Gullfoss or Soothe?

[ManFromGlass]

I’ve been reading SOS reviews on both plugins. I like to idea that they can be used to show areas in a mix that might need more attention. (When the plugins are turned off)
If anyone has both I would love to know your thoughts. I’m about to try the Soothe demo, I already have Gullfoss. Soothe is not a casual purchase but if it helps me become a better mixer then maybe I should take the plunge primarily because my hearing is not that of a 20 years olds anymore and meters can be your high frequency friend!

[yeroc]

I use both for different purposes and use cases, i.e. soothe on vocals, gullfoss on the master bus. If you have Pro Q-4, the Spectral processing might be able to get you where you're aiming with soothe. its hard to say soothe is a must have, but often it gets me what i want very fast, so i think its worth it! Additionally for creative purposes, I like doing weird stuff with the just soothe's delta enabled.

[James Perrett]

As far as I can tell, ReaFIR does much the same thing at a much better price.

[Drew Stephenson]

I think Sam Inglis has tried both so may chip in. I've got Gullfoss and find it a useful 'make it betterer' but I've not tried soothe. I think they're intended for slightly different jobs though.

[Sam Inglis]

Both are great plug-ins, but I don't think they are quite the same. Gullfoss strikes me as lending itself most naturally to 'broad brush' applications, whereas Soothe is more about zeroing in on specific areas of the spectrum and zapping nasties. So if for example your mix was too dark, but brightening it with EQ made it sound harsh and brittle, Gullfoss is probably what you'd turn to first. If your drum overheads have something unpleasant going in the midrange that can't be fixed with EQ, you'd try Soothe. In general I find Gullfoss a bit more mysterious, in that it's perhaps harder to make the mental connection between the control settings and what you're hearing, but the flip side is that it's incredibly streamlined and simple to set up.

There are a couple of decent alternatives to Soothe: the Spectral option in FabFilter's Pro-Q 4 is very effective, and Baby Audio's Smooth Operator Pro (review forthcoming) is a big improvement over the first version. I don't know of anything else that works exactly like Gullfoss though.

[alexis]

I like the sidechain aspect of Soothe 2. Sam or other Gulfloss users, does that have a sidechain capability also?

Thank you!

[Drew Stephenson]

I like the sidechain aspect of Soothe 2. Sam or other Gulfloss users, does that have a sidechain capability also?

Thank you!

It does, but I've never used it.

[SafeandSound Mastering]

If I bring Soothe 2 out a mix is in real trouble usually as it has a sonic footprint that does not go unnoticed even at highest quality settings and moderate use. It is good at times though but you have to be careful, like any tool.

I use it rarely these days but once in a while it excels at a specific task.

Gulfoss I tried and everything I used it on sounded worse and did not buy it and fed that back to the developer.

That might have been because everything I tried it on sounded quite good already which beckons the question : Why is it still doing anything ? No 'intelligence' in most of this stuff because it cannot listen to music.

It deals only with numbers.

I recall having a sea sickness feeling and it was moving something like 1-2dB which I often get with these dynamic spectral adjusters.

This "real time" spectral waggling for want of a better term is not my idea of a good sound. It sounds incredibly unnatural very quickly to my ears.

I don't think having a spectrally moduated mix is a good thing personally.

You probably do not need them, try and make a better mix with your ears.

These are best used as damage limitation tools that cause other, hopefully, slightly less damaging issues as a side affect of correcting a perceived problem.

It is really disappointing to A/B with them on and off at what is deemed appropriate settings even Soothe 2 as it sits heavy on the mix, veil like. More and more I look for other manual solutions instead of putting the spectral wave machines on with their unnatural waggling about business.

To use the technical term.

I much prefer to provide feedback than use these tools or I seek a manual solution of my own choosing.

They get used sometimes but they are no global solution to anything.

[alexis]

For those interested in buying Soothe 2, looks like there is a sale currently.

