Variety of Sound blog: compressor aficionados (1) - Fabien from TDR

[FabienTDR]

Herbert (Bootsie) pounded me with questions today, here's the result:

http://varietyofsound.wordpress.com/201 ... -from-tdr/

Enjoy the read

[3ee]

Hmm.. never imagined you do mastering, anyway...
Enjoyed the read!

+just updated the Feedback II comp to 2.01

[Tricky-Loops]

Without any doubt, the whole dynamics processing field is extremely over-rated. Compressors aren't creative, they don't make sound, they don't make music. Dance floors don't fill because someone compressed his music, they don't make people dance and sing.

Hands up for this! Most dance music is more over-limited-compressed than creative!

[bmanic]

Without any doubt, the whole dynamics processing field is extremely over-rated. Compressors aren't creative, they don't make sound, they don't make music. Dance floors don't fill because someone compressed his music, they don't make people dance and sing.

Hands up for this! Most dance music is more over-limited-compressed than creative!

Very interesting to read how different the opinion on this very thing is. For me the absolute number one function of a compressor is "groove control" or shaping. It's much less about the actual utility of changing the dynamic range.

I also completely disagree with the "they don't make music" bit. They do. You can literally ruin a groovy track with wrong compressor settings (most notably a wrong release) while you can enhance an already groovy track or even a simple drum performance, human or programmed, with the correct compressor and correct compressor settings.

This is my quick opinion on this part.

Cheers!
bManic

[mwaudioprod]

Thanks, very interesting!

[Tricky-Loops]

Without any doubt, the whole dynamics processing field is extremely over-rated. Compressors aren't creative, they don't make sound, they don't make music. Dance floors don't fill because someone compressed his music, they don't make people dance and sing.

Hands up for this! Most dance music is more over-limited-compressed than creative!

Very interesting to read how different the opinion on this very thing is. For me the absolute number one function of a compressor is "groove control" or shaping. It's much less about the actual utility of changing the dynamic range.

I also completely disagree with the "they don't make music" bit. They do. You can literally ruin a groovy track with wrong compressor settings (most notably a wrong release) while you can enhance an already groovy track or even a simple drum performance, human or programmed, with the correct compressor and correct compressor settings.

This is my quick opinion on this part.

Cheers!
bManic

People did already dance and sing in the old Roman Empire, and so they did in medieval times. And they didn't have compressors with attack, release, look ahead, feedback...

So it seems to be possible to make people dance (with good written music) without using a compressor.

IMO the creative aspect in songwriting and arranging is much more important than using compressors & limiters on every track...

And you see this here at KVR, too -- I've rarely read any question about how to write and arrange a great dance song, every nobie asks about limiters, compressor, side-chaining, and then they even have to sidechain the limiters to compress everything even more...

Okay the last sentence is meant as a joke, but it has its truth...

[Aiynzahev]

thermostats or washing machine control circuits



Loved it, glad to have people like you working in audio. Thanks!

[FabienTDR]

bmanic, you know, I had to make a strong point.

I was too vague. My point was more about "blindly adding more compression to make it louder, better, fatter" rather than the subtle sound shaping you have in mind. Tricky-Loops' interpretation is much closer to what I wanted to express, sorry for the confusion.

Glad you all enjoyed the read!

[Tricky-Loops]

bmanic, you know, I had to make a strong point.

I was too vague. My point was more about "blindly adding more compression to make it louder, better, fatter" rather than the subtle sound shaping you have in mind. Tricky-Loops' interpretation is much closer to what I wanted to express, sorry for the confusion.

Glad you all enjoyed the read!

Maybe it's because I worked as a (freelancer) journalist, that I have a preference for "strong" points. You can write pages of tech specifications, and everybody agrees blindly and falls asleep. But wake-up sentences like "A compressor doesn't make people sing and dance!" - this is something that MAKES PEOPLE THINK. And that's why I like it.

[yl]

"Analogue audio is already at its peak, digital audio has just begun!"
allow me to politely disagree
from where I'm sitting.. audio dsp has also reached it's peak and the market (even the free one) is already over-saturated.. with many great audio processors

That said, digitally controlled analog processing, and digital detectors speed are expanding capabilities in the analog world

I also don't agree that no-holds-barred approach to DSP regarding CPU/Memory usage has not been taken
there are many software compressors which take a heavy toll off modern CPUs


I'm intrigued about how you mean to take advantage of RAM in real-time processing in a way that hasn't been done before

nice read..
and nice plugin btw..

[FabienTDR]

Hey yl,

I mentioned the broadcast scene, they still innovate and develop. I'm sure we'll soon see the music scene adopt new "revolutionairy" new Hilbert clippers, super clever IMD free dynamics/saturators, overshoot free band-limited systems, fully harmonic behaviour (in the sense of "now aliasing") and other stuff now only available in expensive broadcast processors. The list of "features" waiting for their "right time" is enormous.

