40 % of midrange distortion at 1W input would indicate serious design flaws
Well, not my intention to come to the rescue, but the MA Alpair 7 are one of the most musical, revealing and best speakers I own. They are Wild Burro Betsy, Vifa TC9FD18 and Lowther DX3 among others.
They say the other MA share a common signature, so I don't care much for the measurements, tough they have been well welcomed. They give us an idea, but the final judge is our ears.
Have you ever heard a Markaudio?
They say the other MA share a common signature, so I don't care much for the measurements, tough they have been well welcomed. They give us an idea, but the final judge is our ears.
Have you ever heard a Markaudio?
Agreed on the sound quality of Mark Audio. I have two generations of the Alpair 7 and also a pair of Quad ESL 57s. The Alpairs stand up well to the Quads.
Looking at the graph shown with the glitch at 1.5khz reminds me of a graph of a BMR driver. Perhaps the Mark Audio driver transitions modes around that point.
Looking at the graph shown with the glitch at 1.5khz reminds me of a graph of a BMR driver. Perhaps the Mark Audio driver transitions modes around that point.
Reply to post 62/63
Yes they can sound good agree but think not it's the sound from MA drivers some have eye on it's the published data and data not published.
To get progress and a day get nearly the same sound as under the production in studio maybe we need data to build on, then get studio plus consumer equipment made on that data, instead of hey this sound good without knowing why.
Rutcho website show FR and HD Mark Audio Alpair 10P-A Loudspeaker Measurements and Information verse published FR http://rutcho.com/speaker_drivers/alpair_10p_a/resources/Alpair_10P_Gen.1.pdf.
By the way for info have tried some listening with minimum phase/linear phase or combination in EQ domain making some of though ragged FR or notches (high Q steep filters) to fair neutral speakers or head phones and a signature of that is the revealing thing in sound picture get lifted.
Yes they can sound good agree but think not it's the sound from MA drivers some have eye on it's the published data and data not published.
To get progress and a day get nearly the same sound as under the production in studio maybe we need data to build on, then get studio plus consumer equipment made on that data, instead of hey this sound good without knowing why.
Rutcho website show FR and HD Mark Audio Alpair 10P-A Loudspeaker Measurements and Information verse published FR http://rutcho.com/speaker_drivers/alpair_10p_a/resources/Alpair_10P_Gen.1.pdf.
By the way for info have tried some listening with minimum phase/linear phase or combination in EQ domain making some of though ragged FR or notches (high Q steep filters) to fair neutral speakers or head phones and a signature of that is the revealing thing in sound picture get lifted.
Last edited:
BYRTT,
Equalizing a wide band driver is probably a very good idea. The frequency response of virtually all full range drivers is very ragged which is highly audible.
But make sure you don't try to equalize stuff that can't be equalized. Non-minimum phase regions for example. You also have to look at the response off axis.
Linear phase filters can introduce pre-ringing which can be audible.
Equalizing a wide band driver is probably a very good idea. The frequency response of virtually all full range drivers is very ragged which is highly audible.
But make sure you don't try to equalize stuff that can't be equalized. Non-minimum phase regions for example. You also have to look at the response off axis.
Linear phase filters can introduce pre-ringing which can be audible.
Those who have played with digital EQ for a while know this, but newcomers may not:
In general, do not try EQ out holes in the FR plot. Your brain will fill in the missing data. EQ the peaks. Use as little little EQ as you can as you will cause aberrations in the phase response, and in the 300-3000Hz range, you will hear it. A rising response above 5kHz or so is best handled with varying the toe in/out.
Bob
In general, do not try EQ out holes in the FR plot. Your brain will fill in the missing data. EQ the peaks. Use as little little EQ as you can as you will cause aberrations in the phase response, and in the 300-3000Hz range, you will hear it. A rising response above 5kHz or so is best handled with varying the toe in/out.
Bob
The brain does not "fill in the missing data". What's not there can't be heard. Get better drivers if the holes are caused by the driver itself.
All scientific studies about audibility of phase shifts that I know suggest the opposite. You would need to abuse an EQ to introduce audible phase errors.
Phase tracking between the left and right speaker is a different matter though. It's of utmost importance to get that right for good imaging.