From an email I received today:

Please visit oeksound.com to get licences at these reduced prices:

soothe2: 139 € (normally 199 €)
bloom: 139 € (normally 199 €)
spiff: 99 € (normally 149 €)

Additionally, all existing rent-to-own subscriptions can be paid off in one go with a 30% discount applied to the remaining amount. Please go to rent-to-own management to pay off the remaining amount. Monthly rent-to-own payments are not discounted.

The sale will end on Monday, May 26th.

[sc1460]

If I bring Soothe 2 out a mix is in real trouble usually as it has a sonic footprint that does not go unnoticed even at highest quality settings and moderate use. It is good at times though but you have to be careful, like any tool.

I use it rarely these days but once in a while it excels at a specific task.

Gulfoss I tried and everything I used it on sounded worse and did not buy it and fed that back to the developer.

That might have been because everything I tried it on sounded quite good already which beckons the question : Why is it still doing anything ? No 'intelligence' in most of this stuff because it cannot listen to music.

It deals only with numbers.

I recall having a sea sickness feeling and it was moving something like 1-2dB which I often get with these dynamic spectral adjusters.

This "real time" spectral waggling for want of a better term is not my idea of a good sound. It sounds incredibly unnatural very quickly to my ears.

I don't think having a spectrally moduated mix is a good thing personally.

You probably do not need them, try and make a better mix with your ears.

These are best used as damage limitation tools that cause other, hopefully, slightly less damaging issues as a side affect of correcting a perceived problem.

It is really disappointing to A/B with them on and off at what is deemed appropriate settings even Soothe 2 as it sits heavy on the mix, veil like. More and more I look for other manual solutions instead of putting the spectral wave machines on with their unnatural waggling about business.

To use the technical term.

I much prefer to provide feedback than use these tools or I seek a manual solution of my own choosing.

They get used sometimes but they are no global solution to anything.

Hi there, interesting, I've not had that impression with Gulfoss but I use it quite subtly and it does in some cases "open up the mix" in the mid-range.

But I have felt the same way about Bloom. In fact I would say I'd sell my copy of Bloom if I could, its algorithm seems to struggle with the intent of the instrument, introduce artefacts, and I think they need to train it a lot more across more genres.

Probably the same with Soothe/Gulfoss, I think these things work better the more data they've been trained on. It's surprising that Gulfoss doesn't have a genre target setting, but as you say it's looking at the numbers where it thinks masking is occurring regardless of producers intent.

Agree with Sam et al, Soothe and Gulfoss have in my view very different use-cases. They'd be in my top 20 desert island.
cheers
H

[SafeandSound Mastering]

I tried Gulfoss demo a long time back just after it was released (subtley), more to see what it was capable of than to use it.

Soothe is a much more useful tool for myself, but it is one of many options.

And sometimes it is better not using them at all.

A human listening is very different than a stream of numbers. I am confident that even with all this number crunching stuff my own work will completely destroy anything so called ay aye "mastering". Although Soothe and Gulfoss may not be necessarily specific to mastering.

It's had 10 years, people will do what they will, cannot change that. Those who eventually understand the difference between what good work is and a ... whatever that stuff is.. I cannot even bothered to know really... no point, for myself personally it is a waste of time. I don't need it for my own music.

I do use Soothe 2, Pro-Q 4 (nothing to do with ay aye) and the like but carefully and I can hear the difference between slapping it on and hoping for the best as I can hear the trade offs well. I hear pretty much everything on this system I built, little escapes attention.

If people simply learnt the basics then all this rather gimmicky, jiggery pokery would be unnecessary.

[Sam Inglis]

I don't think I have ever used Soothe across the mix bus. When I use it, it's generally on an individual source that is badly recorded -- most often vocals or drums. By contrast I do use Bloom on the mix bus, albeit at a low level. I nearly always set the 'window blinds' so that it's doing nothing above 5kHz, as it seems to want to make mixes a bit brighter than I'd like. But it's great for subtly cleaning up the midrange, especially that 300Hz-1kHz region that can be problematic with busy mixes.

My usual approach is to try Bloom on the mix bus to see if I like what it does. If I do, I'll take it off again and tweak individual tracks to get closer to that. Then I'll switch it back on and repeat the process. Sometimes I get to the point where I feel it's not adding anything, sometimes I feel I can't achieve that last 5 percent without it.

I never use the presets in either Bloom or Soothe.

I'm not sure why 'spectral waggling' should be intrinsically more destructive or obnoxious than any other audio process really!

[SafeandSound Mastering]

I believe it is not natural at all and my ears were sensitive to it in my listening situation.

I recall hundreds of times a second from somewhere. So the entire spectrum is adjusted as it calculates fit (not listens), every second.

That's a form of broadband (entire audio spectrum) yet also frequency specific modulation that is superimposed as I would understand it. And not under full human control like as 1 example, a single band dynamic EQ chosen at a specific frequency in the mid channel is.

Informed by the unmatched human ability of listening and also feeling, in context with the whole indivisible presentation of music. Not a stream of data.

I am not aware of any phenomena in nature that would do that to full music mixes, that I can recall right now (cross wind at a outdoor concert ?). When I hear something make audio sound worse I don't use it or I use another approach.

That goes for any audio process, EQ, compression, dynamic EQ, the list goes on etc. It all has very small side effects but generally the pay off is much greater than the trade off. Sometimes under my control, the pay off is vast.

I appreciate synthesis based music (merely instruments/sound sources) is not natural just in case that is brought up, but it is under very precise human creative control in most situations, making it very human as well. There is artistic discrimination, a sonic vision, a back story.

I test these (use, in case of Soothe2) processes on stereo mixes for 2 reasons. 1) I can mix already so do not need it for sources/tracks. 2) Mastering is my craft so I see what tools are doing to sound vs what they promote they can do.

I have not tested anything since because I don't need what is on offer and put my time into continuing skillset/knowledge improvement which is a very good use of time for anyone.

I cannot say more than my own personal experiences with these tools.

Demo's are available, everyone should find out for themselves if they are interested.

[Sam Inglis]

Both Soothe and Bloom have time constants like a conventional compressor, although they aren't calibrated numerically, so I guess they are both frequency and programme dependent. Either way I'm pretty sure they aren't modulating anything hundreds of times a second.

Are they natural? What's natural about anything in a rock or pop mix? There's nothing natural about putting a mic two inches from a snare drum, or hearing the drum kit through ten different mics at once, or about a DI bass signal, or a plate reverb on the vocal. There's nothing natural about transient shaping or parallel distortion or side-chained compression or any of the other tricks that are now ubiquitous in mixing. There's nothing natural about the polar pattern of a microphone, or about listening to music on headphones.

Some audio processes began life as an attempt to recreate natural phenomena. You could argue for example that the use of artificial reverb was an attempt to recreate a familiar effect from nature in a more controllable way. But we have long since left behind the idea that the aim of mixing is to make things sound like they do in nature. (With the exception of classical recording, which I'd argue shouldn't really need mixing as such.)

The lack of full human control does bother me a bit in the case of Gullfoss, which sometimes seems too much like a "make it sound better and don't trouble me with the details" processor. I don't find that with Soothe or Bloom, though. Once you've used them for a bit, you get a clear and predictable understanding of what each control does and what sort of sonic effect it's likely to have. In that respect it's not much different from using an EQ or a compressor.

I totally get that if overused they have a noticeable effect on the sound which I'd perhaps describe as 'uncanny' rather than 'unnatural', and it's easy to fall into the trap of overlooking this. But I think that's mainly because we're trained to notice the side-effects of over-compression or exaggerated EQ, and less familiar with the side-effects of too much Soothe / Bloom / Gullfoss. A part of learning to use these tools is understanding how much is enough, and that goes for pretty much everything.

Everyone hears differently of course and I make no claim to have golden ears. I find that used appropriately these tools can deliver a noticeable improvement without any side-effects that I can detect. But I can well believe that if I gave the resulting mix to @SafeandSound Mastering that his response would be different.

[Aled Hughes]

I believe it is not natural at all and my ears were sensitive to it in my listening situation.

I recall hundreds of times a second from somewhere. So the entire spectrum is adjusted as it calculates fit (not listens), every second.

That's a form of broadband (entire audio spectrum) yet also frequency specific modulation that is superimposed as I would understand it. And not under full human control like as 1 example, a single band dynamic EQ chosen at a specific frequency in the mid channel is.

Informed by the unmatched human ability of listening and also feeling, in context with the whole indivisible presentation of music. Not a stream of data.

I am not aware of any phenomena in nature that would do that to full music mixes, that I can recall right now (cross wind at a outdoor concert ?). When I hear something make audio sound worse I don't use it or I use another approach.

That goes for any audio process, EQ, compression, dynamic EQ, the list goes on etc. It all has very small side effects but generally the pay off is much greater than the trade off. Sometimes under my control, the pay off is vast.

I appreciate synthesis based music (merely instruments/sound sources) is not natural just in case that is brought up, but it is under very precise human creative control in most situations, making it very human as well. There is artistic discrimination, a sonic vision, a back story.

I test these (use, in case of Soothe2) processes on stereo mixes for 2 reasons. 1) I can mix already so do not need it for sources/tracks. 2) Mastering is my craft so I see what tools are doing to sound vs what they promote they can do.

I have not tested anything since because I don't need what is on offer and put my time into continuing skillset/knowledge improvement which is a very good use of time for anyone.

I cannot say more than my own personal experiences with these tools.

Demo's are available, everyone should find out for themselves if they are interested.

Similar to the points that Sam makes - there's nothing 'natural' about converting sound waves to a voltage at all, or converting that voltage to information as binary code or as a magnetic imprint, or processing that information, or converting it back into voltage and then back into sound.

Between the microphone diaphragm and the loudpeaker, it isn't sound.

[SafeandSound Mastering]

You can easily extend my commentary about synthesis based music to rock music. The big difference here is it is under human creative control, music is a natural human art form.

There is creative intention controlling the outcomes, natural or otherwise.

Control at mix stage is discriminatory using the natural phenomena of human hearing. That is a fact. As such this is the natural element of sound production.

Data manipulation is mere computational process, devoid of human connection, emotions or story.

The other and probably most important issue is it did not sound better to me.

As such, and very obviously, I don't use one of the products personally.

One of these tools definitely boasted about how its analysis was 300 x a second. It will not take long for you to find it.

If people think its great after a demo, all the best with it.

No more to say on this.

[Aled Hughes]

You can easily extend my commentary about synthethis based music to rock music. The big difference here is it is under human creative control, music is a natural human art form.

There is creative intention controlling the outcomes, natural or otherwise.

Control at mix stage is discriminatory using the natural phenomena of human hearing. That is a fact, so this is the natural element of sound production.

Data manipulation is mere computational process, devoid of human connection, emotions or story.

The other and probably most important issue is it did not sound better to me.

As such, and very obviously, I don't use one of the products personally.

One of these tools definitely boasted about how its analysis was 300 x a second. It will not take long for you to find it.

If people think its great after a demo, all the best with it.

I agree that the important thing is that it did not sound better to you.

But any processing of an audio signal is computational and done by machines, not humans - be it a fader, a pan pot, a valve compressor, a plugin reverb or Soothe. It's the reaction to what it does that's human. There's no inherent reason that the way these plugins work is any less 'natural' than any other circuit or digital code a signal might pass through,.

[Sam Spoons]

I suppose that the difference is that the human controls the parameters with a desk eq or an eq plugin where the computer controls them with an AI based eq. The fact that both need to do a bunch of clever calculations to simulate an analogue eq is probably not the issue, it is the question of who is setting the parameters for those calculations.

[alexis]

I suppose that the difference is that the human controls the parameters with a desk eq or an eq plugin where the computer controls them with an AI based eq. The fact that both need to do a bunch of clever calculations to simulate an analogue eq is probably not the issue, it is the question of who is setting the parameters for those calculations.

That is a great point, Sam Spoons.

Trying to play the Devil's Advocate on that: I was thinking it might be argued that not all the classic "human-controlled" sound manipulations we're talking about actually do offer much more parameter control than what is available (or maybe more accurately - not available) in AI.

