Its not about what your measurements can do for you but what you can do for your measurements.
I think that any issue of subjectivity calls must start being debated only after all the objective data is there and well sorted out. Assuming completely competent engineering by two different scientists, their speakers are going to have a different sonic fingerprint even if designing under the same general application brief. The secondary objective judgment calls are going to be in different order between them and some chosen material is still going to have some own color be it the cone, box, or capacitor film, etc. Even if all those factors are under objective significance threshold or ABX hidden. Its just annoying that because there is some gray zone of subtle phenomena beyond some established & comprehensive measurement agenda, some non engineering people diss measurements all together as invaluable, mainly because they don't know how to perform some.
I think that any issue of subjectivity calls must start being debated only after all the objective data is there and well sorted out. Assuming completely competent engineering by two different scientists, their speakers are going to have a different sonic fingerprint even if designing under the same general application brief. The secondary objective judgment calls are going to be in different order between them and some chosen material is still going to have some own color be it the cone, box, or capacitor film, etc. Even if all those factors are under objective significance threshold or ABX hidden. Its just annoying that because there is some gray zone of subtle phenomena beyond some established & comprehensive measurement agenda, some non engineering people diss measurements all together as invaluable, mainly because they don't know how to perform some.
And you have to know when to stop correcting things. We don't want to swash the life out of our speaker.
High amplitude but narrow band aberrations of one driver can go un-noticed, but a correction that lowers this amplitude but broadens the bandwidth affected can be more easily heard as a coloration.
High amplitude but narrow band aberrations of one driver can go un-noticed, but a correction that lowers this amplitude but broadens the bandwidth affected can be more easily heard as a coloration.
Re: Re: Re: Testing vs listening
The fact is that this is probably true, but oh so inefficient. Tuning by listening is so prone to going in circles because of the human tendancy to prefering change - no matter what it is good or bad. So while you could design a loudspeaker without measurements its a kin to the monkey typing a coherent sentence - it would happen sooner or later, but I don't have that kind of time to waste. Listening to great speakers is my goal, not the journey getting there.
Measurements help me to decide when I am making real progress and when I have reached a point of diminshing returns. This is very important to getting an optimum product in a short time.
Now if you actually enjoy the "tweaking" then measurements are kind of a hinderence - just go ahead and tweak away.
a.wayne said:
The fact is that this is probably true, but oh so inefficient. Tuning by listening is so prone to going in circles because of the human tendancy to prefering change - no matter what it is good or bad. So while you could design a loudspeaker without measurements its a kin to the monkey typing a coherent sentence - it would happen sooner or later, but I don't have that kind of time to waste. Listening to great speakers is my goal, not the journey getting there.
Measurements help me to decide when I am making real progress and when I have reached a point of diminshing returns. This is very important to getting an optimum product in a short time.
Now if you actually enjoy the "tweaking" then measurements are kind of a hinderence - just go ahead and tweak away.
Re: Re: Re: Re: Testing vs listening
Thats so very true...
but I have also learned that once an improvement is found and though still not perfect, there is usually no turning back...one only have to find the next improvement to make the former valid, and so it goes on and on...its really very very hard
If I get stressed when listening I now something needs to be changed
but inbetween the frustrations its also great fun...isnt that just how life is
I havent tweaked my speakers in a long time, so I reckon they are now finished and as good as it gets, at least with used design/drivers
took me more than 5 years, and many times I was about to give up, but its just not me to quit
now they are sold to a dear friend, and crazy as I am I now look forward to a new and wilder challenge and another 5 years of joyfull tweaking
So yes, music first...but its also great to have fun along the way
gedlee said:
Thats so very true...
but I have also learned that once an improvement is found and though still not perfect, there is usually no turning back...one only have to find the next improvement to make the former valid, and so it goes on and on...its really very very hard
If I get stressed when listening I now something needs to be changed
but inbetween the frustrations its also great fun...isnt that just how life is
I havent tweaked my speakers in a long time, so I reckon they are now finished and as good as it gets, at least with used design/drivers
took me more than 5 years, and many times I was about to give up, but its just not me to quit
now they are sold to a dear friend, and crazy as I am I now look forward to a new and wilder challenge and another 5 years of joyfull tweaking
So yes, music first...but its also great to have fun along the way
Gedlee,
I guess you misread my response, I do favor measuring, would not build /design a system without doing so.
.
Everything, shows up in measuring, good or bad, if you have the technical equipment and know how to deal with it. I also agree that listening evaluation, in conjunction with testing, Is very important . Like everything else , It does requires skill , good sonic memory and repeatability of such. Not everyone can do it.....
Regards.
I guess you misread my response, I do favor measuring, would not build /design a system without doing so.
.
Everything, shows up in measuring, good or bad, if you have the technical equipment and know how to deal with it. I also agree that listening evaluation, in conjunction with testing, Is very important . Like everything else , It does requires skill , good sonic memory and repeatability of such. Not everyone can do it.....
Regards.
I don't think that I misunderstood you, but I think that others might. My point was based on your comment, but not intended as a rebuttle.
The thing is that today measurements are within anyones grasp. There really is no reason NOT to do them. But if we here can help to focus attention on what constitutes a good sound with those measurements then we will be performing an honorable service.
I read D'Appolito's article in Audio X Press and I had to take exception to much of it, so I think that there is a lot of different opinions out there. I want to write in a letter to the Mag, but those take so much time because you have to be far more carefull about what you say and how you say it. Sufice it to say, that Joe and I would agree completely on the kinds of measurements and how they should be done, but we don't agree on how to interprete them as far as audibility goes.
The thing is that today measurements are within anyones grasp. There really is no reason NOT to do them. But if we here can help to focus attention on what constitutes a good sound with those measurements then we will be performing an honorable service.
I read D'Appolito's article in Audio X Press and I had to take exception to much of it, so I think that there is a lot of different opinions out there. I want to write in a letter to the Mag, but those take so much time because you have to be far more carefull about what you say and how you say it. Sufice it to say, that Joe and I would agree completely on the kinds of measurements and how they should be done, but we don't agree on how to interprete them as far as audibility goes.
Here is an odd perspective on this issue -
Would a deaf person reliably and consistently be able to design and build good sounding speaker systems?
Such a person would have to rely exclusively on measurements. But could he have a speaker that measured good, yet sounded like a box of rabid cats fighting?
Also, when you design and build speakers, don't you measure as a way of confirming the flaw that you are hearing. The perception that the Mids don't sound quite right, would be confirmed by measurements as a peak or dip in the response, or perhaps a timing or phase issue. Having measure, you would know precisely where to put your correction and how much to apply.
How long would you have had to guess by using cut and try methods, to make the same correction?
I think you measure to find out where apply correction to problems that you can hear. So, measuring and hearing are intimately tied together. True you can try and to it all by 'ear' alone, but at best, you are just guessing.
We keep measuring and correcting until we like what we hear. It's not one or the other.
Steve/bluewizard
Would a deaf person reliably and consistently be able to design and build good sounding speaker systems?
Such a person would have to rely exclusively on measurements. But could he have a speaker that measured good, yet sounded like a box of rabid cats fighting?
Also, when you design and build speakers, don't you measure as a way of confirming the flaw that you are hearing. The perception that the Mids don't sound quite right, would be confirmed by measurements as a peak or dip in the response, or perhaps a timing or phase issue. Having measure, you would know precisely where to put your correction and how much to apply.
How long would you have had to guess by using cut and try methods, to make the same correction?
I think you measure to find out where apply correction to problems that you can hear. So, measuring and hearing are intimately tied together. True you can try and to it all by 'ear' alone, but at best, you are just guessing.
We keep measuring and correcting until we like what we hear. It's not one or the other.
Steve/bluewizard
I would agree but put the emphasis the other way arround. We should listen only to confirm that we haven't missed something in the design as optimized by measurements.
And yes, I do believe that a deaf person could design a great loudspeaker without listening to it. Once taught about the things that are important they could learn to develop designs that were "optimized" and these designs would sound good to most people - perhaps even great. Measurements tell all and if they are done right they will reveal the truth.
And yes, I do believe that a deaf person could design a great loudspeaker without listening to it. Once taught about the things that are important they could learn to develop designs that were "optimized" and these designs would sound good to most people - perhaps even great. Measurements tell all and if they are done right they will reveal the truth.
How would a deaf person 'voice' a speaker? There's more to speaker desigining/building than charts & numbers can currently show and be interpreted.
How a speaker sounds will depend a lot on the materials of choice. Materials have a great influence on the subtle tonal quality of a loudspeaker that could measure the same but sound different to our most complex test instruments - our ears.
How a speaker sounds will depend a lot on the materials of choice. Materials have a great influence on the subtle tonal quality of a loudspeaker that could measure the same but sound different to our most complex test instruments - our ears.
Correct, but this is a personal preference, I don't sell the speakers that way they are sold perfectly flat - no voicing applied.
The desire to "cool" the top end comes from a lot of people using Summas. They just seem too bright when set flat. The best reason that I have heard is that the power response of the CD design at the top end is far greater than what we are used to - probably double or more of any non CD system, so when all that reflected HF energy reaches you it just sounds bright. A dB or two and its fine. Curtains are just about right.
In almost all sound installations the "house curve" falls slightly with frequency. This just seems to be a characteristic of our hearing.
But again, it would be easy to explain that to a deaf person, you don't need to hear it to understand once you see the data. Yes, a deaf person could not have come to this result by themselves, but everything else could be done with no listening required.
I know this is not a popular position, but maybe there just might be something to this science we call acoustics.
The desire to "cool" the top end comes from a lot of people using Summas. They just seem too bright when set flat. The best reason that I have heard is that the power response of the CD design at the top end is far greater than what we are used to - probably double or more of any non CD system, so when all that reflected HF energy reaches you it just sounds bright. A dB or two and its fine. Curtains are just about right.
In almost all sound installations the "house curve" falls slightly with frequency. This just seems to be a characteristic of our hearing.
But again, it would be easy to explain that to a deaf person, you don't need to hear it to understand once you see the data. Yes, a deaf person could not have come to this result by themselves, but everything else could be done with no listening required.
I know this is not a popular position, but maybe there just might be something to this science we call acoustics.
Analog has more dynamics than digital with better bandwidth.
only loses out with surface noise and convienience. Better than digital for sound.
Voicing of a speaker is very "important" again while everyone can be taught to measure and test loudspeakers , not everyone can build loudspeakers..... Audio is science and art in a strange mix. Hence we can agree or disagree with measuring technique and application, the end results " listening " can be to one man, euphoria and to another apathy .
only loses out with surface noise and convienience. Better than digital for sound.
Voicing of a speaker is very "important" again while everyone can be taught to measure and test loudspeakers , not everyone can build loudspeakers..... Audio is science and art in a strange mix. Hence we can agree or disagree with measuring technique and application, the end results " listening " can be to one man, euphoria and to another apathy .
gedlee said:
I've often wondered why CDs can't be provided with a calibration gain level value (the dBSPL of Full Scale sinewave or bandpassed noise at 1kHz, for example) that one could use to set the system to the level that was used during the mix.
Audio engineering seems all concerned about relative frequency response, but pays almost no attention to whether absolute levels are at all duplicated. It would cost all of the effort of printing one number on the disk (and the effort, for those who wanted to use it, to determine the sensitivities of their systems and to use a volume control marked in dB). We make a big deal about audio information and accuracy, but ignore actual original level and adjust our volumes "to taste"!
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.