Indeed, what did Quad 57, STAX, Acoustat and earlier Soundlab ESL owners do in the 80s and 90s before DSP? Briefly, what problems does DSP solve and how does it do so?
To what degree were frequency response, distortion, and soundstage compromised without DSP?
But in the pre-DSP days, did using active crossovers, multiple subwoofers and/or room analysis and treatment correct most of the problem?
Likewise for open baffle DIYers, whether or not they used active crossovers?
Perhaps this seems naïve, but as I don’t play vinyl I’m looking at used DACs in the $4.5K range, so I’d rather not have lesser DAC in DSP equalizers in the speaker chain compromising my main DAC investment.
I can’t afford new ESLs, at least not from a kit. I’m not interested in Martin Logans. And the least costly full range Soundlab model is >$15K. My speaker budget, including active or passive crossovers, is ~ $5K. I have a pair of Rythmik 12" sealed servo subs and might be getting a second pair.
My room is 14 ft x 19 x ~ 8 ft high, which I think I recall Rob Mackinlay saying is barely big enough for the Acorns. Acorn Electrostatic Loudspeaker Kit
Therefore, consider Acoustat or Quad 57s versus modern three way open baffles with one or two 15 or 18” woofers per side, passively or actively crossed.
Which would tend to suffer less in most respects (particularly soundstage size) in the absence of DSP: These two ESLs or most three-way open baffles?
Why would this be so?
And would the absence of DSP somehow matter much less if the OBs or ESLs were crossed with my pair of powered subs, where the subs take over at between 35Hz and 45Hz?
As for OBs, Pano, a member here at diyaudio.com, while basically into horns, designed his own bi-amped Altec Coax/dual 15” Eminence OBs. http://www.hifizine.com/2010/06/lone-star-audio-fest/ But what really intrigued me was his reply from last year, showing that he’s no stranger to the sound of ESLs.
“As to the midrange, a 12" cone isn't often used for midrange, for various reasons - but I have heard it work beautifully. The example was the Eminence Beta 12 LTA on open baffle above 2 18" woofers. It was crossed to a tweeter at 6K. The crossover was very complex, because the LTA is ragged up top, but it worked. It was open, sweet and light. Sounded nothing like it looks, more clean and effortless - like an ESL.”
I don’t recall if he remembered if it was a ribbon or AMT tweeter (RAAL, Mundorf or Aurum-Cantus).
But here’s Drewan 77’s description of his OB speaker system’s ability to mimic the sound of Jtw’s ESL’s (same page), where “Voices and individual instruments just seem to hang in the air, and stay in a constant…” across the soundstage. Most intriguing! Electrostatic Vs. Open Baffle Speaker Designs | Steve Hoffman Music Forums
However, his open baffles do use DSP.
But if three-way OBs, using whichever woofer size or number and type of crossover-could deliver performance comparable to Rob Mackinlay’s Acorn kit, Quad 57s or Acoustats-without DSP-then OBs seem like the safest bet.
Two other slightly off-topic questions:
I know that H-frame OBs can go lower in the bass, but are there other performance benefits or trade-offs between using them versus a regular flat OB panel?
For a three- open baffle system, if the HF and MF drivers were passively crossed, would this two way active crossover work well between them and the woofers, plus my subs? http://www.firstwatt.com/pdf/prod_b4_man.pdf
I know there’s a lot of variables to juggle here. Needless to say, achieving the all important goal of synergy is rarely simple.
But considering all of the above would Quad 57s or Acoustats (which model, assuming I can find one?)-or would most three way OB speakers be the best bet in the absence of DSP?
Or am I asking for an impossible free lunch? Or can active crossovers and/or my pair of Rythmik 12" sealed servo subs help get me that free lunch?
To what degree were frequency response, distortion, and soundstage compromised without DSP?
But in the pre-DSP days, did using active crossovers, multiple subwoofers and/or room analysis and treatment correct most of the problem?
Likewise for open baffle DIYers, whether or not they used active crossovers?
Perhaps this seems naïve, but as I don’t play vinyl I’m looking at used DACs in the $4.5K range, so I’d rather not have lesser DAC in DSP equalizers in the speaker chain compromising my main DAC investment.
I can’t afford new ESLs, at least not from a kit. I’m not interested in Martin Logans. And the least costly full range Soundlab model is >$15K. My speaker budget, including active or passive crossovers, is ~ $5K. I have a pair of Rythmik 12" sealed servo subs and might be getting a second pair.
My room is 14 ft x 19 x ~ 8 ft high, which I think I recall Rob Mackinlay saying is barely big enough for the Acorns. Acorn Electrostatic Loudspeaker Kit
Therefore, consider Acoustat or Quad 57s versus modern three way open baffles with one or two 15 or 18” woofers per side, passively or actively crossed.
Which would tend to suffer less in most respects (particularly soundstage size) in the absence of DSP: These two ESLs or most three-way open baffles?
Why would this be so?
And would the absence of DSP somehow matter much less if the OBs or ESLs were crossed with my pair of powered subs, where the subs take over at between 35Hz and 45Hz?
As for OBs, Pano, a member here at diyaudio.com, while basically into horns, designed his own bi-amped Altec Coax/dual 15” Eminence OBs. http://www.hifizine.com/2010/06/lone-star-audio-fest/ But what really intrigued me was his reply from last year, showing that he’s no stranger to the sound of ESLs.
“As to the midrange, a 12" cone isn't often used for midrange, for various reasons - but I have heard it work beautifully. The example was the Eminence Beta 12 LTA on open baffle above 2 18" woofers. It was crossed to a tweeter at 6K. The crossover was very complex, because the LTA is ragged up top, but it worked. It was open, sweet and light. Sounded nothing like it looks, more clean and effortless - like an ESL.”
I don’t recall if he remembered if it was a ribbon or AMT tweeter (RAAL, Mundorf or Aurum-Cantus).
But here’s Drewan 77’s description of his OB speaker system’s ability to mimic the sound of Jtw’s ESL’s (same page), where “Voices and individual instruments just seem to hang in the air, and stay in a constant…” across the soundstage. Most intriguing! Electrostatic Vs. Open Baffle Speaker Designs | Steve Hoffman Music Forums
However, his open baffles do use DSP.
But if three-way OBs, using whichever woofer size or number and type of crossover-could deliver performance comparable to Rob Mackinlay’s Acorn kit, Quad 57s or Acoustats-without DSP-then OBs seem like the safest bet.
Two other slightly off-topic questions:
I know that H-frame OBs can go lower in the bass, but are there other performance benefits or trade-offs between using them versus a regular flat OB panel?
For a three- open baffle system, if the HF and MF drivers were passively crossed, would this two way active crossover work well between them and the woofers, plus my subs? http://www.firstwatt.com/pdf/prod_b4_man.pdf
I know there’s a lot of variables to juggle here. Needless to say, achieving the all important goal of synergy is rarely simple.
But considering all of the above would Quad 57s or Acoustats (which model, assuming I can find one?)-or would most three way OB speakers be the best bet in the absence of DSP?
Or am I asking for an impossible free lunch? Or can active crossovers and/or my pair of Rythmik 12" sealed servo subs help get me that free lunch?
Hi,
a DSP is no guarantee for good sound in itself.
Its a tool with great flexibility and comfort in first place and very useful for all, who can't or don't want to build a decent discrete analog filter.
The 'sonic' problem with almost all DSPs are the A to D and D to A conversions.
Even the 'big' DSPs that allow for FIR filtering (besides the IIR filtering that all cheap DSPs allow) feature rather simple ADC and DAC circuits after standard application notes.
These are certainly sonic needle holes.
After my experience the actual DSP core, the 'mathematical machining' is more ore less sonically transparent.
When we 'corrected' our ESLs with FourAudios big DSP it measured close to perfect and sounded very impressive in absolute terms.
Still though the system playing with the analog JFET Buffer crossover sounds more lifelike, authentic and true -of course only, when the room acoustic doesn't spoil too much.
jauu
Calvin
a DSP is no guarantee for good sound in itself.
Its a tool with great flexibility and comfort in first place and very useful for all, who can't or don't want to build a decent discrete analog filter.
The 'sonic' problem with almost all DSPs are the A to D and D to A conversions.
Even the 'big' DSPs that allow for FIR filtering (besides the IIR filtering that all cheap DSPs allow) feature rather simple ADC and DAC circuits after standard application notes.
These are certainly sonic needle holes.
After my experience the actual DSP core, the 'mathematical machining' is more ore less sonically transparent.
When we 'corrected' our ESLs with FourAudios big DSP it measured close to perfect and sounded very impressive in absolute terms.
Still though the system playing with the analog JFET Buffer crossover sounds more lifelike, authentic and true -of course only, when the room acoustic doesn't spoil too much.
jauu
Calvin
Just a thought... Several folks have performed double-blind careful tests comparing "audiophile" DAC with much less expensive products, and found that there were no audible differences. DACs are really not a problem these days. Consider something like the Linkwitz LX521, which will be within your budget if you don't spend all of it on a DAC...
I agree with Calvin, the DSP is irrelevant. DSP is only useful for correcting poor frequency response. If a speaker design is half decent, DSP is unnecessary.
Frequency response is arguably the most important characteristic of a speaker - both in respect of flatness and extent, and a DSP can be a very useful tool to get it right. But once the response is flat, a bunch of other technological issues become important, and at this point, good ESLs have features that cannot be matched by conventional drivers: lower distortion, no crossover, no cabinet resonances, no driver breakup, phase coherence, better controlled polar response. Not so cheap though
Once you have heard a good ESL, you wont want to go back.
Frequency response is arguably the most important characteristic of a speaker - both in respect of flatness and extent, and a DSP can be a very useful tool to get it right. But once the response is flat, a bunch of other technological issues become important, and at this point, good ESLs have features that cannot be matched by conventional drivers: lower distortion, no crossover, no cabinet resonances, no driver breakup, phase coherence, better controlled polar response. Not so cheap though
Once you have heard a good ESL, you wont want to go back.
ESLs can of course sound great without any DSP as long as they're properly amplified. However, that improvement is still available if you want to run subs, or even if you run hybrid bass higher in the range. DSP will provide a huge improvement for integration between ESL panels and a dynamice woofer. Active digital cross overs, room correction dsp, gets you real punch in the bass. I incorporated that way back in 2004. You have a lot of options available today.
DSP is one way to solve this problem, ASP can work just as well but is a much more difficult solution to execute for a multiway speaker. I have Linkwitz Orion 3.3 with an ASP and they sing beautifully thanks to SL's ASP execution. In another room are Linkwitz LXmini, also a pleasure to listen to. I expect only a handful of highly skilled individuals are capable of perfecting the ASP circuitry whereas the miniDSP programming is much more DIY accessible.
With DSP you run into the issues of the effects of digital filters. And the A/D - D/A process. It's far better than it was, but can still be a limiting factor.
The digital filters on DACs remain problematic.
So, the question is which set of compromises will you accept, and which bother your ears?
No free lunch (not yet).
The digital filters on DACs remain problematic.
So, the question is which set of compromises will you accept, and which bother your ears?
No free lunch (not yet).
dac dsp is verry nice to try stuff, but in the end without dac is most useful. when you are going to buy a 4K dac then it is kind of useless to use a DSP in my opinion. it will be **** compared to the dac itself. get a dac and proper passive filter. or dont buy an expensive dac but a dac/dsp . i use my dsp for prototyping , and when it works i just try to get the same slopes using passive components. then when you got that figured out it is easy and cheap to build another pair for instance
You're right, there's no good way to operate either ESLs or OBs without DSP. Behringer DCX2496 does it all for very few dollars.
Specifically, any sub in a room benefits from EQ and no feasible way at low freq except DSP.
You need sharp crossovers to keep stereo image in the ESLs and out of the subs.
The subs need time-alignment since they will be in best room location(s) which is unlikely to be adjacent to mains.
Low freq OBs will have bad freq stretches that need correction.
And so on. I'm sure I speak for many who use the Behringer, hard to image life without it (as OP asked).
As others have said, DSPs are clean as could be.
Ben
Specifically, any sub in a room benefits from EQ and no feasible way at low freq except DSP.
You need sharp crossovers to keep stereo image in the ESLs and out of the subs.
The subs need time-alignment since they will be in best room location(s) which is unlikely to be adjacent to mains.
Low freq OBs will have bad freq stretches that need correction.
And so on. I'm sure I speak for many who use the Behringer, hard to image life without it (as OP asked).
As others have said, DSPs are clean as could be.
Ben
Why not ML? And don't worry about offending me!
I'm running Ascent 'i' using cheapo 90's type opamp based active xo on the bass, but still keeping the stock passive xo and eq in the ESL portion. Powering the ESL is a suped-up single transistor per rail Adcom GFA-5200 which is biased with 0.8A (~20W class A into 8 Ohm), but it goes much higher in class AB.
Sorry; spoke out of ignorance. I haven't even heard MLs. It's just that I have a pair of these Rythmik Audio servo subwoofer 12" F12SE Signature Edition , so I didn't see any point to choose an ESL with a built-in woofer below it, assuming the ML mains go down to at least 70Hz, where my subs would take over.
