Home of the Squeezebox™ & Transporter® network music players.
Page 6 of 6 FirstFirst ... 456
Results 51 to 57 of 57

Thread: SPDIF is evil

  1. #51
    Senior Member
    Join Date
    Feb 2017
    Posts
    372

    Archaeology?

    Quote Originally Posted by arnyk View Post
    False claim. One reason why digital signalling is so valuable is that the digital signal can often be purified by means of regeneration. For example when it is read off the disc, the digital data stream that is read from optical disks is highly jittery, and often has errors. These are generally perfectly removed by the normal functioning of a standard optical disc player.
    Hi Arny!

    I'm fascinated to notice that you've suddenly popped up over here. It's actually quite spooky but disregarding that, are any of the members who previously posted on this thread still active on the forum?

    I recall that somewhere in the Good Book (which I read once in my youth for completeness) it says "there is nothing new under the sun". It seems that a lot of people were using external DACs & the word clock in on the TP back in the day, although no-one seems to have conducted any kind of rigorous testing of the ABX kind. It was interesting to find Sean Adams description of how he derived the quoted measurements for the TP. There didn't appear to be much consensus amongst the earlier posters as to which kind of lab measurements would be useful in the context of external DACs and digital connections...

    I believe that your position on this remains that it shouldn't make an audible difference because the TP's internal jitter is low enough to be inaudible & similarly that its in-built DAC is good enough to be indistinguishable from an external DAC in terms of discernible audible differences in a domestic listening environment with real ears, despite the possibility that minute measurable differences might be detected in lab testing. That's if I've understood your "reliable subjectivist" concept correctly. Please put me right if I've got any of this wrong, I'm not prÚcising your early statements in my own words to irritate you but simply to ensure that I've fully got your drift.

    On the other thread (which has been getting a bit silly, I must shoulder my share of the blame for that!) you commented that you should perhaps write up something on the (or your, I wasn't quite sure) theory of audible perception in homo sapiens. With my deadly serious face on for once, I should be most interested to read that.

    Hope that you're having a pleasant weekend!

    Dave

  2. #52
    Senior Member
    Join Date
    May 2015
    Location
    Grosse Pointe Woods, MI
    Posts
    778
    Quote Originally Posted by Golden Earring View Post
    Hi Arny!

    I'm fascinated to notice that you've suddenly popped up over here. It's actually quite spooky but disregarding that, are any of the members who previously posted on this thread still active on the forum?

    I recall that somewhere in the Good Book (which I read once in my youth for completeness) it says "there is nothing new under the sun".
    Umm yes, Ecclesiastes.

    It seems that a lot of people were using external DACs & the word clock in on the TP back in the day, although no-one seems to have conducted any kind of rigorous testing of the ABX kind. It was interesting to find Sean Adams description of how he derived the quoted measurements for the TP. There didn't appear to be much consensus amongst the earlier posters as to which kind of lab measurements would be useful in the context of external DACs and digital connections...
    If you follow the relevant scientific argument that a DAC can only have 4 kinds of technical problems, all of which are readily measurable and for whom we know quite a bit about the audible thresholds...

    I believe that your position on this remains that it shouldn't make an audible difference because the TP's internal jitter is low enough to be inaudible & similarly that its in-built DAC is good enough to be indistinguishable from an external DAC in terms of discernible audible differences in a domestic listening environment with real ears, despite the possibility that minute measurable differences might be detected in lab testing. That's if I've understood your "reliable subjectivist" concept correctly. Please put me right if I've got any of this wrong, I'm not prÚcising your early statements in my own words to irritate you but simply to ensure that I've fully got your drift.
    Confirmed.

    The TP's internal DAC was tested pretty comprehensively by Archimago in his blog. If I ever get around to writing that article about understanding technical measurements in terms of sound quality, the executive summary can be expected to say that the TP is vast overkill. Of course people say they sound different, but as you say good listening tests confirming that are harder to find than rooster teeth.

