Spectrum Analyzers. Great tool!

[rob aylestone]

Out of curiosity I just looked at the new ABBA album - people who have sold a fair few tracks over the years and the majority shelve gently above 10K, but a few have lots of harmonics up the top - 16-20 actually has content - another track has a hump around 10K. Some have very little bass content, others lots. Very obviously from the waveforms some are heavily compressed and other are not. No attempt here to achieve a flat, or even flattish response - In fact the spectral content is a bit odd. Some, like these have very little energy at all in some areas - time is horizontal 20 20K on the vertical axis.

 

[mjbphotos]

I use Span in the 'Stereo Mix' mode, it gives a nice easy-to-understand 'averaging', nothing like the Abba screenshots above.

 

[Mr. ROUSH]

Izotope Tonal balance is practically a spectrum analyzer with an intelligent hold time, it's great though!

I use ProQ2 on the master as my spectrum analyzer, it's great for that! It has a tilt enabled by default for that meaty low end/smooth highs if you get a relatively flat line. It's a great middle ground, you can go either way a bit but at least the analyzer gets you bloody close.

I don't need to use it anywhere near as much anymore, I use it nowadays mainly just to confirm that I didn't miss the mark, after just using my ears only. But I still find it essential for internalising what a balanced mix should sound like..... it's a way more reliable way of learning your speakers than that old method (listen to lots of music you know well on your monitors). Because let's face it. We can't trust our ears really can we, not when we are learning this stuff. Many a time I have been working too long, think I have a tonally balanced mix and then use the analyzer to find that the 9khz pick attack is leaping out of the speakers. THEN! my ears will hear it as a problem. But the analyzer told me first. Happens quite often.

I'd pick up on that problem straight away if I listened to the mix on my laptop, or if I came back with fresh ears. But to me this is just another win for visual aid.

The Spectrum analyzer has been the single most important tool in helping me understand how to mix, and to what a good mix behaves like within proQ2 only. I used to use SPAN but ProQ2 is where its really at for me.

Fabfilter plugins are absolutely brilliant! Hard to beat! Love the tilt feature!! That's one of a handful of features that don't exist on the plugins that are in the ballpark.

That's a perfect example. Thanks for sharing! My experience has been very similar. It's a huge help to keep my ears in check. It's just kind of a no brainer, really. After listening to music for extended periods of time, the brain tends to enable its own eq to filter out undesirable frequencies. If I hear something on my initial pass that I don't like, but then later it starts to sound better before I've even done anything with it, it's a red flag. In situations like that, it's fantastic to have a visual reference.

Thanks again for sharing!

 

[Mr. ROUSH]

Out of curiosity I just looked at the new ABBA album - people who have sold a fair few tracks over the years and the majority shelve gently above 10K, but a few have lots of harmonics up the top - 16-20 actually has content - another track has a hump around 10K. Some have very little bass content, others lots. Very obviously from the waveforms some are heavily compressed and other are not. No attempt here to achieve a flat, or even flattish response - In fact the spectral content is a bit odd. Some, like these have very little energy at all in some areas - time is horizontal 20 20K on the vertical axis.

I'd say that ABBA is not the best example compared to the bulk of popular modern recordings. I don't think anyone has suggested that a mix always has to be flat or that it can't be dynamic. There are exceptions to every rule. The point is to simply have a visual reference that gives you a picture of what any given sound might look like on the analyzer. If someone wanted their master to sound like an ABBA record, a spectrum analyzer could help a lot to get them in the ballpark. If someone wanted their master to sound like a modern rap record, a spectrum analyzer could help to get them in the ballpark. You can certainly observe very distinct and pronounced differences if you compared the ABBA record to a modern rap record. The list of helpful uses of a good spectrum analyzer is pretty long.

