I have been asked by Chris Werner to open an exclusive thread about Mix & Master. I immediately thought it was an excellent idea so we can gather all our precious ideas, tips and tricks, proceedings, best plugins recommendations, etc., in a single, easy to reach place. First, I thought to open it in the “General Scope Discussion” forum, or in the “Scope Study” or “Scope Creativity”, but as it is so important and so directly related with the musical creative process, I finally decided to open it in here, in the “Music Forum” itself. As we constantly talk about how we mastered our songs and stuff, it is handy to have it in here, so you can simply make the review of the last song, and then add some comets on this thread, that you know it is on top of the Music forum, for your convenience, so I thought it could be a good idea to make it a never ending Sticky post, I hope it is ok for John and the rest of you if I do it this way fellows.
So, let it begin! Bring your Mixing and Mastering expertise or queries to life!
*MUSIC* The most Powerful Language in the world! *INDEED*
I think a tape emu is important in either mixing or mastering. I personally like Voxengo's tapebus, part of their analog flux pack. I'm not a reel to reel connoiseur, but I just like how this thing sounds. Not too much, but definitely not too little color.
Also, the recent console emu type stuff like Satson that does its own summing and some channel crosstalk is an interesting place to explore. It's also not an in your face difference, but the subtle nuances it introduces sound very nice.
On a more general scale (native/scope regardless), I'm a strong believer and overdriving just about everything. My solution to everything that sounds puny is straight up overdrive, or waveshaping. Comp giving you strange peaks? overdrive it. Your eq is too extreme and turning up the volume peaks out that track? Overdrive! You gotta do it just right though, or it'll just sound broken.
For mastering, I'm not a mastering specific engineer so I won't throw it out like the 10 commandments, but seeing that dudes with a bunch of plugins offer mastering services these days, my views might not be too far off.
For me, the playback context plays a big role.. For mp3 and web distribution, whatever. For clubs, I leave a healthy dynamic range. For theater or contexts where it's dead quiet when your stuff's not playing, I almost don't squash at all. It's not so much the playback sound system (provided it's decent) but more of how loud the playback is. If the playback is really loud, then you have that much range to use effectively. The point at which humans perceive "loud" may be well below what you can achieve when your material is peaking out all the time. That'll be too much in a theater, maybe okay in a club if you're just blasting short spurts. Also, an over squashed track played back really loud can also sound very incohesive. You need the dynamic range and contrast to keep sounds intelligible in loud situations. Anyway, just something I picked from having stuff played back in large halls and theaters. My first theater mix was really squashed with T-Racks or something, and it sounded like poop pureed with wood chips.
And for general style.. I wouldn't worry too much about being analog or fat or having to sound like the latest top 20.. There are some criterias though, like for me, it's body, intelligibility (separation), punch, and focus. Body's kind of like low end / mid end, plus the musical content (the right stuff has to happen in the right octaves). The stuff just needs to be substantial in terms of notes and timbre. Intelligibility is a big thing for me. It's sort of high end, but also writing the music so that each part doesn't smear out the other. Sort of like efficient use of the sound space. The mixing also needs to respect that, and keep different parts from eating each other's space. You can do that with very sensitive EQ and dynamics. Which relates to punch. Say your music is written in 4/4 time, and the smallest subdivision is a 16th note (which is most music). For each 16th note, every part that plays is sharing the space. If you want to feature a particular instrument's transient, the other guys need to back off. And it's not "slap on the sidechain and done" kind of thing, it's a smart musicianship sort of thing. This is closely tied to creating a good grove and a punchy mix. You can't make it punchy if it's always punchy (because everything is the same), you need to strategically allocate where the punch is and which part is featured. The featured part brings us to "focus". It can be a particular part, or a particular sound (a voice sample, etc), but if you want to feature it, then feature it. Don't have 15 strange sounds pulling the user's concentration apart. It's also related to intelligibility. If you have 1 sound that you want to feature, work on it, make it stand out in the mix, and go on to the next section where you'll feature something else. The shifting of focus creates drama, and the more subtle control you have, the easier it is to transport the user's attention from place to place. Same concept as orchestration/arrangement, I think applies very well to mixing.
But bottom line is, find a sound you like. I have a very particular sound I like. I sometimes achieve it in my mixes, I don't think I've heard it in anyone else's mixes... and over the years, it becomes more and more consistent. As for me, I think I've gotten to a consistency where many people will recognize stuff I worked on from just the mix itself. (and musical content, if I wrote it, of course) This is by no means saying that I've completed this journey, but it's nice because I worked really hard and long to get that sound I wanted, and people can identify it. Of course, if marketability is important, by all means do the now sound. You kinda have to. I'm just talking about personal growth and self actualization. (and maybe, to a large degree, my sound is just about saturating things, lol)
ah, anyway, there's too much to be said about mixing and mastering I don't even know how to put it together in one cohesive statement. I'm sure once the discussion gets rolling I'll remember thoughts and more explicit tricks/techniques.
