ARRRGGGHHH!!!
Where do I start..........
OK........75 ohm cable has less loss than similar 50 ohm cable. The guy kvetching about replacing coax with CAT 5 is raising the impedance. Twisted pairs start at around 100 ohms, and go higher.
OK.....we aren't talking aboout 1/4 wave dipoles.......75 ohms is used because it has less loss in a DISTRUBUTED cable system, where loss is important.
So......you wanna show me how I can't measure AND hear the difference between 50 ohm and 75 ohm connectors. Bring your $$$$$.........I need it. [joke]
And just what did this Pro Audio guy say???? I think he mispoke.
Type F connectors......like many other connectors designed for high frequency use........do not have a centre pin on the male connector: it uses the centre conductor of the cable. At very high frequencies, it makes a difference.
And don't feed me that crap that Canare has 75 ohm RCAs, 'cuz the Laws of Physics say it ain't so. So there.
Jocko
Where do I start..........
OK........75 ohm cable has less loss than similar 50 ohm cable. The guy kvetching about replacing coax with CAT 5 is raising the impedance. Twisted pairs start at around 100 ohms, and go higher.
OK.....we aren't talking aboout 1/4 wave dipoles.......75 ohms is used because it has less loss in a DISTRUBUTED cable system, where loss is important.
So......you wanna show me how I can't measure AND hear the difference between 50 ohm and 75 ohm connectors. Bring your $$$$$.........I need it. [joke]
And just what did this Pro Audio guy say???? I think he mispoke.
Type F connectors......like many other connectors designed for high frequency use........do not have a centre pin on the male connector: it uses the centre conductor of the cable. At very high frequencies, it makes a difference.
And don't feed me that crap that Canare has 75 ohm RCAs, 'cuz the Laws of Physics say it ain't so. So there.
Jocko
Cable loss is not simply related to the characteristic impedance of a cable.
To quote from a cable website:
How to minimize the attenuation in coax ?
For a line with fixed outer conductor diameter, and whose outer and inner conductors have the same resistivity, and assuming you use a dielectric with negligible loss (such as polyethylene or Teflon in the high-frequency range at least), then you get minimum loss in coax if you minimize the expression:
(1/d + 1)/ln(1/d)
where d is the ratio of inner conductor diameter to outer conductor ID. A spreadsheet or calculator gets you close pretty quickly: D/d = 3.5911 is close. Thr formula was claimed to be derived from the formula for coax impedance versus D/d and a formula for loss that you'll find in "Reference Data for Engineers" published by Howard Sams, on pg. 29-13 in the seventh edition.
The interesting thing to notice is that this minimum loss does not directly yield a line impedance: the line impedance depends on the dielectric constant of the dielectric. For air insulated line, the corresponding impedance is about 76.71 ohms, but if the line is insulated with solid polyethylene, then minimum attenuation is at about 50.6 ohms. So, however it came to be, all the RG-58 we use for antenna feeds and test equipment connections is pretty close to minimum attenuation given the above conditions, and that the dielectric is polyethylene.
But if the line uses foam dielectric with a velocity factor of 0.8, then the impedance of minimum atten would be about 61 ohms. However, that minimum is a pretty broad one, and you don't start loosing a lot till you get more than perhaps 50% away from the optimal impedance.
Note that foam-dielectric line with the same impedance and outer diameter as solid-dielectric line will have lower loss. That's because, to get the same impedance, the foam line will have a larger inner conductor, and that larger conductor will have lower RF resistance, and therefore lower loss.
To quote from a cable website:
How to minimize the attenuation in coax ?
For a line with fixed outer conductor diameter, and whose outer and inner conductors have the same resistivity, and assuming you use a dielectric with negligible loss (such as polyethylene or Teflon in the high-frequency range at least), then you get minimum loss in coax if you minimize the expression:
(1/d + 1)/ln(1/d)
where d is the ratio of inner conductor diameter to outer conductor ID. A spreadsheet or calculator gets you close pretty quickly: D/d = 3.5911 is close. Thr formula was claimed to be derived from the formula for coax impedance versus D/d and a formula for loss that you'll find in "Reference Data for Engineers" published by Howard Sams, on pg. 29-13 in the seventh edition.
The interesting thing to notice is that this minimum loss does not directly yield a line impedance: the line impedance depends on the dielectric constant of the dielectric. For air insulated line, the corresponding impedance is about 76.71 ohms, but if the line is insulated with solid polyethylene, then minimum attenuation is at about 50.6 ohms. So, however it came to be, all the RG-58 we use for antenna feeds and test equipment connections is pretty close to minimum attenuation given the above conditions, and that the dielectric is polyethylene.
