Critical and Analytical Listening

[chazba]

What a great thread...lots of experience reflected in the opinions and viewpoints and much to reflect on...yeah...no doubt.
Does anybody have any suggestions on how to draw out the best performance that a musician or singer is capable of. This one skill will contribute more to a great mix than all the analytic listening you could do for the rest of your life. Some will say that this is the function of the producer but the reality is that most (all) of the groups and artists that show up at my place don't have a producer and expect me to provide authority and guidance. Lets hear from you more experienced guys

Fight Organized Crime
Vote 'em ALL out of office

chazba

 

[masteringhouse]

Did you want us to actually post one of those critical listening evaluations? OR what were you looking for? I filled out a couple of those forms on others' songs... not my own just for my own analysis' sake.

I can post them if you want? I am going to go over and track later today so I might be able to do another one for one of my songs.

Fantastic! Please post.

We will need to hear a version of the songs if they are not commerically available. I was hoping that most tracks would be available on Itunes or other online access to listen and compare. What I would like are good and bad examples demonstrating different aspects of the points mentioned in this thread and the eval sheet. Not everyone is going to agree, but that's where things get interesting.

Next I was thinking that we could take a track from one of the members here and hopefully evalute the strong and weak pointsand ways of correcting it. From there possibly a remix (if needed) and rough masters based on the suggestions.

Sound cool?

 

[gecko zzed]

What looks good in theory doesn't always translate well into practice, and so it was with CLEF, the Critical Listening Evaluation Form, despite its neat acronym. My first hurdle was not being able to figure out how to fill it in electronically. My second was with item 4.2 "Creativity". Both tracks I tried CLEF out on had reasonable creative arrangements, but I didn't discern much in the way of a "use of a new or unusual effect" on either. Nevertheless, the absence of such shouldn't be regarded as a mixing flaw: not every track needs this in the recording or mixing, and some tracks stand out without it. I also found that a problem in one area caused other areas to be affected. This might be okay, but I had gone in with the idea of mutual exclusiveness of the categories.

I reviewed two tracks, both of which I really liked: one that I found to be totally engrossing and deliciously mixed, the other that, though I loved the song, the recording disturbed me. They were "Forget about it" (Alison Krause) and "Straight Lines" (Silverchair).

Because I couldn't fill in the form electronically, here is a summary.

"Forget About it", Alison Krause

Group 1: Frequency balance. All 5. A brilliantly clear and clean mix
Group 2: Sound stage imaging. All 5. A great and credible representation of the band within the stereo landscape. No problems with mono.
Group 3: Dynamics and Distortion: All 5. Clean transients, no audible recording artefacts. I wanted to hear more. Overall RMS was about -18
Group 4: Use of effects: 5 for 4.1 and 4.3 but I didn't know how to respond to 4.2
Group 5: Performance and production: All 5. They are brilliant and savvy musicians

"Straight Lines", Silverchair

Group 1: Frequency balance. All 3. While I think the recording is great, there was a harsh sibilance present about a third of the way in, a problem with 1.3 that affected 1.1 and 1.2. This problem was a manifestation of 3.4
Group 2: Sound stage imaging. All 4. This is a big production, and a lot of sounds are competing for their presence on stage.
Group 3: Dynamics and Distortion: 3.1 is 3, This is a loud track, with an average RMS of about -11.5. The last minute of the song sits at around -10. 3.2 is 4, 3.3 is 4, but 3.4 is 2. While I couldn't find any 0db points, the whole backend of the track sounds like it had digital distortion. This, of course, messed up with results in Group 1, and contributed to me giving it only 2 for 3.5 (Listening fatigue). I didn't want to listen to the end. I am not convinced, though, that this distortion is an intrinsic part of the recording, because I've heard it many times elsewhere without being bothered. But it is certainly present on this CD.
Group 4: Use of effects: As for Alison Krause, 5 for 4.1 and 4.3 but I didn't know how to respond to 4.2
Group 5: Performance and production: All 5. Again, they are brilliant and savvy musicians

 

[masteringhouse]

Thanks!

