• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Why aren't we pushing for more 4-8 channel DACS for a quality Stereo setup

mdsimon2

Major Contributor
Forum Donor
Joined
Oct 20, 2020
Messages
2,610
Likes
3,513
Location
Detroit, MI
I agree that there's a weird gap in MiniDSP's lineup between the MCHStreamer and the U-DIO8. The MCHStreamer supports 8 channel output and is only $125 USD, but you have to bring a chassis and wire up the output channels yourself. (Still, if you're DIY-inclined, it's a good option.) If you want to do audio input, you also have to add an ASRC because unlike the older USBStreamer, the MCHStreamer can't use an input clock. (Or at least it couldn't last time I checked. I need that feature, so it ruled out the MCHStreamer and the U-DIO8 for me.)

I have a MCHStreamer but never really found much use for it. Like all miniDSP products the I2S pinout is terrible (one ground pin!) and the lack of multiple MCLK/BCLK/LRCLK pins means it is really only suited for multichannel I2S DACs which are exceedingly rare. I do have the DIYINHK ES9016 and ES9038 I2S DACs but have always used the DIYINHK XMOS with them as it is perfectly pin compatible.

When playing around with the MCHStreamer I have had some luck daisy chaining clock lines, especially on ESS DACs with ASRC that do not require MCLK but once I went above 2 stereo DACs things got unreliable. I do have an Ian Canada McFIFO/McDualXO which can act as a clock buffer and provide 4x MCLK/BLCK/LRCLK but the added expense puts it above the cost of a U-DIO8 and it is way less convenient as you still need to use I2S DACs or something like WM8805 I2S to AES/SPDIF boards.

Michael
 
Last edited:

JRS

Major Contributor
Joined
Sep 22, 2021
Messages
1,158
Likes
1,007
Location
Albuquerque, NM USA
I so hear you, and have been looking for a single box solution for the past year. But a bit of history--I first heard the all digital tri-amped Meridian speakers in Atlanta--maybe 1994, and having the concept explained, I thought damn, this is so elegant and patently superior, it has to be the future of audio. Notion tucked aside until about 2003 when I read about the DEQX--maybe Stereophile? Curious enough that I talked to the nice folk down under and found a demo for 2/3'rds the asking price of gulp 3800. Wasn't quite sure exactly what I was going to do with it, but thought OK this is going to be the future of audio, and I don't want to be left behind. I'm tired of spending small fortunes for speakers that leave me yearning for even more expensive upgrades of questionable value--here's something to at least get the impulse response close to correct, and maybe, just maybe get off of the upgrade death spiral.

It was a weird happenstance that I happened to have some unassuming JBL bookshelf speakers (300/pr) I had just bought for the bedroom along with a pair of expensive ($5K) Infinity speakers I had on loan from a nearby Best Buy back in the day when they had the Magnolia high end stereo/HT departments, also on closeout. By this time in my audio life, I'd been through some nice speakers including Infinity Gamma IRS,
Thiel Cs 3.5's, Dunlavy SC-IV's and Newform Reseach ribbon/Scan Speak hybrids--all in the 3k to 7k price range, and all very fine speakers but all left me wanting for something better--little less edgy, better focus, whatever. And there was enough of an engineer (hard nosed prove it to me type) left in me that no way I was going down the Krell, Audio Research, etc path. Too many other things to do and other ways to spend $$.

Bear in mind that I had never used the DEQX for anything. It took the better part of a day to get through the manual and get grounded enough to try a project. The baby bookshelves--which sounded pretty good (6.5" two way, the tweeter recessed within an elliptical wave guide)--seemed like a good place to start. Two hours later I had them sounding almost as good as the Infinity's and this w/o bypassing the passive XO's. To say I was dumbfounded is an understatement. The Infinities were returned the next day. This is not to say that one can make a silk purse from a sow's ear, the dynamics were limited and no way could the tiny woofer keep up with a servoed, 400W dedicated amp 11"er. But it was close enough over a range of material and volumes that it sealed the deal for me.

