1. Home
  2. misc-audio
  3. 195864-117078

Here’s how good AI is now. Check it out!

AI Overview

+18

Yes, the threshold for hearing audible phase shift on an analog equalizer can change significantly depending on the audio system and listening environment

. The audibility of the EQ's phase shift is influenced by factors beyond the EQ itself, including room acoustics, speaker design, playback volume, and the complexity of the audio material. 

How the audio system affects audibility

Room acoustics

  • Acoustic masking: In a typical listening room, sound reflections from walls, ceilings, and other surfaces create a complex web of constructive and destructive interference (a "phase soup"). This ambient phase distortion is far greater than what an analog EQ introduces, effectively masking the more subtle EQ-induced shifts.
  • Threshold shift: Because our ears are evolutionarily adapted to filter out the natural phase shifts caused by environmental reflections, the audible threshold for an EQ's phase shift is much higher in a normal room compared to a controlled, anechoic (echo-free) environment. 

Loudspeakers

  • Crossover effects: The crossovers in multi-driver speakers are a major source of phase shifts, as different frequencies are routed to different drivers (woofers, tweeters). The phase shifts caused by a speaker's own crossovers can easily dominate and obscure any shifts introduced by an external EQ. This is less of an issue with headphones or single-driver speakers.
  • Driver alignment: In multi-driver speakers, if the drivers are not perfectly time-aligned, it can create additional, audible phase distortion. Some high-end speakers, for instance, are designed to have all frequencies leave at the same time to ensure better phase coherence. 

Playback volume

  • Audibility increases with volume: Studies show that phase effects, particularly in the lower frequencies, become more noticeable at higher sound pressure levels (SPLs). If you are listening at low volumes, phase shifts are less likely to be audible, but they may become detectable when listening at higher, more intense levels. 

Headphones vs. speakers

  • Headphones offer higher sensitivity: Controlled studies have shown that phase distortion is more detectable on headphones than on loudspeakers in a normal room. Without the masking effect of a reflective room, the subtle changes introduced by an EQ's phase shift become more apparent. 

Complexity of the audio material

  • Musical content vs. test signals: Phase shift is generally inaudible on complex musical and speech signals. It is most easily detected with specific, simple test signals, such as square waves or tone bursts, and only under very controlled conditions. A phase shift that is obvious with a simple test tone may be completely lost in a dense mix. 

EQ-related factors that remain constant

While the audible threshold changes, the physical phase shift caused by a given analog EQ remains constant regardless of the audio system. The EQ's circuit design dictates the filter's minimum phase characteristics, which means the amplitude and phase response are inherently linked. What changes is the listener's ability to perceive that shift based on the overall context of the playback chain. 

tlcocks
audiokinesis
2,800 posts
 

Thanks for your kind and non-AI-generated words, @erik_squires!  I share your skepticism of AI as a reliable primary source.

A few months ago I used AI in an attempt to speed up my search for woofers with a particular set of characteristics, among which was having one or more demodulation rings (also known as "shorting rings" or "Faraday rings"). The answers I received were WRONG more often than right! I found out by checking the data sheet of every recommendation. In some cases I even contacted the manufacturer to double-check whether the data sheets were correct, just in case the AI had a reliable source of information that I was unaware of. Nope.

The AI presented its incorrect recommendations as if they were absolute facts, with no clue in the wording that there was any uncertainty or any possibility of error.

I was, and still am, amazed at how incompetent AI was at finding information that was available to it, and how "dishonest" it was in its presentation of incorrect information as "fact".  This experience makes me wonder how much bad information is being given out by AI in other areas.

Anyway if the answer matters to me, I will do my own fact-checking the old fashioned way. 

Duke

rvpiano's avatar
rvpiano
2,991 posts
 

AI affected my whole listening experience. I’ve suffered throughout the years without the presence of a volume control due the fact that I have a vintage preamp.  Chat GPT suggested a DAC with remote control.  Changed my life. Helped me set things up as well.

It also stroked my ego by asserting my system is in “reference territory.”

toddalin's avatar
toddalin
666 posts
 

AI does not understand the subtleties of humor and will change your question so that it can answer it.

Go to 25:35 to hear the original question as I asked it:

https://youtu.be/lmWFrMq3qNY

 "Why does the porridge bird lay his egg in the air?"

::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::

And the response, which misses the mark...

The phrase "Why does the porridge bird lay his egg in the air?" is not a question with a literal, biological answer. There is no such bird as a "porridge bird" that lays its eggs in the air.

This phrase is a nonsensical riddle or a piece of absurdist humor, similar to "Why did the chicken cross the road?" The humor comes from the fact that there is no logical answer. The purpose of the question isn’t to get a factual response, but to create a moment of confusion or to be a commentary on the nature of truth, reality, and accepting nonsense.

:::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::

This is NOT the response to the question on the disc that the Firesign Theater was looking for and misses the mark.  wink

BTW, I’ve asked this question of AI various times and got various, but similar answers that ALWAYS miss the humor because of the pronoun.

So what is the answer, that AI has never produced?

They Don’t!  The males of ANY SPECIES don’t lay the eggs...,  the females do! 

erik_squires
17,570 posts
 

Like @audiokinesis and @rvpiano -  Using AI for shopping can leave you with very mixed results. 

Even with coding tasks, while sometimes AI can really accelerate my work, at other times it keeps recommending code which just won't work.  The libraries have changed or the programming language is wrong for the syntax.  Ugh. 

drmuso's avatar
drmuso
256 posts
 

I recently asked Google's AI to identify a deck in a photo;  I was wanting to see if a photo of a guitar on that deck had been lifted from another listing for the same guitar.  It couldn't identify/recognize the deck, and it incorrectly identified the guitar as a reissue when it was a somewhat similar-looking vintage guitar that had been refinished.

Any guitar aficionado with good visual memory would not have made the same mistake as AI.  So AI may be good with more left-brain, rational questions, but maybe not as advanced in sensory discrimination.  It would be worthless in describing the difference in sound of two cartridges unless it had incorporated previous reviews of them.

More to discover