A) There are 3 views on this. In theory you can do things digitally you just cannot do in passive. Does it matter? Only you can tell.
B) taps is FIR. In FIR you can do things you cannot do in the analog domain. But with great power etc. But more taps is better to first order
C) MacOS is BSD so effectively linux with a pretty frock! Many music players have ipad front ends so should be workable.
The devil is mission creep. Ask me how I know...
B) taps is FIR. In FIR you can do things you cannot do in the analog domain. But with great power etc. But more taps is better to first order
C) MacOS is BSD so effectively linux with a pretty frock! Many music players have ipad front ends so should be workable.
The devil is mission creep. Ask me how I know...
Thx Bill.....I hear ya with the mission creep, although I do have a pretty good handle on my end game.
I would like to make sure I don’t paint myself into a corner, in other words keep everything upgradable (current) and compatible. Roon seems to be one obvious way (at least from what I can tell) to keep it all cohesive.
I would like to make sure I don’t paint myself into a corner, in other words keep everything upgradable (current) and compatible. Roon seems to be one obvious way (at least from what I can tell) to keep it all cohesive.
Whell I have one of those, first series and I love it.
(The one with ES9028PRO)
Well, the software is a little user unfriendly, but once you get into it, it's just fine.
I'm more like visual person so I would like that it have graphic interface as miniDSP or Behringer DCX or Najda, which sofyware are more user frendly than audiowever.
I used to have behringer DCX, miniDSP 4x10HD, diy Najda DSP.
Any way, I really happy with DSP from analog precision paired with rPi as source, ncore (NC25XMP) ampifiers and satori drivers.
I liked DCX more then miniDSP in terms of sound quality.
Diy Najda DSP did have 6xTDA1541A DACs and P1.7 preamplifiers with relay attenuator integrated but I manage to burn it....
So David's DSP come as perfect substitute
And I like it a lot.
Bob, Be careful about using the word convoluted, since 'convolution' is a mathematical operation sometimes used in DSP.
Regarding (A?) the Behringer, not sure which one it is at this point, DSP can't be done in analog. If a DSP device has analog inputs do not use them! Such DSP units typically have cheap junk analog to digital converters for the input, then a DSP chip, then cheap junk digital to analog converters for the output. There is no point in running music through a dac (or an ADC) than once, and once should be the limit. Anything more than once will just do more damage to sound quality.
Therefore, if you want a DSP unit for playing digital audio, get one that accepts digital input, and if you have a good dac already, then use a DSP that also has a digital output. Run the DSP digital output into your good dac.
(B) More taps means more stages of filter processing. Often, one can more accurately produce a desired filter profile with more taps, but if using a lot of taps you may also want to use more bit-depth (say, 64-bit, rather than 32-bit digital math processing) which may be easier to do in a computer. Many DSP chips have limited numbers taps, and operate at 32-bits only. However, dedicated DSP units may have some advantages in terms of cost, possibly ease of configurability, not tying up an expensive PC, etc.
(C) Don't know if Mac Mini can run linux, a google search should tell you. However, under the hood the Mac OS is lot like linux in many ways. So, if you need to configure things from a text command line you will probably roughly equally comfortable with both OSes. Windows is the one that is the most different, IMHO.
EDIT: I responded to your post before reading the next page of posts. Looks like you already got most of what you were looking for. I'll leave my two cents up in case you find anything useful in it
Regarding (A?) the Behringer, not sure which one it is at this point, DSP can't be done in analog. If a DSP device has analog inputs do not use them! Such DSP units typically have cheap junk analog to digital converters for the input, then a DSP chip, then cheap junk digital to analog converters for the output. There is no point in running music through a dac (or an ADC) than once, and once should be the limit. Anything more than once will just do more damage to sound quality.
Therefore, if you want a DSP unit for playing digital audio, get one that accepts digital input, and if you have a good dac already, then use a DSP that also has a digital output. Run the DSP digital output into your good dac.
(B) More taps means more stages of filter processing. Often, one can more accurately produce a desired filter profile with more taps, but if using a lot of taps you may also want to use more bit-depth (say, 64-bit, rather than 32-bit digital math processing) which may be easier to do in a computer. Many DSP chips have limited numbers taps, and operate at 32-bits only. However, dedicated DSP units may have some advantages in terms of cost, possibly ease of configurability, not tying up an expensive PC, etc.
(C) Don't know if Mac Mini can run linux, a google search should tell you. However, under the hood the Mac OS is lot like linux in many ways. So, if you need to configure things from a text command line you will probably roughly equally comfortable with both OSes. Windows is the one that is the most different, IMHO.
EDIT: I responded to your post before reading the next page of posts. Looks like you already got most of what you were looking for. I'll leave my two cents up in case you find anything useful in it
Last edited:
Yah....we don’t need any more convoluted confusion around here!
Your not following what I was saying about the Behringer,
Digital source>DSP unit> Quality DAC (say D90) for conversion> analog IN to Behringer super X pro crossover (the one that Richard mentioned , all analog unit no dsp or dacs) > amps> speaker.