Now nearly 20 years later, I have designed and put together 4 actively powered and digitally filtered 2 and 3 way systems that are sufficiently satisfying that I rarely even visit audio places. Yes I was able to break free of the addiction. 3 of the 4 systems were because of one theft, two moves, first to smaller digs and then larger quarters, and one which was good, but just never could get quite right, but still use it for a second desktop system. On the way, I learned a great deal about transducers, their various shortcomings, how to measure them, and even to rout a perfect circle with less than a mm of gap. :D And bought countless saws and clamps on the way--need a saw, I now have a jigsaw, backsaw, holesaw, table saw, circ saw, hand and hacksaws, and have made a bunch of sawdust on the way.

What I have discovered is that designing and building a very, very good loudspeaker is not the arcane art of yesteryear. The primary problems are in no particular order: 1) diffraction including baffle step 2) matching sensitivities 3) seamless XO's 4) displacement limits, 5) compensating for driver FR anomalies, and 6) getting a decent power response. All of course should be housed in something non-resonant and pleasing to the eye. The drivers need not be expensive if they are well chosen and mate well with the others. Of course, there is nothing wrong with using the sort of components found in a 10,000/pr speaker: typically these might cost somewhere between 200 and 400 dollars apiece, instead of 50 to 150. So if one is on that kind of quest, a reasonable approximation might be had for roughly 1500 for drivers and another five hundred for cabinets, assuming we aren't too choosy about the finish--i.e. exotic veneers.

The crossover design is the biggest barrier by far. Designing, choosing the components and voicing crossovers are all matters of great expertise, and I tip my hat to those with the time, money, patience and tech savvy to learn the art. I was personally either too lay, too busy or both. But in my experience I have yet to find a passive speaker that outperformed it's digital counterpart. I don't say that they don't exist, just that none of the several kits I have assembled for friends couldn't be implemented actively, sounding as good if not better. As an aside, people point out the cost and complexity of active loudspeakers as they require additional amplifiers, but dismiss the very real cost of coreless inductors (ouch) or the right kind of plastic capacitor. Thanks to baffle simulators and great but affordable measuring equipment and software makes a many month effort doable in a couple of weekends. And amps, buy once, not several times.

Going out on a limb with my trusty hand saw ready, I'd say all but power response and designing/fabricating a wave guide are within easy reach--baffle response is still an approximation using the EXCEL based open source software but all but huge errors can be mostly remedied--or the baffle replaced with a different design (cut and test first before the cabs!) Power response is sometimes hard to discern without having the drivers and making many measurements, but using solid rules of thumb such as not thinking that 15" beast of a woofer will mate well to a 1" tweeter because their FR happens to overlap. And there are always kits (Meniscus, Madisound, for example) that have survived the test of time, and can be sometimes be bought without the XO parts. A very safe way to start with gobs of performance/price.

Wow this was meant to be a post related to finding that single box solution. What I'd love to see: a streamer, multiple digital inputs w/ volume, enough channels to do an active 3 way + sub(s) with both digital and analog outs (the XO filters should not simply emulate analog filters and be min phase up to 102dB/octave), robust measurement and filter design capabilities so as to "fix" the impulse response at 1 meter, and whether it's ATMOS or Dirac Live, a sound correction suite that can average measurements according to users needs, a 10+ band parametric EQ, and have 4 configurations on demand, along with a bunch of preset EQ's for different volumes, types of recording, etc. The DAC's should be of decent pedigree and the SINAID >110dB. And preferably all of this for under $2000.

Anyone know of such a product? I'd buy it in a heartbeat. Anyhow I'd didn't mean to ramble--only to point out that DIY speaker design is becoming more and more practical with an active approach, and maybe get a few thinking about giving it a go. Changed my life. And never looked back. The only commercial speaker I really have to own some day are the Sanders 10E are some other really exceptional planar.
 

audio2design

Major Contributor
Joined
Nov 29, 2020
Messages
1,769
Likes
1,843
What I have discovered is that designing and building a very, very good loudspeaker is not the arcane art of yesteryear. The primary problems are in no particular order: 1) diffraction including baffle step 2) matching sensitivities 3) seamless XO's 4) displacement limits, 5) compensating for driver FR anomalies, and 6) getting a decent power response.

When you say diffraction, can I assume dispersion as well, as in matching dispersion at the cross-over frequency? That is something that still escapes many in the DIY space. To me, the rest really makes little sense to go passive any more. If you extract the lower bass, the power handling of the amp channels is pretty easy.
 
OP
B&WTube

B&WTube

Member
Joined
Feb 17, 2020
Messages
87
Likes
105
I so hear you, and have been looking for a single box solution for the past year. But a bit of history--I first heard the all digital tri-amped Meridian speakers in Atlanta--maybe 1994, and having the concept explained, I thought damn, this is so elegant and patently superior, it has to be the future of audio. Notion tucked aside until about 2003 when I read about the DEQX--maybe Stereophile? Curious enough that I talked to the nice folk down under and found a demo for 2/3'rds the asking price of gulp 3800. Wasn't quite sure exactly what I was going to do with it, but thought OK this is going to be the future of audio, and I don't want to be left behind. I'm tired of spending small fortunes for speakers that leave me yearning for even more expensive upgrades of questionable value--here's something to at least get the impulse response close to correct, and maybe, just maybe get off of the upgrade death spiral.

It was a weird happenstance that I happened to have some unassuming JBL bookshelf speakers (300/pr) I had just bought for the bedroom along with a pair of expensive ($5K) Infinity speakers I had on loan from a nearby Best Buy back in the day when they had the Magnolia high end stereo/HT departments, also on closeout. By this time in my audio life, I'd been through some nice speakers including Infinity Gamma IRS,
Thiel Cs 3.5's, Dunlavy SC-IV's and Newform Reseach ribbon/Scan Speak hybrids--all in the 3k to 7k price range, and all very fine speakers but all left me wanting for something better--little less edgy, better focus, whatever. And there was enough of an engineer (hard nosed prove it to me type) left in me that no way I was going down the Krell, Audio Research, etc path. Too many other things to do and other ways to spend $$.

Bear in mind that I had never used the DEQX for anything. It took the better part of a day to get through the manual and get grounded enough to try a project. The baby bookshelves--which sounded pretty good (6.5" two way, the tweeter recessed within an elliptical wave guide)--seemed like a good place to start. Two hours later I had them sounding almost as good as the Infinity's and this w/o bypassing the passive XO's. To say I was dumbfounded is an understatement. The Infinities were returned the next day. This is not to say that one can make a silk purse from a sow's ear, the dynamics were limited and no way could the tiny woofer keep up with a servoed, 400W dedicated amp 11"er. But it was close enough over a range of material and volumes that it sealed the deal for me.

Now nearly 20 years later, I have designed and put together 4 actively powered and digitally filtered 2 and 3 way systems that are sufficiently satisfying that I rarely even visit audio places. Yes I was able to break free of the addiction. 3 of the 4 systems were because of one theft, two moves, first to smaller digs and then larger quarters, and one which was good, but just never could get quite right, but still use it for a second desktop system. On the way, I learned a great deal about transducers, their various shortcomings, how to measure them, and even to rout a perfect circle with less than a mm of gap. :D And bought countless saws and clamps on the way--need a saw, I now have a jigsaw, backsaw, holesaw, table saw, circ saw, hand and hacksaws, and have made a bunch of sawdust on the way.

What I have discovered is that designing and building a very, very good loudspeaker is not the arcane art of yesteryear. The primary problems are in no particular order: 1) diffraction including baffle step 2) matching sensitivities 3) seamless XO's 4) displacement limits, 5) compensating for driver FR anomalies, and 6) getting a decent power response. All of course should be housed in something non-resonant and pleasing to the eye. The drivers need not be expensive if they are well chosen and mate well with the others. Of course, there is nothing wrong with using the sort of components found in a 10,000/pr speaker: typically these might cost somewhere between 200 and 400 dollars apiece, instead of 50 to 150. So if one is on that kind of quest, a reasonable approximation might be had for roughly 1500 for drivers and another five hundred for cabinets, assuming we aren't too choosy about the finish--i.e. exotic veneers.