 

[rob aylestone]

Sort of, maybe? but it does rather prove that content is the key. I'm not remotely able to comment on rap, it's a genre that will always be a mystery to me. I don't think a spectrum analyser would get anyone remotely in the ballpark as every song is unique - some thick, some thin. Some busy, some sparse. EDM, rap and perhaps thrash metal could be exceptions to the musical rule? I don't really know. Here is an Abba CD - you can easily see the difference in the style - some tracks are compressed and limited with no quiet bits, other up and down a lot. Some very quiet at the start and only match the others at the end, I think this is why I have trouble with a formula for a mix.

 

[Mr. ROUSH]

Sort of, maybe? but it does rather prove that content is the key. I'm not remotely able to comment on rap, it's a genre that will always be a mystery to me. I don't think a spectrum analyser would get anyone remotely in the ballpark as every song is unique - some thick, some thin. Some busy, some sparse. EDM, rap and perhaps thrash metal could be exceptions to the musical rule? I don't really know. Here is an Abba CD - you can easily see the difference in the style - some tracks are compressed and limited with no quiet bits, other up and down a lot. Some very quiet at the start and only match the others at the end, I think this is why I have trouble with a formula for a mix.

Therein lies the point. A spectrum analyzer can show you if a mix is thin, thick, etc, in a given frequency range.. It's about comparing your master to other mastered recordings that you wish to emulate. Obviously, if you put an analyzer on one of the ABBA songs that was heavily compressed, it would look different than the tracks that are very dynamic. If I wanted to emulate one of the more dynamic recordings, I would reference it. If I was trying to emulate the sound of one of the more compressed recordings, I would reference it. It's also a good idea to loop certain sections and to observe the loudest sections. It's not a "formula" for a mix. It's a visual representation of any given mix that can give you a general idea, and yes, can very much get you in the ballpark of the reference.

 

[Massive Master]

I use one (and various other meters) for calibration purposes and then turn as many as possible off while I'm working.

Unless there are clients present. Then I turn them on so they have something to watch (sounds sarcastic, but totally serious).

 

[JamEZmusic]

I don't think you should mix flat at all. I am looking for those bumps and troughs during different sections, but for massive hard hitting chorus's I do think it's good to get a very dense flat/limited/wide frequency response (wall of sound) for maximum impact when those type of songs call for it. This is obviously not the case for every song though. I wouldn't be surprised if the Abba songs looked pretty similar during dense sections if they exist. I would still learn a ton if I used a spectrum analyzer on those Abba songs, I'll bet no key element is jumping out though. and although there are gaps in the frequency spectrum you could probably still draw a line across the peaks and see nothing jump out way above it.

I'm never tempted to fill in holes in the mix if I see one.

But the very common problem in mixing is if you have 1 bright element, then you will have a tendency to mix everything else a little brighter and get a harsh mix, or if you have an out of control kick, you'll might balance to that, not as common because as MrRoush said above, our ears get so used to things being bright, our ears literally EQ that stuff out by themselves after a while, you can't tell me that you have never mixed something too bright! Just based on that fact, how can a spectrum analyzer be such a terrible thing? And no matter how often I successfully mix translateable mixes, I still do fall into that trap. I'm Human, even pros do this. It's a regular discussion I see and is what makes mixing so challenging. Especially for those that do long hours on it, instead of little and often.

A lot of the older music from the Kerrang era can look pretty wild at 10k etc. But.... this is why a lot of those songs can sound like ass on smaller speakers, all you can hear is spshh spshhh spshh from the crash's. But they do have an exciting element to them when listening on speakers with broader response.

The analyzer just helps me pick out glaring faults that I may otherwise miss with the onset of ear fatigue. The analyzer is just a metering tool, and as metering tools go, they get you in the ballpark. I can mix a song that looks flat using a spectrum analyzer but sound muddy/indistinct as hell, so ears are needed for the left to right frequency balance aswel as amplitude because in this regard an anaylzer isn't very useful at all.

To be honest, for an aspiring mixing engineer, the first plugin I would recommend would be Metric A/B, which is an all in 1 referencing tool, it details everything you need to know if you are serious about matching your reference as closely as possible, lufs, stereo imaging, spectral balance, with an automatic level matching function so you can flick between your mix and several other loaded in references in real time. I don't think I would be doing him a disservice at all. Referencing is important, and referencing with amazing metering tools can only help IMO!

