Skip to content

Tamas Dragon Posts

The anatomy of a score mix #5

Today it’s all about reverbs. In some ways it’s interesting how much and how different reverbs are being used in a score mix, on the other hand sometimes it’s quite the opposite, reverbs used very sparingly.

Obviously this heavily depends on the score and the concept. But many times especially on symphonic material reverbs are only used to enhance things a bit, to reinforce acoustics and not for really hear the effect itself. This is because mainly when it comes to a symphonic orchestra, we strive to maintain the real coherent balance, only elevating the whole picture without ruining the delicate bond that makes a real orchestra inimitable.

In a hybrid score that comprise electronic and acoustic material we might use more or even less depending on how we can massage together the two vastly different worlds. Although I’m always surprised how well these two completely different types can live together in perfect harmony sonically if you find the right balance.

This is a hybrid score with lots of electronic stuff in there so let’s see what I’ve used to create space, to enhance acoustics and to create distinct effects.

Altiverb XL

I think this would’ve been everyones’s first guess. Altiverb in my opinion is the very best convolution reverb. High quality impulses and the guys really work hard to grow the already huge library of impulses. Although I’m known to be mostly in love with algorithmic reverbs, if I need anything impulse based, I reach for Altiverb. I have my own favourite halls and rooms I always start out with and then tweak them if needed. The XL version is the complete surround iteration of the plugin which is not cheap, but definitely worth the price.

Exponential Audio R2 surround

This was my very first purchase when I discovered the brand. Absolutely blown away from it I clearly remember that I played with this for days. I liked it so much that I failed to create a favourite preset list as I loved so many presets. Since then I own many of Michael’s plugins and for a good reason. They are superb! The amazing thing is it can be subtle small room of a huge arena, the R2 can recreate it with vibe and feel. The implementation is just perfect. If you want to keep it simple, just call up a preset and probably you’re ready to mix. But if you need more control over the surround field, it’s right in there only a mouse click away. Very well thought out interface that helps you find everything in seconds. Still my first choice for scoring.

Exponential Audio Phoenix surround

The brother of R2 I guess. If you need real spaces, look no further. It’s amazingly clean, many times I like it more than almost any convolution on the planet. I’m not an expert in algorithms so I won’t be able to tell you why this plugin can create more realistic feel than most impulse based one, but it’s true. You should try it. The other nice thing is the Exponential interfaces follow the same basic principle so once you know one, you know all of them. No need to search for things, it’s all very logically placed so during mixing it’s easy if you would like to change any parameter. All of Michael’s plugins are extremely reliable when it comes to automation and they are zipper free so even if you glide from one set of parameters to something completely different, they won’t create that nasty zipper noise many other plugins produce.

Avid Revibe

The old and trusty one. Well, not so old as they updated it to AAX DSP. One thing everyone should know is that Revibe is always running in surround mode, meaning it’s eating the same DSP or CPU power even if it’s only a mono or stereo instance. Otherwise it’s still very popular even in post production circles. It’s a great reverb. For music I have my favourite presets in it, and tweak those a bit to fit my needs. Still, Revibe always find its place in my score mixes. In the last one or two years I tend to use it less and less, but there are certain things that it does perfectly. It’s a huge plus in my book that it is AAX DSP. Sadly very few reverbs supported on this platform.

Waves Abbey Road Plates

The digital recreation of the old and very, very famous Abbey Road plate reverbs. You can check the history and background of these gorgeous plates, how they worked closely with Abbey Road to catch the tiniest details of the original boxes. I can’t compare them to the originals, but this plate is just magic. Every time I use it it makes me smile. Somehow it almost always blends perfectly well with the material. My only negative comment would be that it eats unbelievably huge CPU power. Really, it’s that big of a CPU hog. Not to mention the fact that it is surely loosely optimised as it uses the CPU cores extremely unevenly. Once I had a conversation with Waves’ customer support and they seem to think that it’s fine this way. But hopefully their engineer don’t think the same. At least I’ve never seen any plugin properly coded and optimised using only one core at its extreme while ignoring that there’s another 11 cores would be available.

PSP 2445

It’s the newcomer, at least here. PSP made this based on the EMT 244 & 245 reverberator and all I can say is they did a pretty amazing job. This is also a plate you just insert on an aux, send some signal into it and it’s already sounding gorgeous. Additionally you can switch it to be solely the 244 model or the 245 or the combination of both. Not too much parameter to tweak but it has some under the hood goodies if you open the little box at the bottom of the plugin. While I really appreciate when companies make authentic emulations, the trend that they make it more clever with additional features that had never been available in the original is a great decision.

