[analogsurviver]

   
There is a test with simulated crosstalk in this old thread. No ABX results have been posted, though, either because of the lack of interest, or the difference being much more difficult to detect than people expected.
 
In any case, if you view for example -40 dB crosstalk as 1% narrower sound stage, it is hardly something to be particularly worried about. That figure (and usually much better) can easily be achieved by any decent electronics for typical home listening with headphones or speakers, as long as analog sources like tape, vinyl, or FM radio are not involved.

Dream on - both vinyl and FM can exceed 60 dB of channel separation - but admittedly only by few cartridges and tuners. Both are more limited by the source programme - where vinyl cutting mastering engineers can quickly knock this down to 25 dB - or less.  Better FM tuners http://sportsbil.com/tandberg/tpt-3001-b.pdf ( when tested by Audio in US, it exceeded specs )are equally generally outperforming the specs for the transmitters - but few of the transmitters are good enough to 60 dB and beyond channel separation.   
 
The crosstalk in one of the best vinyl test records - EVER - has been measured by a scanning electron microscope to be on the order of -35 dB. Techniques to cancel this inherent crosstalk to below 60 dB are in place for at least 30 years - and ARE being used. In commercially available products. Today. At a price. http://users.telenet.be/jallaerts/welcome.htm ( check top model MC 2 Formula 1 ).
 
But, yes, -40 dB is where things start to level off regarding improvement of channel separation - and this is the reason why there was recently a resurgence of hardware and software for the correct adjustment of  azimuth for phono cartridges - because -35 dB or better channel separation is  within approximately one third of a degree in angle - which can not be "eyeballed" - even in case of 100% geometrically correct cartridge ( 1 out of 100 - IF you are - lucky ...). You can get more information regarding audibility of even the smallest amount of amplitude AND PARTICULARLY PHASE of the crosstalk signal in the phono cartridge azimuth alignment process descriptions/instructions - then any place else. http://www.hi-fiworld.co.uk/vinyl-lp/37-technology/161-adjust-part-i.html
 
Except that it can be done by the use of only a test record and an ANALOG microscope  oscilloscope - if you know what to do. I have been doing it for nearly 30 years.

 

[StanD]

Seems to be too much effort to get proper separation in a vinyl system when one cannot find much if any content with the requisite separation. Doesn't seem to be worth it, reminds me of BetaMax, only far less available as in almost nothing.

 

[analogsurviver]

  Seems to be too much effort to get proper separation in a vinyl system when one cannot find much if any content with the requisite separation. Doesn't seem to be worth it, reminds me of BetaMax, only far less available as in almost nothing.

Wrong.
 
Go to any dealer prepared to demonstrate to you - on your records - how a properly adjusted quality cartridge can sound. It should result in separation better than 40 dB - which is, for all practical purposes, good enough.
 
Transfering older tape originated analog recordings to any digital will also transfer this channel separation - or lack thereof. Today, both analog and digital can play any channel separation was available on analog tape.
 
As said, getting down to -60 dB separation level is possible - which should be "good enough".
But it is not -100 or more dB.

 

[StanD]

  Wrong.
 
Go to any dealer prepared to demonstrate to you - on your records - how a properly adjusted quality cartridge can sound. It should result in separation better than 40 dB - which is, for all practical purposes, good enough.
 
Transfering older tape originated analog recordings to any digital will also transfer this channel separation - or lack thereof. Today, both analog and digital can play any channel separation was available on analog tape.
 
As said, getting down to -60 dB separation level is possible - which should be "good enough".
But it is not -100 or more dB.

Not enough content + too much hassle = not worth the effort. If one is into it, then it might good for them for the usual audiophile reasons.
Possible separation levels and maintaining this performance is IMO in this case just not worth the hassle.

 

[analogsurviver]

  Not enough content + too much hassle = not worth the effort. If one is into it, then it might good for them for the usual audiophile reasons.
Possible separation levels and maintaining this performance is IMO in this case just not worth the hassle.

Equally, I could write 
 
no content above 20 kHz = not worth the effort with CD.
 
Yet I did chose to use it, making the best of my efforts, to the best of its abilities - although it can never be a match for the best analog has to offer. Learning a few things along the way.

 

[StanD]

  Equally, I could write 
 
no content above 20 kHz = not worth the effort with CD.
 
Yet I did chose to use it, making the best of my efforts, to the best of its abilities - although it can never be a match for the best analog has to offer. Learning a few things along the way.

And I will agree that any content above 20 Khz is not worth the effort on any medium. Of course you are distracting from the point I made by going off on a tangent.

 

[Ruben123]

Thanks all. Is the lower channel separation in db better than higher? Because you're talking about 40 db and I read the Sansa Clip has around -88.

 

[analogsurviver]

Thanks all. Is the lower channel separation in db better than higher? Because you're talking about 40 db and I read the Sansa Clip has around -88.

Good enough

 

[stv014]

Thanks all. Is the lower channel separation in db better than higher? Because you're talking about 40 db and I read the Sansa Clip has around -88.

 
Higher (in terms of absolute dB value) is better, and -88 dB crosstalk is plenty good enough. Although it will likely be worse when loaded with low impedance headphones, it still should not be an issue.

 

[KeithEmo]

Quote:

  False.
 
http://www.earthworksaudio.com/microphones/qtc-series-2/qtc50/
 
Just look at the specs for QTC 30, QTC40, QTC50. There is always a matched pair - with even tighter specs.
 
And there is M50 - a selected basically QTC 50 . Also available as matched pair.
 
There are other makes - but with less info available online.

 
If you're recording something with a stereo pair of microphones, then, if those microphones aren't exactly matched in terms of frequency response, distortion, and phase, the image is going to "pull" to the left or right. Since any product, including microphones, is going to vary over time and production runs, if you want two microphones to be identical in all these characteristics, then they're going to need to be matched (and it most certainly won't hurt anything). Even if you had two microphones that were flat to a fraction of a dB (which would be pretty rare), if one has a peak where the other has a dip, they're not going to image perfectly. (Even if you buy two with sequential serial numbers, which many people seem to consider important, they could be made with the last capsule produced from "batch 11" and the first from "batch 12", which could be made with some slightly different part, or some production run change, or one could have had a part replaced during production testing which could be slightly different, so it does make sense to buy two that are actually matched.)
 
Now, if you're going to do full milti-tracking, where you aren't going to use a pair of stereo mics in stereo anyway, then it shouldn't matter nearly as much.At that point, it's simply a matter of consistency. However, as someone doing recording, even just being able to assume that two of your microphones that are supposedly the same really sound identical without having to confirm it or adjust anything would still make life easier.

 

[Roly1650]

Thanks all. Is the lower channel separation in db better than higher? Because you're talking about 40 db and I read the Sansa Clip has around -88.

The posters have been using shorthand, the negative sign is assumed to be there, as crosstalk wouldn't ever be a positive number, it's the signal level below the opposite channel. So, to us non EE's it's actually -40 dB and the Sansa Clip is a whole magnitude better than that.

 

[analogsurviver]

  And I will agree that any content above 20 Khz is not worth the effort on any medium. Of course you are distracting from the point I made by going off on a tangent.

Sigh...
 
This folly about beyond 20 kHz not mattering has to stop.
 
I found a good sample of acoustic guitar and voice from DPA microphones - NOT stating in which resolution it is. Downloading reveals it is 96/24 - and it has extremely low noise up to 48 kHz limit - with dynamic range between 20 and 50 kHz between 30 and 40 dB. Too bad I did not bookmark it ( it was among few clips posted to Gearslutz - in order to compare various makes and models of microphones - some of which are of interest to me ).
 
From a voice and acoustic guitar... - imagine what would come out with recording in 192/24 or above of some percussion, muted trumpet, violin, etc. 
 