It's almost at the point of this discussion that if the advice is to just (use your ears) then why bother even using references at all. It's ridiculous.

edit: I wouldn't recomment to anybody else other than a beginner, pro's don't need references, metering tools etc. but a lot of us do not have the luxury of mixing 8-16hours per day for 30years.

 

[rob aylestone]

I don't use references - apart from perhaps when I'm working on a show track of a well know song. It's fascinating to see the impact of my evolution in recording compared to other people.
You mentioned this bit

the first plugin I would recommend would be Metric A/B, which is an all in 1 referencing tool, it details everything you need to know if you are serious about matching your reference as closely as possible, lufs, stereo imaging, spectral balance, with an automatic level matching function so you can flick between your mix and several other loaded in references in real time. I don't think I would be doing him a disservice at all. Referencing is important, and referencing with amazing metering tools can only help IMO!

Stereo imagining is something I've really picky about, but I'm not really sure what spectral 'balance' even is? I just want things to sound right, for the piece I'm working on. even worse, practically all my tracks don't get a vocal added at all, apart from a few where the eventual singers might need it in the early stages for a particular tricky bit? So I get better at imagining the result with a vocal. Only a few songs require me to put in the vocal act an early stage to discard it later. I just can't imagine switching between these loaded in references, and with the covers, you would think this would be important, but it's not. I can't see how auto level matching would help me. Levels wise, my monitor settings never change, so I rarely even look at the levels on the meters. I did go through a short LUFS stage when I noticed Cubase could provide me with all , but I've got a standard I base things on and I prod faders till it sounds right and at the final stage I'll tweak the overall level up or down a bit to match the other tracks. Maybe it's because all through from the 70's the quality jumps kept coming fast and furious and I got my changes in process gradually with each upgrade, but then it started getting silly. A young guy I work with has so many plugins that his cubase projects are crazily complex for, I often feel, nothing. He never finishes a recording without applying his standard processing to the bass, guitars, drums and keys - then he starts on vocals. Then he gets his mastering ones going. I just cannot do this. I could get by on the basic ones that come with Cubase. Often I spot things in lists I did not know I had - perhaps being imported from other software, I'm just not interested.

 

[LazerBeakShiek]

guy I work with has so many plugins that his cubase projects are crazily complex for, I often feel, nothing. He never finishes a recording without applying his standard processing to the bass, guitars, drums and keys - then he starts on vocals. Then he gets his mastering ones going. I just cannot do this.

It is confusing to. In the Youtube video's the templates in the DAW have multiple compressors and LPF/HPF filter sets. Not to mention an EQ, an IR loader(EQ), bass enhancer (EQ). What ever happened to enjoying just the sound of the amplifier? Pure clean recording. Nope. Its disappointing when so many VSTs are used.

 

[rob aylestone]

I want to enjoy the music, I don't want to hear anything the recording chain adds - so if an amplifier has a 'sound' I don't want it. I realise people love these things nowadays, but give me a mic, a decent recorder and a pair of speakers and that'll do. Over my history, the biggest issue was always hiss. Once digits came along I was a happy bunny. I'd spend ages finding just the right mic position so it didn't need EQ, and then I'd hand over the finished recording. You really needed those expensive analogue reel to reel recorders and the Quad amps, because they hissed less, and distorted less. Then I got the Sony F1 to go with the betamax recorder and loved digital.

 

[Mr. ROUSH]

I don't think you should mix flat at all. I am looking for those bumps and troughs during different sections, but for massive hard hitting chorus's I do think it's good to get a very dense flat/limited/wide frequency response (wall of sound) for maximum impact when those type of songs call for it. This is obviously not the case for every song though. I wouldn't be surprised if the Abba songs looked pretty similar during dense sections if they exist. I would still learn a ton if I used a spectrum analyzer on those Abba songs, I'll bet no key element is jumping out though. and although there are gaps in the frequency spectrum you could probably still draw a line across the peaks and see nothing jump out way above it.

