Taperssection.com
Gear / Technical Help => Post-Processing, Computer / Streaming / Internet Devices & Related Activity => Topic started by: checht on December 04, 2018, 07:04:28 PM
-
My spouse complained that live recordings are 'just too quiet, then too loud'.
It seemed to me that for her car/work listening, she wants compressed and normalized waves. So I took a recording and compressed it in Audacity with default settings, compressed based on peaks checked.
.
She loved it, so I shared with a few friends, who liked it more than the original.
On first listening, I didn't hear any substantial artifacts, though I need to take a listen on my reference AKG cans.
So, wondering about this in general, and whether others have used compression in their mastering workflow, either as a usual step or just for a show with crazy dynamic range.
I've attached screenshots of the original waves and then compressed version.
Thoughts?
-
I would say I use compression in the vast majority of my recordings. Sometimes more, sometimes less, but it's rare that I don't add at least some compression. I'll also run a small amount of EQ.
Basically I do whatever I want to make the recording sound the best to MY ears. If it sounds good to others, cool, but I make my recordings for myself. I'm also open to sending RAW files to anyone that wants to do their own mastering, but I've never really had anyone take me up on that. I do think many regular folk prefer compression because it sounds more like official recordings.
From my experience I think it's best to sleep on (or at least take a break from) the final product. Frequently my ears get tired from working on a recording and I think it helps to take a break and come back later to confirm I'm happy with what I've done.
-
^Totally.
checht, here's a thread on one way to bring up the quiet parts while not doing damage to transients and loud parts (the loud parts and transients are where the compression distortions are most egregious when standard compression is not set carefully)- https://taperssection.com/index.php?topic=137594.msg1786634#msg1786634
This is even easier to do these days now that many software EQ's offer a Dry/Wet control, which can provides the same "parallel" or "bottom up" compression effect if set correctly. But I still think it's most useful to do it manually at first using the copied channel method to really hear and understand what's going on, then try to duplicate that without channel duplication just using the dry/wet control and other appropriate settings.
-
Funny you should mention it, I've been about to share the news about this new book coming out in February 2019, by Craig Anderton about Dynamic Processing!
https://www.halleonardbooks.com/product/viewproduct.action?itemid=269501
How to Use Dynamics Processing
(This is a new product that is not yet available but may be pre-ordered.)
Series: Musician's Guide Home Recordg
Publisher: Hal Leonard Books
Format: Softcover
Author: Craig Anderton
Release Date: 02/13/2019
$14.99 (US)
Inventory #HL 00269501
ISBN: 9781540024978
UPC: 888680738754
Width: 7.25"
Length: 9.0"
144 pages
(Prices and availability subject to change without notice.)
-
She loved it, so I shared with a few friends, who liked it more than the original.
This isn't really surprising because it's well established that we tend to prefer music that's louder. That's why it's essential to match loudness when comparing audio, and it's what led to the "Loudness Wars."
-
I would say I use compression in the vast majority of my recordings. Sometimes more, sometimes less, but it's rare that I don't add at least some compression. I'll also run a small amount of EQ.
Basically I do whatever I want to make the recording sound the best to MY ears. If it sounds good to others, cool, but I make my recordings for myself. I'm also open to sending RAW files to anyone that wants to do their own mastering, but I've never really had anyone take me up on that. I do think many regular folk prefer compression because it sounds more like official recordings.
From my experience I think it's best to sleep on (or at least take a break from) the final product. Frequently my ears get tired from working on a recording and I think it helps to take a break and come back later to confirm I'm happy with what I've done.
Basically this. I always limit my recordings. I prefer to leave the shows generally with the same dynamic range, but I always pull down the highest peaks by at least a few db
-
I prefer to leave the shows generally with the same dynamic range, but I always pull down the highest peaks by at least a few db
Same here. I usually eyeball the outlier peaks and set compression only to affect them, then do listening spot checks to see if the result sounds significantly different (it almost never does).
And I have at least on one occasion tried making a special "car mix" that was more heavily compressed, though I drive so seldom that I haven't bothered since.
-
I would say I use compression in the vast majority of my recordings. Sometimes more, sometimes less, but it's rare that I don't add at least some compression. I'll also run a small amount of EQ.
