In a nutshell:
No. Jitter is actually worse the way most are implemented.
There are ways around that too.................and the solution is analogous to the copper solution.
Jocko
Jocko,
Check out the 1500.00 audio clock at http://www.performanceaudio.com/cgi/pages.cgi?buildPage=item&ProductID=000109
You think one could de done cheaper.
Guido,
I do Telecom we spent a lot of time look at DS3 cables and connectors 44mbit stuff. In addition, I worked on DWDM optical where we could transmit 2.5 terabit on a signal fiber now that a real hassle. 😉 once you pull the cable from the connector it’s now to dirty to work.
Check out the 1500.00 audio clock at http://www.performanceaudio.com/cgi/pages.cgi?buildPage=item&ProductID=000109
You think one could de done cheaper.
Guido,
I do Telecom we spent a lot of time look at DS3 cables and connectors 44mbit stuff. In addition, I worked on DWDM optical where we could transmit 2.5 terabit on a signal fiber now that a real hassle. 😉 once you pull the cable from the connector it’s now to dirty to work.
Dr. Detroit
" Who do you trust to work on you, a research scientist or a actual doctor with 20 years of working on patients? "
I have been a practicing engineering for 20 years. I also am still learning and thinking, as is any decent engineer. My research into digital cables came about because cables from different manufactures sounded different, even though the "experts" said they shouldn't! The effect of jitter on digital audio signals is well known and AES papers have been written about it. The need for keeping RFI out of audio amplifiers and line level circuits is even more well known. This is not magic, it is just good engineering. It is ashamed that a post production engineer is not very concerned about what something sounds like or familiar with jitter's effect on digital audio. Maybe that is why there are so many bad sounding digital recordings.
PS I can see the difference between video cables on even a 27" Sony
and an inexpensive DVD player. One would hope a post production house would have a higher resolution system than that........
" Who do you trust to work on you, a research scientist or a actual doctor with 20 years of working on patients? "
I have been a practicing engineering for 20 years. I also am still learning and thinking, as is any decent engineer. My research into digital cables came about because cables from different manufactures sounded different, even though the "experts" said they shouldn't! The effect of jitter on digital audio signals is well known and AES papers have been written about it. The need for keeping RFI out of audio amplifiers and line level circuits is even more well known. This is not magic, it is just good engineering. It is ashamed that a post production engineer is not very concerned about what something sounds like or familiar with jitter's effect on digital audio. Maybe that is why there are so many bad sounding digital recordings.
PS I can see the difference between video cables on even a 27" Sony
and an inexpensive DVD player. One would hope a post production house would have a higher resolution system than that........
Since he didn't mention it
Jocko ran the lab for characterization of fiber optics for a very large well know telecom company. He knows about fiber optics.
Jocko Homo said:
Jocko ran the lab for characterization of fiber optics for a very large well know telecom company. He knows about fiber optics.
So I guess that the Sharp / Toshiba Rx/Tx units need to be terminated at their digital(electrical) connections? Same impedance?
Arne K
No.............we are talking about something else.........on the optical side.
The TOSLINK is cheap multimode stuff. Yes, the same problem exists, but always will with that stuff. Crappy rise/fall times, too.
Yes...........Phred is right........we even went as far as measuring the equivalent bandwidth of the fiber. It was called "chromatic dispersion", but there are formulas for getting the bandwidth.
And in case anyone cares..................
The world's first single-mode, long-haul fiber system was put in about.........'84.........between NYC and DC. Ran at 405 Mb/s. Cutting edge stuff back then, and I got stuck testing all of it it in the lab before it shipped to the field.
Yeah, I know......seems dog slow today. The bandwidth ran around 3.2 GHz for a repeater span.........25 km or so.
Jocko
The TOSLINK is cheap multimode stuff. Yes, the same problem exists, but always will with that stuff. Crappy rise/fall times, too.
Yes...........Phred is right........we even went as far as measuring the equivalent bandwidth of the fiber. It was called "chromatic dispersion", but there are formulas for getting the bandwidth.
And in case anyone cares..................
The world's first single-mode, long-haul fiber system was put in about.........'84.........between NYC and DC. Ran at 405 Mb/s. Cutting edge stuff back then, and I got stuck testing all of it it in the lab before it shipped to the field.
Yeah, I know......seems dog slow today. The bandwidth ran around 3.2 GHz for a repeater span.........25 km or so.
Jocko
Must have all been TDM, no ATM layer back in 84.
So can toss link work better if single a chunk of decent single mode is used instead of the Spencer Gifts class multimode stuff? I suspect this would require swapping out the diode and receiver. Humph, makes BNC's seem kind of reasonable.
