How would you go about getting a sound like this?

(I mean in terms of mixing approach / plugins etc. btw. I have to do it ITB..)

I've got a song to mix that's similar in pace and mood to this - with a similar soft-sung (almost half spoken) male vocal.

Particularly interested in the way the vocal sounds (intimate, yet huge..?). That ambience? How?

And the bass (both the bass guitar and the general low end) - how it seems to sit all accross the bottom of the mix, filling the stereo field like a big warm cushion.

Any tips?

Thank You x

 

Regular vocal recording (LDC/condenser, popfilter, highpass at around 100Hz, good preamp, compression, EQ, de-esser if needed, some saturation, etc). Add a lush plate or hall reverb (add Abbey Road Reverb Trick, pre reverb). The vocal pops out because its contrasting to the "darker" instrumentation, but it also has reverberation which makes it both close and roomy (yet intelligible due to the dark/bright contrast).

Bass seems to be a simple bass guitar. Try compression, some saturation and lowpassing (to control the higher harmonics from masking/overlapping). Try adding a microshift or a chorus on the bass to make it a bit wider. Bass is basically playing three notes throughout (from 97 to 135Hz), which doesn't really overlap with the 50Hz kick nor the guitar. It has room to move around in, between kick and guitar.

 

Last edited: Sep 18, 2023

Brilliant! Such a detailed response - exactly what I was hoping for! Thanks Baxter.
Gonna try all this out. (Will have to look up the 'Abbey Road Trick'.. is it an 8th-note delay before the verb, maybe?)
Great info about the bass - thanks for taking the time to really listen to the ref. track, I appreciate it. There's many more bass notes in the track I'm mixing but I'll take the 'principle' that you've laid out here and see what milage I can get applying it in this particular context.
Again, thank you!! Only one response to my request, but you NAILED it, so, one is enough :D

 

Thanks! No worries!

You'll get the hang of Abbey Road Reverb Trick. Watch the Bobby Owsinski video on youtube. Watch it twice and you'll get the idea. The important thing is that the EQ is before the actual (paralllel) reverb.
Yes, you can add some pre-delay on the reverb to create some space in order to increase intelligibility in the vocals/voice.

 

Honestly, I've never understood why - because post or pre doesn't matter, they nil.

Spoiler: proof

Bar 1 & 2: reverb only,
bar 3 & 4: added pre-rev EQ,
bar 5: fade in of rev with post EQ, phase invertred,
bar 6: sound is gone completely

 

I think the idea is to cut the things that overlap/mask, before the reverb, so that the reverb can work as cleanly/transparently as possible with what it has been given. I also place a de-esser before the reverb to essentially kill the sibilance (I'm definitely NOT going for the George Michael sound).
As I've tried both (over my 30 years of mixing and mastering) I find that EQ before reverb suits me better and gives me better results.

 

In theory this sounds convincing, in practice it doesn't seem to matter.

 

In theory it should (and does), because all rooms and reverbs have different reverberation colors (and decay times at different frequencies).Just like you get two different results when using filtering->distortion, compared to distortion->filtering.
EQ after the reverb is good for fine-tuning and tonal correction.
https://www.sonible.com/blog/techniques-for-eqing-reverb/#EQing-pre-vs-post-reverb

 

Last edited: Sep 19, 2023

When you do that null test between pre-reverb eq and post, what sort of low end information is in the original file you are feeding into the reverb? i am wondering because i cannot play the file right now. Is audio passing through the reverb plugin completely unaffected with every reverb related parameter zero'd out?

 

Last edited: Sep 19, 2023

I'm going slightly off-topic but...

From my understanding, I would think the Filter-Distortion example isn't the same as EQ-Reverb. EQ and Reverb are usually linear so it does make sense that the order shouldn't matter: f(x+y) = f(x) + f(y). Distortion however, is non-linear, and filtering before/after is gonna have a significant effect on the final outcome of the sound.

