Eq'ing high end - your thoughts

[osiris]

I love WA Productions Deres. It's very good at what it can do.

[zerocrossing]

I've been solving mix problems these days with high end cut.
It's like my music just got too fizzy. Every effect is grabbing at the top end for "brightness or sheen".
It happens with eq (sonible and isotope),
it happens through distortion a lot (gentle saturation adds up).
It happens from sneaking in mid-side boost. Prob lots more reasons too.
And now the new bandwagon is "a Soother".
Harmonic declutterers abound when we shouldn't really be needing them.
I'm now favouring hi end cut.
Wiggling the curve till "the sharpness" has gone.
If you have ideas on this, please share.

Here are my tricks:

Take a break. This one is hard for me, as I tend to hyper focus and work for a long time. The result is, you stop really hearing what's going on. If you take frequent breaks, it's shocking to come back to your mix and really hear it with fresh ears.

Make sure your room or mixing headphones aren't lying to you. They love to lie. Little liars. Make sure your room is properly treated and/or use room correction software. I can't use monitors in my situation, so I'm using Steven Slate's VSX system, and it's great, but you really have to calibrate it.

Use some sort of dynamic EQ. I'm using Waves' F6. It can mute or boost frequencies based on a threshold set, but it can be from the channel, or from a side chained channel. So for instance, if you have a synth with a lot of high end, you can take a side chain from your hats and dynamically mute the synth at the conflicting frequencies, but only when the hats are happening. I use this a lot for almost everything.

Finally, cool it with the saturation. I use it very sparingly, mostly on a channel, and one on the mix bus, but really subtly. I only put two plugins on my mix bus, and the other is a bus compressor for glue. I see people with giant chains and I have to laugh. If you need more than two, your mix is probably off and you might be making things worse. The chain happens in the mastering pass, and I let Ozone make a guess on the EQ, and it's usually pretty right, though I often mute the amount a bit.

[knob_monster]

chorus phase-rubbing melts the 14khz glass transients, just like hungry tape saturation eats the highs first

[kevvvvv]

@zerocrossing
Side chain fizzy synths off the high hats?
That sounds cool
Will give it a whirl

[mmediaaudio]

AUTO-ADMIN: Non-MP3, WAV, OGG, SoundCloud, YouTube, Vimeo, Twitter and Facebook links in this post have been protected automatically. Once the member reaches 5 posts the links will function as normal.

Hiya - it definitely happens to us, so we wrote this to help, it's technically unreleased but, yet mysteriously still available, but check out Velvet Crush - free demo:

https://www.mmediaaudio.com/plugins/velvet-crush/ (https://www.mmediaaudio.com/plugins/velvet-crush/)

Velvet Crush monitors the high-frequency balance of your signal in real time and corrects when presence and air bands pull too far forward. Not brightness added by an EQ shelf. A spectral balance check that catches harshness before it sets in - and removes it cleanly, without touching what makes the mix exciting. Works on individual tracks, a drum or overhead bus, the 2-bus, or the master chain - anywhere the top end needs keeping honest - blend to taste.

Cheers!
Jeff
M Media Audio

I've been solving mix problems these days with high end cut.
It's like my music just got too fizzy. Every effect is grabbing at the top end for "brightness or sheen".
It happens with eq (sonible and isotope),
it happens through distortion a lot (gentle saturation adds up).
It happens from sneaking in mid-side boost. Prob lots more reasons too.
And now the new bandwagon is "a Soother".
Harmonic declutterers abound when we shouldn't really be needing them.
I'm now favouring hi end cut.
Wiggling the curve till "the sharpness" has gone.
If you have ideas on this, please share.

[Benedict]

It is so easy when starting to think that all the cut and clarity, the detail, all lives above 245,000 Hz, so we boost like lunatics up there. Especially in digital, where there is no hiss penalty. Then we start to wonder why our mixes just don't have the grab or detail that we wanted - that Alan Parsons got 1978 on a Neve (or whatever).

The reality is that most things that matter happen right there in the mids so if something doesn't clarify in the mix, it is unfocused where it matters for that instrument in that song (or even part of the song). It is not about de-resonating or all that garbage, but finding where that sound has its uniqueness, or tells the right Story for this Scene & Song. A dB or so at 345 Hz or 1.2 kHz is probably a far better solution.

That leaves the option to clear the high highs for 'air' when the whole mix is done. The oboe is focused and intimate and the piece overall has this wonderful shine.

[El°HYM]

Lots of good information so far.

I would agree that a simple low pass might be a good option to check out. One of the first commercial available digital synths, the DX7 made use of a higher internal sampling rate (about 50kHz) and then using an analog 12 db Butterworth filter with fixed cutoff at about 16 kHz to compensate. The Signal then got amplified before fed to its Output. Still considered as maybe the warmest sounding FM - Synth ever made, also considering their own follow up DX7II which sounds different, not as warm.

If you imagine to listen to a DX7 before that, like only the raw sound without any filtering, that would be a horrible experience for sure. Now in these days with unlimited bandwidth and software synths providing Aliasing from the very beginning, the use of oversampling is certainly a good Idea, pushing the digital artifacts into the inaudible and smoothly filtering them out of our hearing range. Some developers take good care of this and implement good Algorithms using a proper Anti-Aliasing filter in their plugins. Other might not and that will do more harm to the Audio than any good, so from this point those will already have unwanted Artifacts baked in. Normally I would say that a mirrored Aliasing image folded back into the Audible is indeed something unnatural to our ears, something that causes fatigue or just pierces directly into our most sensitive hearing range.

Like the mentioned DX7 (also K2000) obviously took quite good care of reducing those Artifacts and while still having them already in the Audio, they somehow managed to generate an overall warm yet punchy digital sound. Also a kind of digital sound which has not much to do with our modern, evolved digital sound of today. While still being early digital, it is already considered as vintage and if remodeled into a modern VST, quite some engineering needs to be applied to get anywhere near.

With that said, this would be the DX7 in your Studio before being mixed, recorded, bounced etc. - which would add even more 'warmth' to the Signal, yet there has already happened quite a lot inside the machine before it even touches tape or gets effected in any kind of way. Replicating all of this in modern digital audio is not as easy as it might seem, as pushing your beloved VST synth into your favorite tape emulation, then your Vari - Mu and finally maybe a Pultec, will maybe make it sound a little bit warmer at first, round of the digital edges a bit. Yet, it will also increase all Artifacts already baked in, exponential; especially when using unlimited bandwidth without any filtering in between. There are solutions for this, like the highly recommended TDR - Filter bundle, actually intended to reduce IMD, yet also useful to tame, control Aliasing. Their new Bat Mode in Ultrasonic is also extremely useful in this context, as it shifts back the inaudible into the audible.

This is also the reason we now have tools like Soothe, which was not the case in the early digital age. Listening to the Delta will give you a good impression, of what is maybe too much in there, what is piercing your ears. On the other hand especially those kind of tools also, when overused, tend to suck out the life of the original sound pretty quick. Not all resonances are bad, some often quite essential to the sound. Brian Eno certainly didnt have Soothe or anything similar in his DX7 years, the Albums still sounding great. Which is something to think about and imho something to learn about, something to study and maybe to then apply into the modern digital world we are now living in.

[osiris]

I'll also suggest (a good) Transient Shaper. Can really deal with HH and cymbal bleed.