So can toss link work better if single a chunk of decent single mode is used instead of the Spencer Gifts class multimode stuff? I suspect this would require swapping out the diode and receiver. Humph, makes BNC's seem kind of reasonable.
I use standart av cable to connect my Creative sound card with Dolby digital decoder.
I don't have any problem... I can't hear any distortion!!!
I don't have good speakers and subwoofer (from Germany Clatronic) may be this is reason
But so far as I know digital connection (spdif) works or not
I wanna ask something telecom specialist friends Does SPDIF transmit clock signal and data from different channels(dif. freqency band) If you use not coax or low quality one that cause distortion or jitter effect ?
PS:Sorry my bad eng.
I don't have any problem... I can't hear any distortion!!!
I don't have good speakers and subwoofer (from Germany Clatronic) may be this is reason
But so far as I know digital connection (spdif) works or not
I wanna ask something telecom specialist friends Does SPDIF transmit clock signal and data from different channels(dif. freqency band) If you use not coax or low quality one that cause distortion or jitter effect ?
PS:Sorry my bad eng.
You just can't change optical cables.
Multimode has a much large core and a different size clading size that single mode fiber. So the transmitter, receivers and the single mode fiber cable will not align.
Anyway, it's been proven that optical connection does not sound as good a the copper even on the AT&T fiber audio stuff.
Multimode has a much large core and a different size clading size that single mode fiber. So the transmitter, receivers and the single mode fiber cable will not align.
Anyway, it's been proven that optical connection does not sound as good a the copper even on the AT&T fiber audio stuff.
Re: Dr. Detroit
I use only quality cables and connectors that insure rf and emi shielding, that is not what the question is. We run data rates of 1.485Gb for HD serial data and rf and emi are concerns that we have dealt with. It is your imagination that came up with the idea that I didn't care about shielding or jitter. So stick to the facts and not your desire to insult people. The levels of jitter encountered on the pro equipment used in production is for the most part well under control. Most units have excellent reclocking on their inputs and all the DAs used reclock. All long runs are 75ohm not 110ohm and all power is filtered through a giant UPS for all tech sources. There is much more but I'm sure you know all about it since you are so far ahead of the rest of us.
Fred Dieckmann said:
I use only quality cables and connectors that insure rf and emi shielding, that is not what the question is. We run data rates of 1.485Gb for HD serial data and rf and emi are concerns that we have dealt with. It is your imagination that came up with the idea that I didn't care about shielding or jitter. So stick to the facts and not your desire to insult people. The levels of jitter encountered on the pro equipment used in production is for the most part well under control. Most units have excellent reclocking on their inputs and all the DAs used reclock. All long runs are 75ohm not 110ohm and all power is filtered through a giant UPS for all tech sources. There is much more but I'm sure you know all about it since you are so far ahead of the rest of us.
Mark,
Have you ever had the opportunity to put a TDR on 1.4 Gb signal and check it for reflections. The faster you go the more important it dose become. I am guessing that your work is in a recording studio and not in a lab.
🙂
Have you ever had the opportunity to put a TDR on 1.4 Gb signal and check it for reflections. The faster you go the more important it dose become. I am guessing that your work is in a recording studio and not in a lab.
🙂
???????????
"It is your imagination that came up with the idea that I didn't care about shielding or jitter."
" Analog is much more influenced by impedence, in digital if the signal is there it is there. "
Which is it? I have heard all the rational about why digital cables don't matter before. They do and some research on your part would tell you why. But if makes you happy to not bother to find out, go for it. Others
might make some effort and be very pleased with the results. I was and still am.
"It is your imagination that came up with the idea that I didn't care about shielding or jitter."
" Analog is much more influenced by impedence, in digital if the signal is there it is there. "
Which is it? I have heard all the rational about why digital cables don't matter before. They do and some research on your part would tell you why. But if makes you happy to not bother to find out, go for it. Others
might make some effort and be very pleased with the results. I was and still am.
Try this!
Digital Coaxial Cable by Shark
UKP 1.99 per meter from Maplins
Although this coaxial cable is specifically designed for digital audio interconnects such as those used between DVD players and digital decoders for surround sound systems, it would be an excellent alternative to regular 75 ohms coax where high signal quality is essential. The cable core is 0.5mm diameter OFC Copper plated in 80µin Silver. The Silver plating takes advantage of the phenomenon of “skin effect". “Skin effect" describes the tendency of ultrahigh frequencies to be conducted on the surface of the wire as opposed to through the wire’s total profile. The lower resistance of silver over copper ensures optimum transmission of theses ultrahigh frequencies. This core is covered with a PE insulation. The screen is Mylar/Aluminium covered with a tightly woven Pure Copper braid. The sheath is black PVC.
There are also phono sockets made specially for co-ax cables.
Body containing a split collet with a ridged inner surface, to grip outer screen when folded back over outer sleeve.
Inner conductor can be soldered to centre pin. Insulation material is Teflon.
Digital Coaxial Cable by Shark
UKP 1.99 per meter from Maplins
An externally hosted image should be here but it was not working when we last tested it.
Although this coaxial cable is specifically designed for digital audio interconnects such as those used between DVD players and digital decoders for surround sound systems, it would be an excellent alternative to regular 75 ohms coax where high signal quality is essential. The cable core is 0.5mm diameter OFC Copper plated in 80µin Silver. The Silver plating takes advantage of the phenomenon of “skin effect". “Skin effect" describes the tendency of ultrahigh frequencies to be conducted on the surface of the wire as opposed to through the wire’s total profile. The lower resistance of silver over copper ensures optimum transmission of theses ultrahigh frequencies. This core is covered with a PE insulation. The screen is Mylar/Aluminium covered with a tightly woven Pure Copper braid. The sheath is black PVC.
There are also phono sockets made specially for co-ax cables.
Body containing a split collet with a ridged inner surface, to grip outer screen when folded back over outer sleeve.
Inner conductor can be soldered to centre pin. Insulation material is Teflon.
An externally hosted image should be here but it was not working when we last tested it.
Re: Re: Re: A farmer outstanding in his field
31? Thought it was 24
guido said:
Wake up yourself!
I work in telecoms. We are changing now to digital distribution frames, but the old ones are just coax cabling. Two coax cables are needed for one 2Mb connection and yes it is a 'digital telephone line' (31 times 64k speech/data channels, one channel for synchronisation, at least in europe it's 32 channels).
Actually the coax we are using for measurement equipement makes a nice interlink cable for audio 😀
If you ever saw such a distribution frame (spaghetti) you could imagine EMI matters for the cabling and the connectors.
Loosing one connection for even a short moment means 30 dropped calls.
Regards,
31? Thought it was 24
But jitter is only an issue with systems that work synchronously in real time, without intermediate data storage nor local independent clocks and reclocking/resampling facilities
And unexpected data corruption due to cable reflections is only an issue with systems that don't use CRC to check data integrity and retry packet transfer in case of corruption
Aren't digital telephone lines asynchronous and buffered at both ends?. Isn't the voice data compressed and with CRC on each frame?
There is not much comparison between synchronous and asynchronous systems. In asynchronous buffered digital transmission systems any cable does the job as long as a high enough percentage of data frames or packets reach the other side in an 'understandable' state [ie: valid checksums]
So I think the problem with audio transmission is not the cables, the problem is poor transfer protocols
And about EMI : Computers and custom dsp-based equipment used everywhere nowadays to process digital audio are clocked at 1 to 3Ghz and make nice microwave generators, while digital audio transmission only requires a few Mhz bandwidth [<5Mbps for 24bit 96Khz stereo], and any coax cable properly shielded does a great job in attenuating the already poor EMI radiation at these low frequencies, so I don't think this could be an issue
And unexpected data corruption due to cable reflections is only an issue with systems that don't use CRC to check data integrity and retry packet transfer in case of corruption
Aren't digital telephone lines asynchronous and buffered at both ends?. Isn't the voice data compressed and with CRC on each frame?
There is not much comparison between synchronous and asynchronous systems. In asynchronous buffered digital transmission systems any cable does the job as long as a high enough percentage of data frames or packets reach the other side in an 'understandable' state [ie: valid checksums]
So I think the problem with audio transmission is not the cables, the problem is poor transfer protocols
And about EMI : Computers and custom dsp-based equipment used everywhere nowadays to process digital audio are clocked at 1 to 3Ghz and make nice microwave generators, while digital audio transmission only requires a few Mhz bandwidth [<5Mbps for 24bit 96Khz stereo], and any coax cable properly shielded does a great job in attenuating the already poor EMI radiation at these low frequencies, so I don't think this could be an issue
Well, you could say so indeed, because for async systems jitter is so immense that it is not even on the "issues" list. It is fact of life that is dealt with differently, by means of excessive buffering and feedback flow control.Eva said:
No. No. though yes for CRC. Invalid data is discarded without notice.
This is absolutely not the distinguishing factor between sync and async systems. Either works well with reasonable amount of transmission errors.
Protocols are just fine. If your transmission line is broken, fix it. Much cheaper than dealing with complex protocols in real time.
- Status
- Not open for further replies.