    There are 400 or more different makes and models of audio DACs, each selling to Audiophiles on the grounds of "Better sound". Technically, there aren't that many significantly different designs. The vast majority are based on less than 20 different DAC chips with just a few using proprietary parts. Almost all of those chips are very good and can't be logically expected to audibly change the sound of music conveyed by the analog signals they output. People hoot and holler about differences in power supplies and buffers, but they are even less likely to affect sound quality.
    Last edited by arnyk; 2017-05-14 at 06:51.

  3. #53
    Senior Member
    Join Date
    Feb 2017
    Posts
    372

    Roger that!

    Quote Originally Posted by arnyk View Post
    Umm yes, Ecclesiastes.



    If you follow the relevant scientific argument that a DAC can only have 4 kinds of technical problems, all of which are readily measurable and for whom we know quite a bit about the audible thresholds...



    Confirmed.

    The TP's internal DAC was tested pretty comprehensively by Archimago in his blog. If I ever get around to writing that article about understanding technical measurements in terms of sound quality, the executive summary can be expected to say that the TP is vast overkill. Of course people say they sound different, but as you say good listening tests confirming that are harder to find than rooster teeth.

    There are 400 or more different makes and models of audio DACs, each selling to Audiophiles on the grounds of "Better sound". Technically, there aren't that many significantly different designs. The vast majority are based on less than 20 different DAC chips with just a few using proprietary parts. Almost all of those chips are very good and can't be logically expected to audibly change the sound of music conveyed by the analog signals they output. People hoot and holler about differences in power supplies and buffers, but they are even less likely to affect sound quality.
    Hi Arny,

    Thanks for your prompt & helpful response.

    At least I for one appear to have bothered to actually ingest your posts!

    Relaxing into the music now (Emmylou Harris + Mark Knopfler today, All The Roadrunning + live tour album) then heading into highlights of Spanish GP. Alonso managed to put his hopelessly underpowered McLaren-Honda in 7th on the grid. He's looking like a good prospect for the Indy 500...

    All the best,
    Dave

  4. #54
    Senior Member
    Join Date
    Apr 2008
    Location
    Toronto area, Canada
    Posts
    1,282
    Quote Originally Posted by Golden Earring View Post
    Relaxing into the music now (Emmylou Harris + Mark Knopfler today, All The Roadrunning
    That's exactly what I was doing yesterday, appreciating the technology uncritically with one of my favorite albums.
    LMS on a dedicated server (FitPC3)
    Transporter (Ethernet) - main listening, Onkyo receiver, Paradigm speakers
    Touch (WiFi) - home theater 5.1, Sony receiver, Energy speakers
    Boom 1 (WiFi) - work-space
    Boom 2 (WiFi) - various (deck, garage, etc.)
    Radio (WiFi) - home office
    Control - Squeeze Control (Android mobile), 2 Controllers (seldom used), Squeeze Remote (on Surface Pro 4)
    Touch x 1 - spare
    UE Radio x 1 - spare
    Boom x 1 - spare
    Controller x 1 - Spare
    Duet Receiver (backup)

  5. #55
    Quote Originally Posted by seanadams View Post
    Just to clarify I was talking about AES/EBU _in general_. There's nothing wrong with how I did Transporter's AES/EBU output. The interface is defective by definition, and therefore not possible to execute reasonably by anyone.
    Hope no one is upset I found this old thread and dredged it back up!

    Too bad Sean is not around here these days. You will note what he said..................."The interface is defective by definition, and therefore not possible to execute reasonably by anyone."