Oh course, you can find now suddently compressors allowing to do 4x brute force oversampling, or 8x or 16x even 64x. It's easy to find Multi-band compressors with 3,4,5 or 512 bands. Or the 600th naive copy of a neve EQ - now oversampled.

But I was talking about "relevant" engineering. The creation of something new and great, no matter what the costs are. Something that goes beyond the 50 year old solutions developed to solve a completely different set of problems. And there is a slight difference between the "high-end" philosophy and a bad/lazy programmers asking for half of your CPU without obvious sonic benefit...

Less than 20 years ago, digital audio production was unaffordable to most people. 10 years ago full ITB production became reality, 5 year ago *some* plug-ins began to sound more or less OK. Just a few months ago, Slate, U-He, Fabfilter and others appeared to rule the last isolated analogue nerd (who had very good reasons for it until 2010-2012). Your life span expecation must be pretty short to call the current state of things a "plateau".

Don't you remember the insane Emu/Akai/Sequencer/Mixing-Desk mess little more than a decade ago?! IMHO There's still plenty potential for digital audio, both from the creative, usability and performance point of view. Just look back a few months, the curve still seems to be pretty steep to me!

[yl]

I agree the curve is still there it is a less steeper in the timeframe that you mention (2010-2012)..
(Melodyne polyphonic note analysis is more of an example than any odd compressor to introduce a new parameter for creative tweaking or "aliasing" led)..
while there's still a software product coming out once in a while that creates a bit of excitement.. I'm left uninmpressed in my day to day job..
almost all of the tools I need in the audio world are already made.

as for creative and performance options I completely agree.. this is a golden dawn..
the creative options we have now are tenfolds beyond what was possible before
but.. I see workflow and human interfaces breaking grounds..

but what else more can audio dsp offer from this point aside from the obvious more/better/faster?
(aren't most of the inherent problems solved already? if not I'd really like to know what those are?)
forgive me if I lack the imagination or knowledge

I'm also guessing whatever exciting new products to come out will also be dependent on new hardware technology coming out.
this kind of reminds us of the analog (tubes->transistors) and early digital days

I'm really curious by what you see as 'revolutionary' in the broadcasting industry...
care to share some links?

[do_androids_dream]

great read I use both the TDR compressors all the time - they sound fantastic

[FabienTDR]

almost all of the tools I need in the audio world are already made.

I fully agree. It was possible to produce great records 50 years ago, no doubt. The "main" problems have been solved already.

But the demands are constantly changing, too, and it's extremely difficult to foresee the impact of future ideas, especially in the fast moving software market. Take MP3 as an example, back in 96 nobody in the music scene would have ever thought that a lossy format would conquer the world within shortest time, with dramatic effects on real economy. Or smaller examples such as the hilariously late discovery of the transient designer "trick" which quickly established itself in professional work-flows and virtually replaced most "punch-gates" techniques. It took over 50 years until someone "discovered" this technically nearly trivial circuit! For 50 years, audio gear designers had it right in front of their nose, but couldn't draw the right conclusions! Such things happen! True qualitative improvements. Your Melodyne example is a good one, too. And I'm pretty sure we'll see some extremely powerful speech processors/synths appear soon, able to literally transform the voice of a vocalist. Or maybe even a whole new class of effect processors based on re-synthesis (which could be the basis for incredible dynamics processing). Or maybe someone comes up with a new kind of hosted networked DAW which works! The list is endless. Maybe someone comes up with a perception based dynamic EQ with direct access to convincing acoustic "moods".

Beside the inherent technical problems such as aliasing and precisions issues, there is still a lot to discover and improve.

As mentioned in my previous post, some interesting improvements in the Broadcast field relate to the reduction of IM during dynamics processing and schemes for "true" band-limited overload protection without second clipper/limiter. They already intensively work on solutions for high quality dynamics expansion/restoration (a big demand in mastering nowadays!). Again, they always were 3-4 years ahead of the music industry, no matter if was about intelligent compression, multi-band processing or DSP in general.

I can highly recommend Robert Orban's papers, patents and articles (they are all over the web). You'll quickly see how advanced this scene is. For example, he patented (and implemented) aliasing reduction techniques for dynamics processors far before the music production scene now slowly begins to adopt similar techniques.

For example:
http://www.freepatentsonline.com/6937912.html
http://www.freepatentsonline.com/4249042.html
http://www.freepatentsonline.com/4134074.html

[Tricky-Loops]

Beside the inherent technical problems such as aliasing and precisions issues, there is still a lot to discover and improve.

There even could be some totally new never-heard processors in the future - like Delareverb, Equalcomper, Limitlayer or Saturfilter...