I agree that the absence of a use of a new or unusual effect doesn't mean that it's a bad production, in fact in cases where it becomes distracting it can actually turn it into a bad one. My intent in the form was a tasteful use of an effect that engages the listener and enhances the overall production. Also by "effect" I didn't mean the usual reverb, delay, etc. but it could be something like using the sound of a dripping faucet as a percussion instrument. Personally I gotta give somebody extra points for that. Think of all the engineering breakthroughs that came out of Geoff Emerick and the Beatles.

Regarding and electronic version, this is a good idea if we are going to be doing this over the long term. I think the approach that you used here is fine though to get this party going.

 

[SouthSIDE Glen]

My second was with item 4.2 "Creativity". Both tracks I tried CLEF out on had reasonable creative arrangements, but I didn't discern much in the way of a "use of a new or unusual effect" on either.

I personally feel a bit constrained by sticking literally to that definition there. I'll defer to Tom on this, but I would assume that one could include effects that are not necessarily "new" or all that unusual and don't necessarily fall under 4.1's definition, but rather are common effects used in an uncommon way. Or even the creative *absence* of an effect would qualify in my book.

But I also agree that "scoring" here does not indicate good or bad. This goes back to what I said earlier about how CLEF (I like that!) is almost more of a character study or personality test of the mix instead of an overall score of quality. One should not just add up all the points at the end and decide that the one with the greatest total is the best mix.

Like in most sports and arts, it comes down to a tension between judging the technical vs. judging the creative.

There are some great mixes out there that because of creative decisions purposely made, might "score" fairly low in some key technical categories on the CLEF, while scoring large in other more creative or emotional categories. I named those three tracks earlier for a reason, because those are three tracks that might wind up showing an interesting profile on the CLEF for that very reason. But that wouldn't make them bad.

The bottom line in such cases, IMHO, is how well the mix serves the song. Tom hits on that well with 5.3. It'd be nice to see a little more detail there like on how the mix creates or serves the hooks, how well does the mix balance change to move focus between instruments or parts appropriately, and things like that; i.e. how well do the demands of the song dictate the mix.

G.

 

[masteringhouse]

I personally feel a bit constrained by sticking literally to that definition there. I'll defer to Tom on this, but I would assume that one could include effects that are not necessarily "new" or all that unusual and don't necessarily fall under 4.1's definition, but rather are common effects used in an uncommon way. Or even the creative *absence* of an effect would qualify in my book.

Absolutely, "effect" in this context is a loose definition, not a box. Maybe "treatment" would be a better term.

I'm digging the input on this thread. It's moving toward my favorite here on the forum. I think that this discussion has been long overdue. It gets down
to the real issues in audio production and is much more interesting (to me) than talking about gear, settings to use, and rehashing audio wars.

Rock on ...

 

[masteringhouse]

I'm thinking of combining the free mastering thread with this one now. The song that will be picked from that thread will go through an evaluation here by those willing to participate. With critiques and goals for mastering set, the students will master the track with the tools that they have available and provide the "fixes" to get it closer to the intended goal for the recording. I'll then post for opinions here on what they have accomplished. 15 versions may be a bit much for all of you to listen to, so it may be something like top 5 as decided by the owner of the track, my second engineer, and myself.

Sound cool, or is this a deviation from the direction of this thread?

 

[iqi616]

No question you have developed at least some degree of analytical, or educated, ear. That's not just another factor, but it's what this thread is all about; just how does one GET that educated ear.

For some it comes more naturally than others, but for all with fairly normally functioning ears (even only one of them ) it is a skill that can be learned and developed. To go back to your first sentence, Mike, the question is how did that listener's ears get that education? These methods are key.

Here's how my ears got educated...

I read Bob Katz's book and calibrated my volume knob. For the first time I was hearing reference tracks at the same loudness as my mixes and I could hear the difference between my mixes and the references better and with total consistency and repeatability. My mixes sounded the same the morning after compared to the references as they did the night before.