The crossover design is the biggest barrier by far. Designing, choosing the components and voicing crossovers are all matters of great expertise, and I tip my hat to those with the time, money, patience and tech savvy to learn the art. I was personally either too lay, too busy or both. But in my experience I have yet to find a passive speaker that outperformed it's digital counterpart. I don't say that they don't exist, just that none of the several kits I have assembled for friends couldn't be implemented actively, sounding as good if not better. As an aside, people point out the cost and complexity of active loudspeakers as they require additional amplifiers, but dismiss the very real cost of coreless inductors (ouch) or the right kind of plastic capacitor. Thanks to baffle simulators and great but affordable measuring equipment and software makes a many month effort doable in a couple of weekends. And amps, buy once, not several times.

Going out on a limb with my trusty hand saw ready, I'd say all but power response and designing/fabricating a wave guide are within easy reach--baffle response is still an approximation using the EXCEL based open source software but all but huge errors can be mostly remedied--or the baffle replaced with a different design (cut and test first before the cabs!) Power response is sometimes hard to discern without having the drivers and making many measurements, but using solid rules of thumb such as not thinking that 15" beast of a woofer will mate well to a 1" tweeter because their FR happens to overlap. And there are always kits (Meniscus, Madisound, for example) that have survived the test of time, and can be sometimes be bought without the XO parts. A very safe way to start with gobs of performance/price.

Wow this was meant to be a post related to finding that single box solution. What I'd love to see: a streamer, multiple digital inputs w/ volume, enough channels to do an active 3 way + sub(s) with both digital and analog outs (the XO filters should not simply emulate analog filters and be min phase up to 102dB/octave), robust measurement and filter design capabilities so as to "fix" the impulse response at 1 meter, and whether it's ATMOS or Dirac Live, a sound correction suite that can average measurements according to users needs, a 10+ band parametric EQ, and have 4 configurations on demand, along with a bunch of preset EQ's for different volumes, types of recording, etc. The DAC's should be of decent pedigree and the SINAID >110dB. And preferably all of this for under $2000.

Anyone know of such a product? I'd buy it in a heartbeat. Anyhow I'd didn't mean to ramble--only to point out that DIY speaker design is becoming more and more practical with an active approach, and maybe get a few thinking about giving it a go. Changed my life. And never looked back. The only commercial speaker I really have to own some day are the Sanders 10E are some other really exceptional planar.
Quite the journey. I would love your product for my more serious system. For my other rooms, I still want active, but I want it good and affordable which is ubiquitous is DAC’s, except multichannel.

I do have a question about your tweeter’s digital crossover: Did you have any issue with HF noise on it? There are a couple of excellent speaker builders I know, who have started building active systems based on the Mini DSP HD. They both arrived at active because they could get the speakers to sound better than passives (and the one guy in particular has built custom systems full time for the last 30+ years…which is why I figured if this guy says active is just better- it is). HOWEVER, they both say that the DSP bleeds through a slight high frequency noise. Both guys ended up deciding optimal was to a passive tweeter xover, with everything else active. Did you ever experience that, and if so how did your fix it?
 
OP
B&WTube

B&WTube

Member
Joined
Feb 17, 2020
Messages
87
Likes
105
I do have an Ian Canada McFIFO/McDualXO which can act as a clock buffer and provide 4x MCLK/BLCK/LRCLK
I am very intrigued by the Ian Canada stuff. You are right- it adds up quick. However, cost aside, how do you like his stuff? There is not a lot of evaluation/reviews of all his many boards. I have dug through the forum threads, but it is a lot more ‘how to’, instead of ‘here is why one should choose/not choose Ian Canada’s dac/clock/transport/etc over the one form Allo/Audiophonics.

Anyways would love to hear any feedback/thoughts that could help me weigh the performance of some of the fancy Pi Hats. TIA!
 

garbulky

Major Contributor
Joined
Feb 14, 2018
Messages
1,510
Likes
830
I agree with you... I'd like to see more low-cost options in this space. But it's a very niche area, with a narrow pricing window. The solution would have to have higher DAC performance than a used MOTU 828mk3, but also cost less than an RME Digiface USB + two Topping D10s.
I didn't read the entire thread - but exactly what he said. The fact that we are buying DACs already puts us in a niche area. But most of this stuff is very diy figure it out kind of thing. Which means even more reading and configuration and setting things up from scratch.
Like for instance if there was a thriving market of systems of ready made bi amping software + ready to use passive speakers designed for crossovers etc then MAYBE. Most people haven't even heard of a crossover.
Like me, my mind boggles why they don't have dedicated digital audio transport cards that are PCI or PCI-e. PCI used to be standard for ages and I thought a digital audio only transport options would be plentiful being that it doesn't need to have a DAC/analog stage included making it somewhat easier. But even something like that is just ... not available.
 

JRS

Major Contributor
Joined
Sep 22, 2021
Messages
1,158
Likes
1,007
Location
Albuquerque, NM USA
When you say diffraction, can I assume dispersion as well, as in matching dispersion at the cross-over frequency? That is something that still escapes many in the DIY space. To me, the rest really makes little sense to go passive any more. If you extract the lower bass, the power handling of the amp channels is pretty easy.
Yes, I do--diffraction being an all encompassing sort of term describing when a wave reaches a boundary of some sort causing a change in direction. So even the best baffles, say with a 1 inch tweeter in a rectangular wood panel, introduce a ripple in the frequency dependent on the distance from the tweeter to the edge of the speaker. It may be exactly spaced so that the reflection that occurs at the edge (secondary propagation) reinforces the original wave, or perhaps it partly cancels it. A nearby frequency experiences the opposite, hence the ripple in FR. This is where a large radius curve can be of significant benefit, but other strategies can be used as well--typically the worst case arises when it is symmetrically placed so the distance to the sides and top are all equal and the edges square.

The other form of diffraction is when the frequency is low enough that what was previously 2 pi space (hemispherical) propagation transitions to 4 pi space (omnidirectional). Then a significant fraction of the energy flows backwards and not towards the ear. Both these effects can be modeled fairly accurately in EXCEL and there is some good freeware available to help. But dispersion is also very important--how simple life would be if drivers didn't beam when the wavelength becomes short relative to the moving diaphragm.

Battles have been waged for dozens of pages over the issue of dispersion and I have no dog in that hunt. I think it is safe to say that controlled directivity (dispersion) is a good thing; whether it's focused or dispersed has lots of pros/cons, but in general I'd say wave guides are here to stay. They load the driver more efficiently and provide a more uniform dispersion over a wider range of frequencies. Trouble is making them.

And I couldn't agree more with your last point. To illustrate: with my current rig, my ribbon tweeters are 99dB at 1 watt. The midbass drivers are 89dB. I happened to have good reasons for those choices, but it would be somewhat foolish to use them together with a passive XO, requiring 10dB of attenuation. But separate amps, hmmm, maybe a single ended triode might be the ticket to make the highs even more airy and delish. It's not I choice I'd make, but in any event, isolating it from the mids and bass makes absolute sense. So while I do need 3 amps--the woofers can absorb 1000W peaks and grin--a prosound CROWN or whatever is perfect for the task of covering 20 to 300 Hz, doesn't need to be fancy, just beefy. Midbass/mids need at least 100 better yet 200W--I find myself one of the new Class D amps in whatever particular flavor I favor and I'm good to go. Ribbons, who knows maybe 30W of some purified and spin locked swarm of electrons dancing through space is just the ticket. Obviously the same logic extends to a 2.2 system or even a two way w/o subs.

So yes your point is right on the money. Pick the amps for the function. And as I said in my earlier post, people go on and on about cost of 6 channels of amplification, well good passive XO's aren't cheap--one only has to look at the number of copper crimes that one wouldn't have dreamed of 20 years ago to understand why inductors cost so much. Can I tell a polyester from a polypropylene cap--unlikely, but those who do this sort of thing seriously insist there is a big difference, and so do you want to be a stubborn cheapskate and possibly shoot a great project in the foot, or play it safe, and preemptively spend a whole lot more than you planned?

So yes, I'm sold--feel free to PM me with any specific questions as I don't mean to take this thread off on a tangent. It's just I am very passionate aout building loudspeakers, and the time truly has never been better than the present.
 
