Passive crossovers normally account for their own needed delay. It's interesting that it needs to be made available separately in dsp..
horses account for their own heating and cooling.
it is interesting that AC needs to be made available in cars.
by the way as much as i make fun of your Amish ways your ways are actually the future.
if you go read the stuff put out by the World Economic Forum you will realize in another 100 years we will all live like the Amish.
they don't even have cars in Europe anymore already. they now ride bicycles.
so maybe i'm just backwards.
very wise.
i been reading Christopher Nolan whining about how Digital Cinema is garbage and will never look as good as Celluloid and i was like ... these people are everywhere !
some Cinematographer developed an algorithm to simulate Celluloid look using digital cinema and in a test none of these people like Nolan could tell the difference ...
then just to drive home the point he shot half of star wars in Digital and half in Film completely randomly and nobody can tell which is which ...
yet will any of these people admit this ?
you know the answer ...
we can simulate the sound of vacuum tubes and vinyl in digital. can you simulate the sound of digital using vacuum tubes or vinyl ?
that should tell you the hierarchy of technologies.
there is a theory that we all live in a simulation, in which case whoever is running the simulation is g0d. well if you think about it then digital is g0d because it can simulate other things but not the other way.
and this is not to say celluloid looks bad or vinyl or tubes sound bad - not at all ! i love the sound of vinyl and tubes and i like the look of celluloid.
but digital has no competition.
not in movies not in music not in crossovers not anywhere.
Last edited:
I mean, is there a reason to take sides? I see you're identifying the discrepancies one finds in casual print, ok. Now if you want to know, you'll have to prove it to yourself. You learn what is audible. You master both techniques and do equivalent implementations of each.
yeah i can't believe i respond to these threads. i feel disgusted with myself. it's just i been away from the scene for about 15 years and was caught by surprise to see people still arguing about this. then again, i should have known better. i need to learn to somehow block out the noise.
I compared DSP directly to a matched transfer function passive system on the same speaker in a blind AB, SPL matched test, and I could clearly identify and prefer the DSP setup - night and day difference.
My opinion is as good, or better, than yours. Just saying...
^^ When I asked if there was a reason to take sides, it wasn't meant as a rhetorical question... Is one better, is one worse, are they both the same?
The way beyond this thread is to know. Until then some of the replies are naturally going to be about the challenges faced while in the process of learning just that.
The way beyond this thread is to know. Until then some of the replies are naturally going to be about the challenges faced while in the process of learning just that.
"If a capacitor were JUST a capacitor, then it would make absolutely no difference."
Now, that is one of the few reasonable comments in this thread.
So the question more rightly becomes what other factors?
Capacitance tolerance
C vs frequency
C vs temp
C vs current
Leakage
Inductance
DF
Microphonics
and indirectly HD
ESR is over rated as most of the time it has a series resistor.
All of these factors will change the transfer function. Measurably. If you can hear it is another question. It has been a long time since I was in the FA lab studding cap failure modes, I know I am forgetting some. Most of out problems were with Z5U and X7R, not films.
So a valid test would be to change one parameter at a time. That can be very difficult as the different film materials, termination and physical construction will change multiple variables with one swap.
I think, I can hear the difference between an NPE and Dayton poly, I would not bet more than a pint on it. I only tried a mid grade "audiophile" cap once and heard no difference. I have not tried a $300 copper foil wonder cap. NPEs do have an ageing problem, so I only prototype with them.
As I have been at this for a lot of years and long ago lost all interest in bragging rights, I have trained myself to hear certain kinds of differences. Others can hear other differences. Others only hear what they expect. The other half of the training, or rather discipline, is to ignore all that when you are done so you can just enjoy the music.
There is one place I have no doubt at all about the difference is crossover caps. The bank account of both seller and buyer.
Hey, I just plopped a set of Elac 5.1's in the workshop. I like nice music there too! I could not DIY a pair half as good as the price.
Now, that is one of the few reasonable comments in this thread.
So the question more rightly becomes what other factors?
Capacitance tolerance
C vs frequency
C vs temp
C vs current
Leakage
Inductance
DF
Microphonics
and indirectly HD
ESR is over rated as most of the time it has a series resistor.
All of these factors will change the transfer function. Measurably. If you can hear it is another question. It has been a long time since I was in the FA lab studding cap failure modes, I know I am forgetting some. Most of out problems were with Z5U and X7R, not films.
So a valid test would be to change one parameter at a time. That can be very difficult as the different film materials, termination and physical construction will change multiple variables with one swap.
I think, I can hear the difference between an NPE and Dayton poly, I would not bet more than a pint on it. I only tried a mid grade "audiophile" cap once and heard no difference. I have not tried a $300 copper foil wonder cap. NPEs do have an ageing problem, so I only prototype with them.
As I have been at this for a lot of years and long ago lost all interest in bragging rights, I have trained myself to hear certain kinds of differences. Others can hear other differences. Others only hear what they expect. The other half of the training, or rather discipline, is to ignore all that when you are done so you can just enjoy the music.
There is one place I have no doubt at all about the difference is crossover caps. The bank account of both seller and buyer.
Hey, I just plopped a set of Elac 5.1's in the workshop. I like nice music there too! I could not DIY a pair half as good as the price.
I guarantee your transfer function was not the same. Phase, timing, HD... Using active eq has isolated the driver mostly from the network. You have only matched the frequency response at one level. Now your conclusion, you prefer one over the other, is completely valid for you. 👍
You left out the intermediate step, that of analog active crossover and to emulate the passive and DSP transfer functions. Very hard to do actually.
Don't waste the money for the woofer components, spend all the money you can to buy tweeter cap & resistor.
I guarantee my transfer function was the same - with DSP I can manipulate phase, timing and everything else (I don't know what your "HD" stands for). I matched acoustical frequency response, phase and timing for all drivers, at all levels.
The main point for active system is to isolate drivers from passive network and connect them directly to the amplifier, with obvious benefits.
I left that intermediate and unnecessary step intentionally. Very easy to do (although time-consuming), obviously possible only with right equipment and adequate knowledge.
I know this should work so I can't say anything bad about it in principle, but it is potentially more consistent to match electrical response instead.
I don't understand this at all. Passive xo cannot delay a woofer to match to the distance of a long horn. With analog you can compromise by choosing different kind of filter for the other element, but that doesn't affect timing nor does it create a coherent wavefront. With digital you can just delay the other driver by some μs and enjoy perfect frequency response at XO point without any hacks to phase. Also better aesthetics if you don't need exotic physical shapes for front panel because of distance matching alone.
Delay is not the same as phase and delay doesn't really exist in analog xo. If it does, please tell me how. Delay is a new opportunity in dsp.
Passive can do all the same regular filters, including delay. Wavefront is just as coherent, that is an acoustic concern, not electrical. Waveforms show that phase is like delay.
what if you had fifteen 3-way array cabinets and you wanted to delay each one by one inch more than the next - how would you do this passively ?
Beam steering arrays were invented over 100 years ago. DSP was a little expensive back then 😉
https://en.wikipedia.org/wiki/Phased_array
https://en.wikipedia.org/wiki/Phased_array