interfaces, mixers, and portable recorders, not all are created equal. I would even dare to say neither of them are comparable to each other! There are many reasons why I think this, but the spec we're focusing on today is the preamps, and specifically the noise created by budget quality recording devices.
Let's start at the very beginning... Any electronic device creates noise, maybe on a microscopic level, but in terms of audio recording equipment, they can all sum up to something that's audible and very annoying to have in a recording. The metric we're concerned about when buying audio recording equipment is the signal to noise ratio, and that's dependent on a couple of things. The 2 specs I really look at when shopping for interfaces (which is about every 10-12 years if you buy right!!!) are EIN and Gain. This will determine how low the noise floor is (self noise, preamp noise, hissing and humming), and how high you can turn up the gain in order to get the loudest and clearest signal possible.
Most budget interfaces we are dealing with, the Behringer UMC, the Scarlett 2i2, and things of that nature, only offer 51 to 55 dB of gain, which isn't enough to drive a Shure SM7b to a healthy input signal. What's even worse is when you turn the gain up to 11, it brings up a LOT of noise with it, cause your signal to noise ratio to be smaller, which results in low static mixed in with your voice that is terribly hard to get rid of.
In this episode of the podcast, we'll listen to 3 different samples of audio recorded on 3 different devices (or 4 if you consider my recording as well), and we'll explore WHY there is so much noise in 3 of the 4 records, and what we could have done to record a better signal at the source.. Spoiler alert, we could have bought a better, but NOT necessarily more expensive recording device.
In the episode we have a Shure SM7b recorded on a Tascam DR-40 recorder, and the noise is truly something to behold. I often find that handheld recorders, ESPECIALLY Zoom recorders and really anything made by Zoom has a VERY high noise floor and it is exacerbated by their low gain range, causing the user to max out the gain, and consequently max out the noise. I NEVER recommend handheld recorders to anymore because of their famously high noise floor, but I totally understand the convenience of them. In 2 days I'm heading to California for 2 weeks and then Florida for 1 week. I have a portable recorder that I found on a dumpster years and years ago, but I still opted to murder myself the 2 weeks before leaving town to get all my videos and 8 episodes of my podcasts recorded beforehand because having great audio quality is so important to me. The convenience of a portable recorder still couldn't convince me to use one (unless it's a MixPre 3. Those are dope and totally acceptable. But hey, you get what you pay for).
The next device we demo is a Line6 UX2 that I bought when I was 19 years old because I wanted the proprietary software that came with it, PodFarm, which is a great guitar and bass amp emulator. This device also had a low gain range and a high noise floor, AND minor ground loops that you can both hear in the audio and see in the spectral analyzer. Removing the noise from this device was rather difficult. The worst news was the device was something like $300 back in the day, and SSL now makes an interface for $230 that absolutely blows this thing out of the water. We are truly living in a golden age, but don't take that phrase to mean every device under $300 is acceptable, because most aren't.
The last sample we used is the most unfortunate of all. This podcast is recorded in a professional podcasting studio, and it was an absolute perfect storm of bad mic technique and improper gain staging that made this episode a catastrophe to work on. The guest was 16 inches away from her microphone, which was a Shure Sm7b, so the input levels were SUPER low. To combat that, the engineer cranked the gain on their Allen & Heath mixing board ($1,000>) and it created a TON of noise that was in the same frequency range of the woman speaking, so it was nearly impossible to get rid of. This example highlights that even thousands of dollars spent doesn't always buy you clean preamps, and bad mic technique will only make it worse.
So the main takeaway from this episode is money won't buy you silence, so it's up to you to investigate specs on your interface, mixer, or recorder, but 9 times out of 10, an interface will give you cleaner preamps!
Thank You by Joakim Karud
My Signal Chain
Audio Interface: Apogee Ensemble
Microphone: Shure SM7b
Headphones: Audio-Technia ATH-M50x
Earbuds: Klipsch R6i II
Studio Monitors: Yamaha HS7
Mic Stand: Rode PS1A Boom Arm
IzoTope RX6 De-Mouth Click
IzoTope RX6 Voice De-Noise
Waves Vocal Rider
Waves L2 Limiter
Waves WLM Meter
Waves Durrough Meter
*most of these links are affiliate links*
Hey, what's going on everybody? This is Tom Kelly from Clean Cut Audio and I have a really exciting episode for you guys today. Last week I talked about things to keep in mind when buying audio equipment for your podcast. One of the biggest points I tried to make was investing in a great preamps for your interface, or rather investing in an interface that has great preamps in it.
You know I like my listening exercises. So this week we're going to listen to four examples of different signals being recorded with different quality interfaces. And I'm going to do a before and after of the signal with the noise, and then the signal after trying to remove the noise with Izotope RX.
All right, let's do it.
Before we get too far into this episode, I do want to make a very exciting announcement. I've launched a Patrion for Clean Cut Audio. It's going to go for both the videos and the podcast, and I'll talk about it more at the end of the episode if you're interested, but I think I found a very unique way to kind of partition this as either people who want to support the video, get a shout out, have your name in the videos, but there's also a very interesting coaching and consulting program that's going to be done through this Patreon that's going to be a huge discount from my normal coaching and consulting for people that really want to take their audio to the next level. I'm not going to bother you with that right now. If you're interested in the Patreon, stay tuned to the end of the episode and we'll talk about it more.
The first thing to talk about is why do preamps make noise at all? Like, why are they buzzing or humming or hissing? Well, any electronics just inherently make a noise as you are passing the electricity through these copper wires, through the transistors, and I'm not an electrician or really that savvy with the electronics part, but every time you add a component in this series of electrical flow, we'll call it, you know the signal going in and out of the preamps, the interface, the converters and all that stuff, there's just noise. It's going to dirty up the signal and higher end audio equipment, they take better care to use components that will introduce less noise and that's typically what you're paying for, is just higher quality components.
I mean, think of an Ikea bookshelf. It's not even fiberboard anymore, it's this other weird thing that's not quite wood. It falls apart easier, it flakes off, it's lower quality. But if you buy something from a small retailer, or if you know a woman who builds furniture and she's using a hardwood and really old school joinery techniques, that dresser is just going to last a lot longer, the finish isn't going to start flaking off, the chipboard isn't going to start peeling away. It's just going to stand the test of time. And that's similar across all practices, higher quality stuff, it's gonna look better, sound better, it's going to last longer, and it's going to have more integrity.
So, that is my way of trying to explain preamp noise without really knowing the electronics myself. And it's not all about price either. There's other manufacturers like Zoom for example, I don't know what their deal is necessarily, but they just don't seem too concerned with the noise floor. I was looking at some spec sheets and I saw that, on average, other interfaces, the Scarlet 2i2 has a noise floor or the EIN that we mentioned last week, of -129 dB. I saw a couple of Zoom devices that aren't the cheapest, you're spending $300 or $400 and their noise floor was like -110 decibels. That is 20 decibels louder of preamp noise and system noise than a Scarlet 2i2, which is significantly less money. So they're sacrificing possibly preamp quality for all the features, the internal routing, some of them have onboard compressors and effects, but, you really need to know what you're looking for when it comes to an interface, because this preamp noise will ruin you.
So as I mentioned, I'm going to do a couple of before and afters. I'm going to explain what's going on with the signal and what I did, but just know that in this listening exercise, I used Izotope RX. I'm still on version five using spectral de-noise. I know that Adobe Audition comes with its own noise reduction stuff, I wanted to try it out, but I spent like an hour trying to get all of these examples ready for you and I just am running out of time.
So I use Izotope RX, and again, keep in mind, Izotope RX is like $400 or something. So in order to clean up your signal, you're going to have to pay for it, unfortunately. Audacity has its own built in stuff. It's okay, but some of these 60 cycle hums and really intense preamp noise is difficult. So let's listen to a couple examples.
The first one is from a Tascam DR-40 recorder, which is very similar to the zoom, H4N or H4N Pro. It is a portable field recorder, it has its own built in stereo pair of microphones and just like the Zoom, it features a very, very high noise floor. It doesn't have enough gain to properly drive the Shure SM7b so it came up short. I had to boost it later. This is a situation where you would probably need something like a Cloudlifter to get the signal the rest of the way there and possibly get a cleaner signal. And I realized that all of last week when I was mentioned the Cloudlifter, I was saying they were $100. They're actually $150 per channel. So again, clean signal, you're paying more than you think. Might as well get the SSL2 interface that has plenty of clean gain to go around, or any other interface that has a lot of gain and clean preamps.
