Some thoughts on DSP, and an embarrassing confession

[mattsk8]

Re: Some thoughts on DSP, and an embarrassing confession

The T/I is a circuit board* in which you set the parameters by selecting component values. ( these are fixed )
The Ashly is a complete turnkey solution in which operation can be varied on demand, a bigger feature set.

* http://www.ti.com/lit/ug/tidu035/tidu035.pdf

Your avatar haha. "The Dominator XL" :D

Again, I think we're leaving out the most critical component: the designer. I think the DSP has a far higher chance of sounding better--in the hands of the right designer--because it is almost infinitely adjustable, whereas the op-amp based Signal Processors are more limited.

ASP (analog signal processing), and DSP (digital signal processing) both do exactly the same thing (process the signal before the amplifier, or pre-amplifier). The difference is the analog processing, vs digital processing. Similar to a record vs a CD. Obviously to use either you need to understand how the drivers in multi-way speakers and frequencies all correlate with each other to work and sound good. Both the analog and the digital are infinitely adjustable. Passive is doing the crossover post-amplifier (after the amplifier, between the amp and the speakers). I know all this, I also know how to measure speakers.

What I'm curious about is whether the digital or the analog crossover sounds better. I already think you can get better sound with an active crossover than you can with a passive, you just need more channels of power to go active.

[Sydney]

Re: Some thoughts on DSP, and an embarrassing confession

Your avatar haha. "The Dominator XL" :D.

I enjoy that movie immensely especially the infamous "Salesman finds a Heart" scene.
( Of course Hi-Fi Sales Associates are not like that :D )

What I'm curious about is whether the digital or the analog crossover sounds better. I already think you can get better sound with an active crossover than you can with a passive, you just need more channels of power to go active

I'd say the answer to that lies in the implementation itself.
Most analog crossovers use Op-Amps and there are not all the same.
And I've heard poor A/D and D/A conversion.

[mattsk8]

Re: Some thoughts on DSP, and an embarrassing confession

I'd say the answer to that lies in the implementation itself.
Most analog crossovers use Op-Amps and there are not all the same.
And I've heard poor A/D and D/A conversion.

Just like anything else you can obviously do it wrong. Let's just assume that you have it done correctly. A TM speaker for example, and the midwoofer and tweeter were each measured (correctly). Someone designed a 3rd order passive crossover and the crossover point was 2.2khz and that was perfect for those drivers.

Now lets say the same guy (that knows what he's doing) did an active crossover. (Lets assume he understands level matching output on the drivers, and understands crossover slopes and frequency ranges.) So that guy does an active cross at 2.2khz using a MiniDSP.

Then lets say he uses the analog crossover to do exactly the same thing as he did with the MiniDSP.

Which one would sound better? That's my question.

[ryanbouma]

Re: Some thoughts on DSP, and an embarrassing confession

ASP (analog signal processing), and DSP (digital signal processing) both do exactly the same thing (process the signal before the amplifier, or pre-amplifier). The difference is the analog processing, vs digital processing. Similar to a record vs a CD. Obviously to use either you need to understand how the drivers in multi-way speakers and frequencies all correlate with each other to work and sound good. Both the analog and the digital are infinitely adjustable. Passive is doing the crossover post-amplifier (after the amplifier, between the amp and the speakers). I know all this, I also know how to measure speakers.

I think what Paul is saying is the DSP is more powerful, which is true, and therefore stands a better chance of sounding better. That Ashley unit you pictured is probably limited to certain slopes and maybe has BW or LR selectable. The EQ is a GEQ. Useful, but fairly limited. With DSP you can do biquads, PEQ, shelf filters, LTs, your slopes can be any order usually up to 96db/oct and your choice of Bessel, BW, LR, and more. You can also get DSP units that work before the DAC, so you don't have ADC/DAC conversion, which is the ONLY reason to use an analog signal processor that I can think of.

Oh, and, as for your original question, provided both the passive and active speakers have identical transfer functions, I think you'll have a very hard time hearing the difference. The active DSP can be better because it can do so much more. Getting +/-0.25db is literally a few clicks of the button.

[ryanbouma]

Re: Some thoughts on DSP, and an embarrassing confession

Then lets say he uses the analog crossover to do exactly the same thing as he did with the MiniDSP.

Which one would sound better? That's my question.

It's going to be a pretty impressive analog cross over if it can do what the minidsp can do :eek:

[mattsk8]

Re: Some thoughts on DSP, and an embarrassing confession

I think what Paul is saying is the DSP is more powerful, which is true, and therefore stands a better chance of sounding better. That Ashley unit you pictured is probably limited to certain slopes and maybe has BW or LR selectable. The EQ is a GEQ. Useful, but fairly limited. With DSP you can do biquads, PEQ, shelf filters, LTs, your slopes can be any order usually up to 96db/oct and your choice of Bessel, BW, LR, and more. You can also get DSP units that work before the DAC, so you don't have ADC/DAC conversion, which is the ONLY reason to use an analog signal processor that I can think of.

Oh, and, as for your original question, provided both the passive and active speakers have identical transfer functions, I think you'll have a very hard time hearing the difference. The active DSP can be better because it can do so much more. Getting +/-0.25db is literally a few clicks of the button.

Thanks. Definitely a PEQ is going to give a lot more options to tailor than a GEQ, so there's definitely that gain with the MiniDSP, plus with the MiniDSP you also have the ability for time delay between the tweeter and midwoofer if you want.

But, my experience in the past has been that a digital active crossover made the speakers sound a tad thinner than a analog active crossover; but this was back in around 2005, I haven't played with it since then. I brought a scenario up earlier in this thread when I mentioned the difference between using the crossover in a head unit, verses using the crossover on an amplifier. What I didn't realize until De Focht brought up the ATC statement was that the difference in crossovers that I had used was that one was digital and the other was analog. I'm curious if the same holds true today, that analog active sounds better than digital active (regardless of whether or not the analog is op-amp controlled or a dial).

