Audioman: Subwoofer designers have been putting accelerometers on woofer cones since Arnie Nudell started Infinity (the ServoStatik One) They are part of a feedback loop that makes sure the motion of the woofer follows the (bass part of the) incoming music signal. This circuit does NOT ALTER the amplitude/frequency distribution of the signal IN ANY WAY; and the passive (later active) crossovers all operated in the analog domain; and they DIDN'T MESS with the signal either, except in the case of 2-way, 3-way, etc. speakers where the crossover (A) sent different portions of the signal to the appropriate driver(s) and (B) determined the best way to recombine the divided parts of the signal back together -- sometimes described/referred to as adjusting the hi and lo pass slopes at the crossover point(s).
The DSP (Digital. Signal. Processor.) used in the Ethos is basically a parametric equalizer (remember those?) that RESHAPES the music signal (AFTER FIRST CONVERTING IT INTO DIGITAL DATA to avoid adding noise during processing) -- and THEN converting it BACK into the analog domain to drive the sub amp+woofer,
MartinLogan is using digital hardware and software to "rework" the original music signal in order to compensate for their bass speaker's non-linear response curve. This wouldn't concern me that much (other than I consider it a cop-out from good speaker design) but the part that REALLY bothers me is that once your music signal (or a portion(s) of it) are converted from analog to digital, then THAT'S IT! You can never get it back in its original form, and you are now twice removed from the original music signal coming from your source -- be it a true analog original or a digital-to-analog 'original', doesn't matter. You've sent it on an 'electronic detour' first!
I'm not debating here whether audio systems have to have their response(s) ADJUSTED to compensate for non-linear room or speaker+room response(s) -- that's a different matter, where the object is to achieve a frequency response in (the air of) the room that closely approximates what is on/in the source (and yes, I know; the engineer has already messed with it too -- but that doesn't count, because whatever the engineer did is part of the 'original source')
When the computer world coined the term "GIGO" (garbage in, garbage out) their assumption was that the hardware itself (the computer) didn't alter the data on its way through the machine (be it garbage or not.) But the minute you have to CONVERT the (bass portion of the) music signal into the digital domain at the speaker input, THEN reconfigure it in the digital domain (i.e. alter its original amplitude and frequency profile,) and THEN convert it BACK into the analog domain (D-to-A) in order to send (the now-processed music signal) to the subwoofer amp+driver JUST TO COMPENSATE for a loudspeaker that that is INCAPABLE of providing a reasonably flat response to the original music signal, it seems to me like an engineering COP--OUT and on top of all that, you NEVER get to hear your speaker's response to the original source material as the recording engineer intended (for better or for worse ;--)
Granted, there are instances where it is necessary to alter the signal DURING PRODUCTION (but without converting it into the digital domain,) and according to a predetermined set of rules which allow it to easily be restored to its original profile at playback: two of these are the RIAA curve for making analog records, and the Dolby Noise Reduction System -- originally meant for analog mastertape production, but now digital in most applications; primarily HT digital soundtracks.
To sum up; MartinLogan is NOT using DSP to "accurately control the EXCURSION of the bass driver" -- that is done, and has been done for years, using analog circuits, with a simple accelerometer and electronic feedback loop -- and is very likely part of the Ethos' bass section -- as it is on all their hybrid loudspeakers and subwoofers (Depth and Descent) with self-powered woofer sections. The purpose of the DSP, as used in the Ethos' bass section, is to alter the actual music signal, in order to compensate for a bass driver with a highly NON-linear frequency response curve. So, you might ask, why didn't they just use a woofer (+enclosure) with a nice LINEAR response and call it a day? The answer is simple: because they couldn't! Not with (especially!) the limitations of enclosure size, plus asking the "bass" driver to handle more and more of the lower mid-range -- thus allowing for a smaller, more frequency-limited electrostatic screen. FYI, ML has been raising the crossover point between the woofer and the panel ever since they came out with the Summit (go back and check the specs of the various models if you never have ;--) it's a really poor solution, and if you take it to the extreme, you wind up with a conventional box speaker with a little electostatic tweeter (Jentzen did that in the late fifties!) So now with the Ethos, they've hit the limits of THAT solution, because they couldn't come up with a (cone) driver that could do all the different things they needed it to do (all by itself, that is) so they created a sort of "virtual subwoofer" by combining a small(ish) cone driver with an electronic signal processor. and voila! The fact is that MartinLogan has *NEVER produced a FULL-RANGE, pure electrostatic loudspeaker. Why? Because except for the following three models, they have never produced a FULL-RANGE electrostatic DRIVER (panel.)
Those three models are:
-- the original Statement
-- the CLS (which I've owned for 20 years and just re-paneled ;--)
-- the new CLX
And although all three still require low frequency support below 30 -- 40 Hz, that's still low enough to officially describe them as "full range".