So let's listen to this example here. This is the dirty, untouched signal, just boosted to negative -16LUFS
TASCAM EXAMPLE: Hey there, this is Tom Kelly and I'm recording my Shure SM7b into a Tascam DR40, an old recorder that I found next to a dumpster, tried to return it to its rightful owner, give it about three months, no one claimed it, so I took it for myself. It's a totally decent recorder. I was once stopped in an airport for trying to bring a taser onto an airplane. It is not a taser, alas, it is the TASCAM DR40 portable recorder.
And here's that same signal after running it through Izotope RX. And again, I do want to mention that the gear I'm using is the same. I'm still using the Shure SM7b and the same XLR cable, its just going into the Tascam rather than my Apogee Ensemble. So here is that same signal after running it through Izotope RX.
TASCAM and RX EXAMPLE: Hey there, this is Tom Kelly and I'm recording my Shure Sm7b into a Tascam DR40, an old recorder that I found next to a dumpster, tried to return it to its rightful owner, gave it about three months, no one claimed it, so I took it for myself. It's a totally decent recorder.
Real quick before and after on the Tascam recorder. Let's do it. Hey there, this is Tom Kelly and I'm recording my Shure SM7b. Hey there, this is Tom Kelly and I'm recording my Shure SM7b.
The next sample I'm going to play as a quick one for my cohost Pat from the Reminiscent podcast, my music commentary, pop punk and emo podcast. We were using an old Line 6 UX2 that I've had since probably 2010. I used it for guitar tones back in the day and it only had 50 decibels of gain I think, an insanely high noise floor. And let's listen to his raw signal. Don't mind his room tone. This was testing out the preamp noise. He doesn't have that much reverb in the actual podcast is just for the test, but listen to how noisy this signal is.
LINE 6 UX2 TEST: And my mouth is about six inches away from the mic Tom, Tom, Tom. Sun in an empty room. The weakerthans. Canada. Red, green. Jack black. Black Keys. Jack White. Relient K. Five Iron Frenzy. Switchfoot. Fans that are kind of religious... Religion.
My lord that hiss is bad. And this is after RX.
LINE 6 UX2 and RX TEST: And my mouth is about six inches away from the mic. Tom, Tom, Tom Sun in an empty room. The weakerthans. Canada. Red, green. Jack Black. Black Keys. Jack White. Relient K. Five Iron Frenzy. Switchfoot. Bands that are kinda religious... Religion.
You may have noticed, especially on that last one, that the preamp noise is kind of is rolling in and out along with the voice. And the reason why that happens is we are sampling the noise in Izotope RX. We need to get a reading of what is noise so it can then extract that from the audio. And if you don't have a lot of just pure room-tone or system noise to begin with to sample, you're not getting a super accurate measuring, so it's not pulling it out of the signal very well. So when there is no noise... like that right there, it can easily take all that noise out, but when it is noise and signal, your voice happening at the same time, it has a much harder time ripping the noise out of the signal when there's a lot going on. So you'll hear that the louder the person talks, the louder the noise gets and it's kind of permanently attached it to the amplitude of the signal.
We can maybe make it better, but part of the issue with noise is that the human brain tends to kind of forget about constants. It recognizes changes more than just everything happening in one spot. Think about, you know, you're kind of zoned out and you see out of your peripheral vision something move and you snap back to it. You notice the movement. You didn't notice everything that was there before, but now that there's a change, you're honed in, you catch that. It's kind of like that with noise. So when we have noise coming in and out, up and down. I don't know if I have a firm position on it, but I'm inclined to think that there has to be a tipping point where it's actually less distracting to just have there be a constant drone of noise than having it come in and out every time a person takes a breath or or takes a quick pause.