Basically I do whatever I want to make the recording sound the best to MY ears. If it sounds good to others, cool, but I make my recordings for myself. I'm also open to sending RAW files to anyone that wants to do their own mastering, but I've never really had anyone take me up on that. I do think many regular folk prefer compression because it sounds more like official recordings.
From my experience I think it's best to sleep on (or at least take a break from) the final product. Frequently my ears get tired from working on a recording and I think it helps to take a break and come back later to confirm I'm happy with what I've done.
I need to take this advice more often. I tend to get home from a show and think I can work on it for a bit. Ill upload it and listen to it later in my truck or elsewhere and realize I could have done something completely different. It doesnt happen much with two channel recordings, but the more channels I bring home the more I need a few listens to figure out what I want to do with it. Constant learning process i guess.
-
I distribute my recordings 2 ways:
1) a 24bit filers which is simply the raw, albeit normalized files. This is for purists, or for pro-audio oriented audiophiles like me with the ability to tweak dynamics and eq upon playback.
2) a "mastered" 16bit file set. For this I use a number of AU or VST plugins, usually including a Neve style parametric EQ to adjust to taste, and a LA-2A style compressor to give about 2-3db of gain reduction and make-up gain.
I listen tot he 24bit files on my big stereo at home or on fancy headphones, but with earbuds or in the car the 16bit version is much better.
-
I distribute my recordings 2 ways:
1) a 24bit filers which is simply the raw, albeit normalized files. This is for purists, or for pro-audio oriented audiophiles like me with the ability to tweak dynamics and eq upon playback.
2) a "mastered" 16bit file set. For this I use a number of AU or VST plugins, usually including a Neve style parametric EQ to adjust to taste, and a LA-2A style compressor to give about 2-3db of gain reduction and make-up gain.
I listen tot he 24bit files on my big stereo at home or on fancy headphones, but with earbuds or in the car the 16bit version is much better.
Very interesting.
I FINALLY have started distributing final output at 16 bit, but I still keep my 24-bit raw mixes. "Normalized" as you say; I build my mixes, with volume level changes only, to peak within -1.0 dB FS, before I do any dynamic processing on the final output.
I name those files like
BandName2018-12-05-RAW2448.WAV
Once I process those files, I output the result named like
BandName2018-12-05-LIM1648 or maybe GLIM1648 if I add gain.
If I make different versions I tag it on the end after the depth/rate so I can tell them apart when I'm tracking.
BandName2018-12-05-GLIM1648+2-3.WAV (+2 gain, -3 limiter threshold)
BandName2018-12-05-GLIM1648+3-4.WAV (+3 gain, -4 lim thresh)
-
Really appreciate all the useful input. Spending a chunk of time reading the thread on parallel compression, seems just right for my purposes.
In accordance with Gutbucket's advice, I'll do some manual workflows to get a sense of parameters and their effect, then likely move to using a plug in.
~Chris
-
I distribute my recordings 2 ways:
1) a 24bit filers which is simply the raw, albeit normalized files. This is for purists, or for pro-audio oriented audiophiles like me with the ability to tweak dynamics and eq upon playback.
2) a "mastered" 16bit file set. For this I use a number of AU or VST plugins, usually including a Neve style parametric EQ to adjust to taste, and a LA-2A style compressor to give about 2-3db of gain reduction and make-up gain.
I listen tot he 24bit files on my big stereo at home or on fancy headphones, but with earbuds or in the car the 16bit version is much better.
I like this approach. Best of both worlds, not overly complicated.
"the ability to tweak dynamics and eq upon playback" is what really makes the difference.. even for audiophiles who don't want to admit it (which is generally not the taper-audiophiles)
-
I distribute my recordings 2 ways:
1) a 24bit filers which is simply the raw, albeit normalized files. This is for purists, or for pro-audio oriented audiophiles like me with the ability to tweak dynamics and eq upon playback.
2) a "mastered" 16bit file set. For this I use a number of AU or VST plugins, usually including a Neve style parametric EQ to adjust to taste, and a LA-2A style compressor to give about 2-3db of gain reduction and make-up gain.
I listen tot he 24bit files on my big stereo at home or on fancy headphones, but with earbuds or in the car the 16bit version is much better.
I like this approach. Best of both worlds, not overly complicated.