What I was asking.....
will the dsp originally applied make it through unscathed or will it be a different ‘convoluted’ version of itself?
I notice that some prefer iir for xo duty and eq but fir for phase correction?
Your not following what I was saying about the Behringer,
Digital source>DSP unit> Quality DAC (say D90) for conversion> analog IN to Behringer super X pro crossover (the one that Richard mentioned , all analog unit no dsp or dacs) > amps> speaker.
What I was asking.....
will the dsp originally applied make it through unscathed or will it be a different ‘convoluted’ version of itself?
I notice that some prefer iir for xo duty and eq but fir for phase correction?
Last edited:
In theory the audio will make it through only being filtered according to the crossover settings you make. However, filtering in the analog domain will introduce phase shifts. Of course that also happens with passive crossovers built into speakers, since they operate in the analog domain too. You could go back to your DSP unit and make further adjustments to try to compensate for any effects of the crossover filters you don't like.
The thing I would worry about most is the quality of the electronics inside the box. What do you think they can afford to put in there for $198? The manual doesn't have any distortion specs at all.
When I built the active line level crossover here for RNM's M2s I used Wima MKS 4 caps, Dale RN65 resistors, and OPA1612 opamps. Also used a simple linear power supply that I know sounds good with opamps. It probably cost me close to $198 for parts not including the old repurposed case I used. It sounded pretty nice, actually. Soundstage was not too bad at all, although not nearly up to what the Sound Labs do. The point is that the crossover didn't seem to add any objectionable distortion. Doubt you will be so lucky with that Behringer thingy.
EDIT: Regarding IIR verses FIR, it probably depends on the speakers and other issues. If you have enough taps an FIR it should be fine (I would think). When the number of taps is limited then IIR might be better (again, I would think).
However, you can try both and see what you think. If it sounds right it probably is right, just make sure you listen carefully multiple times over a few days and don't jump to conclusions.
Last edited:
Yes, now we’re cookin with gas, thank you.
I only mentioned that unit because Richard said he bought two but hasn’t tried them yet, your take on it was exactly the same as mine. There are higher quality analog active xo’s available.......
I noticed while voicing my passive xo’s (speaker level) in my last build,I would have to go back and tweak the dsp on every change in component values. Even different type components of the same value.
Another thing I’ve been considering is post dsp line level (analog) passive xo for each but then would have to split The Stereo signal into 6 seperate lines......not real sure how to correctly implement that.
I only mentioned that unit because Richard said he bought two but hasn’t tried them yet, your take on it was exactly the same as mine. There are higher quality analog active xo’s available.......
I noticed while voicing my passive xo’s (speaker level) in my last build,I would have to go back and tweak the dsp on every change in component values. Even different type components of the same value.
Another thing I’ve been considering is post dsp line level (analog) passive xo for each but then would have to split The Stereo signal into 6 seperate lines......not real sure how to correctly implement that.
Last edited:
IMO doing the X-over digitally is a big win. Yes needs more DACs and amps but gives you flexibility. It also lets you do what you want rather than what you should do, but that's just will power. Some people optimise in DSP then build an analog box to provide the required transfer functions. I originally thought I would go down that path then decided to stick with something that worked.
I kinda figured it’s best to keep as much in the digital domain as possible, although it’d be the same amount of amps, still tri amped no matter which way I go.
Wouldn’t be hard to try it out either way I suppose......just get a full digital setup and experiment with hybrid after I get the digital end dialed.
I’ll narrow down a list of possibles and post them tomorrow.
Wouldn’t be hard to try it out either way I suppose......just get a full digital setup and experiment with hybrid after I get the digital end dialed.
I’ll narrow down a list of possibles and post them tomorrow.
Bob, some class D amps measure very well and are very reasonable in terms of cost. Finally, there is a free lunch...well, not exactly. Some people complain of missing low level detail from class D. However, even the pickiest will accept good class D for the sub amps. Best if you can find some to listen to and then decide.
Last edited:
Word is it will not support it
What about this (with Roon to handle the dsp).... sonicTransporter i5 CDR (CD Ripper) – Small Green Computer
Followed up with this... Okto Research
Edit......just dawned on me I need a Roon endpoint to make that work. Ahhh!
What about this (with Roon to handle the dsp).... sonicTransporter i5 CDR (CD Ripper) – Small Green Computer
Followed up with this... Okto Research
Edit......just dawned on me I need a Roon endpoint to make that work. Ahhh!
Last edited:
So I already have an elac ds-s101g elac server/streamer that says it’s a Roon endpoint, does networking that into the computer/server alleviate that problem? And will it all play nice is one of my major concerns, that’s why I was hoping for a ‘all in one’ solution.
Getting the Okto 8 seems to be the biggest problem!
Getting the Okto 8 seems to be the biggest problem!
Bob,
Lots of people use a Nuk for roon, no additional endpoint is needed: nuk i7 - Google Search
The Oktodac is a more or less bog-standard ESS 8-channel design. Not a real keeper, but better than a minidsp box.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.