Does Time alignment and Phase coherency make for a better loudspeaker?


Some designers strive for phase and time coherency.  Will it improve sound quality?

jeffvegas

Lots to unpack in this thread.

Would it not be best to start with how we perceive sound? Our hearing is sensitive to timing over a narrow frequency range, about 200-1500Hz. It would make sense to have time alignment over that frequency range. Many speakers with mid-woofers in the 5-7" range by virtue of where their crossover frequencies are already are time aligned in this frequency range (single driver).

In terms of sound stage, all the other information we use for position, is frequency and volume based, not timing based. With that being the case, is there a good argument for time alignment over the whole frequency range?

@mijostyn , you appear to be advocating that a flat in-room frequency response is the ideal scenario. That is not supported by most people's listening impressions or research into preference, all which suggest a sloping reduced output at higher frequencies.

There is a misconception that in-room frequency should be perfectly flat in order to perfectly recreate the original performance. It sounds great on the surface but it is a flawed premise as you are not trying to recreate the performance, you are trying to recreate what was heard by the recording/mixing engineer, and they have already adjusted the frequency response based on what they were hearing at their workstation which is usually two somewhat near field speakers, but the total response ends up closer to downward sloping at higher frequencies, especially true when they do final mix and test it on larger audio systems and/or headphones which appear to sound best when targeted at a downward slope past about 3KHz.

Some of this may even harken back to the attenuation you would experience seeing a live orchestra or concert hall at typical seating distances and distance to instruments (the front row is rarely where the best sound is).

Speakers that have sharp dispersion limits such as horns, planar speakers and linear arrays have large advantages acoustically in residential rooms. A properly deadened small room say 16 X 30 sounds better than a really big room with high ceilings because these rooms usually have acoustic signatures that are harder to get rid of, they echo.

 

There is more sense and more useful information in the paragraph above than in every thread and post on cable and fuses combined, and you could probably throw in 90% of the threads and posts on DACs and amplifiers too.

I am not saying that these types of speakers are without their own unique flaws as well, but far too much energy is spent painting the pig, not dealing with the little "gifts" the pig has left all over your listening space. Seems pointless to get hung up on things that "may" make a 0.01db, or a few microseconds of difference, and ignore the things that make 1, 2, even 10db of difference, and even 10's of milliseconds of difference.

 

@cindyment , Hi! Thanx for joining this post. I do not believe we have met before.

I am only advocating a flat frequency response as a reference to start because it is the only reference that makes sense.  What else would you use? My own system is boosted below 100 Hz at 2 dB per octave going down and rolled off above 10 kHz at 6 dB per octave. This is the actual measured in room frequency response at the listening position from either channel.  

What most people think of as "flat" is "uncorrected."  Having a flat, in room frequency response with an uncorrected system would be like winning the lottery. In room frequency response can vary wildly. If you tell me a system is rolled of at 6 db/oct above 10 kHz I still have no idea what the system sounds like because I have no idea what the response was to start with. It gets even worse. The individual channels have different frequency response curves and can easily be 10 dB off of each other at certain frequencies even though they are exactly the same speakers. Imagine what this does to imaging! 

Very few audiophiles have a measurement system which in this day and age is a travesty. For $300 you can get an excellent USB measurement microphone and computer program with which you can measure group delays and frequency response. You can mess around with speaker positioning and room treatment or if you really want to dial it in get a digital signal processor.  All I can do when people tell me they can do this by ear is smile.

I saw this ad for a speaker company. It looked like it came out of Better Homes and Gardens. Two black tower speakers on either side of a black fireplace with white walls, a 12 foot ceiling and black granite floors. One side wall was solid window. Imagine what that sounded like. 

 

 

@mijostyn,

If you like your system, I would not necessarily recommend anything you are doing other than what you have right now. However, since you have the equipment and I get the impression like to tinker, I would look at doing a gated frequency response (over the frequencies you can) to see what your on axis frequency response is and how flat it is and see how that compares with the room response which is what I think you are saying is flat. Not as a rule, but as a guide, a flat on-axis with a declining room response after about 3KHz seems to match best average preference across a range of music. Your speakers will have less room interaction which is a good, but it also means one less variable to play with. For most rooms, as you noted, that is probably better. You may have some latitude though, since you have full equalization capability, that you can adjust your toe in to adjust direct/reflected balance, while using the equalizer to flatten the on-axis while achieving a different off axis from what you have now.  Will you like it better? Your two headed coin is as good as mine. I am certainly interested in what happens if you do. I expect a whole lot of worse, but possibly you will find a new sweet spot you like even better. This is not a 15 minute exercise.

Post removed