|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
61.92.98.16
In Reply to: RE: Cd reading errors correction systems impact on sound. posted by beppe61 on June 14, 2015 at 07:46:18
Hi,First, error-correction is exactly that, errors are corrected. Extra information is carried together with the "payload" data and any error that is correctable will be corrected. After this there is no error. Thus error correction in itself cannot cause audible differences.
Second, if an error cannot be corrected the CD-Player will attempt a number of approaches to "conceal" the error, like repeating the previous sample, interpolating between adjacent samples etc., eventually corrupt samples may be muted. This process of error concealment will cause audible artifacts all the way to the sound cutting out.
It should be noted that on a CD in good condition uncorrectable errors are incredibly rare. And one sample worth of error concealment is likely to remain inaudible in practice.
Knowing all of this leaves an interesting question. How do "CD Tweaks" (be it special cleaning, green painted edges, the l'CDFlop mat and so on) possibly make any difference? After all, the data is there 100% complete.
The answer is as so often jitter and crosstalk from circuitry that should be isolated from each other.
If you look at the motors and Focusing electromagnets in a CD Player and at the driver circuitry, you realise you are dealing with a kind of Audio Band amplifier here, which has greatly varying current if for example the CD is not completely flat.
If you look at common commercial CD-Drives, you will see that very rarely is there any serious isolation between the clock ground/supply and the servo drivers. In fact, most of the common PCB's almost everyone uses (no matter if it is a Philips CD-Pro, a Neo something mech etc.) are appallingly bad in this respect.
You can literally correlate clock jitter and servo current (using suitable test gear. And sure, you apply tweaks to the CD and see the impact on servo current and jitter. And unsurprisingly, if you play the CD in the same room where the speakers create vibration, you can see the impact of that vibration in the servo current and so on and if you modify vibrations using (say) damping or different feet - well you can see that too using suitable instrumentation...
The solution? Make sure the clock has an independent supply and and a ground layout that avoids contamination from the servo circuitry. Give the Servo the biggest value local electrolytic bypass cap's you can find and fit and ideally make a separate supply only for the servo Power Stage.
If you then can re-clock the data leaving the transport (be it I2S or SPDIF) against your squeaky clean clock before sending it on, and you will find the resulting transport most resistant to tweaks...
Of course, that usually means you need to either dramatically modify the PCB that comes included with your drive solution one at a time or you design your own, including software and everything.Few HiFi makers actually do that.
Ciao T
At 20 bits, you are on the verge of dynamic range covering fly-farts-at-20-feet to untolerable pain. Really, what more could we need?
Edits: 06/16/15Follow Ups:
I use a device that grinds the edge of the CD so that it reduces the eccentricities of many CDs. You can actually see the CD wobble less when spinning. The audible improvement ranges from subtle to significant depending upon the condition of the original CD.
Which "CD Tweaks" have been measured so far? Do they all produce a measurable impact on servo current and/or clock jitter performance?
Edits: 06/17/15
Hi,
> Which "CD Tweaks" have been measured so far?
Those tests were done over ten years ago, while developing CD-Players. What I tested were the following (memory) and they were tested with the CD playing in a reasonably silent room and with an acoustic signal, common Marantz "Audiophile Tuned" CD-Player:
Ceramic CD-Mat
CD-Flop (basically a 5.25 Floppy Disk cut to CD Size)
CD Tuning Stickers
Green Pen
CDR vs CD
CDRW vs CD
Demagging CD (no measurable result, non audible either)
All had been previously considered offering material changes duriing Audio Club Meetings in tests mostly semi-blind but some sighted (demagg listeing was sighted).
> Do they all produce a measurable impact on servo current
They do unless noted, precise result depended somewhat on the precise CD (excentricity, thickness, flatness etc) and the tweak. But generally two identical CD's one "au naturelle, the other tweaked, reliably measured different and showed recognisable (read correlated) patterns.
> and/or clock jitter performance?
The tools I had than were somewhat limited (AP2 and HP Digital Storage Scope) and I still lack gear that would give me better access to what is happening than I had than (plus, in 2015 I could not give aa flying french coitus about CD), but using enough averages on the AP2 J-Test from CD and a suitable acoustic signal exciting the CD (takes more ime to run the test than to manually grind the coffee in a handcranked mill and traditionally made a decent cup of filter coffee) showed reliable corellation (unsurprisingly, we are dealing with physics here).
Others were much harder to link, but the logical chain of "many CD-Tweaks reliably modulate servo current" and "modulated Servo Current can be reliably shown to exist as correlated clock jiitter in the analogue output from the CD-Player" passes occams razor as does concluding that hence CD-Tweaks can change the jitter pattern of the analogue output and people have been convicted of murder in courts by a jury on less.