I'm never tempted to fill in holes in the mix if I see one.

But the very common problem in mixing is if you have 1 bright element, then you will have a tendency to mix everything else a little brighter and get a harsh mix, or if you have an out of control kick, you'll might balance to that, not as common because as MrRoush said above, our ears get so used to things being bright, our ears literally EQ that stuff out by themselves after a while, you can't tell me that you have never mixed something too bright! Just based on that fact, how can a spectrum analyzer be such a terrible thing? And no matter how often I successfully mix translateable mixes, I still do fall into that trap. I'm Human, even pros do this. It's a regular discussion I see and is what makes mixing so challenging. Especially for those that do long hours on it, instead of little and often.

A lot of the older music from the Kerrang era can look pretty wild at 10k etc. But.... this is why a lot of those songs can sound like ass on smaller speakers, all you can hear is spshh spshhh spshh from the crash's. But they do have an exciting element to them when listening on speakers with broader response.

The analyzer just helps me pick out glaring faults that I may otherwise miss with the onset of ear fatigue. The analyzer is just a metering tool, and as metering tools go, they get you in the ballpark. I can mix a song that looks flat using a spectrum analyzer but sound muddy/indistinct as hell, so ears are needed for the left to right frequency balance aswel as amplitude because in this regard an anaylzer isn't very useful at all.

To be honest, for an aspiring mixing engineer, the first plugin I would recommend would be Metric A/B, which is an all in 1 referencing tool, it details everything you need to know if you are serious about matching your reference as closely as possible, lufs, stereo imaging, spectral balance, with an automatic level matching function so you can flick between your mix and several other loaded in references in real time. I don't think I would be doing him a disservice at all. Referencing is important, and referencing with amazing metering tools can only help IMO!

It's almost at the point of this discussion that if the advice is to just (use your ears) then why bother even using references at all. It's ridiculous.

edit: I wouldn't recomment to anybody else other than a beginner, pro's don't need references, metering tools etc. but a lot of us do not have the luxury of mixing 8-16hours per day for 30years.

I think you're using analyzer plugins very effectively in your process. It's also worth mentioning that the technology has come a long way in a short time. These plugins become more and more useful all the time.

That's a great example of a good use. It seems counter-intuitive to cut in the high end when you're going for a brighter sound. But, you're exactly right. I don't hear strident resonance in the studio on cymbals, etc. It's not until I jump in the car that it becomes truly painful to listen to if I've forgotten to address it. Analyzers just keep me in check at the end of the day. It's definitely effective at identifying those glaring inconsistencies.

Honestly, I just think some people are stuck in their ways and resistant to change and newer technology. There's nothing wrong with that, per se. What works for you and for the sound you're after is all that matters. I think some old school engineers even feel a bit threatened by the advent of some of this technology. It's a bit like the automatic vs manual transmission debate. Many argue that if we were suddenly left with nothing but cars with a manual transmission, most of the younger generations, wouldn't be able to drive. But now, the reality is that manual transmissions are pretty much obsolete. We shouldn't fear a better, more efficient way to do something.

Of course, I like to know how to get the results I want without assistive technology. But, as long as that's the case, I don't mind streamlining my workflow to save time and frustration when there's a faster way to get there. From a business standpoint, time is money. These plugins are useful and just shouldn't be discouraged.

 

[JamEZmusic]

I'm a minimalist when it comes to plugins too. I try to always go for stock and use as little of them as I can. I can't see myself buying plugins in general. Maybe Soothe2, or the equivellent and much cheaper DSEQ one day because I literally have no idea how I could ever achieve similar results as what those plugins can do.

I would typically resent spending £100 on a plugin, but I do not have any problem whatsoever spending £1000 on a mic. So many of the other guys over on the Academy swear by monthly subscriptions for a bunch of plugins, they are off their heads in my opinion. I couldn't imagine all of my old projects becoming obsolete the minute I grow a brain and realise I am being conned and stop paying. Or just that you will never own anything. But that's just me.

I think level matching is really important, if your track is a different volume to the reference then you are going to balance the low and highs wrong. You can of course level match by ear but the plugin does it for you and I think it's a nice touch.

When I say spectral balance I just mean how it's balanced, like what would an analyzer say? Warm/smooth, or thinner/brighter. That Warm/smooth mix with more low end means you'll struggle to hit a certain loudness and may even be impossible. So I do think it's relevant. I personally don't care about LUFS, I will go for warm/smooth almost every time. The problem with that being that my mixes just don't sound as exciting when you listen back to back against a thin/bright mix.

I never mix songs with vocals unless it is multitracks from the academy. My own stuff is all intrumental and I am trying a lot of things out to get a polished sound. The key is definitely in recording, and this is what I am concentrating on at the moment. I wish I had high end mics like the R121 etc to experiment with but at the same time I enjoy using cheap gear, I will purposely make my own life a bit harder to force me to learn.

I do use reference tracks but I don't mix to them, it's more of a curiosity thing and I like to hear how my attempt stacks up to something that was mixed by a professional. I do not care if my mix falls short, I see it as a learning curve and if I end up hating it, the song just ends up in a folder in the pc, probably never to be listened to by anyone apart from me. To looks back on. And I don't mind!

Much respect if you don't need reference tracks and your tracks hold their own! This takes a good level of skill. This is what I am aiming for at the end of the day, I don't want to use any referencing. I am working on internalizing what a good balance mix should sound like in my own studio on my own speakers. I work in audio (broadcast) but not much to do with music so it's not having the time to really get to know this stuff intimately.

 

[JamEZmusic]

I think you're using analyzer plugins very effectively in your process. It's also worth mentioning that the technology has come a long way in a short time. These plugins become more and more useful all the time.

That's a great example of a good use. It seems counter-intuitive to cut in the high end when you're going for a brighter sound. But, you're exactly right. I don't hear strident resonance in the studio on cymbals, etc. It's not until I jump in the car that it becomes truly painful to listen to if I've forgotten to address it. Analyzers just keep me in check at the end of the day. It's definitely effective at identifying those glaring inconsistencies.

Honestly, I just think some people are stuck in their ways and resistant to change and newer technology. There's nothing wrong with that, per se. What works for you and for the sound you're after is all that matters. I think some old school engineers even feel a bit threatened by the advent of some of this technology. It's a bit like the automatic vs manual transmission debate. Many argue that if we were suddenly left with nothing but cars with a manual transmission, most of the younger generations, wouldn't be able to drive. But now, the reality is that manual transmissions are pretty much obsolete. We shouldn't fear a better, more efficient way to do something.

Of course, I like to know how to get the results I want without assistive technology. But, as long as that's the case, I don't mind streamlining my workflow to save time and frustration when there's a faster way to get there. From a business standpoint, time is money. These plugins are useful and just shouldn't be discouraged.

Yeah that's exactly it! I love plugins that help me understand and will speed up the process, my goal at the end of the day is to not have to use them at all, but anything that speeds up the process and gets me to that point quicker is definitely a must have. Time is our most valuable asset.

Everything is bloody counterintuitive in mixing. I came up against a problem the other day where a couple of notes seemed a bit harsh on my acoustic guitar, you would think some warming up EQ in the low end with analogue plugin EQ would have been the answer but instead of that, I needed to actually make the guitar a little thinner, It was the illusion of those notes being harsh because of the warmer notes that were being played inside of an overal warmer mix. I spent a fair bit of time tryng multi band, de-essing, reverb etc. In the end it was just the mix balance was off.

I'm 100% in support of using Spectrum Analyzers as a visual aid. I think it's safe to assume that everybody knows that it's not a good idea to rely on it. I just don't know how I would have ever made sense of what I'm doing without the analyzer telling me that I am heading in the right direction, it's like answering 100questions in a test and nobody telling you how many questions that you got right. You don't know what things you did right/wrong. So it's harder to learn.

I've seen videos pop up recently on YouTube by respected mixers about (using our eyes to mix). It would seem that some of the people I look up to use their eyes to mix all the time!

 

[rob aylestone]

I do think now that we are over analysing. I wonder if it's the old art or craft debate? What exactly are we doing - constantly reinventing things or making real improvements? We should also be careful about the people we listen to? If I read Alan Parsons is using X to do Y, I'd pay attention - always have, and the guys at Spitfire Audio talk my language - but so many of the current crop of names cited as 'respected mixers' are probably people I don't have in my audio collection. I've read Sound on Sound magazine for years but I'm getting more and more distant from the music the 'names' in the articles produce. There certainly are new skills, new processes and techniques, but they're applied to music that I just hate. I really do not need to see the music I'm hearing unless there is a problem I can't solve aurally. So many times we hear dreadful recordings that are fixed, massaged and tweaked to give a good end result. I'm the opposite, I'd rather spend more time producing a decent recording that doesn't need the post production.

 

[snow lizard]

Audio recording follows two different mindsets. The technical side with all the math and physics, and the artistic side which boils down to - is this arrangement of sound appealing or not?

To me, a spectrum analyzer is very useful for the technical side and close to irrelevant for the artistic side. It's a diagnostics tool. A great production isn't required to follow any specific curve. The idea of using one to get some kind of metric standard to mix by can send folks down the rabbit hole.

Being able to get a great mix is an artistic skill that requires experience and a controlled acoustic enviornment where you can hear things accurately. The analyzer can be useful in many ways to get you there, but it's not art.

 

[JamEZmusic]

When I am doing my art, drawing. I analyse how the professionals shade, layer, the pressures, gradients etc.

Just saying

(hope this doesn't come across sarcastic or rude, I'm tired and need sleep now)

 

[Farview]

When I am doing my art, drawing. I analyse how the professionals shade, layer, the pressures, gradients etc.

Just saying

(hope this doesn't come across sarcastic or rude, I'm tired and need sleep.

Do you do that by holding a light meter up to it?

 

[Mr. ROUSH]

Do you do that by holding a light meter up to it?

Ha! As a professional photographer, I appreciate the light meter reference. lol. Actually, I think a light meter is a better argument in favor of spectrum analyzers. Light meters analyze the ambient light, or flash, in order to dial in the correct exposure settings on your camera. Creatively, you might decide to overexpose or underexpose. Nonetheless, the general reference is a great starting point and almost always yields accurate results. I wouldn't say that a spectrum analyzer is quite as accurate as a light meter. However, it's a useful tool. If holding a light meter up to a drawing helps to analyze it, improving your efforts, go for it!

 

[Mr. ROUSH]

Audio recording follows two different mindsets. The technical side with all the math and physics, and the artistic side which boils down to - is this arrangement of sound appealing or not?

To me, a spectrum analyzer is very useful for the technical side and close to irrelevant for the artistic side. It's a diagnostics tool. A great production isn't required to follow any specific curve. The idea of using one to get some kind of metric standard to mix by can send folks down the rabbit hole.

Being able to get a great mix is an artistic skill that requires experience and a controlled acoustic enviornment where you can hear things accurately. The analyzer can be useful in many ways to get you there, but it's not art.

I would more or less agree with all of the above. While it's not a metric standard to mix by, it is a standard practice. Even before spectral shaping was a thing, listening to reference tracks was still common in trying to emulate a target sound. The principal is essentially the same. Personally, I'll take all of the help I can get to reach the target and better understand how I got there. Still, I use it as a diagnostic tool more often than I use it for referencing a target.

I actually did a comparison today after upgrading my Izotope plugins. I compared the EQ match on Ozone to the EQ match in the T-Racks 5 suite. Ozone has really got something. The T-Racks worked pretty well also. However, I think Izotope is leading the competition here. It's going to be interesting to see where the technology goes and how accurate it can become. As it is now, it gets you very close and just requires some minor tweaking.