"the ability to tweak dynamics and eq upon playback" is what really makes the difference.. even for audiophiles who don't want to admit it (which is generally not the taper-audiophiles)
Yeah. I don't understand the audiophile purist approach of shunning eq or dynamic processing on playback. I use Audirvana plus (https://audirvana.com) which allows for all kinds of upsampling in PCM and DSD, and allows for 4 AU plugins. I tend to load up
1. A parametric eq (https://ddmf.eu/6144-equalizer-plugin/),
2. A comp/limiter (https://www.ikmultimedia.com/products/trwhite2a/?pkey=t-racks-single-white-2a),
3. A "tube"/ "transformer" saturation plugin (https://www.soundtoys.com/product/decapitator/) and
4. A just in case brick-wall mastering limiter (https://www.izotope.com/en/products/master-and-deliver/ozone/features-and-comparison/maximizer.html).
I don't always use them all, but having the ability to adjust eq, m/s, dynamic range and saturation really can bring out the best in all kinds of recordings, and I can change on the fly depending to which transducer I'm using (speakers, headphones).
-
Tried it and didn't like the results. Ears told me that some subtlety was lost by compressing. I don't mind quiet/loud that's how the show was played and that's what I want to hear.
-
Tried it and didn't like the results. Ears told me that some subtlety was lost by compressing. I don't mind quiet/loud that's how the show was played and that's what I want to hear.
Well, "compressing" and "compressing well" aren't necessarily the same thing ;D
-
a free "mastering" Compressor everyone here should own is this:
https://www.tokyodawn.net/tdr-kotelnikov/
-
I'd be much more inclined to use volume automation over compression on most show recordings. Combine with a limiter that is hardly/barely touched to set an absolute ceiling.
-
I personally find automation to be significantly less natural than a well set comp/limiter for stuff like this. Just personal preference though.
-
Tried it and didn't like the results. Ears told me that some subtlety was lost by compressing. I don't mind quiet/loud that's how the show was played and that's what I want to hear.
Well, "compressing" and "compressing well" aren't necessarily the same thing ;D
FU Simpson. 😀 but yeah I didn’t exactly put a lot of effort into it.
-
Tried it and didn't like the results. Ears told me that some subtlety was lost by compressing. I don't mind quiet/loud that's how the show was played and that's what I want to hear.
Well, "compressing" and "compressing well" aren't necessarily the same thing ;D
FU Simpson. 😀 but yeah I didn’t exactly put a lot of effort into it.
Haha, no offense intended! It can take some trial and error.
-
I personally find automation to be significantly less natural than a well set comp/limiter for stuff like this. Just personal preference though.
Micro or macro?
I do little bitty (well under one second most of the time) volume fixes with the envelope tool and it's brutal but effective.
I also do broad fixes with the envelope tool (Audacity is my main DAW software) and find it super useful.
I do my final limiting with an automated plugin, but I get things pretty tight and consistent before I do that.
-
I personally find automation to be significantly less natural than a well set comp/limiter for stuff like this. Just personal preference though.
Hmmm, I might guess it's in how you use either one. I'm used to automating volume down to the syllable level on vocal tracks before feeding them into a compressor. Much much less on program material like a live show 2 track. Limiter touching the errant peaks there. There are devices that are AGC's (automatic gain control) which react more slowly than most compressors, and also crossover into expansion from compression, so the threshold is more of a center line. I'm not aware of any software AGC's, but there must be some.
-
I personally find automation to be significantly less natural than a well set comp/limiter for stuff like this. Just personal preference though.
Micro or macro?
I do little bitty (well under one second most of the time) volume fixes with the envelope tool and it's brutal but effective.
I also do broad fixes with the envelope tool (Audacity is my main DAW software) and find it super useful.
I do my final limiting with an automated plugin, but I get things pretty tight and consistent before I do that.
It's more context than micro vs macro. In multis, sure. I just don't like hearing the entire thing go up and down. I find a well set limiter to be way more transparent (I'm also kinda lazy)
-
I find a well set limiter to be way more transparent (I'm also kinda lazy)
Lazy is such a harsh word - perhaps you mean that you are really into time management?
An overweight friend of mine says "If you wanna know the easiest way to do something, ask a fat guy!"
-
I find a well set limiter to be way more transparent (I'm also kinda lazy)
Lazy is such a harsh word - perhaps you mean that you are really into time management?
An overweight friend of mine says "If you wanna know the easiest way to do something, ask a fat guy!"
haha we could say that.