Theoretically, I've thought that the order of Reverb>Delay/Delay>Reverb shouldn't matter either yet people seem to believe that it makes a, keyword here, "huge" difference. It will in most cases make a slight difference since it's not truly linear.

@No Avenger perhaps you can weigh in on this too and correct me if I'm wrong.

 

How did you manage to null it when a single instance of an algo reverb rendered twice won't null with itself?
Not mentioning that with something like Wormhole or any shimmer there'll be a huge difference, since they indtoduce new spectral information. With Wormhole in particular, pre-EQ will not remove lows from the reverb.

Absolutely not, the difference is indeed tremendous.
If you put an absolutely linear delay after reverb, it's gonna repeat the whole reverb tail, making much more ambient sound as well as clearer repeat itself, since it's not muddied by reverb.
If you put it before reverb, it's gonna blend in with it.
I mean, really. Just try it, you'll hear,

 

Last edited: Sep 20, 2023

I mean if you really want me to test it...

Null Test for Stereo Delay<>Convolution Reverb:

The differences are from the modulations and other non-linearities of your particular system of choice. Mathematically, however, it should not make a difference, which is the point I'm trying to make. f(reverb+delay) = f(reverb) + f(delay)

 

Setup is simple:
- Track 1: drums (kick, snare, HH), not routed to main out, FX send to...
- track 2: IR reverb, fully wet, EQ with low and high band, pre Rev (muted for the first two bars),
- track 3: copy of track 2 (without the automation), EQ moved to post Rev (faded in at bar 5).

Whatever I change in this setup, track level, send level, any EQ or Rev parameter won't nil anymore.
You need to do this with wav and IR reverb, with chorused live synth, RR drums or realtime modulated rev this won't work, of course.

The difference for high freqs (shelf/cut) can be inaudible, but they already don't nil. For every lower band the difference is audible. Simple reason, if you saturate the low end it generates harmonics in the (low) mids. If you cut the low end afterwards, these harmonics are still there, if you cut it before, they're not even generated.

 

I think you're stepping into a trap here:
First delay, then reverb means the input of the reverb is delayed.
First reverb then delay means the output of the FX chain is delayed.
These two are indeed almost absolutely identical.

Routing, per FX send, a source to delay and reverb in parallel and then the Dly into the Rev or the Rev into the Dly is something different.

 

Believe? It DOES create a HUGE difference (sorry for the caps). Same as a de-esser before a reverb makes a huge difference, and can not be "reversed" or "cured" with a de-esser after the reverb.

 

Reverb->DeReverb

Everything else is just wrong. You're wrong.

 

That makes sense. I'm not working with those, but I had a thought that IR must be involved if results are so deterministic. Algos can't do that.

Reverb's full wet and delay's 50% wet (as in, both dry and delayed signals are at full), am I hearing this right?
I assumed (without clarification, my bad) a scenario where both reverb and delay are below 50% wet, since it's how you'd realistically use inserts most of the time.

 

Last edited: Sep 20, 2023

So the proof in his next post doesn't count?

I've tried this with two sends aswell. Dly -> IR Rev and IR Rev -> Dly. Phase inverted the difference is ~ -90dB, inaudible.

 

For that example, the delays were at 30% and the reverb at ~50%. But below or above 50% wet it should still null on either order of Reverb/Delay. In fact, you can chain an EQ, Delay and Reverb together and swap the order between them all you want. If there is no other modulation/non-linearity, It will null.

 

Then I don't know what happens. This is not my experience. Whenever I put delay before reverb, it gets washed out, and whenever I put reverb before delay, it gets more dense (and the repeat itself is more audible). I actually stumbled upon it by accident, moving delay to be after reveb in the insert chain inadvertently and noticing a serious change in the sound.
The reverb I noticed it with is Polyverse Comet, and the delay is FabFilter Timeless 3.

Also I can't hear dry snare in the reverb. This is worrying me. Can you please DM me with a full-wet sound of the same reverb on the same snare, so I can check my sanity?