But if the line uses foam dielectric with a velocity factor of 0.8, then the impedance of minimum atten would be about 61 ohms. However, that minimum is a pretty broad one, and you don't start loosing a lot till you get more than perhaps 50% away from the optimal impedance.
Note that foam-dielectric line with the same impedance and outer diameter as solid-dielectric line will have lower loss. That's because, to get the same impedance, the foam line will have a larger inner conductor, and that larger conductor will have lower RF resistance, and therefore lower loss.
Simple & cheap
Hi !
I made my spdif-cable with normal 75ohm-antennacable, soldered
rca-plugs to both sides (nice golded) and ready ! It is 7meters longs,
and never showed up any problems. I don't hear any difference in
comparison with a 50cm toslink optical one...
DTS for example uses checksums to detect defect data (DTS-CD) and
simply drops frames if defect. I had never a single drop with this
7meter cable.
Mike
Hi !
I made my spdif-cable with normal 75ohm-antennacable, soldered
rca-plugs to both sides (nice golded) and ready ! It is 7meters longs,
and never showed up any problems. I don't hear any difference in
comparison with a 50cm toslink optical one...
DTS for example uses checksums to detect defect data (DTS-CD) and
simply drops frames if defect. I had never a single drop with this
7meter cable.
Mike
You are right.....
It isn't that simple. But someone invented 93 ohm coax........because they needed a lower loss one.
(Capacitance goes down as Z goes up. That part is that simple.....)
The point is some guy complained about coax, and bragged how muchj better twisted pair was, without having any knowledge of transmission lines.
I do......and yes, I can measure the difference. And I have demonstrated that people can hear the difference. If you want to believe that all cables sound the same on your SPDIF setup, then that is fine by me, 'cuz you surely won't be buying any cables from me anytime soon.
Right?
Jocko
It isn't that simple. But someone invented 93 ohm coax........because they needed a lower loss one.
(Capacitance goes down as Z goes up. That part is that simple.....)
The point is some guy complained about coax, and bragged how muchj better twisted pair was, without having any knowledge of transmission lines.
I do......and yes, I can measure the difference. And I have demonstrated that people can hear the difference. If you want to believe that all cables sound the same on your SPDIF setup, then that is fine by me, 'cuz you surely won't be buying any cables from me anytime soon.
Right?
Jocko
hmm, maybe i am interested in your cables if my amp is finished ? 
But one thing i never really understood, spdif is a digital connection,
transfering 0s and 1s, as long data is not corrupted how can different
cables change the sound ? Even if they measure different, as long
as no data gets lost, there shouldn't be a difference. But, the only
possibility for affecting the sound would be if the carrier is used as
samplerate, without any correction. But wouldn't this be nuts ?
Just for comparison, i am downloading music from the net, nobody
would have the idea that the quality of the modemcable affects the
quality of this music ?
Mike
But one thing i never really understood, spdif is a digital connection,
transfering 0s and 1s, as long data is not corrupted how can different
cables change the sound ? Even if they measure different, as long
as no data gets lost, there shouldn't be a difference. But, the only
possibility for affecting the sound would be if the carrier is used as
samplerate, without any correction. But wouldn't this be nuts ?
Just for comparison, i am downloading music from the net, nobody
would have the idea that the quality of the modemcable affects the
quality of this music ?
Mike
Re: You are right.....
I do......and yes, I can measure the difference. And I have demonstrated that people can hear the difference. If you want to believe that all cables sound the same on your SPDIF setup, then that is fine by me, 'cuz you surely won't be buying any cables from me anytime soon.
----------------------------------------------------------------------------------
There are some guys over at audio asylum going ga ga about the Benchmark DAC being totally cable independent and jitter free! In my experience, those who claim that cables don't sound different have some weakness in their system. The more sophisticated the setup, the more it exposes cable differences. And direction too.
ps: Jocko; thought you were gone?
I do......and yes, I can measure the difference. And I have demonstrated that people can hear the difference. If you want to believe that all cables sound the same on your SPDIF setup, then that is fine by me, 'cuz you surely won't be buying any cables from me anytime soon.
----------------------------------------------------------------------------------
There are some guys over at audio asylum going ga ga about the Benchmark DAC being totally cable independent and jitter free! In my experience, those who claim that cables don't sound different have some weakness in their system. The more sophisticated the setup, the more it exposes cable differences. And direction too.
ps: Jocko; thought you were gone?
Not gone.....
Just busy trying to keep the doors open. Fixin' to get busy again in a few days. More money going out, but hopefully some will come back this time.
Anyway.........right......1's and 0's....that is all it is.
Until you factor in that you have to recover the clock, with a PLL, from a serial data stream.
Which means jitter......
And jitter has a value, and more importantly, a spectrum.
Once you understand that, it is obvious why cables and interface problems affect sound.
How exactly........well, no one can really explain that part. But the better the interface, the "less digital" the system will sound.
Ideally, if there were zero reflections, and no jitter from the PLL, all cables would sound the same.
But we are not in an ideal world.
Jocko
Just busy trying to keep the doors open. Fixin' to get busy again in a few days. More money going out, but hopefully some will come back this time.
Anyway.........right......1's and 0's....that is all it is.
Until you factor in that you have to recover the clock, with a PLL, from a serial data stream.
Which means jitter......
And jitter has a value, and more importantly, a spectrum.
Once you understand that, it is obvious why cables and interface problems affect sound.
How exactly........well, no one can really explain that part. But the better the interface, the "less digital" the system will sound.
Ideally, if there were zero reflections, and no jitter from the PLL, all cables would sound the same.
But we are not in an ideal world.
Jocko
Still don't believe me?????
Here is a plot from my 40 year old TDR. I know that most of you won't understand it, but you can see how much a 50 ohm part on a 75 ohm cable mucks up stuff.
BTW......most input stages don't have a rho (reflection coefficient) as low as 0.027. Multiply that number by about 10............
Signoro TDR, aka Jocko
Here is a plot from my 40 year old TDR. I know that most of you won't understand it, but you can see how much a 50 ohm part on a 75 ohm cable mucks up stuff.
BTW......most input stages don't have a rho (reflection coefficient) as low as 0.027. Multiply that number by about 10............
Signoro TDR, aka Jocko
Attachments
And PCB traces, and the cables IN the player, huh?
..........
My Sony home theater came with a standard cable with orange plugs for SPDIF. And it's replaced with a Acoustic Research cable, that I haven't decided to buy since it's my cousin who got it + receiver+DVD for my mother. There's NO difference between them, but it remains true that it's
...
Sony
But seriously.....
I don't think that they ever expected that anyone would use it, and if they did............TOSLINK was good enough!
If they had their way in the first place, CDs wold have been limited to 15 kHz response, and 32 Khz sampling rate.
(One of the few things I recieved from all the money I wasted on AES dues.)
Speaking of AES........
The only thing worse than SPDIF was the original version of AES/EBU. I tried to tell them, but no.........................
Eventually, they fixed the obvious problems. Except for XLR connectors..........
Jocko
I don't think that they ever expected that anyone would use it, and if they did............TOSLINK was good enough!
If they had their way in the first place, CDs wold have been limited to 15 kHz response, and 32 Khz sampling rate.
(One of the few things I recieved from all the money I wasted on AES dues.)
Speaking of AES........
The only thing worse than SPDIF was the original version of AES/EBU. I tried to tell them, but no.........................
Eventually, they fixed the obvious problems. Except for XLR connectors..........
Jocko
Okay, but this means, that if the DAC fed by the spdif uses the
samplerate delivered by the spdif-carrier directly, jitter gets audible.
If it uses the carrier only for syncronizing, recreating its own samplerate
with its own quartz, it becomes completely cableindependent, unless
the cable does not destroy bits, right ?
I ask this, because typically these signals are fed into a dsp, (at least
for HT-amplifiers) automatically rebuilding the samplerate...
This would explain why i heard no difference with different digitalcables.
Mike
samplerate delivered by the spdif-carrier directly, jitter gets audible.
If it uses the carrier only for syncronizing, recreating its own samplerate
with its own quartz, it becomes completely cableindependent, unless
the cable does not destroy bits, right ?
I ask this, because typically these signals are fed into a dsp, (at least
for HT-amplifiers) automatically rebuilding the samplerate...
This would explain why i heard no difference with different digitalcables.
Mike
---------------------------------------------------------------------------------MikeB said:
Not necessarilly; even a word clock signal cable on it's own affects the sound. I have a 3 wire sdif (not spdif!) 75Ohm connection to my dCS 954 and not only a5re sonics cable dependent but cable direction dependent.
Measuring the signals correctly terminated give almost perefect wavefroms and better than any dac I have seen!!
I think Jocko is probably right.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.