Because of this folly "everything over 20 kHz is baaaaaaad" - even the DPA decided to "adjust" their specs:
http://www.dpamicrophones.com/en/products.aspx?c=item&category=234&item=24385#specifications
Those who know the real deal from the day the same mic (with a bit lesser electronics ) and the SAME capsule got introduced as Bruel & Kjear 4006 back in 1980 or so can - only roll on the floor laughing at this 20 -20K spec.
 
As I said, this MUST  end. It is exactly the same situation as with HD-800s being accused of who only knows what - on the ground of having too extended frequency response. 
 
NOBODY dared to contest that HD 800 post as having to great frequency response post - because you all know that it was right.

 

[analogsurviver]

   
If you're recording something with a stereo pair of microphones, then, if those microphones aren't exactly matched in terms of frequency response, distortion, and phase, the image is going to "pull" to the left or right. Since any product, including microphones, is going to vary over time and production runs, if you want two microphones to be identical in all these characteristics, then they're going to need to be matched (and it most certainly won't hurt anything). Even if you had two microphones that were flat to a fraction of a dB (which would be pretty rare), if one has a peak where the other has a dip, they're not going to image perfectly. (Even if you buy two with sequential serial numbers, which many people seem to consider important, they could be made with the last capsule produced from "batch 11" and the first from "batch 12", which could be made with some slightly different part, or some production run change, or one could have had a part replaced during production testing which could be slightly different, so it does make sense to buy two that are actually matched.)
 
Now, if you're going to do full milti-tracking, where you aren't going to use a pair of stereo mics in stereo anyway, then it shouldn't matter nearly as much.At that point, it's simply a matter of consistency. However, as someone doing recording, even just being able to assume that two of your microphones that are supposedly the same really sound identical without having to confirm it or adjust anything would still make life easier.

+1.
 
That's why , when recording a super rare Guarnieri (and priceless, estimated to be worth XY M$ for insurance purposes only ) violin in Australia , THE reference stereo pair from Bruel & Kjaer was flown in from Denmark - by the corporate jet. And flew back to the factory in Naerum the minute after finished recording. 
 
It, too, is priceless.
 
In multimiking, it is next to irrelevant to have matched pairs ...

 

[StanD]

  Sigh...
 
This folly about beyond 20 kHz not mattering has to stop.
 
I found a good sample of acoustic guitar and voice from DPA microphones - NOT stating in which resolution it is. Downloading reveals it is 96/24 - and it has extremely low noise up to 48 kHz limit - with dynamic range between 20 and 50 kHz between 30 and 40 dB. Too bad I did not bookmark it ( it was among few clips posted to Gearslutz - in order to compare various makes and models of microphones - some of which are of interest to me ).
 
From a voice and acoustic guitar... - imagine what would come out with recording in 192/24 or above of some percussion, muted trumpet, violin, etc. 
 
Because of this folly "everything over 20 kHz is baaaaaaad" - even the DPA decided to "adjust" their specs:
http://www.dpamicrophones.com/en/products.aspx?c=item&category=234&item=24385#specifications
Those who know the real deal from the day the same mic (with a bit lesser electronics ) and the SAME capsule got introduced as Bruel & Kjear 4006 back in 1980 or so can - only roll on the floor laughing at this 20 -20K spec.
 
As I said, this MUST  end. It is exactly the same situation as with HD-800s being accused of who only knows what - on the ground of having too extended frequency response. 
 
NOBODY dared to contest that HD 800 post as having to great frequency response post - because you all know that it was right.

And you keep beating the dead ultrasonic horse, primarily as a distraction when you can't deal with another issue. This is one thing that might tick people off. Try to stay on topic.

 

[Ruben123]

The posters have been using shorthand, the negative sign is assumed to be there, as crosstalk wouldn't ever be a positive number, it's the signal level below the opposite channel. So, to us non EE's it's actually -40 dB and the Sansa Clip is a whole magnitude better than that.

OK, I would have guessed the ''perceived headstage'' of audioplayers like the Sansa could be explained by that... BUT if the differences are quite small and -88 is already really good I think it has nothing to do with the ''headstage'' differences between audio players.
 
So how could the Sansa's headstage be small then, ignoring placebo.