the dB/octave specification of the crossover?


Hello,
I have a dumb question the dB/octave specification of the crossover. A speaker designer wrote on his Facebook that dB/octave is like the resolution of a television. For example, for the same set of drivers, if we apply a crossover with 12dB/octave, the sound won’t be as detailed as with a 50dB/octave crossover. If this is true, I wonder why there are many speakers with 12dB/octave only. Why isn't there a race in dB/octave just like 4k, 8k TV?
Thank you. 
Huy
Ag insider logo xs@2xquanghuy147
it really depends on the drivers involved.  
a blend between a woofer and tweeter needs to "sum" to fill in the frequencies through the crossover range.  
if both drivers are cut off sharply at the crossover frequency that missing summing could come off as recessed and lacking detail.  
but there are exceptions to the rule, especially if the drivers have a rising response or extended response beyond the crossover.
if a driver has a natural acoustic roll-off at the crossover and you compound that with a steep crossover- e.g. 24 db per octave, then yes the system could sound dull and recessed.  
Or, how about going fully active? Buy a digital cross-over from the very high quality likes of Xilica (without costing a fortune), and make slopes, Q’s, delays, gains etc. on the fly, from the listening position. You’ll need extra amps for each pair of drivers/frequency section, sure, but they needn’t be as expensive being freed of looking into a more or less heavily driven passive XO and instead firing directly into the driver. Slopes up to 48dB/octave, without the added complexity of a passive design, and with the filtering done pre-amplification on signal level there’s no meddling with a full-wattage input (into the filter) and fluctuating filter values.

I'm using a fully active set-up myself with a Xilica digital XO (36dB/octave L-R for mains and 36dB/octave Butterworth high-pass on the subs), and it's a sonic delight. 
How can Kenjit call himself a speaker designer  when he's never actually done it?
Oh yeah.....fb.
@erik_squires  I have been thinking of building speakers with full-range Audio Nirvana drivers (which doesn't require any crossover) for my parents. 

The Sea A26 kit looks to be more room friendly and really nice. I will check it out. 
Thank you. 
There are many who are proponents of 1st order, time coincident crossovers as well. Vandersteen and Thiel among them.

If you are curious, build some. :) How about the Seas A26 kit?  Has 1 capacitor and one resistor for a single order (6dB/octave) slope on the tweeter only. Can easily be converted to 2nd order (12db/octave) as an experiment.

https://www.madisoundspeakerstore.com/2-way-speaker-kits/seas-a26-10-2-way-kit-pair-based-on-the-cla...

Thank you everyone for the valuable input!

On his facebook, he builds speakers with TAD, JBL, Altec drivers. His crossovers looks very massive and impressive. I have been following his page until he made this statement which puzzled me a lot because there exist some hi-end speakers with only 12dB/octave.

I'm not an electrician either so I will check out Danny Richie video clips about crossover. 

@twoleftears :  I won't :)
I have a dumb question the dB/octave specification of the crossover. A speaker designer wrote on his Facebook that dB/octave is like the resolution of a television. For example, for the same set of drivers, if we apply a crossover with 12dB/octave, the sound won’t be as detailed as with a 50dB/octave crossover.

Bullshit. :)  Utter and complete nonsense.

The crossover slopes are chosen to be complementary to the other drivers as well as to minimize interference from one driver to another as you go off axis.

It does not improve your resolution in the way it is stated. There are some benefits to high slope crossovers, but even Joseph Audio, the most well known proponent of high slope crossovers seems to have decided they are not that useful.*


Best,

E

* I'm going from recent measurements of his speakers, which seem to use more conventional crossover designs.

Danny Richie explains it all, in language we non-engineers can understand.

Of course he can,  he's not an engineer. At least, last I heard. He is a speaker designer. So he also designs most of the parts for his and some other speakers he's repaired..

He is an easy listen for sure..

The whole 6,12,18,24,48 db is how shallow or steep the slope is.

EX: 6db first order at 100 hz will still be working all the way into 200 hz or higher, though VERY diminished SPL.
At 48, 125 or so..... Stuff can get real BOOMY, with the SLOW roll off.
A couple answers, control the overshoot, or, quicken the roll off, add a (PR). there is a few ways..

Regards
A lousy metaphor indeed!
Where is the analogy between the resolution of a TV screen and the attenuation slope of a xover???
For a nice explanation of crossover design, watch the GR Research Tech Talk Tuesday videos on YouTube. Danny Richie explains it all, in language we non-engineers can understand. His videos cover a lot of loudspeaker design topics, a few of them specifically on x/o design.
That’s a weird and misleading analogy. A steeper slope will result in fewer competing frequencies which can allow a driver to sound more refined in some cases, i.e. filtering low freqs from a midrange unit that would otherwise be subject to high excursion.

In actuality, the first-order (6db) slope could be considered higher resolution since it’s not filtering out as much bandwidth and is less subject to phase issues. Downside is it might begin to distort at lower SPLs.

A better analogy is that a steeper slope could result in an experience similar to screen calibration, though it really depends on the whole speaker and crossover design in question.