|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
108.169.2.66
In Reply to: RE: Abe, Here is My Take on the REGEN and Jitterbug posted by Mercman on September 22, 2015 at 05:57:47
""PHY chips and processors at the input of every USB DAC are sensitive to packet noise modulation and ground plane noise caused by poor signal integrity and impedance mismatching"Packet noise modulation, ground plane noise, & Impedance mismatching.
Implies one of two things or both: PHY processor/chips are not as "good" as they should be at mitigating the 3 mentioned above: or, - the signal from the USB bus is not all that...
Can't one necessarily draw the conclusion that USB feeds from different transports vary in quality? If not, why "regenerate?"
"Asylums with doors open wide,
Where people had paid to see inside,
For entertainment they watch his body twist
Behind his eyes he says, 'I still exist.'"
Edits: 09/22/15Follow Ups:
Remember my posts about the SUSB for audio ie sometimes universal serial bus with its charge pump power supplies?
John Swenson said then that usb audio with an EMU 0404 was what he was 'comfortable' with. Mercman thought that a 13.5 bit TDA1543 dac with valve output trounced his high end TEAC too. MAC with $1500 player program was hailed here as the 'best, of the bunch. My golly, never use a PC with Windows 7 to play music!.
SPDIF done right then sounded good and better (dCS, Universal Audio, even the Juli@) than usb transfer.
""Remember my posts about the SUSB for audio ie sometimes universal serial bus with its charge pump power supplie""
I am sorry, I do not remember that post.
I am not trying to say SPDIF is better than USB. I am trying to understand why SQ or "signal integrity" issues are the "fault" of the DAC's PHY processor when these devices clearly "re-generate" and manipulate the USB signal. And, some playback devices clearly, (and most everyone admits), produce a "better" USB signal, that renders the USB Regen a "less effective"
Certainly a DAC can and SHOULD much better if the signal coming into it is improved. But a good reason to re-generate it is because of the quality of the variances in transport USB output. A Squeezebox USB output upgrade will be different than the Bryston BDP-2, and different again from a MAC Mini, and different again from a Razberry Pi. I don't think that a DAC manufacturer, can and should "isolate" his/her PHY processor to mitigate, enhance, and repair, issues with the USB on the transport.
Another analogy, we do not "blame the DAC" of an all-in-one CD player when we upgrade a standard Denon drive with a VRDS NEO. There are many factors that go into making the VRDS NEO "better" and the DAC still works great, but the overall sound of the CD Player is significantly better with VRDS NEO transport.
"Asylums with doors open wide,
Where people had paid to see inside,
For entertainment they watch his body twist
Behind his eyes he says, 'I still exist.'"
There is only one way to resolve the issue of what is a better audio transfer method and that is to map by measuring the waveforms through the chain from transport to dac to analog output. This could be a PhD project.
Otherwise, there are just too many things to speculate on, including individual opinions and assertions based on commercial interest that are just not universally true or reliable.
"There are many factors that go into making the VRDS NEO "better" and the DAC still works great, but the overall sound of the CD Player is significantly better with VRDS NEO transport."
I just don't agree with this. A computer drive playing to a memory buffer with the ability to reread errors does just as well for a lot less money.
""I just don't agree with this. A computer drive playing to a memory buffer with the ability to reread errors does just as well for a lot less money."Have you performed this comparison? How could it have into the same exact DAC? I can tell you that I've done it with a VRDS NEO, & a Denon: into the same enclosure.
This is an analogy, different disc transports sound different in one box CD players. Just as different computer playback software and USB implementations sound different: with the same DAC, - you've experienced this...
"Asylums with doors open wide,
Where people had paid to see inside,
For entertainment they watch his body twist
Behind his eyes he says, 'I still exist.'"
Edits: 09/22/15
I have a great deal of experience with VDRS. And yes, I have compared one box CD solutions with DAC/computer. I'll take the latter.
How's about Esoteric SACD players with built in DACs? That's my experience.
I have also compared Playback Design SACD / DAC.
Look, we have a difference of opinion that in the grand scheme of things is a piddly difference.
The rest of your system matters and what you have held as gospel truth does not apply generally. If I remember correctly, you had a Levinson amp which can be 'cool' and earlier Wilson speakers which can be 'upfront'. Some noise dithering in a 13.5 bit dac can help change matters, but the reduction in dynamic range and the increase in distortion would have been there ie 18-19 bits from your TEAC and much lower resolution from your TDA1543 valve dac.
The whole thing build down to system matching and component tuning based on understanding the underpinning science.
