conversion between crosstalk and separation

[Rantenti]

Hello! Hope you are all having a good day.
I was looking at the specs of various headphone amps.

Some specified crosstalk in -dB (eg. Topping A90) and some specified separation in % (eg Burson).

If crosstalk affects the separation and thus the soundstage, is there a formula or chart which I can convert crosstalk into separation or vice versa to compare different head amps?

Thanks.

 

[jcarys]

It's not quite that easy, but you can really just ignore it. All of the Topping amps reviewed here perform far beyond anything you could hear between channels. I don't know the Burson product line, but a quick check of specs shows they also are far beyond anything audible in crosstalk.
I would knock it off of your spec comparison.

 

[Arnas]

Just get Singxer sa1 for the peace of mind. It is as good with better parts and better crosstalk.

 

[voodooless]

Vinyl is touted as having fabulous imaging and soundstage, but crosstalk is only around -30dB. most decent amps have good crosstalk. Not something to specifically look for.

 

[dc655321]

If crosstalk affects the separation and thus the soundstage, is there a formula or chart which I can convert crosstalk into separation or vice versa to compare different head amps?

As already said, cross-talk is not typically an issue unless a device is broken.

I think those are different terms used to express the same idea, though perhaps with opposite sense.
i.e. like SINAD and THD+N. One is expressed a positive quantity (dB or %) the other negative (-dB).

-20dB = 0.1 = 10%
-40dB = 0.01 = 1%
.
.
-100dB = 0.00001 = 0.001%

 

[Killingbeans]

If crosstalk affects the separation and thus the soundstage

The low crosstalk in headphones actually sounds unnatural and fatiguing to a lot of people. That's why crossfeed is a thing.

The brain often favours the miserable channel separation you get from a speaker setup and/or antiquated analogue media.

 

[Rantenti]

Thanks for all your replies!

I guess our own two ears have much worse "crosstalk" than many of the modern headamps we have. It's surprising that vinyl give the perception of such a soundstage we love when its crosstalk is actually so "poor".

I wonder though, is the soundstage theoretically wider (despite being possibly unnatural to the ears) for a machine with eg.99.5% separation versus one with a -102dB crosstalk, or is it actually not determinable just from these numbers and that there are other factors in the design that have more decisive effects on the size of the soundstage/ imaging?

 

[DVDdoug]

And what factors in vinyl give it the perception of that fantastic soundstage we love if the crosstalk is so much "worse" than modern headamps?

Human imagination!

Soundstage (from a pair of speakers or headphones) is an illusion. And, there's a reason for the center channel in surround setups... It tends to create a better illusion and centered sounds can really come from the center.

And I agree, it's not something to worry about. In the vinyl days 20dB was enough to hear the sounds panned "completely right" or "completely-left", and "realistic" recordings weren't mixed that way anyway.

 

[RayDunzl]

I guess our own two ears have much worse "crosstalk" than many of the modern headamps we have

I surprised myself with this:

Make a file in Audacity. Left Channel 400Hz, right channel 405Hz.

Play through speakers, hear 5Hz "beating" as the tones cancel each other in the air. and at your ears.

Surprise Part:

Play back through headphones. One ear gets 400Hz, the other gets 405Hz. No cancellation of the tone in the air nor at either ear.

But I hear the beating, even though neither ear hears a modulated tone.

I suppose this indicates a type of "crosstalk" in your head.

 

[theREALdotnet]

The brain often favours the miserable channel separation you get from a speaker setup and/or antiquated analogue media.

Yes, and that we’re getting in the real world. There is never a sound we hear with just one ear.

 

[theREALdotnet]

Play back through headphones. One ear gets 400Hz, the other gets 405Hz. No cancellation of the tone in the air nor at either ear.

But I hear the beating, even though neither ear hears a modulated tone.

Yes, binaural beats are quite surprising, it is intermodulation through bone conduction (unlikely) or does our brain do the intermodulation?

It’s the first effect demonstrated in part 1 of this fascinating series:

 

[Rantenti]

I surprised myself with this:

Make a file in Audacity. Left Channel 400Hz, right channel 405Hz.

Play through speakers, hear 5Hz "beating" as the tones cancel each other in the air. and at your ears.

Surprise Part:

Play back through headphones. One ear gets 400Hz, the other gets 405Hz. No cancellation of the tone in the air nor at either ear.

But I hear the beating, even though neither ear hears a modulated tone.

I suppose this indicates a type of "crosstalk" in your head.

Very interesting! I guess it's a combination of bone conduction + further brain processing that is giving the "crosstalk" / "crossfeed"

 

[Rantenti]

Yes, binaural beats are quite surprising, it is intermodulation through bone conduction (unlikely) or does our brain do the intermodulation?

It’s the first effect demonstrated in part 1 of this fascinating series:

Ah ha, I just found you said it just before I did. I do have the chance to hear quite a lot of "beats" from acoustic musical instruments (sometimes poorly tuned) and it's surprising when sometimes the sound seem to come from somewhere in the air. And thanks for sharing the audio illusions video!

 

[Chrispy]

Thanks for all your replies!

I guess our own two ears have much worse "crosstalk" than many of the modern headamps we have. It's surprising that vinyl give the perception of such a soundstage we love when its crosstalk is actually so "poor".

I wonder though, is the soundstage theoretically wider (despite being possibly unnatural to the ears) for a machine with eg.99.5% separation versus one with a -102dB crosstalk, or is it actually not determinable just from these numbers and that there are other factors in the design that have more decisive effects on the size of the soundstage/ imaging?

Douglas Self had something about concern about crosstalk for a stereo vs dual mono amps....and it was something along the lines that better than 35 dB (IIRC) it just doesn't matter in that respect.....

 

[formula 977]

Hello! Hope you are all having a good day.
I was looking at the specs of various headphone amps.

Some specified crosstalk in -dB (eg. Topping A90) and some specified separation in % (eg Burson).

If crosstalk affects the separation and thus the soundstage, is there a formula or chart which I can convert crosstalk into separation or vice versa to compare different head amps?

Thanks.

I am not sure which one is separation and which one is crosstalk.

Signal in Left output R level
Signal in Right output L level

L-R: Signal L+R- output R-L
R-L: Signal L+R- output L-R

In need of some additional input from the members.