This question and others related top it seem to keep popping up all the time. The first thought I have is that there is no answer. we are all different and all hear differently. Rooms are different and especially in modern open plan houses, as so many are now in Australia, they differ from day to day. I think I would be a fool if I built something that measured really well, as presumably most audiophile speakers do, but found it dull and lifeless but continued to live with it because it cost a lot of time/money. This is a frequent criticism of so called audiophile designs in may threads I have read over the past few years.
My thought is that if the music is of a serious enough nature as to actually convey feeling and emotion, no matter what type of music, and I become involved in the emotion and anticipation as to where it is going when listening through speakers, the speakers have done their job no matter how they measure. By this I mean music that is not just jingles/musak. Another thought is that if it does not strain or tire me to listen, providing that emotion is there, then again the speakers are doing their job. Submitted for what its worth.
jamikl
My thought is that if the music is of a serious enough nature as to actually convey feeling and emotion, no matter what type of music, and I become involved in the emotion and anticipation as to where it is going when listening through speakers, the speakers have done their job no matter how they measure. By this I mean music that is not just jingles/musak. Another thought is that if it does not strain or tire me to listen, providing that emotion is there, then again the speakers are doing their job. Submitted for what its worth.
jamikl
formalism vs. aesthetics - the definitive conclusion...
The formalism and the aesthetic are completely different but inter-dependent enterprises.
I think it is generally agreed that by following the formalism one has a better chance of acheiving the aesthetic - that is why the formalism exists - tried and tested over many years as a method of getting "acceptable" results...
However, once the basic requirements of the formalism have been met, the rest is an art form that is wholly dependent on the intersection of the intentionality of the creator and that of the person for whom the creative enterprise is undertaken.
For example, if I serve a perfect bowl of spagetti bolognese to someone who doesn't like it - it will inevitably fail to please...
I do believe that the subject deserves plenty of discussion but it needs to be realized that a concrete conclusion just ain't possible.
Sound of one hand clapping right?
Regards,
Tom
The formalism and the aesthetic are completely different but inter-dependent enterprises.
I think it is generally agreed that by following the formalism one has a better chance of acheiving the aesthetic - that is why the formalism exists - tried and tested over many years as a method of getting "acceptable" results...
However, once the basic requirements of the formalism have been met, the rest is an art form that is wholly dependent on the intersection of the intentionality of the creator and that of the person for whom the creative enterprise is undertaken.
For example, if I serve a perfect bowl of spagetti bolognese to someone who doesn't like it - it will inevitably fail to please...
I do believe that the subject deserves plenty of discussion but it needs to be realized that a concrete conclusion just ain't possible.
Sound of one hand clapping right?
Regards,
Tom
madass said:
There are lots of things we don't see measurements for... for instance a speakers ability to do downward dynamic range.
The most common tool that gets used for measuring speakers is FR. It is only one dimension of a many dimensional space... i have heard far to many speakers where flat FR was the goal, and they ended up with the life strangled out of them.
dave
phase_accurate said:Or a coherent multiway !
I have a JAES paper by the late P.W. Klipsch who used this term (also ?).............
A coherent multiway would be 'way cool'. I am the only guy in our HiFi group that is trying to make his system sound like his AM car radio!
Briggs may have repeated that line.
On Klipsch (nice guy) I want his dynamics AND the coherence of a one way. I am still searching.
If you are measuring things that you can't hear or hearing things that you can't measure, then something is wrong. When the two things become one and the same then you are on the right track. Its not a matter of which one is better or worse - they are both absolute requirements, but it is essential that they agree!! Otherwise you are lost in the woods.
The main problem is habit taste of a listener!
All of as have listened bad speakers, and If you not try to listen other speakers you gonna have psycho habit keen on that sound or you'll get isolated.And when you hear different sound you gonna hate , or perhaps you'll like that new sound.
So all new made speakers should be compared with some reference speakers with same equipment same room and at the same time.
Measure?- Yes that is the best way to tune some speaker to sound neutral, of course.I mean flat speakers, that is the rule no. 1 when designing a new speaker, and listening second thing.
Why flat?
Because the most of the monitors in recording studios are very flat, and that is the start and reference point.
If you like painted sound than use a tone controls, I finished with them long time ago.
Of course that freq. resp. is not one and only thing, but there are many other important things like THD second and third harm. waterfall etc., but all they can be measured extremely precise.
And what you gonna do whit speaker which have pretty flat freq. resp, and you don't like the sound, probably some eq curve on freq. resp, and that is wrong.
Listening is just to confirm perfect result from the measurements.
The listening test should be done not by one person, but at least 5 good experienced "golden ear " listeners.
All of as have listened bad speakers, and If you not try to listen other speakers you gonna have psycho habit keen on that sound or you'll get isolated.And when you hear different sound you gonna hate , or perhaps you'll like that new sound.
So all new made speakers should be compared with some reference speakers with same equipment same room and at the same time.
Measure?- Yes that is the best way to tune some speaker to sound neutral, of course.I mean flat speakers, that is the rule no. 1 when designing a new speaker, and listening second thing.
Why flat?
Because the most of the monitors in recording studios are very flat, and that is the start and reference point.
If you like painted sound than use a tone controls, I finished with them long time ago.
Of course that freq. resp. is not one and only thing, but there are many other important things like THD second and third harm. waterfall etc., but all they can be measured extremely precise.
And what you gonna do whit speaker which have pretty flat freq. resp, and you don't like the sound, probably some eq curve on freq. resp, and that is wrong.
Listening is just to confirm perfect result from the measurements.
The listening test should be done not by one person, but at least 5 good experienced "golden ear " listeners.
planet10 said:
This is interesting. A lot of expensive speakers sound awful (dead) at low volume. As you turn them up they start to sound good. I believe this is to do with suspension stiffness and or damping which is common in speakers that try to get rid of all resonances. On the other hand some cheap speakers and most professional compression drivers sound alive even at low levels.
In a shop you listen at fairly high volume where a speaker may sound good. But once home in a quieter enviroment and for late night listening they can often disappoint.
Anyone else noticed this?
Re: listening levels
It seems self evident that absolute volume matters(in speakers), else why would designers struggle to minimize excursion throught use of multiple drivers. Keep the devive in it's sweet spot!
Electronics I'm not sure about(within reason). "Micro-detail: is the term I've heard before and I believe the Fletcher-Munson curve may have more to do with preceived "detail" than actual fidelity...
Keep the device in it's sweet spot - I mean your ears.
Re: Late night fidelity
I've heard it said the power lines are cleaner - maybe. And I had first hand experience with power in my previous dwelling. Low supply voltage caused all kinds of "anomolies".
But late night is definitely better because of less ambient noise. (I never notice the trains in the distance during the day...). And by listening at lower levels (you ready for this?), I believe the room acts in a more benign way because total nearfield reflections are minimized(read absorbed) proportional to direct signal...
It seems self evident that absolute volume matters(in speakers), else why would designers struggle to minimize excursion throught use of multiple drivers. Keep the devive in it's sweet spot!
Electronics I'm not sure about(within reason). "Micro-detail: is the term I've heard before and I believe the Fletcher-Munson curve may have more to do with preceived "detail" than actual fidelity...
Keep the device in it's sweet spot - I mean your ears.
Re: Late night fidelity
I've heard it said the power lines are cleaner - maybe. And I had first hand experience with power in my previous dwelling. Low supply voltage caused all kinds of "anomolies".
But late night is definitely better because of less ambient noise. (I never notice the trains in the distance during the day...). And by listening at lower levels (you ready for this?), I believe the room acts in a more benign way because total nearfield reflections are minimized(read absorbed) proportional to direct signal...
fredex said:
I have definately notice this. But I tend to think that there is a great deal of a psychological aspects to it. I find, even with great speakers, that every song has its "sweet spot" of playback level where it just seems to come alive. But this is different for every song and sometimes even the listening session. Thats what makes me think that there is some psychological aspects involved. I think that there may be a level at which the "mix" is coret because that is the level that the original was mixed at.
That said there are low level effects in loudspeakers. I once measured a loudspeaker where it resonance was nearlt twice as large at low signal levels than it was at higher signal levels. Clearly a crossover effect in the suspension. But not much work has been done at very low levels in loudspeakers, but I do test power amps by droping the signal level and noting the distortion as the level goes down. Its the low level stuff in an amp that is important - perhaps this is so in the loudspeaker too.
the theme of this thread has presented itself in different forms at DIYaudio. Here are my thoughts:
1. if the measurement is not capturing the differences you hear, then then it is likely you are using the incorrect measurement.
2. it is the responsibility of those proposing that they cannot measure something they hear to at least describe it very well. Better yet, come up with an appropriate measurement.
3. any individual analysis technique (ex frequency response) is not going be informative about all aspects of sound reproduction. frequency response is a higher level analysis designed to tell you something specific. It must be considered in a broader context.
4.For those who cannot come up with a measurement that captures your hearing impressions, i strongly encourage you to look at the original data , not higher level analysis (FR THD etc), first. For example, it has been proposed in this thread that one thing measurements do not capture is low level detail that occurs at the same time as a higher volume passage. What i would do is record the output of the speaker at the same time as recording the input and compare the two waveforms. This will allow you to see that: 1. the speaker is receiving the low level detail in the first place and 2. that it is or is not accurately reproducing it. 3. Also, make sure that you include a passage with nothing but low level detail to make sure that it is the high level background that is the cause of the filtering (if some is found).
1. if the measurement is not capturing the differences you hear, then then it is likely you are using the incorrect measurement.
2. it is the responsibility of those proposing that they cannot measure something they hear to at least describe it very well. Better yet, come up with an appropriate measurement.
3. any individual analysis technique (ex frequency response) is not going be informative about all aspects of sound reproduction. frequency response is a higher level analysis designed to tell you something specific. It must be considered in a broader context.
4.For those who cannot come up with a measurement that captures your hearing impressions, i strongly encourage you to look at the original data , not higher level analysis (FR THD etc), first. For example, it has been proposed in this thread that one thing measurements do not capture is low level detail that occurs at the same time as a higher volume passage. What i would do is record the output of the speaker at the same time as recording the input and compare the two waveforms. This will allow you to see that: 1. the speaker is receiving the low level detail in the first place and 2. that it is or is not accurately reproducing it. 3. Also, make sure that you include a passage with nothing but low level detail to make sure that it is the high level background that is the cause of the filtering (if some is found).
fredex said:
For a loudspeaker designer to exclude the psychological aspects of this work would be grossely negligent. Getting past these aspects is one of the hardest things that I do. Looking at your "baby" with truthful eyes and ears is very very difficult - whether that baby is a child or a loudspeaker, its the same thing.
Re: Re: Testing vs listening
One could design a good speaker if you have the necessary experience without measurements , but it will never be a great speaker without solid technical application and measurements.
Everything we hear good or bad can be seen from measuring . These measurements are a picture window to what we hear , not necessarily what we like.
A.Wayne
One could design a good speaker if you have the necessary experience without measurements , but it will never be a great speaker without solid technical application and measurements.
Everything we hear good or bad can be seen from measuring . These measurements are a picture window to what we hear , not necessarily what we like.
A.Wayne
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.