Do I have a smoking gun, Photo of defendent firing said gun, photo (full frame) of bullet from defendents gun hitting the victim? No I do not.
Using measurements and blind listening tests again correlated measurable changes with audible ones. Paradoxically, subjective preference was not always for lowest jitter/least servo modulation, but easily measurable differences generally made it past the listening test noisefloor.
I think there are in principle a few dissertations in this, but as said, CD is SO 1980's, just as the Mullet Haircut and Duran Duran, so who cares?
Ciao T
At 20 bits, you are on the verge of dynamic range covering fly-farts-at-20-feet to untolerable pain. Really, what more could we need?
There might be a small to medium-sized group of middle-aged people playing CDs for some time to come.
I'm not saying anyone cares about this, but I'm one of those touchy-feely people who like physical things and who plans on keeping some of those great sounding old vinyl and polycarbonate records out of the landfill.
Hi,
Love Vinyl. I cut my Sound Engineer teeth on Tape and Vinyl.
Many great recordings that despite the 75dB (appx.) limit of LP put most Seedee (CD) recordings/remasters to shame for realism and dynamic range. Always wonder what these recordings would have sounded like on a Pacific Microsonic Model 2 in digital (176.4/24, no oversampling)...
Never got attached to Seedee's (CD's), ruddy bad sounding coffee coasters... If analogue goodness re-mastered for Seedee (CD), worse, I mean they managed to screw up Madonna, fer chrissake.
Though the original Dire Straights Tunnel of Love Album on Seedee beats the LP, narrowly, even early mother and stamper numbers.
The remastered versions are uniformly awefull (including LP) but the DSD sounds great (I usually dislike DSD a fair bit), so I kept the DSD and deleted the rest.
Yes, this is the album with the album where Mark Knopfler famously "makes a Schecter Custom Stratocaster hoot and sing like angels on a Saturday night, exhausted from being good all week and needing a stiff drink."
Which makes two of us, me and all the angels that can dance on the point of a needle. Okay three, my Girl also needs a drink, make it Wolf Blass (shoutout to a fellow ex-east german) directors selection Red all 'round.
If music has to be Deegeetaal (sounds like an aweful curry), can I have 176.4/24 recorded using Pacific Microsonic Model 2?
Seedee?
Who ever cared? I have around 30% of my music in Deegeetaal HD, the rest selected early Seedee pressing rips, Vinyl Rip etc. Of course it resides in a custom PeeSee (PC) with 9TB Raid 5 for music.
The PeeSee never sounds better than the source, but over a decade plus I figured how to make sure it sounds no worse.
Ciao T
At 20 bits, you are on the verge of dynamic range covering fly-farts-at-20-feet to untolerable pain. Really, what more could we need?
I never like them until my current DAC (Octave).
Hi,
I liked them since I got a Pioneer "Legato Link" DAC Equipped DVD Player, that played CD's in effect as Data disks (async) read and liked them even better when I made my first non-oversampling DAC (read a much earlier version of the same concept your DAC implements) using stacked TDA1543, which was like '98 or so...
I found switching to suitable software for playback (back then Winamp with Ochtan ASIO Plugin and an Asio Cable soundboard with SPDIF out) and ripping all my CD's to a Media Centre PC I had build primarily for TV/Movie use had no negative impact on sound quality. Moving from SPDIF to USB was the next step up.
Never bothered with seedee's ever since, except to rip.
Ciao T
At 20 bits, you are on the verge of dynamic range covering fly-farts-at-20-feet to untolerable pain. Really, what more could we need?
That's experience that I can trust, since you heard what many have not in terms of the NOS DACs. But I'm not ready to swap my disc spinner for a computer.
I suppose that after an EMP I will have those discs and records (but nothing with which to play them other than a pre-amp).
I guess I'm one of the lucky (unlucky?) ones who has never known a well-recorded standard CD to sound that much worse than any other format. I still prefer good old vinyl though.
Hi,
> I guess I'm one of the lucky (unlucky?) ones who has never known
> a well-recorded standard CD to sound that much worse than any
> other format.
Good on you. What you miss though is my point about re-mastering. But never mind.
> I still prefer good old vinyl though.
Well, forgive me please for being such a prick.
But if you never known "a well-recorded standard CD to sound that
much worse than any other format."
How could vinyl be preferrable?
Ciao T
At 20 bits, you are on the verge of dynamic range covering fly-farts-at-20-feet to untolerable pain. Really, what more could we need?