Actually, the Levinson amps had a warm sound; not remotely cool. I think I was using a Wilson Watt / Puppy 8s in those days.
"The whole thing build down to system matching and component tuning based on understanding the underpinning science."
Lucky for most of us ignorant bastards that the equipment seems to work and provides some pleasure even though we don't understand the underpinning science.
I was making an analogy between two different types of CD transports only.....Not trying to compare computer audio to CD spinners.
I was trying to make the point that what comes out of the computer via USB can vary in SQ and there are differences between the USB signal with different computers....and playback software... no DAC manufacturer can "fix" these differences inside the DAC.....
"Look, we have a difference of opinion that in the grand scheme of things is a piddly difference.""
Yes, that's truly a given. We both speak from our experiences. Although I have not yet heard anything that beats an APL NWO spinning discs with a VRDS-NEO, - I suspect that I will, and soon.
I was likely unclear that I was not talking about computer/digital file playback in drawing the disc transport analogy. That is prolly easily confused as I've been advocate in comparing digital file playback with optical disc spinners. I should've used turntables in the analogy instead.
Without even hearing it, the USB Regen just makes a lot of sense....
"Asylums with doors open wide,
Where people had paid to see inside,
For entertainment they watch his body twist
Behind his eyes he says, 'I still exist.'"
Edits: 09/23/15
The TD1543A ? I was talking about was a ladder DAC I used in 2006. Compared to the sound of the Esoteric of that day I would gladly take the the Wavelength DAC of that period.
The software you are referring to was Amarra-and it was $1000. Sonic Studio made it up to us by giving us the updated newer software with the Dirac.
The Esoteric DAC I reviewed for AudioStream last year sounded nothing like the UX-1 under discussion.
an individual's view and system cannot be taken as generally true. With a 13.5 bit dac, one cannot expect 'best' digital replay thru a high performing audio chain.
PHY chips do signal processing to make sense out of the incoming waveform. The amount of processing required depends on the quality of the waveform, which will vary depending on the source and the cabling. With a very high quality input signal, little or no signal processing will be required, just sampling the waveform at the appropriate times and looking at the values to slice the amplitude into 1's and 0's. With a low quality waveform there will have to be additional processing to make sense out of the waveform before it can be sliced and diced into the bits.
Signal processing takes energy, and using this energy creates noise. This is an inevitable byproduct of the processing. The cure is not "unobtainium" PHY chips, it is additional isolation. It is possible to isolate the PHY chip from the critical portions of the DAC (clock, and analog circuitry), but if this is inadequate then the problem is with the lack of isolation, not the PHY itself. Alternatively, and this is what the REGEN does, the cure is to feed a cleaner USB signal to the PHY chip, so it has less work to do.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
Tony,
Since you have had the Regen for a bit longer now, and further listening opinions or feelings?
I am very happy with the improvement I am getting with it.
Regards
Bob
I've not felt the urge to remove the REGEN and conduct further tests. I've just been enjoying the music. (Or not, when it comes to a few poor recordings, such as the off-pitch 96/24 EMI transfer discussed in the thread linked below.)
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
I have been doing the same with mine. Just enjoying the music. :)
As you say... there will always be a few clunkers out there.
Regards
Bob
Actually, the poor recording was of a superb performance of the Beethoven Violin Concerto. This created quite a bit of clash, and the clash was musical as well as sonic.
The good news is that it only took 10 minutes to repitch the recording, once I decided it was definitely needed and how much was required. I don't normally bother to fix poor recordings as the time consumed is too large relative to the benefits. In this case, however, the performance was so wonderful that the effort was worthwhile.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
Thank you for that explanation......
It makes really good sense.
And, again, - depending on what you mean by isolation of the PHY processor, this indicates NOT a problem with the DAC, - but with the integrity of the USB signal coming from whatever transport. And dwelling on the PHY isolation for a moment, - are we talking about a physical shielding of something like ERS paper? Moving the PHY chip physically away from the rest of the circuit....
I intend neither crassness, nor sarcasm: but from most DAC reviews showing pictures, - the PHY transceiver chip isn't being physically isolated. Yet the DAC manufacturer seems is often "blamed" for what is a "problem" at the transport. When in fact, (and/or beyond the choice of which PHY processor), there isn't a lot a manufacturer can do.
Without going into the fact that we agree that running SPDIF has issues, - just different kinds: an external USB to SPDIF IF may not have the same issues if reclocking & converting is done well. In that case, the Regen theoretically would be less effective.