Then I started to go beyond Bob's concepts because I don't believe we should always monitor that loud. The ear's frequency response varies at different loudnesses and so you need to monitor at various loudnesses to make sure the mix/master works whether played back quietly or loudly. The key concept I retained is that the reference tracks are always adjusted to be at the same loudness as my mix. I also discovered that it's easy to detect over-compression at lower monitoring levels. I can hear the sound hitting the limit. With further practice I can now hear compression and limiting in other people's work.

Then I started using a pair of cheap computer speakers as a limited-bandwidth alternative reference. This really helped me hear the difference between my mixes and the references. I could hear why my kick drums became click drums through the computer speakers.

This combined with extensive experimentation with 4-band parametric EQ and compression has trained my ears. I've spent considerable time listening to the difference between professional recordings and my recordings and recordings made by other people. I have known for a while now what a professional recording sounds like and since then it's been a question of refining my skills so that I can bridge that gap between what I have and what I want.

For phase issues, I'm not sure how I picked up the ability to hear that. I've always been pretty sensitive to that when listening to stereo TV - sometimes they screw up so bad it's just sadly unprofessional.

 

[SouthSIDE Glen]

I'm thinking of combining the free mastering thread with this one now. The song that will be picked from that thread will go through an evaluation here by those willing to participate. With critiques and goals for mastering set, the students will master the track with the tools that they have available and provide the "fixes" to get it closer to the intended goal for the recording. I'll then post for opinions here on what they have accomplished. 15 versions may be a bit much for all of you to listen to, so it may be something like top 5 as decided by the owner of the track, my second engineer, and myself.

Sound cool, or is this a deviation from the direction of this thread?

I don't have a problem with it myself, Tom; just being unfamiliar with your course, I'm just not sure quite where you're going with this. I know you don't want me exposing what's behind the curtain and all that, just saying that I don't quite understand asking folks to jump into the water quite this early. I'm not disagreeing with it, just confused by it.

G.

 

[masteringhouse]

I don't have a problem with it myself, Tom; just being unfamiliar with your course, I'm just not sure quite where you're going with this. I know you don't want me exposing what's behind the curtain and all that, just saying that I don't quite understand asking folks to jump into the water quite this early. I'm not disagreeing with it, just confused by it.

G.

Since we're getting such an overwhelming response on evaluations from commercial recordings , I thought that bringing in something a little closer to home[recording] might get us more feedback. Also it corresponds with what the students will be doing, so once the evals are in here, we can compare to the mastering results in about 2-3 weeks (this class is once per week so it will take time as well).

There's no reason why we can't continue with other examples in the meantime though if folks want to give them up.

 

[SouthSIDE Glen]

Since we're getting such an overwhelming response on evaluations from commercial recordings , I thought that bringing in something a little closer to home[recording] might get us more feedback. Also it corresponds with what the students will be doing, so once the evals are in here, we can compare to the mastering results in about 2-3 weeks (this class is once per week so it will take time as well).

There's no reason why we can't continue with other examples in the meantime though if folks want to give them up.

Sounds good to me.

I don't know what it is about people being afraid to deconstruct and analyze commercial recordings. I ran across the same roadblock in ATC. But, unless is an obvious crushing or something, there seems to be some reticence to critiqe the commercial stuff the way they do in MP3 clinics.

And in the meantime while we're waiting for the mix, I wouldn't mind continuing a little further discussion on some further ear training techniques.

G.

 

[Yonce N Mild]

Some awesome stuff here. Thread bookmarked.
I'm gonna dig into this stuff a little more when I get time.

Thanks guys!

 

[masteringhouse]

Sounds good to me.

I don't know what it is about people being afraid to deconstruct and analyze commercial recordings. I ran across the same roadblock in ATC. But, unless is an obvious crushing or something, there seems to be some reticence to critiqe the commercial stuff the way they do in MP3 clinics.

And in the meantime while we're waiting for the mix, I wouldn't mind continuing a little further discussion on some further ear training techniques.

G.

One of the typical ways of learning frequency recognition is to run pink noise through an EQ and have someone both cut and boost frequencies at random then have you tell them which one. After getting somewhat familiar with pink noise run various types of music through the same process.

I also recommend that you do ongoing testing like this yourself. Throw away the spectrum analyzer. Whenever you are adjusting an EQ, before you go sweeping around, try to determine the frequency that needs to be boosted or cut. Then try the boost/cut and see how close you came while sweeping. If you do this religiously I can guarantee you'll get better at it with practice, especially when you have a client behind you looking at his watch.

 

[SouthSIDE Glen]

One of the typical ways of learning frequency recognition is to run pink noise through an EQ and have someone both cut and boost frequencies at random then have you tell them which one. After getting somewhat familiar with pink noise run various types of music through the same process.

Yup, especially the music part of it. Once one gets used to listening to the *sounds* of the instruments behind the music and paying attention to detail, it's time to start associating those aural memories with names and numbers.

I like to suggest two or three reference CDs, one we are very familiar with in a genre we're comfortable with, and two that are outside our comfort zone that are recommended as having wide and balanced frequency usage, but in fairly different genres from each other and from our favorite.

Take those and a 2/3rds octave (i.e. 15-band) graphic EQ or better. I like graphic for this because it's graphic-ality will help those who claim to be visually-oriented ID frequencies with position in the spectrum, and because we're guaranteed to have each band actually labeled by frequency, and not just guessing by being between hash marks on some hardware parametrics.

Sit down with those CD playing random tracks from them through the EQ, starting with all bands set flat (dead in the middle), and, one at a time, move the sliders up and down during each track to hear what that general frequency area (remember there is Q there) actually sounds like. Use the same kid of listening technique we used when listening to the storm, the ballpark, and the live acoustic combo. Listen to the sound, and not just the music. Listen to each instrument in a mix and listen how it's timbre changes when that frequency is added or taken away from it, and how that change makes you feel.

Do this for an hour a night for a week or two, and we should be ready to bring a friend in and test us with a different set of CDs than we practiced with, and without our being able to see the EQ controls, have them move them around and test our ears until we can get within at least one slider away 90% of the time.

Keep going, of course, but then one is at the point, IMHO, when they are ready to start reading things like frequency charts and EQ tuning tips, because only then will that information and tips actually start making any sense, and only then will we have the basic skills to know if, when and how literally any given tip actually would apply to our situation.

All of a sudden now we find ourselves not only listening to our ears, but directing them as to what to actually listen to, and actually starting to somewhat make heads or tails out of what they are actually telling us in terms we read in the books and on these forums every day. Our ears are now starting to become an actual piece of studio gear.

But we're not there yet. Getting close, but now it's time to start taking our newfound analytical discrimination skills and start applying them to specific characteristics of the music and the mix.

This is where Tom's CLEF comes in to the mix (obvious pun intended) by identifying an categorizing specific attributes and goals of a mix that a good engineer usually uses their listening skills to listen for.

I think the CLEF is an excellent guide and listing of these attributes. As something not meant to replace or change it, but perhaps as an extra perspective of sorts to view its items through, I'd like to offer up that perspective to go along with and, if desired, to be discussed here.

But that's for the next post.

G.

 

[cakewalkKaKed]

And in the meantime while we're waiting for the mix, I wouldn't mind continuing a little further discussion on some further ear training techniques.

G.

regards ears training ,one idea would be to supply a complete un-effeted song .. drums bass vocals guitars e.t.c in separate .wav files, to load into preffered sequencers... you could then reference tweaks with track and time markers and call the thread "clef pallete"



(ill get my coat)

 

[michaelst]

More for ringing out a system in a live setting, but helpfull nonetheless for ear training:

Simple Feedback Training

 

[SouthSIDE Glen]

More for ringing out a system in a live setting, but helpfull nonetheless for ear training:

Simple Feedback Training

Nice link!

If I had the time, I'd like to develop an app that does the same kind of thing, but instead of using tones (BTW, anybody else notice that many of the tones in that SFT app do not seem to be pure?), to use selected WAV or MP3 music files and then adjust the frequencies like I explained in my last post.

It wouldn't be difficult at all to program, really; I'll put it on my To Do list but there are so many things in front of it right now, I don't know if/when I'd get to it.

G.

 

[gecko zzed]

I would say that this is definitely discriminating listening, but more in the area of listening for informational purpose. Now if you felt that the sound of the car brakes ahead has too much in the area of 8K and is 3 db too high then were talking critical listening.

Okay . . . while doing (of all things) the dishes just now, my mind went daydream mode and I started thinking about listening modes in mixing.

To my way of thinking, there are four listening abilities needed to be able to create a respectable mix.

These are:
1 discriminatory listening - in which we listen to identify problems that may cause us grief later on. This involves listening beyond the obvious and seeing what's lurking in the background - is someone shuffling paper, is that a cough, is that a popped 'p', is that some unwanted hiss, is that unwanted distortion?

2 diagnostic listening - in which we identify and fix problems with wanted sounds, e.g. is that voice too honky, is that snare sounding like cardboard, id that guitar too brittle, is that vocal track too wayward, is the mandolin competing with the fiddle? This is where we use tools such as EQ and compression to bring these things into line.

3 discernment listening - in which we arrange the sounds on our palette to create an integrated, spatially satisfying, acoustically credible whole. Is the vocal buried, does the kit sound like it's in a different room, have the instruments been spread so much that there is a big hole in the middle? Here the main tools are levels, panning and other spatial treatments (e.g. reverb).

4 creative listening - in which we listen to the song and unearth what it is telling us about itself and then decide what extra we can do to reveal that story. Does this song have a critical lyric line that needs to be clear, does this song suggest a particular mood that we can bring out with some special effects? This is where we can add things that are not part of the original instrumentation, and may not have been considered by the performers, but nevertheless will add to the song; things such as introduced sound effects and noise, the range of effects we can use, various manipulations of the recorded tracks and so on.

I once likened mixing to flower arranging, and I mentioned this elsewhere in this forum. The analogy is remarkably robust:

Given a bunch of flowers, what can you do with it?

1 discrimination: pull off the bugs, dead leaves and faded petals of individual stems

2 diagnostic: selecting from that bunch the stems that provide a pleasing balance of colour and height.

3 discernment: arranging these stems in a vase so that the colours and heights form an integrated display

4 creative: adding something that may not have been present in the original bunch, e.g. some contrasting greenery or other colur, or a different texture of leaf.

This is only part of the story though, because still unanswered here is how we learn to know what a 'wayward vocal track' is, or what constitutes an 'acoustically credible whole'.

Perhaps I should wash the dishes less often, but I put this up as an idea for others to consider and pull apart if necessary.

 

[rayc]

Gekko - Have you rewritten daddy Cool - "Doin' the washing UP."
I've only just stumbled onto this thread and I'm peeved as I've been on hols & could've deveoted some time to working my way through it at a relaxed pace as it developed.
then again, my tinnitus is soo loud these last few days (particularly my left ear) that it'd probably have been a waste of time. I'll have to get to a quack & be referred for a hearing test & HOPEFULLY some remediation.
Well done Tom & Glen, I love learning & just by reading I've put a couple of things into the broader understanding corner of my brain box.
Oh, I down loaded the sft to do a little self diagnosis but, if it's just tones downloaded as mp3 type files the spectrum will be truncated. Is it a series of tones or a tone generator?

 

[SouthSIDE Glen]

To my way of thinking, there are four listening abilities needed to be able to create a respectable mix.

These are:
1 discriminatory listening

2 diagnostic listening

3 discernment listening

4 creative listening
...
Perhaps I should wash the dishes less often, but I put this up as an idea for others to consider and pull apart if necessary.

Hey, zz, I do some of my best thinking while washing dishes, or doing laundry, or (my fave) taking a shower. Something about the menial tasks of keeping things clean that lets one's mid go?

I really like your cataloging by general listening function. It's jet another way to look at it and get one thinking about how to recognize and parse the details of what we're listening to. What's interesting to me about that is that although it's rather different than Tom's CLEF and my categorization perspectives coming up, none of these perspectives clashes, and they all contribute together to an understanding of what critical listening actually listens for and why.

G.