Last edited:

audio2design

Major Contributor
Joined
Nov 29, 2020
Messages
1,769
Likes
1,843
I am more than "confident" that a relatively low cost amp and DAC with a digital crossover is superior to even the most expensive passive. 100W is more than adequate for mid-highs (IMHO).

I am a bit beyond "diy", so I get more out of low cost purpose built amplification than I could ever get off the shelf. What I am doing in processing and amplification is simply impossible in a traditional amp/crossover/driver setup.
 
  • Like
Reactions: JRS

JRS

Major Contributor
Joined
Sep 22, 2021
Messages
1,158
Likes
1,007
Location
Albuquerque, NM USA
Quite the journey. I would love your product for my more serious system. For my other rooms, I still want active, but I want it good and affordable which is ubiquitous is DAC’s, except multichannel.

I do have a question about your tweeter’s digital crossover: Did you have any issue with HF noise on it? There are a couple of excellent speaker builders I know, who have started building active systems based on the Mini DSP HD. They both arrived at active because they could get the speakers to sound better than passives (and the one guy in particular has built custom systems full time for the last 30+ years…which is why I figured if this guy says active is just better- it is). HOWEVER, they both say that the DSP bleeds through a slight high frequency noise. Both guys ended up deciding optimal was to a passive tweeter xover, with everything else active. Did you ever experience that, and if so how did your fix it?
So my bad--I must have been unclear--I am still using the DEQX and am considering the miniDSP DDRC-88A as a replacement. As so many else have found with the miniDSP product offerings, I am befuddled. What I want is a DDRC-88a and DDRC-88d rolled into one, and please upgrade the dacs and processors to those found in the top of the line. I get engineering compromises and economics, but damn it, they peed all over themselves on these choices--just because I want analog outputs doesn't mean I want analog inputs! And just because I need analog outs doesn't mean I might find a good use for digital outputs. Am I being perverse?

I suspect when the next product line is rolled out there may be just the ticket, but along with you, I'm still in the hunt. As to your specific question: I will say that with the DEQX which has S/N ratios maybe 108dB I hear a slight whisper of a hiss, but that is when set at considerable volume with ear at tweeter, and certainly nothing that requires fixing in my view. But what you describe sounds like a legit concern. No expert here by any stretch but a parallel feed to an analog high pass XO for the tweeter is not a huge deal, especially with all the FR tricks at ones disposal. But yea, who wants to be fussing with it. I wonder, were the tweeters like my ribbons (99dB/watt) and super efficient--if so it may be a non issue entirely if you were to use an ordinary dome.

I'm so frustrated at this point I might just get either the 4x10 or the and be done with it for now, and concede that there isn't an affordable all in one solution. Or the DDRC-88a or have my DEQX repaired--that seems really weird, now nearly 20 years later one would think there would be a plethora of choices. I shouldn't knock REW but in a way that may have steered enough potential buyers such as ourselves (but younger) into doing something nearly free as everyone has a PC and even adding a good soundcard won't break the bank. I don't know. Certainly didn't think that such a device would be so nearly unobtanium.
 

JRS

Major Contributor
Joined
Sep 22, 2021
Messages
1,158
Likes
1,007
Location
Albuquerque, NM USA
I am more than "confident" that a relatively low cost amp and DAC with a digital crossover is superior to even the most expensive passive. 100W is more than adequate for mid-highs (IMHO).

I am a bit beyond "diy", so I get more out of low cost purpose built amplification than I could ever get off the shelf. What I am doing in processing and amplification is simply impossible in a traditional amp/crossover/driver setup.
Just so long as you don't say that out loud on certain forums. A lot of guys I know elsewhere have literally spent a lifetime honing their craft, and wouldn't cotton to the concept of even analog active systems using op-amps, discrete transistors maybe--but at least there you could have wirewound resistors and whatever caps smell the sweetest and not go broke. People are funny (I'm a psychiatrist and that's all I have to say).
 

mdsimon2

Major Contributor
Forum Donor
Joined
Oct 20, 2020
Messages
2,610
Likes
3,513
Location
Detroit, MI
I am very intrigued by the Ian Canada stuff. You are right- it adds up quick. However, cost aside, how do you like his stuff? There is not a lot of evaluation/reviews of all his many boards. I have dug through the forum threads, but it is a lot more ‘how to’, instead of ‘here is why one should choose/not choose Ian Canada’s dac/clock/transport/etc over the one form Allo/Audiophonics.

