twisted or warped?
I call it an advantege.
That's true. You call 6MHz high?
If that's the case, then 50% of what you listen to came the "wrong" way!
Why would you want to? Outputs-is-outputs, inputs-is-inputs.
I call it an advantege.
That system never got off the ground. It certainly wouldn't have travelled any distance.
The high signal levels make it more tolerant to interference, electrical and mechanical.
Re: twisted or warped?
For serious interfacing I'd require 100 MHz (remember squarewaves contain high frequency components
On longer cables it is more easy to induce RF currents, which in turn induce jitter (any clue what you can induce in 200 meter ?).
In addition the impedance mismatch creates problems too, in the jitter area.
ABX testing on a decently designed interface, compared to a 110ohm XLR twisted pair based system where very clearly in the advantage of the first one.
The data may be transfered, the jitter and other RFI problems will spoil the rest
best regards,
dhaen said:
For serious interfacing I'd require 100 MHz (remember squarewaves contain high frequency components
On longer cables it is more easy to induce RF currents, which in turn induce jitter (any clue what you can induce in 200 meter ?).
In addition the impedance mismatch creates problems too, in the jitter area.
ABX testing on a decently designed interface, compared to a 110ohm XLR twisted pair based system where very clearly in the advantage of the first one.
The data may be transfered, the jitter and other RFI problems will spoil the rest
best regards,
God I love arm chair engineers.......
The signal itself will introduced vibration. The really good cables address mechanical damping which is audible. I have discussed this with Jim Aud at Purist Audio, Chris Sommivigo of Illuminati, Ron Hedrick at Marigo Audio Labs, Scott Nixon, and a few lesser know cable designers. I designed about a dozen commercial digital cables (Audient Technologies) including an AES EBU cable with a matching network that sold for $650 and took several hours to make by hand. I have also listened to and measured a dozen highly respected cables by the above designers. I have built several hundred cables( many with networks). I have done TDR, Spice models, and hundreds of hours of listening test as well as hac others who involved in listening test. Digital cables are directional. I have listened to and measured enough of them to know. The bandwith of the digital interface is in the 30 to 200 Mhz range depending on the rise time of the drivers which are usally HC CMOS or band with of the pulst transformers ( 20 to 100MHz) when used. I designed pulse transfomers too and was mentored by John Marshall. He designed the Scott transformers recomended in the Crystal Semi data sheets for their Interface ICs. Maybe that would make at least a little qualified on the subject?
And your professional (or otherwise) experience with digital cables is?
The signal itself will introduced vibration. The really good cables address mechanical damping which is audible. I have discussed this with Jim Aud at Purist Audio, Chris Sommivigo of Illuminati, Ron Hedrick at Marigo Audio Labs, Scott Nixon, and a few lesser know cable designers. I designed about a dozen commercial digital cables (Audient Technologies) including an AES EBU cable with a matching network that sold for $650 and took several hours to make by hand. I have also listened to and measured a dozen highly respected cables by the above designers. I have built several hundred cables( many with networks). I have done TDR, Spice models, and hundreds of hours of listening test as well as hac others who involved in listening test. Digital cables are directional. I have listened to and measured enough of them to know. The bandwith of the digital interface is in the 30 to 200 Mhz range depending on the rise time of the drivers which are usally HC CMOS or band with of the pulst transformers ( 20 to 100MHz) when used. I designed pulse transfomers too and was mentored by John Marshall. He designed the Scott transformers recomended in the Crystal Semi data sheets for their Interface ICs. Maybe that would make at least a little qualified on the subject?
And your professional (or otherwise) experience with digital cables is?
That's true, but we get into an argument about whether the reflection of those edges causes any degradation, and where, temporaly they fall.
A hell of a lot (haven't calculated), but since it's balanced, its common mode. Since the standard strongly suggests transformer coupling, and most manufacurers comply, the CMRR will be very good.
I think this is covered in point 1.
So how do you explain how the music originators get away with it? True, most re-clock.
Cheers,
Getting away with it..........
"So how do you explain how the music originators get away with it? True, most re-clock."
I could write a book on how poorly most profession digital interfaces are done after designing them for three years and taking apart a few dozen DACs. Wadia and Audio Alchemy were two of the worst in my opinion.
"So how do you explain how the music originators get away with it? True, most re-clock."
I could write a book on how poorly most profession digital interfaces are done after designing them for three years and taking apart a few dozen DACs. Wadia and Audio Alchemy were two of the worst in my opinion.
You seem a very boastful person
Fred,
I've designed and supervised the installation of studio's, both production and broadcast. I follow industry and AES / EBU recommendations. It works.
I've designed SDI interfaces for studio applications working at 270MBs at board level.
How much of these have you done?
And one question:
Where did the music come from that you used for tests?
Cheers,
Fred,
I've designed and supervised the installation of studio's, both production and broadcast. I follow industry and AES / EBU recommendations. It works.
I've designed SDI interfaces for studio applications working at 270MBs at board level.
How much of these have you done?
And one question:
Where did the music come from that you used for tests?
Cheers,
There is a large difference between "it works!" and "it sounds really great!". Hell, even TOSLINK works. Big deal.
The AES kludge does not sound good. Go measure the jitter on one sometime.