For example, if one uses hardware advertised as applying "color" or "analog sound", it often is baked into the sound, without the ability to modify it much, if at all.

Apologies if this isn't quite what you were referring to!

[Drew Stephenson]

For what it's worth my experience of Gullfoss is a bit different to SASM's.
I use it primarily for its reveal function rather than the taming side of things. A bit of subtle use on the master bus (trimming the effective area to the mid-range because I think it can get a bit weird on HF content) can just open up a mix a bit and provide a bit more separation.
But I suspect I'm closer to the target market as I'm not an experienced engineer with 20+ years before the mast(er)*, I'm a keen beginner / amateur with about 12 years of self-taught, part-time fiddling around with things.
Different skill levels require different tools sometimes. As always, try the demo and see whether it's something that works for you.

* Sorry, couldn't resist.

[Sam Spoons]

I suppose that the difference is that the human controls the parameters with a desk eq or an eq plugin where the computer controls them with an AI based eq. The fact that both need to do a bunch of clever calculations to simulate an analogue eq is probably not the issue, it is the question of who is setting the parameters for those calculations.

That is a great point, Sam Spoons.

Trying to play the Devil's Advocate on that: I was thinking it might be argued that not all the classic "human-controlled" sound manipulations we're talking about actually do offer much more parameter control than what is available (or maybe more accurately - not available) in AI.

It's not so much the number of parameters but who/what is controlling those parameters. A basic desk eq may have very few parameters to adjust but the choice to use it or not is down to me*.

For example, if one uses hardware advertised as applying "color" or "analog sound", it often is baked into the sound, without the ability to modify it much, if at all...

Sure but it's my choice to use it and how much of it to add.

TLDR explanation :-

Sort of but not quite, it's not the way the processor/plugin achieves whatever changes it achieves, or indeed how detailed those parameters are but the decisions made by the entity controlling the parameters, eg (in the simplest sense) I want to increase the bass so I turn the bass gain control clockwise (or click and drag the mouse in the appropriate direction) I'm deciding how much I want the bass to increase using my ears. It doesn't matter if the knob is on my old analogue mixer or on a screen in my DAW, it's my choice to add so much bass and to use that particular eq whether it is a shelf with a fixed turnover frequency or a fully featured eq with options for everything from LPF to parametric with almost infinitely variable Q, frequency and gain (and maybe even phase) controls. The extra parameters allow more detailed sculpting the sound but it's me that decides. With an AI based eq you are, to some degree, allowing the plugin to make those choices for you and that's where I feel like it's cheating. Put more simply, I don't care what happens in the box (be it hardware or computer) as long as when I turn the LF eq knob clockwise the bass gets louder.

[SafeandSound Mastering]

"There's nothing natural about transient shaping or parallel distortion or side-chained compression or any of the other tricks that are now ubiquitous in mixing. "

I will just add a few more points. At the end of the day anyone can do whatever they want to, including myself of course.

With regard the above. Personally speaking for my own music, even the most technically advanced versions of it (modern in style) I don't use very much side chaining, deep compression, parallel anything, transient enhancement (in the sense of adding transient enhancers etc.)

Others might but I don't, so the comparison is lost on me.

And yes all of that will create an unnatural sounding mix. This production or mix style is usually employed for music that ends up at -5LUFS integrated where every achilles heal is pushed into submission before it is is then clipped (often on channels) then limited to have an extreme perceived volume.

It is merely one style, a style that locks your music into sounding one way. Which might be precisely what you desire.

That does not make any given piece of music better, more memorable or more interesting.

It is just a style and can be done very well or for someone attempting it without the correct techniques, tools or monitoring, an anxiety inducing impediment to a good sounding mix. Detracting your attention from making a good track, in the worst case.

Some genres (and the associated labels) expect high volumes, that is an additional challenge. Others not so much.