    Yeah, it is! Which is too bad. By using a balanced cable, it could have been better than plain ol' SPDIF. Especially when some jerks down in Texas try to isolate things with a transformer, which makes the shield "hot", and..................well, it isn't the best approach, but it works. But, they (AES et al) messed up. (In fact, the original version was MUCH WORSE. I had a long phone conversation with the guy who headed up the AES working group on that, and without ever seeing an implementation of it, I 'splained to him what was obviously wrong with it. I can only imagine the look of horror, on his face, as I told him what was wrong and why. Funny thing is.............................less than a year later, they came out with a new version of it! With all of my suggestions. Too bad I didn't nail him down on the output voltage, and some of the other dumb things that were in it. Just goes to show how hard it is to make things idiot proof when the idiots are so dang ingenious. And the worst piece of gear I ever saw, with AES/EBU outputs came from Philips! And cost several grand. Go figure. Guess they never thought putting DC on the output would be a problem. Worked for them, as it was designed to go with something that needed DC on the output........................oy vey.)

    OK, where was I?

    Ok, TOSLINK..............

    A few years back, I measured the "jitter" of the old Toshiba TOSLINK parts, and compared them to the flavor-of-the-month Chinese brand. Since someone paid me to do this, I can not share the details. But, I can give you a rough idea.

    All of them have a very high noise floor. That is the part that gives the "jitter number". (See other posts on why that number is useless.) The Toshiba is a bit better. The FOTM Chinese one performed better if you used the "high-speed" one. You could see the effects caused by using a crappier phototransistor in the regular speed one. It was bad. And right where you should expect a noisy transistor to have noise.

    In all cases, all of the "jitter" contribution was caused by the noise floor. IOW, they DID NOT affect the "close-in" jitter, which is what is really important. (No, I am not going to into that. I have done so numerous times over the last decade or two, and I am not going over it again. If it annoys you, keep it to yourself. I am not interested in hearing why you think I am wrong.)

    The fact that none of these methods (SPDIF, AES/EBU or TOSLINK) do not totally wreck the sound, even though they are highly flawed and have a ton of problems, is because it does not do anything to the signal in the most critical region. IOW, down around 1 Hz, and lower.

    And back to the question of return loss..............

    Yes, it does make a difference. I find it odd few ask about it. I'll just add if you get the return loss of the source, the cable and the receiver all correct (iow, below a certain level we put a lot of work into figuring out what that number is), then guess what?

    They all sound the same. Which would put a lot of cable companies out of bidnis. Where would this industry be without something to sell the punters that will make their system sound "different"/"better"?

    Peace. Out.

  6. #56
    Senior Member sfraser's Avatar
    Join Date
    Oct 2005
    Posts
    249
    I never understood why they just can't put a decent clock and jitter buffer in the DAC . If they did , the jitter and clocking issues should disappear. A decent size jitter buffer will remove any jitter which may occur during the unidirectional transmission of the data stream from the transport device/LMS server etc. Reclocking the jitter buffer playout to the DC chip locally would also help solve the problem.
    2 Chan. System
    SB3->Benchmark DAC-1-> Bryston(BP-25,3B)->PMC TB2
    Home Theater System
    SB2-> Bryston(SP1,4B,4B,2B,2B)-> PSB Stratus Goldi
    Basement System
    Duet-> Parasound Preamp (carver M1.0t) ->Klipsch La Scala's
    Bedroom System
    SB2-> Sony BoomBox
    Rear Deck/Patio
    Duet-> Yamaha Reciever-> PSB Mini's,
    Office
    Squeezebox Boom
    Kitchen
    Squeeze Radio
    Ensuite
    Squeeze Radio

  7. #57
    Senior Member Julf's Avatar
    Join Date
    Dec 2010
    Posts
    2,502
    Quote Originally Posted by sfraser View Post
    I never understood why they just can't put a decent clock and jitter buffer in the DAC . If they did , the jitter and clocking issues should disappear. A decent size jitter buffer will remove any jitter which may occur during the unidirectional transmission of the data stream from the transport device/LMS server etc. Reclocking the jitter buffer playout to the DC chip locally would also help solve the problem.
    Most modern DACs do, and use an ASRC as well.
    "To try to judge the real from the false will always be hard. In this fast-growing art of 'high fidelity' the quackery will bear a solid gilt edge that will fool many people" - Paul W Klipsch, 1953

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •