Is this a trick question? IIRC the SPDIF standard provides 6 - 8 bit bites per time slice (44,100Hz).
No,
But if you have the reference spdif is based on AES/EBU.
The issue is how do you synchronize clocks in a serial data transmission system.
Richard, this is the main page: High Resolution Music DOWNLOAD services .:. FLAC in free TEST BENCH
What I've looked at so far is Haydn: String Quartet in D, Op. 76, No. 5 - Finale - Presto Engegårdkvartette, the 192 and 96 versions, which differed, wrongly IMO. Then, in the last round: B.Britten: Simple Symphony, Op. 4 - Boisterous Bourrée, TrondheimSolistene / Øyvind Gimse / Geir Inge Lotsberg / Anders Nilsson, here the 352.8 vs. the 96 examples. The latter has been badly resampled, or a different encoding method has been used...
The 352.8 track is stinking with ultrasonic rubbish, the waveform looks like you've got a nice little LC resonance in the neighbourhood, going along for a merry ride.
Frank
What I've looked at so far is Haydn: String Quartet in D, Op. 76, No. 5 - Finale - Presto Engegårdkvartette, the 192 and 96 versions, which differed, wrongly IMO. Then, in the last round: B.Britten: Simple Symphony, Op. 4 - Boisterous Bourrée, TrondheimSolistene / Øyvind Gimse / Geir Inge Lotsberg / Anders Nilsson, here the 352.8 vs. the 96 examples. The latter has been badly resampled, or a different encoding method has been used...
The 352.8 track is stinking with ultrasonic rubbish, the waveform looks like you've got a nice little LC resonance in the neighbourhood, going along for a merry ride.
Frank
Last edited:
Everyone, I hope that I do not give the impression that I, and I alone, designed everything useful for quality audio. No, I didn't, BUT I have found that reaching beyond the typical and usual, can bring derision from others who don't want to be bothered, and 5 years (typical) will find the SAME people claiming that they made a great engineering breakthrough based on what we originally argued over as being 'too far out' 'not worth the effort' etc.
To be clear: Scott Wurcer put out a significant amount of effort to provide a pretty good microphone for amateurs to build. When I mentioned that I had not learned anything new, it was essentially because I read ONLY part 1 of his tutorial. I still don't have part 2, so it is entirely possible that he has made an exceptional breakthrough that I am not yet aware of. I hope this settles the matter.
To be clear: Scott Wurcer put out a significant amount of effort to provide a pretty good microphone for amateurs to build. When I mentioned that I had not learned anything new, it was essentially because I read ONLY part 1 of his tutorial. I still don't have part 2, so it is entirely possible that he has made an exceptional breakthrough that I am not yet aware of. I hope this settles the matter.
Part 2 does have more interesting circuits, I make no claim to exceptional breakthrough just applying things from other applications to the problem. The idea for using photo-voltaics for high voltage has been around. The military has used it for remote detonation by laser for a long time.
You can put the 200V bias supply in the mic and run it off of a couple of mA. I tried to present only things a DIY'er could easily reproduce.
Please tell us. Nice to have you back ED I miss these side conversations, but for the record I find 16/44.1 produced as intended with unmodified equipment usually sounds just fine to me. In fact I have a few CD's played on an ordinary Philips CD player and a T-Amp into my Sequerra MET-7's that sound wonderful.
Gasp! Surely you JEST!
Hallelujah! HALLELUJAH!
Scott, at last fame and adulation as befits someone with our Lord JC's blessing
Even though its obvious he didn't even read the 1st article. Maybe he didn't read those B&K & AES articles either. Did he really explain kT/C to Harry Nyquist?
Gasp! Surely you JEST!
Hallelujah! HALLELUJAH!
Scott, at last fame and adulation as befits someone with our Lord JC's blessing
Even though its obvious he didn't even read the 1st article. Maybe he didn't read those B&K & AES articles either. Did he really explain kT/C to Harry Nyquist?
Come on Richard, as a friend, lighten it up a little. Let's all have a happy holiday season.
Assuming you are capturing the word entirely and then retransmitting it with a equal or higher quality clock, the jitter is reduced when you go back to the base rate.
That was the issue that was raised and left unanswered.
Now what happens if you write the CD's files to a hard drive and then use a parallel DAs?
Almost all the chatter here is about equipment. Most of the issues that matter for my work are about the acoustic environment.
I find most CDs and other recordings fail miserably at realism. The exception is when you play just about any decent recording in an appropriate performance space.
While you remain in the digital world there are never any problems. Zero, zilch. The only thing that matters is how well the DA does its job, and how well protected, how impervious is the following analogue, audio circuitry to the digital goings on.
And part of how well the DA does its job, is how stable the clock pulse is, yes, the jitter thing, at the precise point where it feeds into the "real" digital to analogue conversion area. Everywhere else, it can be as jittery as hell, do a merry dance as much as it wants, and it matters not at all. Provided, also, that it doesn't spray RF interference in all directions such that it can interact with the analogue areas.
That is because of weaknesses in the replay chain. Digital sound is a twitchy, high performance sports car compared to vinyl's boulevade cruiser; meaning, if the tyre pressures aren't right then it will fly off the road on the first fast corner.
Frank
Last edited:
1- I don't remember it was so expensive at this timen comparing to my professional ewuipments (a middle priced consumer product ? ), but it was log time ago and my memory is old too ;-)
1-You are probably right. Something was original and different from others on the market at this time, over sampling ? I was experimenting too on delta sigma DACs at this time, i probably had confused. Sorry for the confusion.
Can't-we pretend it add more informations in the digital domain, as each added sample is an average between the two adjacent original values ? But, yes, it does not add *real* information to the original signal.
As you know, it smoothes the signal in the digital domain and free us from a brick wall near the original sampling frequency. Each time we double the frequency, we can remove a pole from the analog filter.
Numerous ways. The simplest is to have a very low phase noise, ie. jitter, oscillator, crystal right next to the DA with the most direct and shortest path into that circuitry; a complete CD player, well engineered, should do it easily. If the source is external, then asynchronous operations where that clock adjacent to the DA controls everything will do the trick. If that is not available then smart buffering will get you through, and finally if the clock absolutely has to come from some external area then there are now chips available that can regenerate clocks with quite staggeringly low levels of jitter, way better than digital audio needs ...
Frank
Last edited:
Do not worry, John, Scott knows how to deserve everybody's respect by the relevance of all what he publishes, his scientific approach, his modesty.
He is well known as Scott 'guru' Wurcer.
In those methods I mentioned, the data rate is locked onto the DA clock for the CD player, and asynchronous ways. Smart buffering means that the circuitry doesn't have to care if clock and data rate keep varying with respect to each other: there's lots of padding so to speak between what comes in and what the DA circuit needs. The ultra-low jitter clock regenerators create the clock signal from the data rate, removing any significant wobble as they do so.
Frank
- Status
- Not open for further replies.