|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
12.72.208.21
In Reply to: RE: By the way... posted by Charles Hansen on January 07, 2008 at 09:50:51
I think we have already established that the data is getting read from the disc well enough, so it seems like maintaining much better isolation between the clock and the "dirty" digital section that is presently corrupting it with jitter components would be a more cost effective approach. Of course, that would mean using D/A convertors without built-in digital filters since you want the DAC chips in the clean section, and would necessitate re-syncing the data to the new word and bit clocks generated on the clean side.Just a thought. I've done it in DACs using BB ISO150 type high speed digital couplers to good effect, only bringing the data across the barrier to the clean side. Much easier in a CDP since you have a fixed crystal and don't have to worry about synchronization.
Gets a lot more complex (as I know you've addressed in your company's products) when designing multi-format players.
Edits: 01/07/08 01/07/08Follow Ups:
That big magnesium clamping mechanism, the overdone, hot-rodded motor and super heavy, overengineered stable base result in much better reads; yes more data gets transferred resulting in less read errors. The difference between identical players, one using the VRDS, and one not, - is dramatic, I've heard it directly.
Thin man, in a powder blue suit, with eyes that slice you right through. The cut of his clothes are strange indeed, a hundred years too soon.
From what I've seen it is close to zero with most decent transports. Couldn't be enough to change the sound unless it was constant errors, could it? Has to be the the result of the PLL clocked datastream coming off the disc interfering with the timing of the system clock at the read FIFO, and at all points after, aka data correlated clock jitter. Many electrical mechanisms in play.
whatever...
To my knowledge, everyone who's done a test has come to the same conclusions as me, - the transport affects the sound of the overall player dramatically, and yes, sounds much better than the other transports.......
Thin man, in a powder blue suit, with eyes that slice you right through. The cut of his clothes are strange indeed, a hundred years too soon.
The sonic difference between transports is *not* due to "fewer data errors" or anything like that. If the player is properly designed, then jitter shouldn't be a factor either. Basically it something that nobody fully understands. All we can do is make changes and note the sonic differences produced by those changes....
But seriously, do you really believe there is some X-factor beyond data integrity and clock jitter that isn't yet (or can't) be measured but that affects digital to analog conversions? I realize there are many other factors that contribute to the ultimate sound, everything from RF noise to power supply capacitors, but time shift the digital datastream from two different transports by storing it to hard drive and you lose the transport sound signature when the files are played back, provided the data is read correctly. That would imply to me that the difference is all clock jitter when you are in the digital domain, and that nobody knows how low it has to be before inaudible.
< < do you really believe there is some X-factor beyond data integrity and clock jitter that isn't yet (or can't) be measured but that affects digital to analog conversions? > >
Absolutely.
In fact there are thousands of "x-factors" that affect all sorts of different aspects of sound reproduction. We don't really know how any of them work.
The best analogy is gravity. For example, we understand the *effects* of gravity so well that we can send a space probe to Pluto, which is literally 5 billion miles away with unbelievable accuracy. But nobody has the slightest idea how gravity actually works.
Similarly, we can build massive optical drive transports that sound better than lightweight plastic ones, but nobody knows why. All we can do is observe the effects.
"Basically it something that nobody fully understands."
What a breath of fresh air it is to hear someone like yourself say this!
With the VRDS, is it the case that the over-engineering and superior disc clamping is resulting in superior information retrieval from the disc, or is it that there is physically less vibration affecting the rest of the analogue circuitry?
As you suggest, it will probably be the case that we'll find out only by constant sucking and seeing rather than head scratching.
It's interesting that the Memory player with it's 'read until right' technology has received excellent reviews in terms of providing the highest CD resolution and that the VRDS Esoteric players use a similar technique and buffer so that the same CD pits are being read many times during playback.
Best Regards,
Chris redmond.
< < VRDS Esoteric players use a similar technique and buffer so that the same CD pits are being read many times > >
That's the first I've heard of that. I don't believe it to be the case.
And the "read until right" thing is just a myth. It's only required when a computer drive is ripping the disc at some ultra-high speed like 24x or whatever. Any ordinary audio CD drive will get all of the bits perfectly also.
< < is it the case that the over-engineering and superior disc clamping is resulting in superior information retrieval from the disc > >
Well, if that were the case, I would assume that it would be something easily measured -- jitter, cleaner "eye" pattern, et cetera. So I don't think that's it.
< < or is it that there is physically less vibration affecting the rest of the analogue circuitry? > >
I also don't think that this is the case. In the first place, if you stop and think about it, solid-state circuitry shouldn't even be vulnerable to vibrations. (But we know that it is, because different "footers" will affect the sound of solid-state components!) And in the second place, I don't think that ordinary transports create all that much vibration in the first place -- certainly not a lot more than the VRDS. And if vibration were a problem, then SACD would sound bad since the disc rotates at roughly 10x the speed of a CD.
Like I said, nobody knows...
"That's the first I've heard of that. I don't believe it to be the case.
And the "read until right" thing is just a myth. It's only required when a computer drive is ripping the disc at some ultra-high speed like 24x or whatever. Any ordinary audio CD drive will get all of the bits perfectly also."
I'm not sure if it's the Esoteric players in general then or just the APL player which uses the VRDS NEO, but below is text from a review of the APL describing the benefits of the VRDS NEO;
1. Since the reading speed is much higher, the audio data is being stored in large SDRAM memory buffers. The audio data is being clocked out of the SDRAM which effectively eliminates jitter coming from the transport.
2. If reading error occurs (imperfection on the disc), the laser assembly can be sent back to the problematic passage to re-read the data and try to correct the error. This happens while you are listening to uninterrupted music coming from the SDRAM buffers. In other words, the VRDS-NEO is capable of extremely powerful error correction, including the so called RUR (Read Until Right).”
Best Regards,
Chris redmond.
Alex is a really smart guy, so I'm hesitant to disagree with him. But in this case I think I'll have to. I've never seen a dedicated (ie, non-computer) DVD player play CD's at anything besides 1x. Obviously the mechanism is capable of reading higher speeds, because DVD runs at higher speeds. But if you take the cover off while playing a disc, every DVD player I've seen runs at normal (1x) speed while playing CD's.
All of this is controlled by the electronics. The Esoteric UX-1 uses the same chipset as the Pioneer players (Mitsubishi MPEG decoder chip.) When it plays a CD, it goes into CD mode and uses what are essentially standard CD algorithms that have been used for decades. I doubt there is a practical way to change the behavior of the Esoteric DVD player in this regard. I have a UX-1 sitting around somewhere, if I had enough time I could take the top off and verify the speed.
I could be wrong, but I don't think so.
"I've never seen a dedicated (ie, non-computer) DVD player play CD's at anything besides 1x. Obviously the mechanism is capable of reading higher speeds, because DVD runs at higher speeds. But if you take the cover off while playing a disc, every DVD player I've seen runs at normal (1x) speed while playing CD's."
Below is some more text from the review by Constantine Soo, but I'm not sure if the information is actually from Constantine or Alex.
It does state that ALL digital discs rotate at a higher than normal rate though:
"Besides the mechanical superiority of the VRDS-NEO transport, its electronics companion, the Digital Signal Processing (DSP) board, is also of the highest possible quality. The DSP uses the best devices by Sony Electronics ensuring unbeatable processing quality. The disc rotation speed for all digital formats being played on the VRDS-NEO is much higher than normal."
Best Regards,
Chris redmond.
Like I said, I could be wrong.
I have the schematics for the unit at work (only in printed format, not electronic). The machine is here at home, but it is too heavy for me to pick up from my wheelchair. Maybe I'll see if I can get someone to help me take it apart and look at it.
Again, I would be very surprised if it spins *any* disc at greater than 1x speed. Normally that is the province of computer-based machines. The first time I heard of that in a stand-alone player was the Denons. They used a computer drive and spun the disc just slightly faster than normal. I believe it was 1.2x or something. The idea was to fill up a video buffer so that the layer change on the double-layer DVD was very quick. Normally the picture will "freeze" for 1 to 2 seconds during the layer change. A buffered design (which includes all of the cheap players such as Oppo and low priced Pioneers) based on the MediaTek MPEG decoder chip also have a very quick layer change and must read the DVD at slightly higher than 1x speed.
But if they do this for CD, it would be the first I've heard of it. The Ayre CX-7e and Resolution Opus 21 use a computer-based CD-ROM drive running at 2x. The data feeds into a buffer. When the buffer is full it stops making requests for data. When the buffer is less than full it requests more data from the drive. But even in this case the data is never *re-read*. There is no reason to, as there are never any errors unless the disc is scratched so badly that it skips.
All newer Esoteric DVD based transports (VRDS and VOSP) spin the CD at x4 and buffer in SDRAM. The only exception is the older Pioneer based DV-50 that spins CD at x1.
To my knowledge, there is nothing better than Sony RF, CD/DVD/Digital Servo and DSD (SACD) digital signal processors. This is what all current Esoteric transports are using.
Regards,
Alex
Howdy
The hassle is that RUR as advertised from the Memory player folk doesn't pay attention to whether there's a read error indicated from the drive, they read more than once using their own idea of error detection and correction... It confuses matters to use the term "RUR" in other contexts since this is quite different than what essentially everyone else does. (And FWIW I agree with Charles that the number of uncorrected errors encountered in practice for most disks is so small that all of the RUR is just hype.)
-Ted
nt
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: