Reference tracks in the age of spotify

The Lounge General Discussion
#1

I was advised on another thread that a good way to better mix my own stuff was to get a reference track and compare/contrast.

I used Soundflower to record Daniel Avery’s “Lone Swordsman” into Ableton Live 11, and put Swiss Army Meter on the track. Integrated LUFS -15, peak -5. Well UM, isn’t that exactly the mumber from Massive Attack’s Inertia Creeps?

OK, so Spotify normalizes volume, does that effect using a track for a reference? Probably not, as it is the relative balance of elements that I’m looking at, right?

But it made me wonder, so I paid for the AIF of “Lone Swordsman” from Band Camp (which I should’ve done anyway, cos I love that track) and dropped that In Ableton. Immediately it LOOKS totally different. LUFS at -8 integrated and > 0 for peak. It sounds way dirtier too, and the kick has more resonance and more release(?!) A lot of sonic differnces.

EDT: I also tried chucking a Utility on before and after a loundness meter to get the volume about the same as the Spotify version (couldn’t manage to get Peak and Integrated to match) and it still as a different character (to my ears.)

I guess my question is, how do you mix/master when the end product will be consumed sounding so different from your own finished version? (I admit this is moot for me, cos I’m still learning how to even make a basic track, but it has me thinking all the same.)

Is sound design itself different in the age of normalization? Cos the kick on Spotify and the one in the AIF are quite different sounding things.

Cheers
Russell

2 Likes
#2

…ableton is not the right thing for such comparison jobs…

reference tracks are a good thing just for a basic guide line…
u listen to that track, not within ur same mainframe daw lined up to ur mix in progress…
that’s way more irritating than helping…as u are about to find out right now…
don’t bother here…it’s for the better…

once u really wanna go down to details and REALLY wanna see what made the trick in that mix and u wanna get as close as possible THERE to match up…

u gonna need proper gain compensation…and by no means additional inner daw gain and pan laws or even further warp modes again…
especially not rerecorded stuff…of course ur lufs end up all over the place…
and lufs is final treatment exclusively…nothing to keep an eye on during mixing…
leave somer overall headroom and ur always good to go…
otherwise ur brain will endlessly play tricks on u again and again…an endless playitagain sam desaster…DON’T!

and that can only happen in one of this masteringsuites sofwares…
most common these days to name a few…ozone or finalizer…

while i always recommend…finsh ur mix first…
and yes listen to ur favourite example to see how u get along…
but DON’T try to match for real here…it’s just ur reference guide line over the thumb for now…
so finish ur mix…
take THAT final mix to master…don’t do both at once…
that are two different things…
and THEN go for a STAND ALONE mastering suite…drop ur mix there…drop ur reference there…switch on gain compensation!!!..and take it from there all the way to the finish line…

ozone is most common…but it’s fooling a lot again…give finalizer a try…that’s stand alone ONLY for too many good reasons…

1 Like
#3

Using reference tracks is normally advised to evaluate your speakers and listening room and is mainly useful when you are interested in translation (i.e. your music sounding good on a wide variety of playback systems/environments).

If you play a professionally mastered track (preferably a CD) and compare the sound to the sound of your own music it can help to identify any problems with your system and listening environment which have to be compensated for.

3 Likes
#4

Thanks. I am doing it wrong. I only really have headphones at the moment. I guess then listen to a refrence track on headphones, listen to my track, try to level match, make notes, comparison?

#5

in the age of spotify

that’s the problem.
but the solution is easy — i never use anything modern for reference.

3 Likes
#6

Reference tracks are good from two perspectives:

A. To match the tonal balance of a professionally produced piece

B. To match the loudness of a professionally produced piece

However you don’t do the two things at the same time.

When mixing, you go for A, not giving a flying f about anything B related.

When mastering you go for B, and if you did your job during mixing, not giving a flying eff about A

Now, in order to do this successfully, you have to think of the routing so that you don’t process the reference track, so you make sure it bypasses the master channel. An easy and convenient way to use references is by using a plugin called metric AB. It is the best solution I’ve found, you just place it at the end of your chain, it takes care of matching the loudness (you’ve noticed that peak and loudness aren’t equal and you either match one or the other during mixing. I say not to bother matching either, just get on the ballpark on the loudness and stay under 0dbfs true peak and you should be good) so you can focus on the other qualities and can compare the two in a more efficient manner.