Unless, the ESL's or OB's frequency response is poor, some here claim DSP is unnecessary. Why would they be wrong?
My Rythmiks subs don't use DSP, though designer Brian Ding is certainly not opposed to it, as shown here Rythmik Audio • Servo subwoofer products However, he does imply that if room modes (bass peaks and nulls?) are the chief problem that acoustical treatment and parametric EQ may work as well as DSP.
I'll likely be adding a second pair of Rythmik sealed subs, where I hope to place one pair within time alignment of the mains, while the other placed so that both pairs can minimize room modes, as per these observations
https://www.harman.com/sites/default/files/white-paper/12/11/2015 - 06:12/files/multsubs.pdf
Really? John Janowitz assured me that a pair of his Dipole 15s per side can go down to 30Hz at 93db.
Acoustic Elegance Dipole15 woofer for Open Baffle Applications
Are a pair each of Dipole Subs & Front-Firing Subs at Odds?
Here’s my situation: I have a 14 ft x 19 room, with one side open to another ~ 9 ft x 10 room. There’s a two-way horn/cone midwoofers system along the 14 ft wall, and a pair of Rythmik 12" sealed servo subs placed not far from them.
And along the opposite 14 ft wall is where I want to add a pair of ESLs or OBs with another pair of subs. Naturally, I’d only be playing one pair of mains at a time (though 4.4 surround sound is a nice fantasy).
The question is would a duplicate pair of Rythmik 12" sealed servo front-firing subs work fine with those dipole mains, or must I use dipole subs?
If so, what design works best for either ESLs or OBs?
One potential advantage with OBs: A pair of AE Dipole 15s per side can go down to 30Hz at 93db or so. Acoustic Elegance Dipole15 woofer for Open Baffle Applications Thus, the subs wouldn’t have a whole lot of work. Might that make it less of a priority to use dipole subs?
Or if a dipole sub is still essential for the OB mains, is there a design that would also work best with my Rythmik 12" sealed servo subs to simultaneously minimize room modes, regardless of which pair of mains I am playing?
Admittedly, this is probably asking for a very nice free lunch.
Here’s my situation: I have a 14 ft x 19 room, with one side open to another ~ 9 ft x 10 room. There’s a two-way horn/cone midwoofers system along the 14 ft wall, and a pair of Rythmik 12" sealed servo subs placed not far from them.
And along the opposite 14 ft wall is where I want to add a pair of ESLs or OBs with another pair of subs. Naturally, I’d only be playing one pair of mains at a time (though 4.4 surround sound is a nice fantasy).
The question is would a duplicate pair of Rythmik 12" sealed servo front-firing subs work fine with those dipole mains, or must I use dipole subs?
If so, what design works best for either ESLs or OBs?
One potential advantage with OBs: A pair of AE Dipole 15s per side can go down to 30Hz at 93db or so. Acoustic Elegance Dipole15 woofer for Open Baffle Applications Thus, the subs wouldn’t have a whole lot of work. Might that make it less of a priority to use dipole subs?
Or if a dipole sub is still essential for the OB mains, is there a design that would also work best with my Rythmik 12" sealed servo subs to simultaneously minimize room modes, regardless of which pair of mains I am playing?
Admittedly, this is probably asking for a very nice free lunch.
Are a pair each of Dipole Subs & Front-Firing Subs at Odds?
Studies into the use of multiple have clearly weighed in favor of doing so. https://www.harman.com/sites/default/files/white-paper/12 /11/2015%20-%2006%3A12/files/multsubs.pdf and AudioKinesis Swarm Subwoofer System | The Absolute Sound
However, Earl Geddes claims that the multiple subs should NOT be identical. The Pros and Cons of Multiple Subwoofers
Please share your thoughts on Dr. Geddes’ theories on this issue.
Indeed, might his assertions support using my pair of Rythmik 12" sealed servo subs and a pair of dipole subs?
In any case, if the room needs at least some degree of correction (as most probably do), would this be among the best ways to proceed? Fill out this form and mail it back Auralex Personalized Room Analysis Form , after which Auralex (or another firm that you might suggest) evaluates the room and suggest absorbers and diffusers to quell ring nodes and other aberrations?
Studies into the use of multiple have clearly weighed in favor of doing so. https://www.harman.com/sites/default/files/white-paper/12 /11/2015%20-%2006%3A12/files/multsubs.pdf and AudioKinesis Swarm Subwoofer System | The Absolute Sound
However, Earl Geddes claims that the multiple subs should NOT be identical. The Pros and Cons of Multiple Subwoofers
Please share your thoughts on Dr. Geddes’ theories on this issue.
Indeed, might his assertions support using my pair of Rythmik 12" sealed servo subs and a pair of dipole subs?
In any case, if the room needs at least some degree of correction (as most probably do), would this be among the best ways to proceed? Fill out this form and mail it back Auralex Personalized Room Analysis Form , after which Auralex (or another firm that you might suggest) evaluates the room and suggest absorbers and diffusers to quell ring nodes and other aberrations?
Thanks TNT; point taken on the OB format. Now to decide on ESLs or OBs. Not so easy, though maybe the OBs extended bass might help simplify matters, at least if my above assumptions about using multiple subs is correct.
you are running in circles since you lack a definition for objectives. a notion that an expensive DAC is a must is not a goal.
for example I start with:
goal 1: controlled (narrow-ish) directivity to avoid early reflections and off axis tonal balance to be maintained over the dispersion angle. (got hooked a while back when listening to it at Earl Geddes's home).
question 2: is front firing the goal (like Earl's horns) or do you want a dipole action?, (consider speaker placement in the room! and room reverb); if dipole then over what frequency range? there are several threads here where people more or less established in what freq range the dipole is truly beneficial, and where it is really optional and transitions from monopole through cardioid etc. are perfectly acceptable. this is also a judgment call on what suits the acoustics of a particular room. your speaker placement on the long wall also raises the question: is there a wall right behind your listening position? (because you definitely do not want one there or no speaker choice will do).
question 3: what kind of music is targeted? is there a preferred speaker sensitivity in regards to common listening levels? higher sensitivities will often sound better at low levels. what is the available amp power? is multi-amping an option if a lower power class A amp was to be used on a critical mid range for example?
this eventually brings you to a particular speaker design which in turn answers the question if EQ is required or not. for example, SL's approach is to combine narrow baffle with driver sizes where both lower and upper midrange are used just to get the same tonal balance over 10deg increments off axis while the flatness on axis is sacrificed, which in turn mandates the use of EQ to get that back. additionally SL wanted to use low Qts bass drivers on OB so the EQ with a boost down low was a must. only then one decides if an analog or a digital solution is better for them.
the bass is all about the room size and shape (irregular is better) and bass unit placement and timing. one in a corner can bring about excellent results just like two or three around the room can (I made a whole bunch of subs with Earl's B&C drivers and ended up using only one; maybe I would have come out better on multiple subs if I had more freedom choosing their locations).
in the end it is all about experimentation and matching a room with audio gear, and hopefully coming up with a combo which works for the kind of music played most often.
for example I start with:
goal 1: controlled (narrow-ish) directivity to avoid early reflections and off axis tonal balance to be maintained over the dispersion angle. (got hooked a while back when listening to it at Earl Geddes's home).
question 2: is front firing the goal (like Earl's horns) or do you want a dipole action?, (consider speaker placement in the room! and room reverb); if dipole then over what frequency range? there are several threads here where people more or less established in what freq range the dipole is truly beneficial, and where it is really optional and transitions from monopole through cardioid etc. are perfectly acceptable. this is also a judgment call on what suits the acoustics of a particular room. your speaker placement on the long wall also raises the question: is there a wall right behind your listening position? (because you definitely do not want one there or no speaker choice will do).
question 3: what kind of music is targeted? is there a preferred speaker sensitivity in regards to common listening levels? higher sensitivities will often sound better at low levels. what is the available amp power? is multi-amping an option if a lower power class A amp was to be used on a critical mid range for example?
this eventually brings you to a particular speaker design which in turn answers the question if EQ is required or not. for example, SL's approach is to combine narrow baffle with driver sizes where both lower and upper midrange are used just to get the same tonal balance over 10deg increments off axis while the flatness on axis is sacrificed, which in turn mandates the use of EQ to get that back. additionally SL wanted to use low Qts bass drivers on OB so the EQ with a boost down low was a must. only then one decides if an analog or a digital solution is better for them.
the bass is all about the room size and shape (irregular is better) and bass unit placement and timing. one in a corner can bring about excellent results just like two or three around the room can (I made a whole bunch of subs with Earl's B&C drivers and ended up using only one; maybe I would have come out better on multiple subs if I had more freedom choosing their locations).
in the end it is all about experimentation and matching a room with audio gear, and hopefully coming up with a combo which works for the kind of music played most often.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.