I don't agree. It's best handled with EQ as most wide band drivers have a frequency response that changes dramatically fast off axis. See Rutcho's measurements.
About an octave above the impedance peak that marks the transition from pistonic to resonant behaviour.
dave
I knew you would fall for this trap. It's called "phycoacoustics". Google it
sometime. You might actually come to understand something about speaker design.
Bob
You seem to be confused what the term "psychoacoustics" means. Did you ever read Blauert "Spatial hearing"?
From wikipedia:
"Psychoacoustics is the scientific study of sound perception. More specifically, it is the branch of science studying the psychological and physiological responses associated with sound (including speech and music). It can be further categorized as a branch of psychophysics."
So how does this relate to what you've said? A driver that has "holes" in its frequency response is just not a good driver just like a display with dead pixels isn't a good display. You won't notice it with certain pictures but with others you will. Get a display without dead pixels is the cure. It's not like all displays would have dead pixels.
Last edited:
A driver with narrow holes (dips) is passable as sounding good but one with narrow peaks is noticeable. Well trained ears (like mastering technicians) can detect narrow dips but most people won't. It has to do with how hearing is similar to freq response graph smoothed by 1/6th octave. Psychoacoustics research indicated that most people can't hear any more frequency content than 32 frequency bands (similar to a 33-band graphic EQ used in pro mixing). This is proven millions of times a day when people listen to a song from mp3. MP3 uses 32 discrete frequency banks corresponding to 32 FFT coefs. That is how the compression works: take FFT of signal and store only the uniquely changing first 32 FFT coefs. Playback then uses those coefs and an inverse FFT is performed. So if you have content that was in between one of those 32 bands, it is probably missing but most people's brains don't care
Where you will hear the dip is when it lands in the middle of a sliding scale like a trombone or slide whistle.
Where you will hear the dip is when it lands in the middle of a sliding scale like a trombone or slide whistle.
Last edited:
A bit OT but since the last few comments were related to rutcho's findings with the MA 10Ps graphs, may I know which MA drivers have you owned & pls share your experience.
a3cervo,
Yes that's off topic and you should take this to another thread or PM. I do have a couple of wide band drivers and boxes including Markaudio. I also have multi-way speakers.
Wide band speakers can be good value within a single seat solution if equalized correctly and paired with a subwoofer (most wide band drivers heavily distort at low frequencies). Stay away from expensive full range drivers as they don't represent good value. Go for multi-way instead.
In any case measuring equipment is required because the speaker is just one part of the equation, the other part is room acoustics.
Yes that's off topic and you should take this to another thread or PM. I do have a couple of wide band drivers and boxes including Markaudio. I also have multi-way speakers.
Wide band speakers can be good value within a single seat solution if equalized correctly and paired with a subwoofer (most wide band drivers heavily distort at low frequencies). Stay away from expensive full range drivers as they don't represent good value. Go for multi-way instead.
In any case measuring equipment is required because the speaker is just one part of the equation, the other part is room acoustics.
Yes, my listening experience shows, that the dips with width of 1/6 octave or less is hard to hear. Something more - if the dip can be observed in a narrow dissipation angle, in a normal living room is not audible at all, even to trained ear.In the case with Alpair 10P-A this is the dip at 6KHz. The another one - at 14KHz, is on the edge of the hearing range, when musical instruments don't have basic tones, but only overtones of their spectrum. Of course, the sound image will be better without this hole, but it does not affect the overall tonal balance too much.
I continue with 4" fullrange measurements. A friend ordered from Madisound the Fostex P1000E DIY Kanspea kit with both available drivers. Here they are:
Fostex FE-103 En
Fostex FF 105 WK
The problem with those high frequency peaks and dips seen in lots of wide band drivers is that there are so many. This will lead to reduced loudness and spectral coloration. A countermeasure would be to apply a broad boost but then the peaks will become audible again. It's a catch-22. All in all drivers with smoother response are better.
By the way, do you plan on doing more off-axis measurements? CharlieLaub has a fantastic tool for plotting these.
http://www.diyaudio.com/forums/soft...ns-wanted-directivity-plot-2.html#post4111473
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.