Now there's something that we might be able to do, which is essentially, err we're not going to get into dithering, but you can mask an unpleasant noise by adding a more pleasant noise to it. It sounds very counterintuitive, but think about it this way. With that last signal, we had noise coming in and out when my buddy was talking, and you can hear the noise coming in and out. So what if we had a recording like a five, even a two minute recording of a little bit of noise, just a little bit, and we looped it throughout anytime the one speaker is audible. So what we're doing is essentially trying to reduce the obviousness of the bad noise coming in and out, and we have a very, very low drone of noise that's kind of masking it. This is something that is probably meant for a YouTube video on why we shouldn't have moments of absolute silence in a vacuum in our podcasts and why when I'm cutting one track of audio, I always make sure that something else is overlapping it so that there's never a pure silence cause that doesn't really sound natural. But it's something to keep in mind that when you have noise in your signal, it either needs to be all the way gone or it needs to be still a little bit there, not moments of being there, moments of being gone. It's more obvious when it's in and out. Just human instinct, I think most animal instinct is to notice the changes, so we don't want that in our recordings, especially.
And the last sample that I have, it was a client who is in a podcasting studio. They have an Allen & Heath mixing board, and they had a client... This is like the perfect storm of things that could have been done very well. A few minor things added up to make a very, not a bad recording, but something that I definitely had to take time to repair. And I'm an idiot, and in that contract, I didn't have a surcharge for audio restoration, so I kind of ate it on these two episodes before I had them fix it at the source. Because again, we need to fix these problems at the source. Sure, we can buy Izotope RX and try to do spectral repair and spectral de-noise, voice de-noise, but what if it was just never there to begin with? Then we're not trying to take things away, and when you're trying to take away noise, you're just going to be degrading the signal a little bit. Maybe it won't be noticeable. Maybe it will, depending on how bad the noise is.
But with this client in particular, there were three things. There was a very soft speaker who was nervous, so she was very far from the microphone, like a foot and a half, two feet. So the engineer had to crank the gain so loud in order to get a good input signal that the mixing board, the analog mixing board introduced a ton, and I mean a TON of preamp noise. These boards are over $1,000 and I've mentioned it and a few episodes, a few videos why I'm not a fan of board mixers, analog mixers. I know they have their time. I know they have their place, but I typically see that most people, unless you're doing a live situation, can get away without a mixer. And when you use an interface, typically you gets a cleaner signal.
So again, $1,000 and it still wasn't a clean signal. So just, you know, whatever. Take that as you will. I just thought it was kind of funny, but either way, we had a lot of gain. We had someone with bad mic technique, and then what was even worse and bringing up a lot of noise, it also made the microphone bleed from the other two hosts, it made it become much, much worse in her microphone because we're increasing the sensitivity of the preamps, it's picking up more of all the unwanted stuff in addition to what we, the only thing we wanted, which was her voice.
So I've mentioned it in YouTube videos, in one of my most recent episodes, get close to that microphone. I mean, get real close to that microphone. I'd say probably no more than four inches is what I would say. But you have to experiment for yourself and find out what you like the best. So I'm going to roll this audio. And again, this is Shure SM7b on an expensive board, and without mic technique, you will not get good results. Doesn't matter what the gear is, you have to do things right on your end. So let's roll this.
MIXING BOARD EXAMPLE: Well, in practical terms, I only discovered this particular UK rule and interpretation under the English companies act because the board I was on was involved in a very heavy duty international transaction.
And after RX. Not a ton better, but definitely better.
MIXING BOARD AND RX EXAMPLE: Well, in practical terms, I only discovered this particular UK rule and interpretation under the English companies act because the board I was on was involved in a very heavy duty international transaction
And what I'm hearing after the RX is a degradation in the higher mid-range of her voice, which are very important frequencies. They were kind of diminished and attenuated as it tried to remove noise that was occupying those same frequencies. So we can kind of get the noise out, but at the expense of the quality of the signal. So, if you're listening to this show, you're probably one of the people that already know this. We need to be getting good signal at the source. Yes, we can do a lot of work in post, but the goal should not be to fix it in post. When I was a very young audio engineer working on music... I don't really like recording. That's the thing. I didn't like recording bands. I liked editing. I loved quantizing drums so that every hit was damn near perfect, and all the guitars and the bass were matching up with the kick drum to make it rhythmically powerful. So I was always drawn to post.
