DIY hifi source

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Slightly OT, but thanks for the René Magritte link, I have been wanting to do a set of photos of the family with mirrors involved and its given me some clool ideas, just need somone who can use PS, as I never seem to find time to master masks. Cheers. :)
Though I better be careful where I store the pictures.........
 
An illustration an anecdote may help. Once I got called up to be in a musical. It was a BSO (Boston Symphony) concert production of Gershwin's Girl Crazy. It has songs you may be familiar with: Embraceable You and I Got Rhythm. To get to the original sound they distributed the original scores. They were stamped with Property of the Gershwin Family on it. At one point in the rehearsal the horns made a godawful sound. Turned out Ira forgot a sharp on one of the trumpet scores. The sound of a minor 2nd interval by trumpets is not pleasant! The conductor brought the rehearsal to an immediate halt. The trumpeteer defended himself, showing the conductor that he was playing the music as written. The conductor had him pencil in the sharp. There was a moment of discussion. These scores were quite used, dog-eared and about the color of a brown paper bag. How could such a mistake have gone unnoticed before? The musicians could only guess that the musicians way back when must have known Ira made mistakes like this, and knew that he wouldn't write something like that, and so corrected the mistake on the fly, never bothering to correct it (it occurs to me now that a pencil marking from 75 years ago may have worn off anyway.)

In a way, the digital file is like sheet music. It doesn't change with copying. In fact that old score would have been vastly improved in legibility if only it had been scanned, and touched up like an old photograph to original contrast. But the musicians could read it no problem. It would take worse for them to actually give up. Professional pride and and bright stage lighting.

The musicians are analogous to the analog stage. The sheet music is never going to move from la to ti. But the musician might take liberties, slide a bit if the instrument, say a trombone, is made for it.

This pretending that the notes can change with copying, that is, that digital files can be the same and yet have discernable sonic differences is not credible. It's imaginary, and one thing audiophiles do seem to have in abundance is imagination. Maybe that helps.

Scott

Oh and by the way, the production manager was very sure as we took our bows on the last night to collect each and every one of those scores. Gone. Would have fetched a pretty penny at Sotheby's.
 
Has anyone read the Mark Porzilli interview on his memory player in positive feedback? memory palyer 64

I don't know enough to truly critique what he says and am only now trying to get myself up to speed with servers usb DACs and the like. As I understand it, he says amongst other things that jitter problems can be built into ripped digital files because the bits are stored in caches, which I interpret as being packets of bits. Thus the individual timing of the bits within the cache may be corrupted from the start and is not going to be improved by later reclocking, since this will only affect the timing of the cache itself.

Don't shoot the messenger!

I may also have got Mr Porzilli's reasoning wrong.
 
Sorry, that thinking, as you've just presented it, doesn't sense: within a media file the contents are stored as logic states. It's only when the pure data, which is what it is, is being recovered, read from the media, is being processed, that things become more complicated ...

Frank
 
Sorry, that thinking, as you've just presented it, doesn't sense: within a media file the contents are stored as logic states. It's only when the pure data, which is what it is, is being recovered, read from the media, is being processed, that things become more complicated ...

Frank

Maybe so, but have you read the article? I might also have muddied the water by renaming what Porzilli calls "packets" into "caches".

Thus, his point can be discussed rather than my possible misinterpretation of it.
 
Last edited:
... individual timing of the bits within the cache ...
... timing of the cache itself ...

No, both things I have quoted do not exist. The "timing" occurs when sending
the bits out (via USB, SPDIF etc.) at a known and constant rate, there is no "individual" timing of samples stored within the file (no matter if it´s read from
harddisk, RAM or whatever).
 
Last edited:
thats correct cache is a small piece of memory, it speaks of the action of accumulating information for 'later' in the short term.

Master clock is usually ticking 8->64 per bitclock cycle and 128-2048x FS (sample rate/word clock cycle)

but MCK is not part of the i2s standard, MCK is assumed to be on the convertor side, either on the USB->i2s convertor, or the local DAC clock, or its both the convertor and dac clock, with without some sort of division with logic parts. again this is running synchronously using its own clock at the receiving end, nothing to do with the incoming clock.
 
Last edited:
thus there may not even be audio clock timing for i2s in USB, there will be a separate USB data clock (usually 12MHz or something), but master clock (MCK) is not part of the i2s standard, it assumes a local clock and that is normally where the sample timing is imposed.

some dacs may have a PLL or reclocker, but the master clock does not need to be there to drive timing. sometimes SPDIF will recover an embedded clock, AES can have a separate Wordclock. thats very rare these days and easily avoided.

in all of that though, a buffer is simply where the data builds up in memory and spat out again in a burst, but again the clock on the USB bus is not imposing any timing on the DAC local master clock. the sample rate can be identified by the Wordclock (LRCK) or the bitclock and bitclock is preferred to sync to if any. but there is no room for a super-secret clock. mostly the USB receiver chip will have its own fifo buffer as well.

these are not places for evil to build up, quite the opposite....
 
I just read the article linked, besides highlights like "Impulse Discharge of Events in Atemporal Space"
it´s a lot of marketing BS made to sound "technical" and "knowledgeable".

The main claim made is the dicovery of a previously unknown (?) type of jitter:
"...the second type is jitter which we can predict, but have not been able to directly measure, called "packet jitter"."

The he continues to mix that up with the term "packet jitter" sometimes used in
in conjunction QoS (which is completely unrelated here).

Finally one can notice where the journey goes to:
"We are planning to offer "IDEAS for Audio", an internet-based service which will be available by subscription. One will be able to log on and reduce the jitter of his music library dramatically. Owners of music servers or even computer audio collections, may log on and run IDEAS on their music collection, with often astounding drops in jitter."

I´m pretty sure this fantastically service can´t be cheap...
 
More highlights:

"ECC will raise jitter to intolerable levels."

"The idea behind memory playback was, in its most generic of definitions, that if we copy all of the bits to a piece of memory large enough to not fragment the music files, we'd have the time and accessibility to clear it of jitter."
 
I read it through!!!!! Audiophile based pseudo science to sell somthing for $17,950 that you could do with a $5,000 or less PC.
Fas42, thinking of playback and noise, when a PC plays back music from any source uit is not constantly accessing the data like on a CD, but will use burst mode to read the data into a cache, thus the activity of the computer will not be that different from one source to another. Hard drives can achieve read speeds of 50+mbs so the activity during a song playback may be a few seconds at the most.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.