I always have my mastering chain on from the start. The traditional way is to do mastering last, but reversing that means for a series of similar tracks (eg an album of the same band playing the same instruments) you can start the mixer with no EQ whatsoever, and once the recording is done, only EQ individual tracks where neccessary.
With Psy-Q, MasterIT and Brickmaster always enganced through the whole process, very few individual tracks need much processing at all.
interesting. I've tried that, but unfortunately since I'm in the native realm, and since some of the plugins in my mastering chain introduces latency, I can't leave them on. But on scope, yes, that's an interesting approach. It saves you from doing excessive sound processing. Though, from a traditional perspective, the mastering chain shouldn't change your signal that much. But it is an interesting approach.
Back when I had a scope card i would ALWAYS use the free DAS plugin digitalanalogsound (tape emu with bias control) and set the preset to "2" and tweak the bias. - i used to do this on real tape machines as well (studer/otari) on the end of the mastering chain and on the drum bus). It sounds much more natural than using an eq, and it smoothens out and smears the hi freq content in a very pleasing way! I have not yet found a plugin that does it as good as the das plugin, to me it sounds exaclty like the real deal!
-I used this plugin everywhere - the drumbus, the bass, the vox bus and the master.
try it out!, and let me know when scope supports mac, and i'll be back again
Wow! I see the thread is already going on! How cool! All in all, very interesting points in general. Of course I will contribute to this one whenever I have some time left, which is very little right now.
Ken, that one was a master class without you willing for it to be
*MUSIC* The most Powerful Language in the world! *INDEED*
There was a period when I spent a lot of time with using reverb units as a bass enhancer. I still do when I want a boomy kick. I've used all sorts of reverbs for this, but settled on SIR convolution with a custom built impulse. It really makes a difference.
The impulse or reverb setting is to give the sound some space to move around before it goes out. The extra distance and interaction really allows the bass to mellow out and develop if done well. Typically, I apply an impulse that I created with Voxengo's impulse designer to simulate a packed club. I don't know what the perfect ultimate dimensions are, but I know that sound when I go to a packed club and the DJ's got it going. I mix a bit of the wet with some low pass with the dry signal to get a more natural sound rather than "out of your sequencer directly to your ear" kind of a sound. Sometimes that sound is cool, for very clinical digital, chippy tunes, but most times it feels like you got gadgets stuffed up your head and feels strange. You can also fix the "in your head" effect with early reflections too. I know scope had it, I've been looking for a native equivalent, but nothing quite the same. ER used to be a standard in hardware synth effects but I guess with convolution a lot that stuff fell out of style.
An 808 kick is really not that impressive with a direct recording, but with a room, it can really make things rattle. I also give rhodes tracks some space too, since it usually sounds too direct and you get this flat sound coming from a flat phantom dimension. Give it ER or a light ambience with convolution, and you get a sound that's coming from a distinct place, a source. I guess effectively it's like mixing in the ambient mic in a recording situation to push back the track a bit in space.
BTW, ER is extremely effective in bringing out vocals or helping it sit above the tracks. It's easy for vocals to get buried in synth tracks or loud busy material, so you can beef it up with ER. It's not hard to take the effect too far and have the vocals totally pop out of the mix. Very effective!
Anyway, using reverb to mellow out bass and avoiding the "direct from the box" sound.
Last edited by kensuguro on Tue Aug 28, 2012 1:11 pm, edited 1 time in total.
Over the years, I'm sure many of you have developed this.. You hear a sound, identify what's wrong with it, and should be able to point to a frequency you want to change. Maybe a low, midlow, mid, midhigh, high resolution is good enough for start, or if those are already hard wired in your EQ. Eventually you want to be like, 100-300 area overpowering 4k zone, a bit of a notch at 3k, and absolutely nothing above 8k. You create a map to work with. You can drop 100-300 zone, neutralize the 3k notch, maybe bring up 4k zone, and try to pull up 8k, if not see if an exciter can help. If you're searching around for the "sweet spot", that's fine, but eventually it should become second nature. Then in a mix, you can say, "hey, the pad's root note is eating into 200 where I want the bass to be", and make precise adjustments without getting into a tug of war with messing with the bass, messing with the pads, messing with the kicks to clear up the 200 zone. You make a choice. Do you want the pads to be there? Bass? Make a choice, and create the sound you want. Not what the sound wants to be (you can respect it at times) or what the EQ/equipment wants you to do. (you can also respect that at times)
At the same time, you want to become very, very sensitive to volume. The smallest increments you can hear and control, the better. Most of the time you don't need to adjust that much, maybe 0.1-0.2db to clean up a sound. (provided the sound's not broken in the first place) I see a lot of less experienced people with crazy EQ curves that create very distinct sounds, but they also spend a lot of time getting it back down to something more subtle and effective. If 16 tracks are all bent to the extremes, it takes a long time to fix them all, so it's much faster to keep things subtle and precise in the first place. Sometime it's not even the EQ that's the problem, but just bad track volume balancing. Subtle volume control can go a long ways.
BTW, you may be wondering, "how about sound sculpting?". So in the sound design phase, do whatever you want. I'm just talking about final touchup in the mixing phase where you want everything to sit together well. If all you did was a 0.2db boost at 10k to bring up the rip in a supersaw.. that's a shitty supersaw. When you're making your sound, be bold, very bold. Bolder than your equipment will allow.
EQ just comes with a track so it's like, "that's just the EQ".. but it's a really hard part of the signal flow to get right.. it takes years and years to be even moderately effective at it, so my suggestion is to use it critically at all times, concentrate, and always make sure you're applying the least amount possible to achieve the effect. Always find the point at which you achieve what you're looking for, and what amount makes you lose the effect. Finding the critical points will help you internalize where they are, and also forces you to be very clear about what you're looking for. You want to remember why it worked so you can make it happen again, in a different context. None of this "I did it some how, but I can't do it any more" business. That's useless. It's same as writing music, hear in your head first, then go make it. Everything else should help make that happen.
I've helped lots of DJs get all their songs homogenized. If you've done it before, it can get pretty tedious. the interesting thing about DJ mixes is that you get lot of songs, 20-30 of them and their sources can range from bad mp3, CD rip, tape, to good LP, old EP... And each song has its own mastering, sometimes different genres.. so it's inherently a tough group to homogenize.
I usually do a pass faders first, to get them all in a similar volume zone. Then move on to a 12 band eq on each song to roughly get them to have similar characteristics. I focus on getting the 20-300 area to have about similar power, and also 8k-16k area. Inbetween will vary a lot from song to song. I usually try to get the vocals to sound similar and prioritize comping lower. A sudden change in vocal tonality will hit the listener much quicker. After eq, I check for phase problems. Since the material is from all sorts of sources, you can sometimes end up with strange phase shifted channels. Correct those with a stereo imager or mess with channel delay. I just did a raggae thing the other day, and just remembered now that I forgot to this step. oops.
There will always be a handful of songs that are squashed beyond belief, and are much louder than all other tracks. If bringing those tracks down in volume doesn't do it, then I compress all the other tracks to get within range with the squashed tracks. It's kind of sucks, but it does homogenize. If the loud one is distorted... well, there's not much you can do about it. I've used declippers but they don't really work too well, what's broken is broken.
Then you can send everything through your mastering chain. All the tracks are already compressed, so just light compression to tame bad fades and scratches if any.
Firstly, I find this tends to 'widen' the stereo image. For example, having left and right speakers 3 metres apart at volume 5 might be similar to having left and right speakers 10 metres apart at volume 10.
Make adjustments to make your mix bigger at lower volume. If you can achieve that, then when you turn up the volume, its really gonna sound huge.
Firstly, I find this tends to 'widen' the stereo image. For example, having left and right speakers 3 metres apart at volume 5 might be similar to having left and right speakers 10 metres apart at volume 10.
Make adjustments to make your mix bigger at lower volume. If you can achieve that, then when you turn up the volume, its really gonna sound huge.
Unusually interesting point to be checked about..., I will try it out.
*MUSIC* The most Powerful Language in the world! *INDEED*
Try putting the stock Stereo Overdrive S|C plugin in the Master FX BUSS. Set the OD amount to 2 and save yourself money on Tape Plug Ins.
Try using a SpaceF FAT II-S in an AUX just for shits and grins.
Or any plug in that doesn't make sense. Just see how it acts on certain channels.
Even running a Drum BUSS ( VCA Group ) through the SSB Modulator is fun.
Sidechain Vinco using the SBX on the chosen sidechained source.
Buy the DAS Drawmar Gate emu and use the Keyed frequency tricks.
It's the best virtual Gate I have ever tried, so many have noisy decays, this one is smooth.
It's hard to tell others how to mix, so I prefer sharing tricks I learned that I like.
dawman wrote:Buy the DAS Drawmar Gate emu and use the Keyed frequency tricks.
It's the best virtual Gate I have ever tried, so many have noisy decays, this one is smooth.
Yeah due it has the attack/release envelope rather than having it completely closed or completely open, good stuff !!
There is Voxengo EQ and tape saturation in Cubase 8. I will try it but my current philosophy is unless you have some very expensive hardware, it's best not to master anything. I try to get the gain up as high as possible without clipping and may raise the gain in low spots with a cross fade.