Eventide 2016 Room

It is an old-new love for me. The first moment I had the chance to try the original hardware I knew we would be great friends. And our friendship is stronger than ever. It is a perfect room if you want walls around any source or even if you’d wish to have bigger rooms for horns and percussion. The amazing thing is that it also works on strings. Honestly I know it’s not a plugin with a zillion parameter to adjust, but it really works. Works on any instrument in any genre. And it is really light on CPU which is a great thing when you’re mixing a huge score.

I don’t know the proper background but the thing is, these old reverbs, or the emulations of them many times seems to sit better in the mix. Reverb plugins became unbelievably great in the last few years, yet I often reach back to an emulation that is based on some old hardware. Maybe it’s because back then they’ve spent more time to develop one algorithm instead of rushing to release something, maybe it’s just my taste but for me one of the serious points in using or not using a reverb plugin lies in its ability to blend in the mix without tweaking it for 30 minutes. As you can see I’m in love with a few very new ones but also don’t want to uninstall the oldies as they really not only get the job done, but do it beautifully.

Comments closed

The anatomy of a score mix #4

As previously we took a look at the various EQs I’ve used during this particular score mix, now let’s see the dynamics arsenal that helped me shape the final mix.

McDSP AE600

I know now you think I’m crazy because it was also featured in the EQ article, but this beast is also a dynamic EQ, and as such I think it fits in the dynamic processing category also. With this, you can selectively process different frequency bands with great precision. The main point of the dynamic EQ is that it only works when needed. If you cut or boost with a regular EQ, the chosen frequency will be altered continuously. But with a dynamic EQ, the processing affects the frequency range only if the signal is above the threshold.

For example I had a cue where the high strings were getting a bit harsh, instead of cutting with a regular EQ, I just set a dynamic band for them which treated the harsh frequencies only in the loudest parts, otherwise I left them as they were.

Avid channel strip comp section

The second plugin that got mentioned in the EQ part too. As you probably already know this, it is still the channel strip from the famous Euphonix console. The dynamics section is a workhorse, if someone would force me to choose only one plugin for a full mix I’d happily choose this as my all-around tool. One more positive thing is that it maps perfectly on the S6 surface, which is a huge plus in my book.

Avid ProCompressor

Probably one of the most underrated compressor in the plugin world. Actually it is a very clever and great sounding beast, offering a complete Swiss army knife feature set. With the multiple algorithms – smart being a truly smart one actually – , dry-wet, DSP support, clear and informative GUI it always has a place in my mixes. Honestly if you need a great all around compressor, you should consider this as a serious candidate. The algorithms are working truly well, I even use them to experiment and test things with it without inserting many different plugins. You can even listen to the part of the signal that is being affected (that is over the threshold).

Avid ProMultiband

Actually it’s two plugins but mostly people think this is merely a multi-band processor. Well, that is indeed a multi-band compressor with a great interface, with nice options to control your sound, but it also has a truly great part. That is it can split up your audio into different parts based on frequency. So you can split up the signal into different frequency bands adding your own processing for the highs, adding something different to the low-mids while applying a third approach when it comes to real subs. It is all up to you. Or this way you can create interesting parallel chains for different areas of the spectrum.

Softube TLA 100

A true classic with additional features. This is a real know-it-all type plugin. For the lack of a better term it’s really musical sounding even if you hit it harder than the normal. The additional dry/wet mix knob makes it an excellent choice for bus duties also. The other great add-on is the saturation knob, so we can control if it should be absolutely clean or not. I’m a huge fan of adding tiny bits of “analogue” distortion throughout the chain so sometimes I even you this only for the sound, without really using the compressor section.

Brainworx Bx console comp section

What can I say? It’s a Neve compressor with some clever add-on like, dry/wet, side chain HPF and most importantly with the sound of a real gem. Be aware of that it’s gain reduction meter is a bit strange, at first you might end up hitting the compressor harder than you would like to. But once you find the sweet spot, it’s gorgeous. Another thing you have to get used to is the buttons and their behaviour. If you’re familiar with some real old-school Neve gear, then there won’t be any problem. Otherwise it’s not rocket science but certainly requires a little attention here and there.

SSL channel strip comp section

I don’t think I can tell you any more good things about the old classics. Without further explaining my fanboy attitude, this thing just works. It’s really that twist the knobs until it sounds right. No secret sections, no big tricks, it’s mixing on a visceral level.

API 2500

In my opinion this is the king of bus compression. I fell in love with it the first time I’ve tried. The real deal is that this API truly works on any material from drums through electronic stuff to symphonic material. Do not miss the different possibilities this circuit offers. The thrust control, the ability to act like a feed-back or feed-forward type compressor, the unique partial and adjustable stereo link. In my experience if this is not working or doing harm to the sound, then there’s a 99.9% chance that the settings are wrong.

That’s it, although I have more dynamics processors than I can list here, I always use the ones I feel I need. Especially with compression you must be cautious, it is very easy to overdo it and kill the dynamics of your material.

Comments closed

The anatomy of a score mix #3

By popular demand, let’s take a look at the plugins in the session. As you can imagine there’s plenty of different brands and types represented in this mix. I admit that sometimes I suffer from plugin fetish in a way that I fall in love with some and maybe use them many times, but try to actively control myself in this regard.

During the mix I have the luxury to use both AAX DSP and AAX Native plugins as I don’t need to overdub right into the supersession. If we need some additional recording then I prepare another rec session instead of dealing with this huge monster. It would not only be extremely hard to do everything in the supersession during recording but would put immense burden on the recording machine too.

Let’s see the plugins divided into different categories.

EQ

Probably the first one will raise some eyebrows, but I still use it and actually like it. It is very reliable even when you automate the hell out of it.

EQIII

Yes, the old built-in EQ in Pro Tools, available to anyone. It’s versatile, trusty and requires virtually no DSP or CPU power so if you can have zillions of instances.

Avid Channel Strip

This is also a very good choice, a huge plus is it maps on the control surface almost perfectly. Honestly I could mix a whole score only with this. Don’t get me wrong I still love some other EQs, but I’m happy to use this anytime.

BX console

It’s a beautiful Neve emulation and a very good one in fact. Even if I only use the equaliser part of it, I often have it at certain places just for the sound. This EQ always make me smile, it’s smooth and for the lack of a better term, it’s always musical sounding no matter what you do with it. I don’t use it for surgical stuff but for general ‘console’ EQ. Also maps great on the Avid S6.

Maag EQ 4

I really rarely say this, but this thing is magical. Although by today’s standard this is not a flexible tool as it doesn’t offer variable frequencies. But frankly, it is still one of my all time favourite tone shaping instrument. Even half a dB can make huge difference and the Air band is amazing.

Waves API 550

The 550 B and A also falls into the emulations category, and I confess I’m in love with them. For me, these pieces just works. It doesn’t matter what the application is, you can be sure about that if you insert an API on it, it’s going to sound awesome. It has fixed frequency bands, but the bands are overlapping so it is possible to achieve anything you want.

EMI TG 12345

Yes, I know, another vintage emulation… But believe me, they offer vastly different vibes and sound. These oldies are somewhat limited when it comes to features, especially when you compare them to the newcomers, but they still has some very special magic you can’t really achieve with the modern tools.

SSL EQ

I know this is a big fetish in some circles, and for a very good reason. This is for me a very liberating tool. Just insert it, twist the knob and enjoy the result. Not saying that I’m thinking too much when using any other EQ, but this one always makes me smile.

McDSP AE600 Active EQ

This is a very special one. As this masterpiece can be a spectacular EQ and a Dynamic EQ at the same time. I use it for cleaning up things and for general tone shaping. It also sits on all the stems. As much as I love vintage emulations, this EQ offers all the flexibility you’ll ever need. It has 6 fully overlapping fixed and active bands and you can choose from many different curves depending on your needs.

Well, believe it or not, that’s it, I’ve used these EQs during the score mix. I know many think that using these vintage emulations are useless because you can achieve the very same thing with a flexible and modern tool as are able to have many different curves. I won’t say that sometimes it’s impossible to almost or completely match the curve of a vintage EQ. But! And this BUT is bigger than you’d think. The main point here is to achieve what you want rapidly without overthinking and without loosing perspective. If you start to be an engineer with a white lab coat analysing and reproducing different special curves, you surely won’t have a great mix. This is the reason why we still love and use these emulations.

In my opinion during a mix the bottom line is to use any tool that helps you achieve the result you’re after. Different EQs leads to different results and mixing is definitely more than the sum of its parts. Use whatever you like, use it wisely and trust your ears.

Comments closed

The anatomy of a score mix #2

In this second post I share the basic layout of my supersession. But first things first, let’s start from the beginning.

Basics

This film is in 5.1 so the music supersession is 5.1 too. As I’ve had a conversation with the re-recording mixer, he’s going to get 5 5.1 stems and one 5.1 fullmix.

  • String stem
  • Brass stem
  • Atmospheric stem
  • FX inst stem
  • Percussion stem
  • Full mix

I use the Centre channel sparingly constantly double-check that I’m not hurting the dialogue. As this is a modern hybrid score with lots of atmospheric and synth instruments, we agreed on that I’ll use the LFE for certain things where we think it has added value. The main point is, the score is complete without the LFE, it only adds weight at certain points.

If I feel I might have gone too far, after checking the temp dialogue and fx I received from the dub stage I ask about the specific scene, how likely my LFE would hurt anything. I’m a bit old-school in this regard as I sat in their chair a few times and I know exactly how hard it can be to decipher all the messy stems, music and effects when you end up with a constantly pumping LFE in the theatre. It’s really better to double-check and make sure you’re going in the right direction.

Up to this film score mix I’ve always used a dedicated LFE-maker per stem. Now I didn’t feel the need of it so I abandoned that approach in this case. Probably because the electronic part of the score was filled with pretty high quality samples and analogue instruments I didn’t have to generate anything, just had to treat what’s already been there.

Without sounding overly obsessed, one more thing about the use of LFE in a music mix. It’s dangerously easy to overdo it. Because as you start adding things into the LFE channel, you instantly feel the satisfaction as the additional powerful sub enhances the low-end of your mix. It’s not only sonics but also a real physical sensation as we’re talking about deep lows here. And we can’t deny the fact that it’s very, very entertaining. And this is the reason why it’s such a huge trap. It’s like a child in a candy store, really. So when I really use the LFE, I always double and triple check if I’m in the right ball park, and always have a coffee break before I print the stems for delivery. This is a kind of reality check which keeps me on the right path. I know in our industry it’s already a well known cliche, but it’s even more true here, less is more!

Colour coding and organising

Basically every stem has its own colour. This is the primary sign which helps me find everything rapidly. Generally I avoid using bright red colour because that is used on the record stems and fullmix so whenever I see bright red I know it’s a record channel.

Other than that there are really no rules here, except that every stem must have a unique colour. If you still need more separation, then use an inactive track to separate them. If I find I always searching for things in the session then I set up these separator tracks that always stay inactive and their only purpose is to guide my eye. It’s quite amazing how these small things can help and guide your brain and eyes so you’ll find things much easier and faster. Combining the separator tracks with vividly different colours are the best for me. Remember, your session will only grow as you delve deep in the mix so use whatever can help you to keep track of everything.

Use brighter or darker colours inside a stem to differentiate things. I try to follow my own rule of using one main colour throughout a stem, only varying the brightness of it. If the stem is huge with lots of tracks, then I might split up to two different main colours. The only really serious rule is to not mix colours between stems.

The reason for this is rooted deep in our brain. If you use kind of similar colours, your brain will automatically recognise things without searching through many tracks or reading the track name. If you mix colours through multiple stems then your brain loose track of things. The point is to find a system that works for you, and no matter how hard it seems to stick with it, you must to stick with it. You’ll thank yourself later. Remember, we’re talking about hundreds of tracks and in any given moment you must find certain elements in a huge mix in seconds to deal with them. It is not wizardry, but certainly requires preparation and diligence during the mix.

At last, I always leave the internal audio buses and stem fx tracks at the end of each stem. I’ve been doing this so long that even the stem fx tracks signal me the end of a complete section. Combining the colour coding, the separator tracks and the routine to have the effect tracks at the bottom of every stem I can quickly fly through 300-500 tracks without being lost.

Groups

I usually create groups to have certain kinds of instruments or sounds treated together. For example if I have eight kind of similar synth pads then I create a Edit-Mix group for them. This way I can adjust them in bundle, but it’s also very easy to adjust only one if needed. But once I have a pretty good internal balance, it’s rare that I need to adjust them absolutely in isolation from the others. If it happens, then I can grab two faders in the group so it temporarily let me adjust them separately, or I can quickly suspend the group to have them separated completely for the time being.

The other trick is to don’t use the globals but create your own rules for the group. For example controlling certain inserts or send together or have their panning controlled by the group. If you choose wisely this can save you tons of time.

This kind of grouping is a double edged sword. If you need to edit a lot, than you might consider using mix only group or make a separate edit group so you can selectively have the needed groups enabled. Obviously huge editing shouldn’t happen during the mix stage but honestly many times today you can’t eschew that. Since picture lock seems to be non-existent, the constant changes forces the composer and you to re-edit things to really fit the picture.

One thing I always do is to create a CUT group. That includes all the audio from all the stems, except the mockups, the dialogue and effects track, the video and the final prints. If I need to readjust-copy-move an entire cue for some reason, this group takes care of the process. Just be sure to really include all the necessary things but nothing more. Double-check it, otherwise you might get into trouble using this global group.

Marking the groups properly is another must do thing. All the edit groups start with capital E, all mix groups start with capital M. VCA groups get a small v prefix. For special things I usually create so called Temp groups. These can be anything from special group panning to similar plugin adjustment. If I know it’s not going to be an absolutely necessary group in the mix, I mark it TEMP before the name. Of course this is my system which works perfectly for me. You can use this, or make your own system. The point is, you need something that works for you. At the end you need something that is informative and easy to recognise.

Comments closed

The anatomy of a score mix #1

As I’m completely snowed under with the current score mix I’m working on, I thought I share the anatomy of this mix. Have to break it into parts as I frankly have no functioning brain at the end of a long day.

Format

It might be different where you live, but here 5.1 is still the major format. Occasionally we encounter other formats, but that’s so rare I wouldn’t mention all of the possibilities. Immersive formats are far away things we only dream of, but never ever met. It might change in the future, but honestly in Europe I think there’s still only very few cinemas and dub stages offer true immersive workflow and experience. There’s some news that the biggest post houses started to upgrade their stages to be able to mix in Dolby Atmos or Barco’s Auro, but I don’t see it’s spreading that fast, so for the time being we seem to stick with 5.1.

It’s important to know your delivery format before you even start to build your mix session because it can lead you to very different bus and session layout that would be extremely hard to change during the mix. Wait, I mean it, really! Starting with an inappropriate format can ruin your work and drag you into an endless re-route, re-do, re-mix cycle. I know it seems like the most obvious thing, yet still so many fall into this trap.

Principles

It’s all about what to do and what not to do. The best is to eliminate any guesswork and have a discussion with the re-recording mixer. This is the only advice I can honestly give. Really. You can endlessly browse every audio forum, try to decipher anyone’s workflow, watch every tutorial videos on Earth and still, the only person who really knows the answer is the re-recording mixer. And to further confuse you, it might be different on every film. So this must be on your checklist.

What you need to know is: are you allowed to use the LFE for certain moments? How to treat the Centre channel? Leave it empty or use it sparingly? These are very important questions which can affect your final mix on the dub stage.

LFE stands for Low Frequency Effects, hence it’s not there to create ample lows for your orchestra. Treat it as a special thing. If you put everything in it, you have a good chance that it’s going to be muted, and for a good reason.

Almost the same with the Centre. This is where the dialogue is the king, period. If you happen to ruin their chances to clearly hear the dialogue, they will eliminate your Centre, again, for a good reason. In a film mix there’s tremendous amount of things they have to balance in order to achieve a nice, good sounding mix, don’t make their life harder, especially if you can make it easier.

This doesn’t mean that you should absolutely avoid using the Centre channel, but my advice is to make sure you’re not putting anything truly important Centre only. If unsure, consult the re-recording guy, even ask for some time on the stage to test your mix, but never ever rely on it as your main anchor.

Stems

Let me start with a common misconception. I’ve heard this numerous times and the misbelief is still very wide spread even amongst professionals. Here it is: “I don’t deliver stems, fullmix is my mix and they shouldn’t change that…

Well, stems are not there to alter your precious mix. Quite the opposite! They are there to preserve your mix as much as possible. Think about this example: you delivered only a fullmix, where a huge taiko drum hit just happen to be at the same spot as an explosion. What would the mixer do with this? Lower your music fullmix in order to carve space so that the explosion is clearly audible and have its intended impact. Now let’s see what happens if you deliver stems. Now they lower only your Percussion stem, which means most of your mix is left intact, only that taiko drum hit suffer, but given the circumstances, that’s understandable. Which is better? Obviously I’d choose the second one.

Delivering stems are the only way to make sure your mix will be preserved as much as possible. Also, this should be discussed with the re-recording mixer. How many stems does he/she want? Too many might put too much on their shoulder, too few or badly organised ones might cripple their chances to affect only the problematic areas.

Don’t forget, this is a collaborative vocation, it’s not about your ego, it’s about the film.

Comments closed