AUDIO STARTS WITH RECORDINGS OF PERFORMANCES

[Jim Shaw]

For audio playback enthusiasts, it all starts with a performance and its recording. Obvious. But from the performance, the sound path can be cooked, tortuous, and hijacked. But for us, that recording is what we may be stuck with as a starting point.

I find it interesting to look at recording techniques with an eye toward uderstanding (especially) what might sound really bad in our listening rooms. And it can also help explain why some systems -- especially speakers -- can be flawed but sound rather OK or really terrible. No speaker is perfect, and each model is different. All of the inherent issues of masses in motion and suspension, resonances, driver breakup, shortcomings of multiple drivers and crossovers, and mismatches of transfer functions between amplifiers and speakers can go largely unnoticed -- or sound just awful when reproducing music. Is there a way to study why one set of issues sounds 'good,' and another 'annoying'?

I recently watched this short YouTube video, where a recording engineer speaks of the character of certain frequency bands and how they can grossly affect the average ear. I found it interesting, You might too. If we were to overlay these bands with empirical playback system test data, would we find an advantage?

I'm interested in enlightened replies. (And the opposite is absolutely true.)

Here's the video. There are, I know, tons more.

 

[escksu]

For audio playback enthusiasts, it all starts with a performance and its recording. Obvious. But from the performance, the sound path can be cooked, tortuous, and hijacked. But for us, that recording is what we may be stuck with as a starting point.

I find it interesting to look at recording techniques with an eye toward uderstanding (especially) what might sound really bad in our listening rooms. And it can also help explain why some systems -- especially speakers -- can be flawed but sound rather OK or really terrible. No speaker is perfect, and each model is different. All of the inherent issues of masses in motion and suspension, resonances, driver breakup, shortcomings of multiple drivers and crossovers, and mismatches of transfer functions between amplifiers and speakers can go largely unnoticed -- or sound just awful when reproducing music. Is there a way to study why one set of issues sounds 'good,' and another 'annoying'?

I recently watched this short YouTube video, where a recording engineer speaks of the character of certain frequency bands and how they can grossly affect the average ear. I found it interesting, You might too. If we were to overlay these bands with empirical playback system test data, would we find an advantage?

I'm interested in enlightened replies. (And the opposite is absolutely true.)

Here's the video. There are, I know, tons more.

I would say alot has to do with personal preferences...

From the posts in this forum, there are pple who find the speakers/music too bright. Some find it flat (need tone controls to change the sound), wanted more bass. Some wanted it to be as close to the recording as possible. Environment also plays a part.

 

[Jim Shaw]

I would say alot has to do with personal preferences...

From the posts in this forum, there are pple who find the speakers/music too bright. Some find it flat (need tone controls to change the sound), wanted more bass. Some wanted it to be as close to the recording as possible. Environment also plays a part.

If you read the research done by Dr. Floyd Toole (among others), one of the revelations is that quite a large majority of experienced listeners agree rather than disagree about carefully crafted listening tests of many speakers. This agreement became one founding and enabling principle of Toole's research.

The common wisdom (?) says what you say -- that listening tests are all scattered by personal tastes. The evidence shows otherwise -- most listeners tend to agree rather than disagree about the quality of music reproduction. Read Toole, if you haven't. Toole's colleague, Dr. Sean Olive, has done similar research into headphones with general acceptance of the principles of agreement in large numbers of tested listeners. The statistical analyses have been extensive, and they agree.

I have little interest in persuading you to change your opinion. But perhaps you might not be quite so certain that you're right.
...
What interests me is the potential correlation between what this YT video says as it might be overlayed on the test data we are collecting. What does it mean if a speaker emphasizes the frequency band that contains the annoying sound of fingernails on a blackboard? What if there's a slump in the speaker's spectrum in the 2.5 kHz region? How do listeners like the sound of an underdamped speaker that offers up resonant peaks in the 80 to 150 Hz region?