Effectiveness of DSP at passive crossover

[tinnitus]

Is that a trick question?

No, i have a dsp with xo and room correction. Every chassis is connect with an power-amp channel. Nothing inside as cables and chassis.

 

[voodooless]

No, i have a dsp with xo and room correction. Every chassis is connect with an power-amp channel. Nothing inside as cables and chassis.

Sounds like an active speaker to me..

 

[sejarzo]

A DSP speaker takes a digital signal, performs the crossover digitally, does multiple channel D/A conversion into multiple amps, which power the drivers directly.

Also, note that a great many DSP speakers do not even have digital inputs. Not everything is the same.

I took Keith's original comment "takes a digital signal" to mean the signal after it performs it's initial A/D conversion of an analog input signal.

Is that substantially different from feeding a digital signal to a DSP speaker that has a digital input? Yes, there is an additional D/A then A/D conversion in the total chain before the speaker DSP implements the digital crossover, but if the added conversions are transparent, is that material?

 

[voodooless]

Yes, there is an additional D/A then A/D conversion in the total chain before the speaker DSP implements the digital crossover, but if the added conversions are transparent, is that material?

If it is transparent, it’s also no longer an argument

But yes, usually this is transparent. The main arguments here would be cost and practicality. You don’t need to buy a DAC, just something that gives you Spdif or AES. On the other hand, good luck hooking up your AVR…

 

[Keith_W]

I took Keith's original comment "takes a digital signal" to mean the signal after it performs it's initial A/D conversion of an analog input signal.

There are all sorts of DSP speakers out there. I owned a pair of Dynaudio Xeo's. These had multiple digital inputs (AES/EBU, HDMI, Toslink) and from memory, two analog inputs (RCA and 3.5mm). If you fed it a digital signal, it would simply send it on for processing. Give it an analog signal, and it would obviously have to do A/D conversion first. ASR can correct me if i'm wrong, but I am not aware of any DSP speaker that does not include an A/D stage given that it is inexpensive to implement and increases versatility for legacy devices.

Is that substantially different from feeding a digital signal to a DSP speaker that has a digital input? Yes, there is an additional D/A then A/D conversion in the total chain before the speaker DSP implements the digital crossover, but if the added conversions are transparent, is that material?

Different audio religions have different beliefs about whether A/D conversion is transparent or not. I myself am not sure, sometimes I think I hear a difference, sometimes not. I think the difference has more to do with the impedance ratio between the output of whatever analog source you are using and the input impedance of the A/D converter rather than A/D conversion itself. Then of course you need to be sure you are not driving the A/D converter into clipping, or feeding it a signal so low that you increase the noise floor. On balance, if it is properly implemented, I would lean towards so transparent that it is immaterial.

 

[Keith_W]

BTW, I thought I would share a little anecdote A few years ago, I sat down to listen to music on my system. I noticed that somehow the soundstage was pulled to the right. That's strange, so I went and put my ear over every driver. They were all working fine. I inspected all the connections (not easy when you have 8 DAC channels, 8 amplifier channels, and 8 drivers), they were all connected fine. Nothing was connected out of phase (again, not easy to check because ALL my interconnects are the same no-name IC with no red/white indicator to indicate polarity). I pulled out a mono recording, and the soundstage was definitely pulled to the right. Finally in exasperation, I pulled out the measurement setup and ran sweeps and overlaid my left and right speaker traces. To my astonishment, they were exactly the same!

So how could speakers which objectively measure the same from the listening position have soundstage pulled to the right? I think I must have driven some forum members nuts with my questions because none of them believed that there could possibly be a difference.

Finally someone suggested the problem was with my hearing. Bingo. Answer: earwax.

 

[TimW]

Some passive speakers measure better than some active speakers. Every speaker measures slightly different, there are no identical speakers with one being active and the other passive. You could say the best speakers currently available are active but how many of us can actually afford the best? If you look at speakers within a certain budget, there may be passive speakers that measure better than the active options. Room correction using DSP can be applied to any speaker system. You do not need active speakers to use room correction and every speaker system can benefit from room correction.

 

[Tangband]

Some passive speakers measure better than some active speakers. Every speaker measures slightly different, there are no identical speakers with one being active and the other passive. You could say the best speakers currently available are active but how many of us can actually afford the best? If you look at speakers within a certain budget, there may be passive speakers that measure better than the active options. Room correction using DSP can be applied to any speaker system. You do not need active speakers to use room correction and every speaker system can benefit from room correction.

You will have a hard time finding a speaker that measures better than a Genelec 8361, overall. Regardless of price.
10 000 dollars for a pair is a lot of money, but there are some terrible passive speakers out there for a lot more.

 

[TimW]

You will have a hard time finding a speaker that measures better than a Genelec 8361, overall. Regardless of price.
10 000 dollars for a pair is a lot of money, but there are some terrible passive speakers out there for a lot more.

I know the Olive score is not the end all be all for performance, but it certainly takes into account many factors important to sound quality. You can compare the spins with your own eyes and judge of course.

According to the Olive score, the $1500 per pair Ascend Acoustics Sierra LX measures as well as the 8361.

The $4000 per pair March Audio Sointuva measures slightly better.

And the $19,000 per pair KEF Reference 5 measures significantly better.

 

[Tangband]

Measurements of multitone distortion and compression - Genelec 8361 , outstanding results.

Genelec 8361A Measurements and Multitone comparison with other speakers.

Hi, I measured a pair of Genelec 8361A. And I have attached multitone distortion comparison data with other loudspeakers, so I hope enjoy it. Frequency response This is 3-way loudspeaker. I'm using near-field measurement(LF) and window gating(5ms, 1m distance) to get a frequency response...

www.audiosciencereview.com