Mixes can sound great in many different ways, some brighter, some thicker, some with more bass, tonally. However, many subjectively "good" mixes will not go extremely loud easily (or at all without serious side effects.). But many times it is often the case that many of the attributes of mixes that do go very loud have many good attributes of a generally good mix in addition to those that allow higher loudness. (and the fact is some genres release very loud masters, and some sound very good.)

The effort to do this can be extreme, pull your hair out extreme and can make people feel inferior even if they have written amazing tracks. That is a real shame.

Now if this is the goal, great ! Mission accomplished. However, you have made a rod for your own back using these "ultra processed" techniques, which may be a better term than "natural". (I know, as I have been there done it and got the T-Shirt, I understand the techniques as I have used them (broadly worked them out myself), to degrees, but never in an ultimately extreme sonically defeating way so my own music sounded bad. I appreciate quality too much.)

When you hear these tracks on a superb system you can sometimes hear the trade offs that have been made, clearly.

I think the point is people who are using various tools let's say that automate various processing fixes are clearly not entirely confident of some aspects of their ability.

As such it seems to me great amounts of trust are put in these tools. One had better hope they are really doing good things rather than harm. And if you can hear something is let's say... clearer... then why not work that out for yourself and improve your own listening ability ?

There are so very many factors to a good end result.

Auto modulating look ahead EQ's attempting to hit some kind of ref target is merely one thing, there are so many more issues that comprise a good mix.

Like choosing high fidelity quality sources at the very start of making a piece of music for example and many other reasons.

Using auto everything, the first quality that will disappear I suspect is definition (clarity, leading edges, solidity, punch, power) as everything is moving about at the will of DSP hitting whatever it thinks will sound good.

Imagine what all these dynamic phase changes are doing, unless linear phase. (or maybe try and hear them.) A swirling soup before it is even on a streaming platform with perceptual encoding filters all over it.

Mixing using basic fundamental principles is actually not that hard when you have your room and monitors accuracy in check. It's the creative sonic aesthetic that is a little more tricky and for that you must 100pct have the basic room and monitors sorted out. Otherwise you won't know where the creative fine lines between style and mere error lie.

I appreciate this post is rather all over the place, and most here will not be planning their next release at -5LUFS but I think modern techniques (there are so many, so it is difficult to be specific it is just a generalization) are possibly making sound quality suffer even more.

The best mixes and sound quality I hear daily are using classic mixing, not all the tricks that all seem like a compromise. False stereo width, all the hazy phase auto-tronics stuff etc.

Anyone can do what they want, but there may be reasons why your tracks do not sound as good as those using classic techniques where they honed their skills for years.

[Sam Inglis]

Those are all good and thoughtful points.

One thing I would mention in response to some of the earlier posts is that as far as I'm aware, none of the plug-ins that has been mentioned in this thread uses AI or machine learning. They're all algorithmic in nature, so although the algorithm may be complex, it's not a 'black box' and will deliver predictable results.

Personally, there are two reasons why I use this sort of processing. One is corrective, when a source is compromised and nothing else can help it. It must be lovely to be Spike Stent and to only ever get to work on music that's amazingly recorded by the best engineers in the business. I don't have that privilege, sadly, and quite often things just need fixing before they can be mixed.

The other is as a sort of bootstrap process described earlier. It helps me diagnose where my mix could be better, then I remove it and try to make the mix better, then I repeat the process. Sometimes I'll get to the point where I feel it no longer needs to be there at all. Other times I think it still adds a small improvement and keep it in. But in this case it's never doing very much.

(Incidentally, I hate false stereo width as much as @SafeandSound Mastering and never, ever use stereo widening plug-ins. We have this thing called a pan pot...)

[Aled Hughes]

"There's nothing natural about transient shaping or parallel distortion or side-chained compression or any of the other tricks that are now ubiquitous in mixing. "

I will just add a few more points. At the end of the day anyone can do whatever they want to, including myself of course.

With regard the above. Personally speaking for my own music, even the most technically advanced versions of it (modern in style) I don't use very much side chaining, deep compression, parallel anything, transient enhancement (in the sense of adding transient enhancers etc.)

Others might but I don't, so the comparison is lost on me.