Any questions about mixing and mastering, just hit me up.

9 Likes
#7

Oh yeah don’t use Spotify, get the lossless files and use that for referencing.

1 Like
#8

Ah, that is a really clear explanation of the difference, thank you.

Thanks! I have so many questions :smiley:

1 Like
#9

I use this method for referencing in Ableton Live:

I’m sure it’s not ideal, but it definitely helps. Obviously you have to manage gain control yourself (generally dropping the reference track by a couple of dBs works ok for this, at least in terms of my requirements in this capacity)

What I’m interested to know is:

What tracks do people use for their A-B reference?

I’m struggling to find an appropriate track at present. I’ve just been trying out “Nobody Speak” (DJ Shadow & RTJ) and it’s really nicely mixed, but I rarely use vocals as such, so…

2 Likes
#10

Thanks! That is a great step-by-step article, I’m going to try it today.

1 Like
#11

Personally I am not pandering to the whims of online streaming platforms. Fuck them. If they cant leave my shit alone, they dont deserve to host it.

If people want to hear music as it was intended to be heard, they’re not listening on spotify.

Keep it real, mix your music the way you want it, not the way some bullshit algorithm tells you its supposed to be.

Which reminds me, I must delete my distrokid and get the fuck off Spotify. Total waste of money.

4 Likes
#12

It’s a difficult point to argue against. I’m not sure if my music is on Spotify - I’ve never used it - but I suspect it is as one of the labels I release music with uses Distrokid. I’d be intrigued to hear how it sounds compared to an uncompressed wav… but probably not intrigued enough to actually bother using Spotify if I’m completely honest.

1 Like
#13

Usually tracks that are similar in terms of gerne, arrangement, instrumentation etc. a track that makes sense comparing against basically.

1 Like
#14

Do you have specific go-tos?

#15

A few artists/albums yes, but I’d have to check my computer to tell you exactly. What kind of music do you make?

Btw, Plugin Alliance are currently giving out vouchers. Maybe you can get the before mentioned Metric AB for cheaps. I bought it last year for 29,99 IIRC. Great software! You don’t need it when working with reference tracks, but it makes everything easy and at that price well worth it imho!

2 Likes
#16

Although (wireless) data speeds are increasing and increasing and storage p/Mb is cheaper and cheaper i can’t understand why lossy audio is still used on subscription audio platforms. Sometimes i feel like Spotify is gettin worse soundwise. Music is presented totally soulless and i see people walking alround with €600,- headphones listening to crappy versions of the final product producers spend sweat and tears and used tons of gear to make it. Why my lord? Whyyyyy. It’s like someone made an excellent wine with 500 years of secret family knowledge and the consumer get it presented in a paper cup to review it’s taste

2 Likes
#17

Just a reflection of how much many (most?) people value music…

#18

Did a quick forum search for this (slightly OT, maybe I should start a new thread) but what music players are you using on your phone to listen to high quality recordings? I download AIFFs from bandcamp when I buy music, and listen to them on my laptop…on my phone I use spotify for that same old terrible reason, convenience…I need a music player for AIFF/WAV etc

#19

Avery might have used a different master for the Spotify release. Not that uncommon.

1 Like
#20

I use foobar2000 on my iPhone. It’s pretty easy to load songs to it over a network and it plays just about anything.

Also, I second the recommendation for Metric AB above if you can get it for <$50. it has in loudness compensation, tons of meters, buttons to isolate different frequency ranges, and the ability to load a bunch of tracks and rapid fire switch between them. As others mentioned, you do have to keep in mind that your reference tracks will be mastered and therefore likely louder and more polished than yours.

My general advice with mastering is to not try to match Spotify LUFS (or those of any other service). It’s no big deal if Spotify lowers your volume but keeping your tracks that low will make them comparatively low for anyone who downloads them off Bandcamp (or anywhere else that doesn’t do loudness compensation, e.g. SoundCloud). You could do a different master for your distributor but you’ll still have quiet tracks if anyone buys them off iTunes or other stores. I usually roll with roughly -10 to -12 LUFS and things end up fine everywhere.

Not worrying about loudness, Spotify, whatever at all is fine if you don’t care about competition, but fighting for listens means playing the game. That said, a great master of a great mix will sound great everywhere, even if it’s not identical from service to service.

3 Likes