Is transient response the most important thing for the perceived audio quality in a system ?

[Tangband]

This is an interesting article :

Understanding Transient Response in Audio: A Beginner’s Guide – HiFiReport

www.hifireport.com

”Why is transient response important? Well, music is full of transients. If your audio system can’t reproduce them accurately, the music will sound muffled, lifeless, or distorted. On the other hand, a system with good transient response can make the music sound dynamic, detailed, and vibrant.”



( A Scottish audio company concentrated during the -90 in good transient response in their loudspeakers, but also possible in their other products. There is no whitepaper about this and it might be a company secret or just a rumour )

…so are we concentrating on the wrong things when we measure our loudspeakers?

Is a good directivity and frequency response in a loudspeaker not good enough for a really good perceived sound, if the transient response is bad ?

Discuss !

 

[HarmonicTHD]

This is an interesting article :

Understanding Transient Response in Audio: A Beginner’s Guide – HiFiReport

www.hifireport.com

”Why is transient response important? Well, music is full of transients. If your audio system can’t reproduce them accurately, the music will sound muffled, lifeless, or distorted. On the other hand, a system with good transient response can make the music sound dynamic, detailed, and vibrant.”



( A Scottish audio company concentrated during the -90 in good transient response in their loudspeakers, but also possible in their other products. There is no whitepaper about this and it might be a company secret or just a rumour )

…so are we concentrating on the wrong things when we measure our loudspeakers?

Is a good directivity in a loudspeaker not good enough for a really good perceived sound, if the transient response is bad ?

Discuss…

The Transient response is covered already by the FR and does not need to be measured separately (see Fourier).

This has been discussed a gazillion times and you find many threads on this already here.

 

[sergeauckland]

Exactly. If the amplifier can output 20kHz at full power, and the loudspeakers output 20kHz at sufficient loudness, then there's nothing more that needs to be done. A 'Transient' can't be faster than that, as the rise time of a transient on a recording is band-limited just as everything else is. In real-life, if a transient is faster than that, it won't be captured on the recording and/or won't be audible.

S.

 

[amirm]

Bandwidth of audible band is so low that you can't remotely get anything close to what people think is a "transient." Digital music at 44.1 kHz will produce just a pure sine wave if you give it a 12 kHz step signal! Indeed, the jitter tests I run are squarewaves but due to the limited bandwidth, it winds up looking like a sine wave at one frequency.

People mistakenly throw very high bandwidth impulses at audio gear and go and worry about the output. Since per above such signals are "illegal," the results of such tests is not meaningful. Sure, having extra bandwidth is good but not so much as to resolve very high bandwidth transients.

 

[Josq]

Is damping a factor? On a piano you can use a damper pedal to stop the oscillations of the strings. Are some speakers better than others in stopping the oscillations of the drivers?

 

[Tangband]

The Transient response is covered already by the FR and does not need to be measured separately (see Fourier).

This has been discussed a gazillion times and you find many threads on this already here.

Bandwidth of audible band is so low that you can't remotely get anything close to what people think is a "transient." Digital music at 44.1 kHz will produce just a pure sine wave if you give it a 12 kHz step signal! Indeed, the jitter tests I run are squarewaves but due to the limited bandwidth, it winds up looking like a sine wave at one frequency.

People mistakenly throw very high bandwidth impulses at audio gear and go and worry about the output. Since per above such signals are "illegal," the results of such tests is not meaningful. Sure, having extra bandwidth is good but not so much as to resolve very high bandwidth transients.

What about speakers then ?

 

[Waxx]

It's visible in the step response of a system, that is like said before an derivate of the FR by the system of a Fourier Transform. It can be measured seperate, but the result will be the same when deducted from that FR by the fourier transform method. And transient response depends on a few factors, the damping and MMS/Bl balance of the drivers used, the alignment and the amps used are a few of them.

 

[HarmonicTHD]

Is damping a factor? On a piano you can use a damper pedal to stop the oscillations of the strings. Are some speakers better than others in stopping the oscillations of the drivers?

You would see that in the distortion measurements which Amir does with every review (usually at 86 and 96dB SPL). Usually it is quite benign (for reasonably well engineered speakers) for higher frequencies and for lower frequencies mainly woofer size and SPL are the limiting factors. Either way, no secret here.

 

[FrankW]

This is an interesting article :

Yet another article with zero references to blind listening (audio) tests? Rather uninteresting IMO

 

[Tangband]

Here is what happens if a harpsichord music piece is played backwards - the caracter of the instrument is clearly defined by its transient envelope in the beginning of each note, but playing it backwards completely ruins the caracteristic of the instrument . ( listen to the whole short piece, first it is correct, then played backwards )

Timbre and envelope: From Physclips

Timbre depends strongly on the time envelope of amplitude and of spectrum. Transients (attack, release, decay, etc) are very important in determining timbre

www.animations.physics.unsw.edu.au

 

[Trouble Maker]

Backwards played harpsichord! What are you trying to do, invoke the devil?

 

[theREALdotnet]

With single driver speakers the transient response is completely described by the frequency response (magnitude and phase). With multiple drivers you also need to make them coherent, so as to simulate a single source. When taking a frequency response measurement of a speaker, only one driver it typically playing at any point in time (except for the crossover regions), however, with most acoustic or musical transients, all drivers play together.

 

[tomtoo]

3) For me the most grey area in Hifi.

 

[Tangband]

Good transient response in microphones are very important, this is common knowledge for music engineers. Is a loudspeaker a reverse of a microphone ?

 

[TheZebraKilledDarwin]

IMO transient response is not important for consumers. It takes well trained ears, btw, to be able to judge transient quality (which is not about hearing a plucky ADSR-curve, but the sound character of the attack itself). It is important in the production and mixing process, especially with ITB mixing: a good transient response of the speakers is not only necessary to be able to judge the right amount of compression and the attack time, but also the necessary amount of artificial saturation and other transient softening processes, to make the mix sound like a record.

IMO for the listening pleasure of consumers, the biggest factor is room acoustics (time domain) and a well balanced overall frequency response and within especially a well balanced midrange.

 

[Cbdb2]

Good transient response in microphones are very important, this is common knowledge for music engineers. Is a loudspeaker a reverse of a microphone ?

Wrong, most mics are not flat and roll off the HF. This means an inaccurate transient in many cases.
If you want to test for transient response use an impulse.

 

[PatentLawyer]

If you can reproduce 20kHz, would imagine the transient response is good enough.

 

[Tangband]

IMO transient response is not important for consumers. It takes well trained ears, btw, to be able to judge transient quality (which is not about hearing a plucky ADSR-curve, but the sound character of the attack itself). It is important in the production and mixing process, especially with ITB mixing: a good transient response of the speakers is not only necessary to be able to judge the right amount of compression and the attack time, but also the necessary amount of artificial saturation and other transient softening processes, to make the mix sound like a record.

IMO for the listening pleasure of consumers, the biggest factor is room acoustics (time domain) and a well balanced overall frequency response and within especially a well balanced midrange.

Why are a good transient response important with a microphone but not in a home-loudspeaker ?

 

[TheZebraKilledDarwin]

Why are a good transient response important with a microphone but not in a home-loudspeaker ?

I am not sure, the claim, that a good transient response is important for a mic, is true. Measurement mics are not used for recordings and many well known and highly regarded mics do quite the opposite: they soften the transients and make the sound more gentle and pleasing.

But it is important for studio monitors, because you need to hear, if the the sound is really too bright, or if the attacks are too hard (additionally in the studio, where the transient response of speakers really matters, also the acoustics are on a level, that make it possible to hear it).

Consumers at home, listen in a much less acoustically controlled environment, with much higher levels of indirect sound, usually don't have the trained ear to hear transient quality even in optimum surroundings, and want a pleasing, gentle but punchy sound, but not the most accurate one, where for example, too hard consonants or guitar plectrum starts to distract or annoy.

 

[Tangband]

I am not sure, the claim, that a good transient response is important for a mic, is true. Measurement mics are not used for recordings and many well known and highly regarded mics do quite the opposite: they soften the transients and make the sound more gentle and pleasing.

But it is important for studio monitors, because you need to hear, if the the sound is really too bright, or if the attacks are too hard (additionally in the studio, where the transient response of speakers really matters, also the acoustics are on a level, that make it possible to hear it).

Consumers at home, listen in a much less acoustically controlled environment, with much higher levels of indirect sound, usually don't have the trained ear to hear transient quality even in optimum surroundings, and want a pleasing, gentle but punchy sound, but not the most accurate one, where for example, too hard consonants or guitar plectrum starts to distract or annoy.

Thanks for the response, though I dont agree. I sometimes do recordings for myself and others and good microphones are essential when recording real instruments in a concert hall. I also prefer a studiomonitor sound in my livingroom.