To be totally clear, on the basis of all the testimony, I am NOT saying that the Regen isn't a supere bloody awesome device: I am just trying to weed through some of the information, and get an indication which application of it is optimal...
Cheers,
"Asylums with doors open wide,
Where people had paid to see inside,
For entertainment they watch his body twist
Behind his eyes he says, 'I still exist.'"
The isolation would need to be outside of the chip, specifically on the data lines, the power lines and the clock lines. There would be packaging issues, e.g. the way the ground planes were laid out on the circuit board(s) and possible need for shielding, etc...
The DAC manufacturer is selling a digital to analog converter. It is the function of this device to take a digital signal and convert it to analog. This includes dealing with the reality that all digits are not equal, but reading them as if they were: 0 101110110 111 101 is the same number as 0101110110111101.
With an asynchronous USB connection, there is no transporation or timing provided by the so called "transport". This terminology goes back to analog tape recorders, where the transport did control the timing of the music. This is not how modern DACs work. There is no "transport". There is a digital storage device. In the absence of overrun errors or CRC errors on the USB cable, any sound quality differences that depend on the source are entirely the fault of the DAC and can be reduced just as effectively by removing them inside the DAC as removing them inside the computer. (Or removing them in between, as with the REGEN.) It may be more expiditious to improve the transport than redo the DAC, or it may not. However, if the DAC is sensitive to its transport it shows that the DAC is not properly acting as digital to analog converter.
This is made obvious by John S's explanation of noise being created by the PHY chip. That chip is inside the DAC and it is the DAC designer's fault if he has not taken into account the characteristics of the chips he specifies.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
Thanks for the explanation.....
But a "DAC" is still a DAC if it uses USB, SPDIF, AES, all in the same device.
Indeed we have seen issues where the "not so great" implementation of USB sounds way worse than SPDIF. In the case of an external USB to SPDIF converter, we've seen varying degrees in SQ depending on device philosophy/configuration.
""There is a digital storage device.""
Yes, sometimes it's the same as the "player" software processing the digital file to be played back. Sometimes that digital storage device is an external hard drive or NAS: a container.
Because the computer is a different "type" of transport, doesn't mean that it's not a transport: it still sends the digital signal to the DAC to be processed.
""However, if the DAC is sensitive to its transport it shows that the DAC is not properly acting as digital to analog converter.""
The only way that that can be the case is with a partial or complete failure of DAC output to amplification. Clearly not the case, when a DAC performs better with it's AES input then it does with its USB.
What seems to me to make the above even more incorrect is that all DACs vary in sound quality whenever different USB ports are used on some USB buses: even well designed DACs. Or all DACs sound worse with ITUNES than they do with JPLAY. If the fault was in the DAC, (and not the quality of the waveform/signal it receives), then why does player software vary even on good DACs?
"Asylums with doors open wide,
Where people had paid to see inside,
For entertainment they watch his body twist
Behind his eyes he says, 'I still exist.'"
As far as I am concerned, when a DAC is used with different input circuitry it is, for all practical purposes, a different DAC.
My comments in this thread were with respect to USB DACs, specifically those that operate asynchronously. (If a DAC receives audio timing information from the same "box" as its digital samples, as is the case with SPDIF and many AES/EBU configurations, then indeed, I would agree that the device sending clock and data is a transport, since then the partition of functions between the two boxes is conceptually similar to the traditional two box CD players or for that matter a two box tape deck with heads and mechanicals in on box and electronics in the other.)
For a DAC to pass noise on its digital input through to its analog output is similar to a DAC passing noise on its power input through to its analog output. In both cases, if this noise is excessive (relative to the price point of the DAC) then the DAC is not operating properly. Another possibility is electrical noise from nearby devices. (I've not experienced this with a DAC, but of course it is common with analog sources if equipment is not located properly.) Here, by "noise" I include subtle changes to the music as well as clearly separate background sound.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
""Can't one necessarily draw the conclusion that USB feeds from different transports vary in quality? If not, why "regenerate?"""
My take on it is, "why not"… If you look at it on a larger scale i.e. a transmission line, by the time the information is sent down the cable whether it be 1 foot, 3 feet, or thousands of miles, why not clean up the information and get everything as clean and quiet as possible right before it gets to its destination. It is an excellent idea which does make sense IMO and cannot really hurt anything.
And… As we see from the majority of people it is clearly an improvement. I also have a Regen and find it an improvement. Yes it can vary from system to system, so you have to try and see what it would do on an individual basis. My findings match Steve's excellent review. I posted my opinion on CA.
Regards
Bob
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: