I was answering your question about "shouting", sorry for the misunderstanding.
I did not know that, i thought it was calculated based on many measurements mostly done off axis
A quote from Alan Shaw of Harbeth;
"A perfectly flat speaker response is a very poor design choice. It will most likely sound unnatural and far too bright."
Yet earlier he posted a plot of one of his speakers, his equivalent of the 3/5a, and it was incredibly flat, +/- 2dB from bass to top.
"A perfectly flat speaker response is a very poor design choice. It will most likely sound unnatural and far too bright."
Yet earlier he posted a plot of one of his speakers, his equivalent of the 3/5a, and it was incredibly flat, +/- 2dB from bass to top.
It depends how he defines flat i think, and how the speaker is intendent to be used. i can imagine that a speaker intended for nearfield use can have a flat on axis response, but used in the farfield and maybe mounted close to the Wall behind it and the same speaker may sound a bit dull, if the speaker has a downward slope freq response on axis it will sound even more dull, it all depends how speakers and room summs together in the listening spot, some eq might be helpful here
I can only imagine speakers with very controlled dispersion and used more or less in the open field, like pa speakers
Just look at dome tweeters, those without waveguides, see how different they perform on and off axis and you will hear these differencies, so you need to balance the frequency response on and off axis to make the speaker sound coherent
Just look at dome tweeters, those without waveguides, see how different they perform on and off axis and you will hear these differencies, so you need to balance the frequency response on and off axis to make the speaker sound coherent
Here is a good set of measurements of a harbeth loudspeaker
Harbeth Monitor 30 Speaker Review | Audio Science Review (ASR) Forum
Harbeth Monitor 30 Speaker Review | Audio Science Review (ASR) Forum
It is description of a flat power response, I presume. On the other hand, flat anechoic response is a good thing.
Last edited:
I'm not wishing to be dispiriting or cynical, but it seems that we as a race, have reached a definite limit with speaker design, slight improvements only being made over the years, but which excites us because of the relative stagnancy.
We also seem to be facing philosophical questions which raise the issue that we may well be at a stage of limits in progress as a result.
Suppose we had ten pairs of speakers which were perfect from a FR and distortion POV, but with different driver configurations on each. I'm sure that they would all sound very different, and that may be so even only on a presentation basis; we are still stuck with our drivers on boxes.
There is obviously a massive amount of work being done by guys on this site, with high quality craftsmanship evident on several threads. But this appeals to the eyes, and we should be judging with our ears.
The vast number of iterations and permutations surely must be good at revealing some science about design, and I can't help but wonder if all of the members got together and worked on design, if could they develop a world beater.
We also seem to be facing philosophical questions which raise the issue that we may well be at a stage of limits in progress as a result.
Suppose we had ten pairs of speakers which were perfect from a FR and distortion POV, but with different driver configurations on each. I'm sure that they would all sound very different, and that may be so even only on a presentation basis; we are still stuck with our drivers on boxes.
There is obviously a massive amount of work being done by guys on this site, with high quality craftsmanship evident on several threads. But this appeals to the eyes, and we should be judging with our ears.
The vast number of iterations and permutations surely must be good at revealing some science about design, and I can't help but wonder if all of the members got together and worked on design, if could they develop a world beater.
A fine conclusion if you are advocating one pair of speakers for each stereo recording, for each stereo listening environment or for each listener's preference - and if you don't mind central images being deliberately linearly distorted.
For the same reasons as above, you will end up with many different preferred target responses depending on the source material and the listener's expectation of what it should sound like. The targetting of a uniform magnitude, minimum-phase response with a well-controlled, monotonically falling power response remains the sensible goal in loudspeaker design.
Stereo reproduction does have its limitations, but "fixing it" via questionable loudspeaker design targets is not the best way to produce improvements in stereo reproduction. Where it is possible, fixing problems at their source is always the best approach. Your cynicism is not out of place.
It does in places where it is appropriate, but not in places where it is not. There appear to be no new truths to be uncovered in this particular matter.
that is very true, we seems to focus on things that looks good, like wide baffle or narrow baffel, hitech materials or not, flat or not flat freq responses, low dist or little more 2nd harmonics, instead of designing a speaker from how our hearing works and how we and the speakers interact with the listening room and the recordings. this topic seems so complex that we rather consentrate on easier things as looks
The scientific side of the hearing vs. speaker design is pretty well covered in the Floyd Toole book 
And the more one spends time in the forums and trying things out, the more sense posts from well established characters like Earl Geddes make. But, in the end it is all about having fun and I can imagine hands-on woodworking is more fun to some folks than maths or thinking the effect of a room or Fletcher-Munson curves or whatnot...
And the more one spends time in the forums and trying things out, the more sense posts from well established characters like Earl Geddes make. But, in the end it is all about having fun and I can imagine hands-on woodworking is more fun to some folks than maths or thinking the effect of a room or Fletcher-Munson curves or whatnot...
Last edited:
That has some truth and I refer you back to my earlier post in this thread that discussed how we might need to consider the ear and brain as (at least) a bispectral analyser if we want to better interpret the audible significance of the measurements we make. Until then we are stuck with second-order analyses that, with good reason and long-known, repeatable findings, lead us to target a flat magnitude loudspeaker response, for example.
This thread concerned a departure from that target that has been shown to have a real and measurable audible effect. That is to say, its subjective impression can be predicted from an objective measure. It is not, however, the optimum way of compensating for the defect in stereo reproduction it targets and imparts further compromise in most cases. I would suggest that it is the offering of some individual's subjectively-preferred target response different from the ideal that is the source of most confusion.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.