Of course, most studios that I have been in are populated by twits who can not hear. So who cares about a high-performance interface. But since performance was not the primary goal, what can you expect.
Jocko
The AES kludge does not sound good. Go measure the jitter on one sometime.
Of course, most studios that I have been in are populated by twits who can not hear. So who cares about a high-performance interface. But since performance was not the primary goal, what can you expect.
Jocko
"How much of these have you done?"
"How much of these have you done?"
I have worked on even faster stuff in Telecom and was the Signal Integrity guy in my group as well as the EMI guy. I would think someone who worked on 270 MHz would take impedance matching more seriously and not state that the AES EBU interface is 6Mhz. I have lots of well recorded and lots of poorly recorded CDs. My wife plays in the Dallas Wind Symphony and I have watched Keith Johnson from Reference Recordings record the group. I have all the recordings. It's nice actually hearing a group play in the hall live that the recordings were made from.
http://www.dws.org/main.htm
http://www.dws.org/records.htm
http://www.referencerecordings.com/DallasWind.HTML
I also have some recordings done by a friend with modified tube Neuman mics with $3000 in custom diaphragms. I have heard the live feed from them on several ocassions.
May I recommend a good signal integrity book by the Guru of the industry? I don't know if you have read it. His seminar was great!
High-Speed Digital Design: A Handbook of Black Magic
http://www.sigcon.com
"How much of these have you done?"
I have worked on even faster stuff in Telecom and was the Signal Integrity guy in my group as well as the EMI guy. I would think someone who worked on 270 MHz would take impedance matching more seriously and not state that the AES EBU interface is 6Mhz. I have lots of well recorded and lots of poorly recorded CDs. My wife plays in the Dallas Wind Symphony and I have watched Keith Johnson from Reference Recordings record the group. I have all the recordings. It's nice actually hearing a group play in the hall live that the recordings were made from.
http://www.dws.org/main.htm
http://www.dws.org/records.htm
http://www.referencerecordings.com/DallasWind.HTML
I also have some recordings done by a friend with modified tube Neuman mics with $3000 in custom diaphragms. I have heard the live feed from them on several ocassions.
May I recommend a good signal integrity book by the Guru of the industry? I don't know if you have read it. His seminar was great!
High-Speed Digital Design: A Handbook of Black Magic
http://www.sigcon.com
Actually I said 270MB per second (that's 135MHz) maybe my syntax confused you
It's where the reflected edges fall that affects whether it's worth considering as higher.
Hmm, so do I, but I can't believe too much of it can be blamed on interconnects.
As has been mentioned previously, there are a lot of poor operational staff.
So what are you really saying? That most of the recordings are dross? That's not my experience. True, there are some exceptional ones.
If it was truely as you and Jocko say, then 90% of them bould be rubbish.
I'm certainly intersted in the book. I'll investigate.
Cheers,
dhaen said:
Hi
There should not be reflections, but with AES that is hard to achieve.
Induced current can be balanced out, as long as the transmitting and receiving system are truly balanced, which they are absolutely not, at least not at 100 MHz, I measured on commercially available "balanced" transformers and got tears in my eyes.
To avoid being depending on the symmetry of interface, I long ago suggested the use of balanced screened cables. All the measurements I did (but never published due to the lack of time) point toward that direction.
Reclocking at the transmitting end helps, but it does not reduce induced jitter afterwards.......
Like Jocko says: It works, but it sounds lousy
regards
Grazie....
Back to the original topic.....
Frank, would you care to elaborate on the types of coax that you used??? Some of the poorest sound SPDIF cables that I have tired were coax; but it was not so much the type, but the length. Too short, and it will sound rotten. Here is one case that longer is usually better.
Jocko
Back to the original topic.....
Frank, would you care to elaborate on the types of coax that you used??? Some of the poorest sound SPDIF cables that I have tired were coax; but it was not so much the type, but the length. Too short, and it will sound rotten. Here is one case that longer is usually better.
Jocko
Re: Grazie....
Too short, and it will sound rotten. Here is one case that longer is usually better.
Jocko [/B][/QUOTE
-------------------------------
Jacko
How long in your experience? It amuses me that hi end retailers in UK are selling super expensive cables in 0.5 m lengths to make the whole thing cheaper!
Too short, and it will sound rotten. Here is one case that longer is usually better.
Jocko [/B][/QUOTE
-------------------------------
Jacko
How long in your experience? It amuses me that hi end retailers in UK are selling super expensive cables in 0.5 m lengths to make the whole thing cheaper!
Most anything 1 m or under is too short. 2 m might even be too short.
I setttled on 6 m because of the yield from a typical reel of Belden cable.
The rational is:
Think how long it takes for the first reflection to hit the RX end, bounce back to the TX, and end up at what point in the waveform at the RX end. And what it will do if it is in the wrong place at the wrong time, i.e., the decision point.
Assume a propagation time of 66%. Close enough for DIY work.
Jocko
I setttled on 6 m because of the yield from a typical reel of Belden cable.
The rational is:
Think how long it takes for the first reflection to hit the RX end, bounce back to the TX, and end up at what point in the waveform at the RX end. And what it will do if it is in the wrong place at the wrong time, i.e., the decision point.
Assume a propagation time of 66%. Close enough for DIY work.
Jocko
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.