Genelec on audio science

[sergeauckland]

With the warm weather here, I'm glad for my right to bare arms.
S.

 

[Thomas savage]

With the warm weather here, I'm glad for my right to bare arms.
S.

Don’t bare legs though, no amendments would cover that crime.

 

[TBone]

Don’t bare legs though, no amendments would cover that crime.

OT, but it's been a really hot spring in these parts, the fish are being pushed to depths that I usually only encounter early to mid summer. And, every year at this time, I re-develop an annual hatred for insects that love to feast on blood, brutal this year. But what's worse ... the Tick population, not long ago rare, is now getting out of control. Deers can carry thousands of 'em, sucking em bone dry. Never mind taking your dog into the deep bush without fear of being mauled by the lime-disease carrying vampires.

So indeed, don't bare legs ...

 

[pirad]

If only those Genelecs were used properly...

 

[tomeh]

Wow, thanks for all the thoughtful discussion and questions. I can only add that there are a ton of people out there that are supposed to be professionals but have no business being in their positions as recording engineers, mixers and masters because they ignore these "facts" discussed here and choose playback systems based on their personal preferences for bass especially, brand names, marketing BS and inflated senses of personal worth. I blame most of this on a dangerous statement, but I'll do it anyway, "most musicians are terrible technically." Their ignorance of the technical side leaves them open to all sorts scams, loss of control of the preservation of their music, and loss of money. I have seen/heard people mixing and mastering on terribly unbalanced playback systems and everything below 500 hz is what they like the "sound of" on those speakers, in that room. I am now searching for Dr. Toole's book.

 

[tomeh]

Dr. Toole's book will be here Oct 19th. Thanks Dr. Toole

 

[Speedskater]

Dr. Toole's book will be here Oct 19th. Thanks Dr. Toole

My copy arrived about a year ago. Still haven't read the entire book.

 

[tomeh]

My copy arrived about a year ago. Still haven't read the entire book.

Wow Just got the book and I'm enjoying like nothing else in the past 30 years. I'm reading it straight through and it's every consideration, experiment, results, logical evaluation, reasoned direction and anticipated "next question" that I've either experienced or wished I had the time and resources to explore. He's brought together everything in such a logical manner. It's fantastic. And I only at page 75.

 

[svart-hvitt]

For our members in NYC, AES session with highly respected luminaries of audio science and production, moderated by @Thomas Lund this morning:

AES New York 2018
Recording & Production Track Event RP07
Thursday, October 18, 9:15 am — 10:15 am (1E07)
Recording & Production: RP07 - Circles of Confusion
Moderator:
Thomas Lund, Genelec Oy - Iisalmi, Finland
Panelists:
Bob Ludwig, Gateway Mastering Studios, Inc. - Portland, ME, USA
Sean Olive, Harman International - Northridge, CA, USA
Floyd Toole, Acoustical consultant to Harman, ex. Harman VP Acoustical Engineering - Oak Park, CA, USA
30.000 year old cave paintings are among human beings' most impressive cultural heritage, while it's unknown how excellent composers sounded even 300 years ago. We recently acquired technical skills to record sound, but that asset is now degenerating because of cognitive limitations and the circles of confusion, addressed by this panel.
Without proper anchoring of spectral balance and level, drifting over time is inevitable in self-referenced systems, thereby putting legacy recordings at the risk of sounding dated for no good reason, or causing irreversible distortion to be added to pieces of art. The panel will discuss baseline listening requirements for in-room and headphone spectral balance and level that stand the test of time, putting our interests as a species above commercial trivialities.

Source: http://www.aes.org/events/145/recording/?ID=6327

 

[Cosmik]

For our members in NYC, AES session with highly respected luminaries of audio science and production, moderated by @Thomas Lund this morning:
We recently acquired technical skills to record sound, but that asset is now degenerating because of cognitive limitations and the circles of confusion, addressed by this panel.
Without proper anchoring of spectral balance and level, drifting over time is inevitable in self-referenced systems, thereby putting legacy recordings at the risk of sounding dated for no good reason, or causing irreversible distortion to be added to pieces of art.

Absolutely. And preference-based listening tests are a major mechanism by which this drift can come about. Only by sticking with objectively-accurate systems and paying no attention to purportedly scientific listening tests can we avoid that drift.

 

[andreasmaaan]

Only by sticking with objectively-accurate systems and paying no attention to purportedly scientific listening tests can we avoid that drift.

What do you define as an objectively accurate system in respect of the one parameter that can't be ascertained objectively - dispersion? I suspect something quite different from the systems used to mix and master those legacy recordings.

If so, how does such a system avoid this drift?

(Will get back to your very interesting pm this evening btw!)

 

[Cosmik]

What do you define as an objectively accurate system in respect of the one parameter that can't be ascertained objectively - dispersion? I suspect something quite different from the systems used to mix and master those legacy recordings.

If so, how does such a system avoid this drift?

The point is that if you fix the characteristics of the transducers/room/human system objectively (and that's a subtle thing to understand - it might still be 'wrong'), the drift is eliminated. As soon as you start changing that system in response to what people say they prefer, drift occurs. One 'round' of preference-based testing might be allowable, but more than that, and it's obvious that drift can occur!

 

[andreasmaaan]

The point is that if you fix the characteristics of the transducers/room/human system objectively (and that's a subtle thing to understand - it might still be 'wrong'), the drift is eliminated. As soon as you start changing that system in response to what people say they prefer, drift occurs. One 'round' of preference-based testing might be allowable, but more than that, and it's obvious that drift can occur!

I'm not sure I understand In @svart-hvitt's post I took "drift" to mean drift from spectral balance/levels that predominated in the era of legacy recordings. Is this how you're defining it here too?

 

[Cosmik]

I'm not sure I understand In @svart-hvitt's post I took "drift" to mean drift from spectral balance/levels that predominated in the era of legacy recordings. Is this how you're defining it here too?

As the post says, "anchoring" the spectral balance and levels objectively is essential to avoid a continuous drift as listeners buy systems that emphasise, say bass, and then recording engineers begin changing the bass content to get their recordings to sound what they think is good on those commercial systems, but the public then start buying even more bass-heavy systems and so on.

It is easy to see how the average system as it was in the 1950s versus the 1970s will have influenced the balance of those respective recordings, and how headphone-wearing today could affect our current recordings. It also depends on how recording engineers monitor the mixes (near field/far field). Only today's speakers with uniform dispersion at all frequencies will sound spectrally the same in the near field as they do in the far field.

 

[andreasmaaan]

As the post says, "anchoring" the spectral balance and levels objectively is essential to avoid a continuous drift as listeners buy systems that emphasise, say bass, and then recording engineers begin changing the bass content to get their recordings to sound what they think is good on those commercial systems, but the public then start buying even more bass-heavy systems and so on.

It is easy to see how the average system as it was in the 1950s versus the 1970s will have influenced the balance of those respective recordings, and how headphone-wearing today could affect our current recordings. It also depends on how recording engineers monitor the mixes (near field/far field). Only today's speakers with uniform, relatively narrow dispersion at all frequencies will sound spectrally the same in the near field as they do in the far field.

Yeh, more or less agree with all of that

But "anchoring" to only recently-available (and still not widely used in mixing and mastering) uniform dispersion systems would seem to avoid the question of how to properly/best spectrally represent recordings of the past, would it not? Should we just forget about the dispersion characteristics of past mixing/mastering (and still overwhelmingly of the present, for that matter) and set the anchor where we think best, now?

I'm inclined to say yes to this question myself btw - I'm of the view that standardisation in mixing and mastering would improve the audio experience of everyone. I just don't think it solves the problem of properly representing these past recordings.

It also begs the question, which I've raised before, of what the "correct" uniform dispersion characteristic (i.e. directivity index) is.

Indeed, consensus would seem to be the only way to answer this question, and this consensus may be best determined by preferential listening studies and/or with reference to normalised dispersion characteristics of legacy systems.

 

[Cosmik]

I'm not sure I understand In @svart-hvitt's post I took "drift" to mean drift from spectral balance/levels that predominated in the era of legacy recordings. Is this how you're defining it here too?

I could imagine that there's considerable latitude in what sounds good if the 'drift' is due to a change in fader levels e.g. how the bass guitar fader is set versus the vocals etc. rather than a change in frequency-selective EQ or other invasive techniques. I would put the pinnacle of recording quality to be during the 70s, with a lot of those mixes sounding straight and pristine compared to today's. They still sound excellent on a good modern system.

 

[Cosmik]

It also begs the question, which I've raised before, of what the "correct" uniform dispersion characteristic (i.e. directivity index) is.

Indeed, consensus would seem to be the only way to answer this question, and this consensus may be best determined by preferential listening studies and/or with reference to normalised dispersion characteristics of legacy systems.

I would guess that even though people might have a preferred uniform dispersion angle for some recordings, and probably different ones for other recordings, it might not affect how an engineer would mix the recording. I would say it's mainly a 'focus' or 'ambience' issue rather than a spectral or level issue.

(I removed the bit about "relatively narrow" from my earlier comment but you caught a snapshot of it in your reply .)

 

[andreasmaaan]

I would guess that even though people might have a preferred uniform dispersion angle for some recordings, and probably different ones for other recordings, it might not affect how an engineer would mix the recording. I would say it's mainly a 'focus' or 'ambience' issue rather than a spectral or level issue.

(I removed the bit about "relatively narrow" from my earlier comment but you caught a snapshot of it in your reply .)

Even if we define it as a "focus" or "ambience" issue (which seems very ambiguous and vague to me btw), how would you propose the question be answered?

 

[andreasmaaan]

I could imagine that there's considerable latitude in what sounds good if the 'drift' is due to a change in fader levels e.g. how the bass guitar fader is set versus the vocals etc. rather than a change in frequency-selective EQ or other invasive techniques. I would put the pinnacle of recording quality to be during the 70s, with a lot of those mixes sounding straight and pristine compared to today's. They still sound excellent on a good modern system.

Which 70s recordings would you say exemplify this?

 

[Cosmik]

Which 70s recordings would you say exemplify this?

I would cite Rocket Man by Elton John as an example (I think the one I listen to is re-mastered or is it re-mixed), and the first albums by Kate Bush.

Of course there's always the danger that someone will say "I was there, and I know we used a graphic equaliser out of a car on maximum smiley setting and DBX compression without the expansion. The mix you're listening to comes off one of the cassette copies we made on a dictation machine next to one of the monitor speakers..." etc.

In fact, we do know that the Elton John came through a certain type of 1970s op amps in the mixing desk, but the damage was minimal I think .