And yes all of that will create an unnatural sounding mix. This production or mix style is usually employed for music that ends up at -5LUFS integrated where every achilles heal is pushed into submission before it is is then clipped (often on channels) then limited to have an extreme perceived volume.

It is merely one style, a style that locks your music into sounding one way. Which might be precisely what you desire.

That does not make any given piece of music better, more memorable or more interesting.

It is just a style and can be done very well or for someone attempting it without the correct techniques, tools or monitoring, an anxiety inducing impediment to a good sounding mix. Detracting your attention from making a good track, in the worst case.

Some genres (and the associated labels) expect high volumes, that is an additional challenge. Others not so much.

Mixes can sound great in many different ways, some brighter, some thicker, some with more bass, tonally. However, many subjectively "good" mixes will not go extremely loud easily (or at all without serious side effects.). But many times it is often the case that many of the attributes of mixes that do go very loud have many good attributes of a generally good mix in addition to those that allow higher loudness. (and the fact is some genres release very loud masters, and some sound very good.)

The effort to do this can be extreme, pull your hair out extreme and can make people feel inferior even if they have written amazing tracks. That is a real shame.

Now if this is the goal, great ! Mission accomplished. However, you have made a rod for your own back using these "ultra processed" techniques, which may be a better term than "natural". (I know, as I have been there done it and got the T-Shirt, I understand the techniques as I have used them (broadly worked them out myself), to degrees, but never in an ultimately extreme sonically defeating way so my own music sounded bad. I appreciate quality too much.)

When you hear these tracks on a superb system you can sometimes hear the trade offs that have been made, clearly.

I think the point is people who are using various tools let's say that automate various processing fixes are clearly not entirely confident of some aspects of their ability.

As such it seems to me great amounts of trust are put in these tools. One had better hope they are really doing good things rather than harm. And if you can hear something is let's say... clearer... then why not work that out for yourself and improve your own listening ability ?

There are so very many factors to a good end result.

Auto modulating look ahead EQ's attempting to hit some kind of ref target is merely one thing, there are so many more issues that comprise a good mix.

Like choosing high fidelity quality sources at the very start of making a piece of music for example and many other reasons.

Using auto everything, the first quality that will disappear I suspect is definition (clarity, leading edges, solidity, punch, power) as everything is moving about at the will of DSP hitting whatever it thinks will sound good.

Imagine what all these dynamic phase changes are doing, unless linear phase. (or maybe try and hear them.) A swirling soup before it is even on a streaming platform with perceptual encoding filters all over it.

Mixing using basic fundamental principles is actually not that hard when you have your room and monitors accuracy in check. It's the creative sonic aesthetic that is a little more tricky and for that you must 100pct have the basic room and monitors sorted out. Otherwise you won't know where the creative fine lines between style and mere error lie.

I appreciate this post is rather all over the place, and most here will not be planning their next release at -5LUFS but I think modern techniques (there are so many, so it is difficult to be specific it is just a generalization) are possibly making sound quality suffer even more.

The best mixes and sound quality I hear daily are using classic mixing, not all the tricks that all seem like a compromise. False stereo width, all the hazy phase auto-tronics stuff etc.

Anyone can do what they want, but there may be reasons why your tracks do not sound as good as those using classic techniques where they honed their skills for years.

I do agree with the bulk of what you’re saying - it’s just the notion that some processing is more ‘natural’ than others I object to.

I’ve never used Soothe, but I have got Gulfoss, and sometimes I find it can make a track sound a tiny better, and sometimes not so much.

[Martin Walker]

The other is as a sort of bootstrap process described earlier. It helps me diagnose where my mix could be better, then I remove it and try to make the mix better, then I repeat the process. Sometimes I'll get to the point where I feel it no longer needs to be there at all. Other times I think it still adds a small improvement and keep it in. But in this case it's never doing very much.

Same here. My mixes tend be iterative, slowly approaching the best I can make them. I typically use the SmartOps feature of TDR's SlickEQ M on my master buss to even things out, but by the time the mix is finished its frequency response ripples are rarely larger than +/-2dB.