So I was one of those little 19 year old, thinks they know everything guys who said, "dude, it doesn't matter. We can fix it in post." But what I always ended up with then were lifeless, artificial sounding songs. And it's because I wasn't taking care. And with podcasting, I know that it's, to say it's simple, is, um, might be... What's the word? Not super respectful or aware of people that have never done anything like this before. But in terms of audio recording, the simplicity of a podcast is pretty much unparalleled in the rest of the recording industry. I mean, most bands and musical performances have a hundred plus layers, and when you think you're hearing one voice, it's actually seven layered together, and it's never just one guitar, it's six, but they make it sound like one. There's such advanced principles that when it comes to capturing a voice, especially for a podcast, we're not trying to do anything crazy. We're not adding these crazy effects. We're just trying to make the voice sound as natural as possible. That should be the goal, natural.
And our listeners are already listening in noisy environments, so we're already competing with noise. We don't want to be adding, then, more noise into our signal so that we're competing for even more, audibility... Is that a word? Either way. More signal, less noise at all times. And when we are not investing in quality preamps, we get noise. That's just kind of a fact, really. I mean, there are interfaces that are affordable. $180-$230, the Motu M2 which does have some 60 hertz cycling in it. Kinda like some, sounds like some very minor buzzing that is very easy to take out with Iotope RX, especially when my cohost gives me 10 seconds of absolute silence so I can sample it and take it out.
And by the way, when I say silence, I've had to train some of my clients on this. It's not breathing and scratching and rubbing your hands and shuffling papers. I mean, it's almost like leave the room silence. Take a deep breath and hold it for as long as you can and do not move a muscle. Because if you're crumbling a piece of pape, now you're sampling a crumbling piece of paper, and it's now removing those frequencies from your signal. And that's not what we wanna do. We wanna just remove the noise. So if you are a service provider, you know, as am I, where I edit shows for clients. I know I know that it can be frustrating to feel like you are offering advice to these people and they're just not taking it, but I think it is our duty to educate our clients. If for no other reason, then selfishly it makes your job easier.
I wish a lot of times that I were paid for some of the consulting work that I do. And the clients... It's funny, the clients that pay the most respect your time the most and are willing to pay you for consulting, but I still have some clients from very early on when I was just getting started out. They're already severely underpaying me just because they've been around for a long time. But there's, there's just things you got to hold their hand through, and setting the gain appropriately is one of them. Or not buying a terrible device is another. And I think as a service provider, as a podcast editor, or engineer, or whatever you want to call yourself, I think you owe it to your clients to teach them the best practices. It's up to them whether or not they want to put them into practice, and then it's up to you whether or not you want to continue working with them because their podcast quality is a reflection, unfortunately, of your quality. But I think we need to be learning as much as we can ourselves and then passing all that information or as much of it onto our clients as they're willing to accept.
And as you get busier, as you can get more picky and take on only the clients that you want to work with, you can then choose to only work with clients who will do what you want, or buy the equipment you think is great, or sit down for an hour on the last Friday of the month and pay you to go over best recording practices. Hey, these are the things I noticed in the month of April, how can we fix them in the month of May? And sometimes it comes down to, hey, we might need to buy you a better interface because all this noise is degrading your voice, and I want what's best for you. And I would be ecstatic if you would invest into your sound because it's helping your reputation, your perceived authority, and it's helping me do my job easier so that I could focus more on making what's good sound great, rather than making was bad sound just okay. That's, I think, the important lesson in all of this.
So whether it's your podcast or you're working on a podcast for a client, just know that your interface and the preamps within, we're not even talking about conversion yet. The pre-amps will greatly affect your sound. So we need to be looking at EIN on our spec sheet. And if we can't find them since not all interfaces, it's not always easy to find that information. And sometimes information can be very misleading. We should be trying to find audio samples online. I know that Podcastage, I'm not like... I don't want to say anything, but either way. There have been some great demos and reviews of interfaces to where you can hear the noise of an interface before you buy it. But we should be buying equipment with these things in mind and understanding why recorders like the Zoom, if you've ever worked with those, why they might sound a little bit different than something else in the similar price range. It's because those preamps are garbage and noisy. Sorry, Zoom.