Anyways would love to hear any feedback/thoughts that could help me weigh the performance of some of the fancy Pi Hats. TIA!

The only things of Ian's that I have used are the McFIFO / McDualXO and some of clock and U. Fl adapters, all of which are excellent. That being said I really don't have any interest in pursuing any of his exotic super capacitor power supplies or relocking solutions. I imagine his Pi HATs are actually pretty good performers as his gear does seem well designed (if a bit esoteric at times).

Of course the McFIFO / McDualXO is a reclocking solution but that was not my primary motivation. My main motivation was to use it as a clock buffer and in that function it works flawlessly assuming you can tolerate the delay from the FIFO. While most of the time the FIFO really does not do much, in DACs with relatively poor jitter rejection like the Schiit Modius I have measured improvements on the J Test when using the McFIFO / McDualXO upstream. It really is a godsend if you are integrating I2S DACs or WM8805 I2S to AES/SPDIF boards with something like a miniDSP miniSHARC. I no longer use this setup but here is a writeup of the miniSHARC + McFIFO / McDualXO project I did -> https://www.minidsp.com/forum/hardware-support/18550-minisharc-8-channel-digital-output.

Michael
 

mdsimon2

Major Contributor
Forum Donor
Joined
Oct 20, 2020
Messages
2,610
Likes
3,513
Location
Detroit, MI
Quite the journey. I would love your product for my more serious system. For my other rooms, I still want active, but I want it good and affordable which is ubiquitous is DAC’s, except multichannel.

I do have a question about your tweeter’s digital crossover: Did you have any issue with HF noise on it? There are a couple of excellent speaker builders I know, who have started building active systems based on the Mini DSP HD. They both arrived at active because they could get the speakers to sound better than passives (and the one guy in particular has built custom systems full time for the last 30+ years…which is why I figured if this guy says active is just better- it is). HOWEVER, they both say that the DSP bleeds through a slight high frequency noise. Both guys ended up deciding optimal was to a passive tweeter xover, with everything else active. Did you ever experience that, and if so how did your fix it?

There are a lot of variables that come in to play with hiss, but I will agree that the miniDSP 2X4HD is relatively noisy and depending on tweeter sensitivity and amplifier gain this can cause issues with hiss. With Hypex NC252MP amplifiers (26 dB gain) and an Okto DAC8 pro I have not experienced any hiss with conventional tweeters, it is dead silent. I made some measurements of various DACs (including the 2X4HD) and tried to correlate them to perceived hiss level in this thread -> https://www.audiosciencereview.com/...-topping-performance.24768/page-2#post-849537 which might be of interest.

A lower gain amplifier on your tweeter is also a solution. If you are running a higher sensitivity tweeter with a higher gain amp and are cutting level in DSP you are effectively unnecessarily amplifying the noise from the DAC (which will stay constant regardless of attenuation applied in DSP).

I guess to me you can boil it down to starting with a low noise DAC and using a low noise, low gain amplifier on sensitive tweeters.

Michael
 

dualazmak

Major Contributor
Forum Donor
Joined
Feb 29, 2020
Messages
2,934
Likes
3,167
Location
Ichihara City, Chiba Prefecture, Japan
Last edited:

dualazmak

Major Contributor
Forum Donor
Joined
Feb 29, 2020
Messages
2,934
Likes
3,167
Location
Ichihara City, Chiba Prefecture, Japan
Just for your reference and info...

A few days ago, one of my dear ASR Forum friends informed me that Greg Timbers, past designer/engineer for JBL, is/was DIY-building similar "fully active 4-way" multichannel multi-way "stereo" system using JBL Everest DD67000;
https://positive-feedback.com/interviews/greg-timbers-jbl/
WS002444.JPG


It is really interesting for me to know that he uses rather small (and reasonably priced) Pioneer Elite A-20 for UHF compression drivers (super tweeters); as shared here, similarly I use rather small (and reasonably priced) Yamaha A-S301 driving my metal horn super tweeters, Fostex T925A.
 
