Hi and welcome back. And let's start with the question, why should
I care? Who listens in mono in 2019? Well actually there are still quite a few
scenarios where you might hear a mono downmix... Mobile devices, Bluetooth speakers, Background
music systems, Many PA systems (especially auxiliary fills). … and many of these are associated with
music discovery: the first time someone hears your mix might be while scrolling through
their social media feed, or propping up the bar at the club, so you don't want the version
they hear to suck too badly. Of course a mono mix will never sound as good
as a good stereo mix: stereo can add depth as well as width and will always sound bigger
and more impressive. But if you're careful you can encode within
your stereo mix a mono version that still hangs together as a song and makes musical
sense. Let's go back to the beginning: Alan Blumlein
pioneered modern stereo recording techniques back in the 1940s. The Blumlein pair he gave his name to is a
pair of figure 8 microphones, placed as close together as possible, but angled 90 degrees
apart. This means that signals to the left or the
right of the array will end up louder in the left or right channel… but they arrive at
both microphones at the same time, so there are no timing differences between the signals,
and the left and right channels are perfectly in phase. The same happens when we use the pan pots
in our mixers: panning left makes the left channel louder and the right channel quieter,
so the signal seems to move to the left… and vice versa… but we haven't introduced
any timing or phase differences between the left and right channels. This is important: phase differences can result
in dramatic filtering or EQing effects when the signals are added together, and these
are usually the most egregious mono compatibility issues. If there are only level differences between
the channels, they'll add together nicely without any unexpected filtering side effects. Many people think this means perfect mono
compatibility, but in fact its a bit more complicated than that. When you press the mono button on your master
channel to check for mono compatibility, this simply adds the left and right channels together
to create a mono sum, as you probably already knew… Well actually that's not quite true: adding
two identical signals results in a doubling of signal level, which is equal to a 6.02dB
level boost… but pressing the mono button doesn't make a mono panned signal get louder,
because in fact the mono sum is left plus right divided by 2. If a signal is panned to the middle it gets
added to itself then halved, so you end up back where you started. But if a signal is hard panned so it's only
in the left or right channel, it gets added to silence and then halved in level, so hard
panned signals will end up almost exactly 6dB lower in level in the mono downmix. Here's a practical example: drums and bass
are mono and centred, while guitar is panned hard left, and keys are panned hard right. When I drop this mix to mono… the guitar
and keys parts seem to drop in level, while the bass and drums don't change… so we end
up with a different balance. In practise of course it's not really a 6dB
drop in loudness, as when it's mono'ed the hard panned signals are now playing on two
speakers instead of just one. But the signals from the two speakers don't
add together as perfectly in the air as they do when summed electronically or digitally,
and this generally only results in a perceived level boost of about 3dB. So in reality your hard panned signals will
seem to get about 3dB quieter when you listen in mono. How big a problem this is depends on a few
things… does the song still work if those panned elements are a bit quieter? If so go ahead and pan them as wide as you
like. But if the choruses lose their impact when
the guitar is too quiet you might want to consider panning that guitar part closer to
the centre. At this point you may be thinking: “just
change the pan law!”. Actually this won’t help. Let's look at why: Here we have pink noise panning left to right
and back, with a zero dB pan law. This means that when the signal is panned
to the middle its at unity gain… when it pans left the right channel is turned down
while the left channel remains at unity, and vice versa. This causes a problem when we’re listening
in stereo: when its panned in the middle its playing at unity on both speakers, so seems
about 3dB louder than when its panned to the sides. And it's an even bigger problem in the mono
down mix: now it gets 6dB louder when its panned centrally. So lets switch to a -3dB pan law. This is ideal for stereo mixing: the signal
now stays at about the same perceived level as it pans across the stereo field. But it's still not quite right in the mono
downmix: now it gets 3dB louder whenever its panned centrally. We can fix that by switching to a -6dB pan
law instead… now the mono downmix stays at a constant level… but when we listen
in stereo the level seems to drop by 3dB when the signal is panned to the middle. A -6dB pan law is most appropriate where the
mono downmix is more important than the stereo version, perhaps in broadcast contexts. But -3dB is more normal, to work well in stereo. And the -4.5 dB pan law is simply a compromise
between the two, intended to be only slightly wrong in either case. But here’s the thing: it doesn’t really
make any difference unless you’re automating the panning of the part. When a part is panned to one static position
the choice of pan law will just change the overall level slightly, which you will compensate
for with your channel fader to get the balance you want. When the mix is mono’ed this doesn’t pan
all your channels back to the middle: it just adds left and right and divides by two. It doesn’t care about your pan law. In the real world the mono sum won’t happen
in your DAW anyway: your stereo file will be mono’ed by whatever mono device its being
played on, which doesn’t know or care what pan law you were set to. So anyway, how do we make a mix sound nice
and wide, without those wide panned parts getting too quiet in mono? One factor is, how good is the mix? I've deliberately made the mono elements in
this example quite dense and thick, so there's little space in between the parts for the
guitar and keys to poke through. This means a relatively small drop in level
causes the hard panned parts to be masked by the mono elements. Now I'm going to simulate a much better mix
by creating a much sparser arrangement… because of course those two things are intimately
related… now the drum and bass parts are much less busy… and they're much simpler
in terms of spectral content… so there's much more space for the guitar and keys to
fit around them. When there's more separation and space between
each part, there's more leeway for the balance to change a bit without it all falling apart…
now when we listen in mono those hard panned elements still get quieter… but they don't
get buried in the same way they did before. So, one way to ensure good mono compatibility
is just to create a really awesomely good mix! A popular strategy is to build your mix in
mono to begin with: make all your main EQ decisions with your parts panned centrally,
and make sure each part sits in it's own space, without relying on panning to create that
separation. When the mix works well in mono, then you
can start to pan parts left and right, and you'll be able to pan them further before
mono compatibility becomes a problem. However, panning things hard left or right
is not the only way to create a wide sounding mix. In fact our ears are incredibly sensitive
to subtle differences between left and right: its possible to have the most important elements
of the mix panned much closer to the middle, but use subtler cues to create that sense
of width and depth. Textural elements like pads, or this rhythmic
noise patch, are rarely the backbone of the song, so you can pan them as far left or right
as you like: if they get a bit quieter in the mono downmix its probably not going to
matter. In fact, as the mono version has less space
available for these kinds of parts, if they get a bit quieter in mono that might be a
good thing. In this case, there’s no need to stop at
hard panning: its actually possible to pan signals further left or right than that. Here’s how a mono signal looks on a traditional
goniometer display: just a straight up and down line. If I pan this signal left… the line rotates
anti-clockwise, but remains a straight line… and when its hard panned it sits at a 45 degree
angle. And likewise, panning right rotates the line
clockwise, till its at 45 degrees when hard panned. Now, what happens if I take that hard panned
signal, flip the polarity to create the opposite signal, then add a little bit of that to the
left channel? The answer is: that hard panned signal rotates
even further over, and is now further right than hard panned… By far the easiest way to achieve this is
by mid side processing. Here I’ve routed two different parts to
the same stereo sub group, each hard panned in opposite directions, and I’ve loaded
a Pro-Q3 on the group. Lets click to open the output panel… and
notice the panning ring around the output volume knob. In fact strictly speaking with a stereo signal
it acts as a balance control: turning it to the left attenuates the right channel only…
and vice versa… but if I click the little button below and to the right I can switch
to mid side mode instead. Now if I pan right I’m actually turning
down the mono sum of both channels, while leaving the difference signal at unity…
if I also turn up the overall gain to compensate… I’ve now just boosted the level of the side,
or difference channel. And this is literally the same as what I described
earlier: we’ve taken a little bit of the left channel, flipped the polarity and added
that to the right channel, and we’ve taken a little bit of the right channel, flipped
the polarity, and added it to the left channel. If we look at each part individually on the
goniometer… we can see that toggling the EQ on and off widens that part from hard panned…
to even harder panned.... Now, when we’re listening in stereo, toggling
the EQ on and off results in an audible volume change… when the EQ is on and boosting the
side channel, those parts seem to get a bit louder. But what the EQ is doing is literally going
to disappear in mono, as the side channel will just cancel out. So this kind of widening means these parts
will drop much more in apparent level when you mono the mix… but thats ok, because
there’s not much room for these parts in the mono mix anyway. If you want you could go even further and
remove the mid component completely… the result will look like a horizontal line on
the goniometer, and might seem like its coming from somewhere behind you when you listen
in stereo… but this is going to literally disappear completely in mono, which is a bit
too extreme in most cases. Lets extend this concept to the use of effects. I treat my delay and reverb return channels
as just another set of mix elements: I’ll go ahead and process them however is required
to make them fit in the mix, just like any other channels. And this also applies to how they're panned. Lets add some delay to the guitar, via a mono
send… but then consider where we can pan that delay: I can hard pan it in the opposite
direction to the dry signal… and the guitar part now bounces from left to right in a pleasant
manner, and we’ve added a great sense of depth. Or I could hard it pan it the same way as
the dry signal… now that guitar part seems to be panned further to the left, even though
the dry signal is only panned halfway… and because the delay is hard panned, but the
dry signal isn’t, the mono version will be a little bit drier, with relatively less
delay. This is often a good thing, as the mono mix
has so much less space available for these kind of effects. Reverb is just as critical. In fact, simply sending a part or two to a
stereo reverb, with no special trickery involved, can add a great sense of width and depth. If you turn Pro-R’s Stereo Width knob up
past 100%... this is like widening the 100% setting with some mid side trickery… this
helps to enhance the sense of space and width the reverb adds in stereo… but the reverb
will seem to get quieter in mono… which again, is often a good thing. Conversely however, and perhaps counterintuitively,
one of the most powerful ways to create a wide sounding mix is to turn the stereo width
all the way down for the reverb… we can now pan it hard left or right as we did with
the delay… or alternatively we could pan it to the same position as the dry signal. This places the part very firmly at that location
in the mix. The part itself is less wide, but the contrast
between its placement and other elements of the mix can sometimes make a more compelling
stereo image. And of course, this way the dry / wet mix
won't change at all when you listen in mono, so this is a good choice when the reverb is
really important to the sound of that part. Ok, before I leave you, lets pull up the goniometer
again… here’s a mono sine wave test signal, panned centrally… and panned left… and
right… and back in the middle. Now look what happens if instead of panning
the signal, I instead create phase differences between the channels… Phase differences between left and right channels
kind of break the rules that Alan Blumlein set out in the 1940s, but nevertheless can
create a really wide and spacious feel… and if you do it very carefully, it can still
be mono compatible. That's going to be the subject of part 2. Thanks for watching.