The only current production, FULL RANGE (down to 20Hz), all-elecrostatic loudspeakers, are the TOTL Soundlab models, and they are truly magnificent . . . . IF you have the money, and room to set them up properly. And they definitely don't have to screw with your music before you can listen to it!
.
The DSP (Digital. Signal. Processor.) used in the Ethos is basically a parametric equalizer (remember those?) that RESHAPES the music signal (AFTER FIRST CONVERTING IT INTO DIGITAL DATA to avoid adding noise during processing) -- and THEN converting it BACK into the analog domain to drive the sub amp+woofer,
MartinLogan is using digital hardware and software to "rework" the original music signal in order to compensate for their bass speaker's non-linear response curve. This wouldn't concern me that much (other than I consider it a cop-out from good speaker design) but the part that REALLY bothers me is that once your music signal (or a portion(s) of it) are converted from analog to digital, then THAT'S IT! You can never get it back in its original form, and you are now twice removed from the original music signal coming from your source -- be it a true analog original or a digital-to-analog 'original', doesn't matter. You've sent it on an 'electronic detour' first!
I'm not debating here whether audio systems have to have their response(s) ADJUSTED to compensate for non-linear room or speaker+room response(s) -- that's a different matter, where the object is to achieve a frequency response in (the air of) the room that closely approximates what is on/in the source (and yes, I know; the engineer has already messed with it too -- but that doesn't count, because whatever the engineer did is part of the 'original source')
When the computer world coined the term "GIGO" (garbage in, garbage out) their assumption was that the hardware itself (the computer) didn't alter the data on its way through the machine (be it garbage or not.) But the minute you have to CONVERT the (bass portion of the) music signal into the digital domain at the speaker input, THEN reconfigure it in the digital domain (i.e. alter its original amplitude and frequency profile,) and THEN convert it BACK into the analog domain (D-to-A) in order to send (the now-processed music signal) to the subwoofer amp+driver JUST TO COMPENSATE for a loudspeaker that that is INCAPABLE of providing a reasonably flat response to the original music signal, it seems to me like an engineering COP--OUT and on top of all that, you NEVER get to hear your speaker's response to the original source material as the recording engineer intended (for better or for worse ;--)
Granted, there are instances where it is necessary to alter the signal DURING PRODUCTION (but without converting it into the digital domain,) and according to a predetermined set of rules which allow it to easily be restored to its original profile at playback: two of these are the RIAA curve for making analog records, and the Dolby Noise Reduction System -- originally meant for analog mastertape production, but now digital in most applications; primarily HT digital soundtracks.
To sum up; MartinLogan is NOT using DSP to "accurately control the EXCURSION of the bass driver" -- that is done, and has been done for years, using analog circuits, with a simple accelerometer and electronic feedback loop -- and is very likely part of the Ethos' bass section -- as it is on all their hybrid loudspeakers and subwoofers (Depth and Descent) with self-powered woofer sections. The purpose of the DSP, as used in the Ethos' bass section, is to alter the actual music signal, in order to compensate for a bass driver with a highly NON-linear frequency response curve. So, you might ask, why didn't they just use a woofer (+enclosure) with a nice LINEAR response and call it a day? The answer is simple: because they couldn't! Not with (especially!) the limitations of enclosure size, plus asking the "bass" driver to handle more and more of the lower mid-range -- thus allowing for a smaller, more frequency-limited electrostatic screen. FYI, ML has been raising the crossover point between the woofer and the panel ever since they came out with the Summit (go back and check the specs of the various models if you never have ;--) it's a really poor solution, and if you take it to the extreme, you wind up with a conventional box speaker with a little electostatic tweeter (Jentzen did that in the late fifties!) So now with the Ethos, they've hit the limits of THAT solution, because they couldn't come up with a (cone) driver that could do all the different things they needed it to do (all by itself, that is) so they created a sort of "virtual subwoofer" by combining a small(ish) cone driver with an electronic signal processor. and voila! The fact is that MartinLogan has *NEVER produced a FULL-RANGE, pure electrostatic loudspeaker. Why? Because except for the following three models, they have never produced a FULL-RANGE electrostatic DRIVER (panel.)
Those three models are:
-- the original Statement
-- the CLS (which I've owned for 20 years and just re-paneled ;--)
-- the new CLX
And although all three still require low frequency support below 30 -- 40 Hz, that's still low enough to officially describe them as "full range".
The only current production, FULL RANGE (down to 20Hz), all-elecrostatic loudspeakers, are the TOTL Soundlab models, and they are truly magnificent . . . . IF you have the money, and room to set them up properly. And they definitely don't have to screw with your music before you can listen to it!
.