OP
B&WTube

B&WTube

Member
Joined
Feb 17, 2020
Messages
87
Likes
105
Hello from Japan, OP @B&WTube,

Although quite be lated, I just came across with this thread.

How would my multichannel multi-driver multi-way multi-amplifier "stereo" project and my latest system configurations fit in your perspective?
Greetings! You have a nice setup. The Okto DAC8/Pro is a high quality, measures well, and does what I would want. However, there are a number of issues with it:
1) Primary Issue: You can’t buy it “due to large number of orders” and apparently, it has been this way for a while
2) Customer service: They are bad at this, as they don’t respond to emails, and take months to ship out products even when in stock.
3) Price: At this time, considering all the great measuring DAC’s for $100-$200- we should be able to get 4-8 channels for between $400-$600, IMO.

That being said, I would still be interested if they ever become available again- but would prefer decent customer service and decent price. Thanks for chiming in!
 
OP
B&WTube

B&WTube

Member
Joined
Feb 17, 2020
Messages
87
Likes
105
Just for your reference and info...

A few days ago, one of my dear ASR Forum friends informed me that Greg Timbers, past designer/engineer for JBL, is/was DIY-building similar "fully active 4-way" multichannel multi-way "stereo" system using JBL Everest DD67000;
https://positive-feedback.com/interviews/greg-timbers-jbl/
View attachment 155774

It is really interesting for me to know that he uses rather small (and reasonably priced) Pioneer Elite A-20 for UHF compression drivers (super tweeters); as shared here, similarly I use rather small (and reasonably priced) Yamaha A-S301 driving my metal horn super tweeters, Fostex T925A.
Active is the way to go, but in his system the DEQX HDP4 that is controlling the crossovers covering DAC duties is $6,500, with a $4,000 DEQX PREMATE (pre-amp). With a Rasberry Pi, or other computer, there is free software than can accomplish the same crossover control and DSP, for a fraction of the cost, IF we have a good and multichannel DAC.

To me, this really is a big deal, and should be a big deal to all audio enthusiasts. Your buddy’s setup is like a $20k+ setup. I think I can likely match it for around $2k- IF we get a good multichannel DAC for around $500, which should be extremely doable.
 

AudioJester

Addicted to Fun and Learning
Forum Donor
Joined
Jan 7, 2020
Messages
980
Likes
1,324
Active is the way to go, but in his system the DEQX HDP4 that is controlling the crossovers covering DAC duties is $6,500, with a $4,000 DEQX PREMATE (pre-amp). With a Rasberry Pi, or other computer, there is free software than can accomplish the same crossover control and DSP, for a fraction of the cost, IF we have a good and multichannel DAC.

To me, this really is a big deal, and should be a big deal to all audio enthusiasts. Your buddy’s setup is like a $20k+ setup. I think I can likely match it for around $2k- IF we get a good multichannel DAC for around $500, which should be extremely doable.

That allready exists in the pro world, something like Motu is crossing over into home audio for people like you.
By the way, one of the new DEQX models is 8 channels which means you no longer need 2 boxes for an all in one 8 channel system.
 
OP
B&WTube

B&WTube

Member
Joined
Feb 17, 2020
Messages
87
Likes
105
What a coincidence with this thread, at 54:15, Laurie Fincham answers Erin’s question about the future of speaker improvements, saying, “I think they’ll get better when people find a ways of doing active speakers economically.”
 

jtwrace

Major Contributor
Forum Donor
Joined
May 31, 2017
Messages
1,227
Likes
1,413
Location
Orlando, FL
What a coincidence with this thread, at 54:15, Laurie Fincham answers Erin’s question about the future of speaker improvements, saying, “I think they’ll get better when people find a ways of doing active speakers economically.”
It already exists at a very high level. Having just completed implementation using Audiolense with the help of @mitchco, I can say it's there. :)

Now we just need @Geshelli to make a MCH dac to simplify things. :p

IMG_5831.jpg
 
Top Bottom