I realize if someone doesn't understand how to do it, it's going to sound bad; but I don't understand the relevance of that in this situation, or why the original question couldn't just be answered without shedding so much light on that fact. Obviously you can screw it up, but is that something that doesn't apply to pretty much everything you do in life???

[mattsk8]

Re: Some thoughts on DSP, and an embarrassing confession

It's going to be a pretty impressive analog cross over if it can do what the minidsp can do :eek:

I guess I can't explain it without going around in a million circles and getting a lesson in my A/B/C's first. Not entirely sure why the original question is so difficult to grasp though.

Edit: Forget the examples and let me ask a simple question (I don't need to know that there's a possibility to make it sound bad with either, I understand how this works)... Does analog active sound better than digital active (does anyone with experience using both have an answer to that?)?? ATC obviously thinks so, I'm curious if anyone here has tried both.

[Sydney]

Re: Some thoughts on DSP, and an embarrassing confession

Undeniably once in a digital domain as bits - these numeric values can be manipulated with great flexibilty

[mattsk8]

Re: Some thoughts on DSP, and an embarrassing confession

Undeniably once in a digital domain as bits - these numeric values can be manipulated with great flexibilty

So you think the digital is going to give better sound? When comparing <JUST THE CROSSOVER> part of the MiniDSP to the crossover in a analog crossover, you think the digital would sound better? Analog has pretty much infinite control just the same... where is the digital one going to have more control over the crossover function?

[Sydney]

Re: Some thoughts on DSP, and an embarrassing confession

So you think the digital is going to give better sound?

No sir, not merely by virtue of being converted to a digital representation.
I'm not sure I can provide a simple description of ALL the issues involved in converting an Analog signal to a PCM based digital stream* and then the conversion back to a usable Analog signal**.
There were a lot of technical hurdles that had to be overcome; A major one was the implementation of the error control coding utilizing the Reed Solomon code.
There are others - But this isn't a Digital design forum.

* http://www.ti.com/general/docs/lit/g...2&fileType=pdf
** http://www.ti.com/lit/an/sbaa055/sbaa055.pdf

[bcodemz]

Re: Some thoughts on DSP, and an embarrassing confession

Just want to give some thoughts on this as at one time I spent a considerable amount of time and money trying to build a small speaker with good bass.

In short, if you can deal with the extra amps and cables, active is the way to go. There is so much unused potential in a speaker that can be brought out by using DSP. Let me explain.

Most people listen to music at moderate levels, but the drivers we use are capable of playing much louder. As long as you don't exceed the maximum SPL capabilities of the drivers you're using in your listening, then using DSP can let you get away with using much smaller speakers. For example, I have a speaker with dual 4'' woofers that I use in my kitchen. By using DSP, and applying a ~20dB boost (depending on the frequency) in the bass, it is capable of sounding like there is a very large subwoofer in the room when playing at normal volumes. It has shocked many people that a tiny, 0.15 cf speaker is putting out this kind of bass. Even though it can only sustain this much of a bass boost at a normal volume, that's pretty much all I need. If I wanted this time of bass performance without DSP, I would either need a *very* large floorstanding speaker or a subwoofer.

On the topic of crossovers, besides cost and more amplifiers, an active crossover is better than a passive one in *every* way.

- Higher efficiency because there are no losses from resistors.
- Lesser load on amplifiers because you don't have to worry about impedance and phase changes from using inductors and capacitors (which also improves sound quality).
- The ability to use steep slopes, which helps the transition between the crossover region, reduces distortion, and can better protect the tweeter.
- Micro adjustments to the frequency response of the driver to achieve closer to the perfect response
- And of course, way easier to design and experiment.

There are caveats. As Paul mentioned, it is not as simple as entering a crossover value and set slopes. You still need to listen. And be careful of over-EQ'ing speakers. While you can tinker and tinker and get a very flat on axis frequency response with DSP, it doesn't necessarily mean it'll sound good in a room. That's because when we're in a room, we hear the on axis and off axis reflected sound from the speaker, and that is determined by the directivity of the drivers. You can EQ the frequency response, but not the directivity because that's a property of a driver.

People shouldn't be ashamed of using EQ. Instead, it should be embraced. Smart DSP use will bring out the most out of a speaker in ways that passive crossovers simply can't. If you want the best speaker possible, active is the only way to go.

[mattsk8]

Re: Some thoughts on DSP, and an embarrassing confession

I actually had a good answer to my digital active verses analog active question before I even asked it, but I must've skimmed over it.

Be aware that ATC (Billy Woodman, Bob Polley) do not like DSP crossovers. To them active is still analog all the way through.

I think DSP from Lake, XTA and BSS offer some excellent DSP crossover algorithms. The Digmoda dsp works very well also with 3886 based amplifiers and the B&O amplifiers.

Recently I have been torn with the result of some Op amp based crossovers versus the dsp based crossovers I have available. Mostly I agree with ATC's Woodman that inexpensive dsp gives up resolution to the op amps and can add artifacts.

Having owned many ATC loudspeakers. The active version are superior to the passives.

DSP versus Op amp active.....still not sure.

[Sydney]

Re: Some thoughts on DSP, and an embarrassing confession

I actually had a good answer ...

I just started a related resource thread you might find of interest.

[mattsk8]

Re: Some thoughts on DSP, and an embarrassing confession

I just started a related resource thread you might find of interest.

Dang it Sydney. Thankfully I have a 3 day weekend :p

[Sydney]

Re: Some thoughts on DSP, and an embarrassing confession

don't read them all in one setting
( have fun )