But thanks for listening of through these listening exercises. I don't know if people want this kind of content, so reach out. Let me know if you like this. And I don't know why I'm like apologizing for being really niche. I think that's one of the first marketing tips that anyone teaches, but this is some really niche stuff, and I think it's interesting. I wish I could test the noise floor on every interface I could ever think of and put it on this show, but that's not really a possibility since I don't have any sponsors yet. But now might be a fun time to talk more about the Patreon because there's some kind of exciting stuff going on in there and I've got some pretty big ideas for it. As my dog walks through my room and steps on pieces of paper and makes a bunch of noise in the back. Hi puppy. Hi puppy. It's one in the morning and he's wondering why I'm still awake.
So let me go over the five tiers that I have available for the Patreon. At the very bottom, I have the supporter tier, which is $3 a month. This is the biggest thank you in the world for helping me continue to do what I do with the podcast and with the videos. I really want to be in this kind of educational space and use my platform to help others, not just myself or my clients. I feel like it can help more people this way. So that $3 a month really helps me make up the difference of "the money lost while not producing", but also just kind of a thank you for, you know, the effort that I'm putting into the shows and to let you know that I've really, really appreciate it.
The next year is the fan at $5 a month, and with this you get listed on my wall of thanks on my website, I'm going to have a membership page where I'll list your name and hyperlink it to anything that you want. So if you want it to link to your podcast or your podcast production, I can totally do that for you. I'll put that on my wall of gratitude on my website.
The next tier is the co-executive producer for $10 a month. You get all the perks from the ones before listed on my wall of gratitude, and you'll get your name in text in my videos, on all of them moving forward in the membership part of the video. So you'll get your shout out in the videos and I'll shout you out on the podcast as well. So two places for that $10 a month. And not only that, in the Patreon, I'm going to be doing biweekly AMAs. So it's a really good, that stands for "ask me anything" really good chance to, any problems that you're having, you can get an answer from me and other people in the group who are really looking to do better audio. And you'll be able to use that to suggest upcoming videos. And I will do my best to do all of those for all of you guys, use the AMAs as content for the videos and answer your questions on the spot in the Patreon.
The next tier, and this is kind of slightly different, rather than being a supporter, this is being more active in the educational space. I have the student tier at $50 a month. And this is for people that are really, really looking to up their stuff in podcast production. So, with this, you get all of the perks from before, the AMAs, the wall of thanks, the name in the videos, the name and the podcast. But also once a month I'll be doing mix critiques for all students in this tier. So I'll listen to your show, I'll put it into Pro Tools, find out what it needs, maybe if there's anything over done, and provide a written report for you every month with tips and advice for making your most recent episode sound even better. This will be mixing. It'll be editing. It'll be all of that, and it'll be very actionable. So I can recommend things for you to do to make your show better. And then the next week your show can sound better than the one before. So that is the student here that's limited to 20 people.
And then the next tier is the master. So for $100 a month, this is more like a coaching program, is how I'm trying to fit this all into the Patreon. You get all the perks from before, the mix cits and the wall of thanks and the AMAs, but this also features a monthly, a 60 minute one-on-one coaching or consulting call with me.
This is a 33% off of my normal rates. I usually charge $150 for a 50 minute coaching or consulting call. So if you're looking to just really, really, really up your game, this is a way to get, basically a pretty big discount on coaching. And not only that, if you want to have more than one 60 minute call a month, you'll receive 25% off all additional consultations and coaching calls moving forward.
So that's the Patreon. I really wanted to have two different functions for it. People to just support the work and people who are like really looking to do some serious coaching. You can do all those by contributing to the Patreon. If you go to patreon.com/cleancutaudio, you'll find these five tiers and you can sign up for any of them that you'd like, but just know that no matter what tier you choose, I'm so incredibly grateful that a) that you're here, and b) you like it enough to want to take this relationship to them the next level. I really appreciate it. And for those of you who either can't contribute to the Patreon or just like don't want to, that's totally fine. I'm going to continue to make these shows for free for everyone. Just those are the options that are there that are available if you're looking to take this relationship to the next step. I really appreciate it. I appreciate all of you no matter what.
So that is it for this week. Please subscribe to the show, and if you're loving it, rate and review it on Apple podcasts and Podfest is coming up soon, March 6th through the 8th. Finally got my flights the other day. If you're going to be in the area in Orlando for Podfest, please make sure to say hi. I be ecstatic to say hey to anyone out there that listens to the show and has learned anything from my videos or my podcast. It'd be really cool to meet you, so I will talk to you all soon.
All right. Bye everyone!