I said that I've "...never known a well-recorded standard CD to sound *that much worse* than any other format".I've listened to originals and some re-masters on vinyl, CD, SACD, DVD-A, HDCD, etc... Once again vinyl is often preferable to all others, but only slightly.
PS: You're not a "prick", or penis. Not a vagina or an anus either.
Edits: 06/17/15 06/17/15 06/17/15 06/17/15
Isn't this why the "memory players "were created?? The tonal saturation and lower distortion levels produced by these units was a revelation -to my ears.I wish that I had converted earlier.
Tom:cat
Of course, not everyone would consider "tonal saturation and lower distortion levels" to be on the order of a "revelation". "Slightly better", perhaps?
.
Dear Mr. Thorsten,
Thanks a lot indeed for the extreme interesting explanation of the cd reading process.
I am quite lost in all this discussion but I cannot help but asking: could it be done differently and more easily ?
I remember reading that playing back wav files in an asynchronous way (I think) could be a much less critical task than reading redbook from an optical medium.
If this is really true converting the cds in wave files could be indeed the solution, the way to follow.
On the basis of what you very well explain the optical cd reading is a very challenging task, with so many variables at play.
It seems that almost everything counts.
And the sonic differences between players and media confirm
The huge number of tweaks appeared on the market are a nightmare for many listeners.
I end saying that today I am more focused on getting the best out of wave files than playing cds. They are just a PITA.
I spent many hours trying to get a good sound out of them and failed.
Thanks a lot again.
Kind regards,
bg
Edits: 06/16/15 06/16/15
Hi,> but I cannot help but asking: could it be done differently
> and more easily ?Yes, or differently and more complex.
You must understand that CD as technology harks back appx. 35+ Years. When CD as system was designed, it represented the ultimo ratio, the sine pari of technology. The 680MB of Data stored on a CD where so cutting edge in 1982, when big size hard drives were measured in MB and 2 Digit or less as well. Computers would take a decade to catch up on storage (also needs).
> I remember reading that playing back wav files in an
> asynchronous way (I think) could be a much less critical
> task than reading redbook from an optical medium.You might have read that from me. I might have mentioned an early Pioneer DVD Player that did that.
> If this is really true converting the cds in wave files could be
> indeed the solution, the way to follow.I have been doing that since the late 90's...
> On the basis of what you very well explain the optical cd
> reading is a very challenging task, with so many variables
> at play.Not really. One merely must understand the variables and find "the knot of Gordias" in this, or what precisely produces the linkage between these variables. Then, like Alexander the Great, we may strike at the linkage and dissolve the knot.
Or we may instead not do it and commit the cardinal mistake, according to Vizzini, of getting involved in a land war in Asia.
BTW, simply ripping CD's to a computer SHOULD be trivial. I would like to submit that it often is a little less so than we would wish.
Ciao T
At 20 bits, you are on the verge of dynamic range covering fly-farts-at-20-feet to untolerable pain. Really, what more could we need?
Edits: 06/17/15
" Hi, Yes, or differently and more complex.
You must understand that CD as technology harks back appx. 35+ Years. When CD as system was designed, it represented the ultimo ratio, the sine pari of technology. The 680MB of Data stored on a CD where so cutting edge in 1982, when big size hard drives were measured in MB and 2 Digit or less as well. Computers would take a decade to catch up on storage (also needs). "Good morning Mr. Thorsten,
yes everything should be put in the right context." You might have read that from me. I might have mentioned an early Pioneer DVD Player that did that "
i do not know why the good ideas do not have more followers.
" I have been doing that since the late 90's...
Not really. One merely must understand the variables and find "the knot of Gordias" in this, or what precisely produces the linkage between these variables.
Then, like Alexander the Great, we may strike at the linkage and dissolve the knot.
Or we may instead not do it and commit the cardinal mistake, according to Vizzini, of getting involved in a land war in Asia.
BTW, simply ripping CD's to a computer SHOULD be trivial.
I would like to submit that it often is a little less so than we would wish.
Ciao T "this also i do not understand. The musical tracks on a cd are not just plain .wav files but are converted to another format. Why ?
I mean what comes out from a recorder is a wav file.
Why convert it again to another format ? this is perversion ?
One day i would like to make a decent wav file and play it back and listen. I am sure it will sound at least quite good. Bearable.
Maybe i am wrong but i think that if they were just plain .wav files the reading process would be much much error safe. Am i wrong ?
Instead ripping tracks from an audio cd involves format conversion and this could lead to errors.
I use usually Exact Audio Copy the with check sum feature.
I think that the ripped tracks can sound quite good ... but it is up to the playback system.
Thanks a lot again.
Kind regards,
bg
Edits: 06/17/15 06/17/15 06/17/15
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: