Skip to main content ↑ ↓

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: On "24-bit digital remasters" and remastering in g (Read 26083 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

On "24-bit digital remasters" and remastering in g

Reply #75 –
:-)

(how did you know I was thinking about some of the music I heard via these forums? ;-) )

anyway, Armor All doesn't seem to rot 78s - I wouldn't advise using it generally, but it can help some poor condition ones.


Quote
Edit -- OK, I confess I'm basically out of touch... I wouldn't know a song by N'Sync from a song by Brittney... 


Be proud Sir!!! and maintain this ignorance at all costs!

D.

On "24-bit digital remasters" and remastering in g

Reply #76 –
Quote
Originally posted by fewtch

Clearly, people still aren't understanding how CD's work and what *digital* really means.  Perhaps a course at a local tech college... ?

How do you think a carbon brush is going to improve the sound of a CD?  As long as the laser can read the bitstream, the music sounds fine.  Even a lot of dust on the CD should have minimal effect... at most, it will cause "skipping" or clicking sounds due to uncorrectable errors (fingerprints are far worse than dust). 

Static on the surface of a CD will have *zero* effect.  Magnetism will not erase a CD (yes, maybe that has to be explained too!!!)

I'm not specifically referring to you Pio, but are people really thinking that bass, treble, stereo separation, instrument placements, etc. can be affected by imperfections related to the CD? 


A flawless CD can certainly sound better than a scratched CD.  Granted, you'd need more than a single scratch to make a hugely noticeable difference, but either way, what a normal CD player plays would not be identical to the original recording pressed onto a flawless CD.

The issue is really the error correction algorithms in a CD player.  What you get is not an exact replication of an unreadable section, but a best guess at what used to be there.  In most cases this result is un-noticeable, but I don't find it far-fetched that somebody could A/B a heavily scratched CD.

As for the pen being discussed here, it sounds like total bunk.  That is unless it evenly coats the surface of the CD and changes some of those ones and zeros into .5's and 1.5's  Waxing the surface of a heavily scratched CD will reduce the unreadability of the scratches, but you're probably better off taking good care of your CD's to begin with.

G

On "24-bit digital remasters" and remastering in g

Reply #77 –
Quote
Originally posted by gdougherty

The issue is really the error correction algorithms in a CD player.  What you get is not an exact replication of an unreadable section, but a best guess at what used to be there.  In most cases this result is un-noticeable, but I don't find it far-fetched that somebody could A/B a heavily scratched CD.


That's not exactly true. Redbook CD, has, as any other digital storage media, redundant information stored in it. That means that the player can "perfectly" correct (or better say recover) ocassional errors. Only when there are too many errors to be recovered from the redundant information, the player can resort to interpolate missing samples. But if there are too many errors, and depending on the player, it will make clicks, or will softly fade out and in the sound over the damaged area.

These error-recovery correction algorithms are part of the standard redbook cd specifications, and any decent cd player should implement them. It's also true that there is less redundant information in redbook cd in comparison with DAT or cd-rom formats, it is a less robust system because it is not so critical to lose a few samples from time to time, as these can be interpolated.

On "24-bit digital remasters" and remastering in g

Reply #78 –
Quote
Originally posted by KikeG

you can argue that you did blind tests, and I don't doubt that you did, but I'd like to know more details about these tests. What was the exact procedure? Were they double blind, single blind? How many trials? How many correct identifications? Same content at painted and non painted cds?


Let's see if I can remember that long back... It all began in high school 1992 - 1993. Me and two other guys tried every tweak in the book. This includes sorbothane(rubber) and spikes under cd player and other units, several different expensive cables, green pen, wax for bottom side, metal plate stabilizer on top, cd-lens cleaner, cable-plug cleaner, garden tile under speakers and other hifi-units, foam rubber on the walls, wadding on the walls and ceiling, various carpets and more. Actually much much more.

Anyway, we tested the green pen on these three cds:

Meat Loaf: Bat out of Hell
Alanah Myles: Alanah Myles
Ozzy Osbourne: No More Tears

We all owned these cds in same versions.

The first thing we did was to perform sporadic tests to insure that the cds did indeed sound equal. Neither of us detected any differences.

Then we applied green paint to one of each cd. One of us played the disc jockey and played the three discs (2 ordinary and one green) in random order on the Denon DCD-580. Then we repeated the tests with another disc jockey. Test results were first discussed after completed testing. The green discs were identified with 100% accuracy. It was really easy, and the DCD-580 generally sounds horrific.

Then we moved to another location to the Denon DCD-695. Again we performed the tests and could detect the discs with 90%+ accuracy (I don't remember exactly).

We did not alter any other factors during the tests.


I remember the green sound to have firmer bass (higher resoultion). But I might not remember correctly. It is afterall around 10 years ago.

And unless you are 30+ years old and worked as an electronics engineer some 10-15 years ago, I would be careful with my statements. Technology has changed a lot. 10 years ago CD-ROM drives were spinning 2X. Most DACs sounded terrible - people talked about how much better vinyl sounded than cds. Blue lasers/DVDs were science fiction.

So to round off: We all know how the digital media works: We transfer a stream of bits. We also know that there are too few error correction bits in the CD format. CD players use oversampling as well - they read the same data several times and compares to compensate.
How does the player read the bit stream? A laser emmits a beam onto the rotating disc. The disc is a piece of plastic with some "bumps" covered with a reflective layer of aluminium. When the laser beam hits the reflective layer it gets reflected back and can be read by a sensor (the cd lens). Depending on whether the laser hit a "bump" or not it reflects light differently. Do we only get reflections back to the sensor or do we have reflections all over the place? Will some of these hit the other and inner rim (also recommended by the pen manufacturer). I'll stop now.

On "24-bit digital remasters" and remastering in g

Reply #79 –
Quote
Originally posted by KikeG

These error-recovery correction algorithms are part of the standard redbook cd specifications, and any decent cd player should implement them. 


The error correction can only recover ONE bit error.
If there are two incorrect bits, the error correction can only detect the error but not do anything about it.
You can even produce enough bit error to make make the error undetectable. I don't remember the math.

That is very little and that is why audio extraction is such a huge problem on computers.

On "24-bit digital remasters" and remastering in g

Reply #80 –
Quote
Originally posted by fewtch
How do you think a carbon brush is going to improve the sound of a CD?  As long as the laser can read the bitstream, the music sounds fine.  Even a lot of dust on the CD should have minimal effect... at most, it will cause "skipping" or clicking sounds due to uncorrectable errors (fingerprints are far worse than dust). 
I think he was kidding. At least, I'm pretty sure he was...

On "24-bit digital remasters" and remastering in g

Reply #81 –
Quote
Originally posted by fewtch
As long as the laser can read the bitstream, the music sounds fine.  Even a lot of dust on the CD should have minimal effect... at most, it will cause "skipping" or clicking sounds due to uncorrectable errors (fingerprints are far worse than dust). 


I've posted a link to the best explanation of jitter in the 8th post of this thread, but since people don't always take the time to go and see (and the paper is long), I then posted an explanation in the 10th post.
Last post of page 2, 2BDecided posted an explanation about how badly isolated DAcs were affected by interferences. I confirm this : in an SB64 V, you can hear the CPU computing : the noise increases when MP3 encoding starts.

Since people kept saying that it was impossible for 1 to sound otherwise than 1 and for 0 to sound otherwise than 0, I repeated, first post of page 3, that the bits have to be played in order for the music to run, and that if they are played at the wrong time, the sound will be distorded, be them lossless.

And in Bob Katz's tests about jitter (http://www.digido.com), he clearly states that the tested CDRs had been analyzed and had exactly the same data on them.
I checked the SPDIF output of my Yamaha CD Player : not a single read error over 4 minutes of playing in the worst conditions : cheapest and worst sounding non-75 Ohm coax cable, player on the top of a stack of magazines.
http://pagesperso.aol.fr/lyonpio2001/proof.htm
Therefore differences in sound, if they exist, are not caused by read errors.



I wasn't kidding. A lens is made of plastic, as is the CD, and it is very sensitive and very close to the CD.
You know that different signums attracts each other, while same signums push themselves aside.
Therefore if the electric charge is not evenly distibuted on the CD, the lens will be alternatively attracted and pushed. Alledgedly, the carbon brush solves this problem.

Therefore I ran the test :

Source : YamahaCDX860 CD Player, 1991, 450 €
CD Talvin Singh, OK, track 2, 40 firsts seconds.

Analog output, RG179bu cables, Sony DTC 55ES recording at 48 kHz, optical output, Fostex COP-1 optical/coaxial converter, Marian Marc 2 coaxial input, settings : master clock=digital input, SoundForge 4.5, options/recording device=Marc2 digital in, new file=48 kHz, 16 bits stereo.

One recording, eject/close/optical plugged into the SPDIF output of the Yamaha/SoundForge, new file@44.1, recording
Back to first setting and recording again.
Brush the CD, recording again.
Eject, brush, close, Yamaha digital out, recording @44.1
Eject, brush, close, Sony digital out, recording @48

Summary : two analog and one digital recordings without brush, same with brush.

Subjective impression during recording (listening to Sony analog output) : first analog recording with brush sounded a little better, second analog recording with brush sounded much better.

EAC - compare wavs : no differences between the digital copies (with and without brushing) and a secure extraction : there was no read errors.

Listening to the recorded wavs : 
No audible differences.

Analysis with soundForge : the problem with pasting wavs inverted in order to find their differences, is that the sampling is not made at the same time each recording. The timing of the samples can vary up to 1/88200th of a second, and there are big differences in level from one sample to the following.
Therefore I set the sample rate (without processing) to 6000 Hz, then resampled to 48 kHz with highest quality (4) no antialias.
It allowed to adjust the offset of the pasting with 1/8th of sample accuracy.
The RMS resulting power (=difference between the wavs) rises from -55 db to -25 db 40 seconds later.
It means that the files were not played or recorded at the same speed.
The comparison is not possible.

Conclusions :

No audible differences were recorded before and after wiping the CD with the carbon brush.
The recorded files are so different from each other that an analysis with cut'n paste inverted is not possible, even the two files recorded one after another without brushing the CD.

I can send the files to anyone willing to resample them until they match in lenght, but I don't think anything will come from the analysis.

My opinion, reading annuka's posts, and having myself tested the sound of CDRs on a JVC CD Player, and not getting the expected results (I expected worse sound from the copy, and I got better, so placebo effect is unlikely) is that old bad CD Players were sensitive to the quality of the media, while good CD Players are not.
It is possible with the JVC that it came from read errors. That player didn't have digital output so it was not possible to check.

On "24-bit digital remasters" and remastering in g

Reply #82 –
OK then, well one final thing remains to be resolved:

How is it that green marker pen results in "better bass" and treatment of the surface results in "better treble" (as per annuka's claims) and how does bass and treble relate to jitter?

P.S. fwiw I've known for years about how badly isolated audio components can pick up interference from the CPU.  I used to turn up the volume on my Commodore 64's speaker (back in 1985),  and listen to the CPU make different types of sounds as it ran different program loops, etc.  Actually it was fascinating, the variety of sounds depending on CPU activity.  It seems with a 1MHz CPU, there was much more available interferences in the audio range.

On "24-bit digital remasters" and remastering in g

Reply #83 –
Quote
Originally posted by fewtch
OK then, well one final thing remains to be resolved:

How is it that green marker pen results in "better bass" and treatment of the surface results in "better treble" (as per annuka's claims) and how does bass and treble relate to jitter?


I don't trust words.
I qualified a Rega ampli as "agressive", and the salesman said "strange, most of the times it is said very soft".
I don't know what "open", "soft", "fast", or "accurate" mean in a subjective listening.

"Better bass" certainly doesn't mean that there are more or less bass, it can mean anything. It can mean that the treble was different, so that the listener can focus his attention to the bass part...anything...

On "24-bit digital remasters" and remastering in g

Reply #84 –
Quote
Originally posted by fewtch
OK then, well one final thing remains to be resolved:

How is it that green marker pen results in "better bass" and treatment of the surface results in "better treble" (as per annuka's claims) and how does bass and treble relate to jitter?


It is actually quite easy to explain, but the explanation may be false.

The green pen was among the first tests we performed back in high school. We all had pretty cheap hifi systems and bad accoustics. The systems with the Denon DCD-580 and DCD-695 were quite alike. Cheap Denon AMP, cheap B&W Speakers, same cheap speaker cable. It is quite possible that the ordinary cd signal provided a much higer resolution treble and mid-tones than the system could reproduce. But the system could reproduce a much better bass than the ordinary cd signal. So the green pen might have provided a general improvement, but we only noticed the bass as the rest of the systems were crap.

Many months later the wax was released. Our systems had changed a lot (new components and modifications) and we did not apply the same thoroughness as we did in the beginning (it is very hard work). This time we noticed mostly treble. I believe we tested the DCD-695 and a DCD-980 (way better than the DCD-580).

It is also possible that my memory has degraded over the years.

Audiomusings.com describes the green pen difference like this:

>>CD Stoplight, the fabled green pen, is used on the inside and outside edges, and the raised rim at the center of CDs. An easy and effective way to enhance the performance of the digital medium. Increases detail and musical cues due to less laser scatter, as shown by actual tests that imply a reduction of sampling errors.<<
c.f.: http://www.audiomusings.com/Tweaks.htm

On "24-bit digital remasters" and remastering in g

Reply #85 –
Interesting page.  But the guy forgot:
[Sarcasm]
* Using audio equipment in outer space, to minimize effects of Earth's gravitational pull on mechanisms
* Custom-built clean-room, always wear a white suit and air-lock decontaminate before playing any CD.
* Soundproofing the room with the same material as used in laboratories, to eliminate noises from outside
* Full-body wash in 10 disinfectant solutions of gradually increasing strength, before approaching within 100 meters of audio equipment
* Always wear a hat near CD-players, so a stray hair won't fall into the mechanism and destroy the sound
* Disinfect a CD with hospital-grade germicides, because bacterium and virii block the laser light
* Brush teeth before listening, as coating on teeth/tongue can affect hearing in extremely subtle ways...
 
[/Sarcasm]

On "24-bit digital remasters" and remastering in g

Reply #86 –
Quote
Originally posted by fewtch
* Brush teeth before listening, as coating on teeth/tongue can affect hearing in extremely subtle ways...
 


LOL, I'm sure you'd love this device called ionizer (in French) that cancels the electric charges in the air in order for the sound to be better transmitted from the speakers to the ears 
This did existed ! Audiophile shop sold (or at least tried to) some !

On "24-bit digital remasters" and remastering in g

Reply #87 –
Quote
Originally posted by Pio2001


LOL, I'm sure you'd love this device called ionizer (in French) that cancels the electric charges in the air in order for the sound to be better transmitted from the speakers to the ears  
This did existed ! Audiophile shop sold (or at least tried to) some !

I believe it... 

So who wants to buy the first tube of "ClearAudio" toothpaste, with the special brush... the toothpaste is just $30.00/tube, and the brush is only $25.00.  A great bargain, decreases hearing losses due to plaque!  Not only that, but dentist drills are shown to ruin hearing.

For a limited time, a version with fluoride is only $15.00 extra per tube. 

On "24-bit digital remasters" and remastering in g

Reply #88 –
fewtch wrote:

Quote
OK, I confess I'm basically out of touch... I wouldn't know a song by N'Sync from a song by Brittney


Hehe, if it's a boy singing, it's Nsync, if it's a girl, it's Britney.  I really wouldn't consider either of those "artists" as dance groups, though.

Today's equivalent of Depeche Mode or New Order would be maybe groups like Orbital, Daft Punk, etc, which rarely get played on MTV or the radio.  And, that 80s synth sound is huge in a kind of underground way, right now.  The Felix Da Housecat album (which won several awards for album of the year in Europe, I believe) was so 80s, it gave me flashbacks.

On topic, I remember hearing about those green pens back in the 80s even, but never tried it.  Also, lately I've seen a few places that sell plastic rings that you stick on the top outer edge of a cd that are supposed to balance the cd while it's spinning and give you better playback.  And, I've also recently seen for sale clear plastic covers that you stick on the data side the cd for extra protection from scratches.  It seems to me that there is more risk of unbalancing the disc, rather than balancing it.  That can't be good, can it?

On "24-bit digital remasters" and remastering in g

Reply #89 –
Quote
Originally posted by Pio2001
I wasn't kidding. A lens is made of plastic, as is the CD, and it is very sensitive and very close to the CD.
You know that different signums attracts each other, while same signums push themselves aside.
Therefore if the electric charge is not evenly distibuted on the CD, the lens will be alternatively attracted and pushed. Alledgedly, the carbon brush solves this problem.

Therefore I ran the test :
Okay, thanks for proving me wrong again. I won't make another presumptive post in this thread, or else I might make a bigger fool of myself. :insane:

So let me make sure I understood the results of your test (there's a good chance I misinterpreted them, so please correct me):

    On "24-bit digital remasters" and remastering in g

    Reply #90 –
    When performing listening tests, it is vital that the temperature stays constant during the sessions. A normal climate system isn't suitable, as it is noisy and disturbs the air. The solution is to install a water cooled aluminuim ceiling (around $250,000)... And so they did at the Technical University of Denmark.

    So $20 a pen is next to nothing to the true audiophile :)

    On "24-bit digital remasters" and remastering in g

    Reply #91 –
    Quote
    Originally posted by Annuka
    When performing listening tests, it is vital that the temperature stays constant during the sessions. A normal climate system isn't suitable, as it is noisy and disturbs the air. The solution is to install a water cooled aluminuim ceiling (around 0,000)... And so they did at the Technical University of Denmark.

    So  a pen is next to nothing to the true audiophile

    Heheheheh... Ok, how long have you been pulling people's legs?

    Sir, you are mad.  So are all "true audiophiles."  After all this talk, that's the only sane conclusion!

    Now I know the favorite hobby of people diagnosed with obsessive-compulsive disorder. :idea:

    On "24-bit digital remasters" and remastering in g

    Reply #92 –
    Quote
    Originally posted by fewtch
    OK then, well one final thing remains to be resolved:

    How is it that green marker pen results in "better bass" and treatment of the surface results in "better treble" (as per annuka's claims) and how does bass and treble relate to jitter?


    Well, I think you have to start with some really good drugs. The kind that I haven't had in many years. Once you've estabslished a good drug induced baseline, it's time to start testing
    flac > schiit modi > schiit magni > hd650

    On "24-bit digital remasters" and remastering in g

    Reply #93 –
    Well, well, let's be serious

    Quote
    Originally posted by SometimesWarrior
    • There was an audible improvement when you took the CD out of the player, used a carbon brush on it, and then put it back in when listening to the analog playback. Repeated brushings continued to improve the sound quality.


    It was a subjective ephemeral quality improvement 

    Quote
    Originally posted by SometimesWarrior
    • The two digital recordings were bit-identical, before and after brushing. (Question: did they sound the same during playback? Or was your playback chain set up so that you couldn't monitor the audio until it was already recorded to the hard drive?)


    I didn't listen carefully enough, they sounded more or less the same.

    Quote
    Originally posted by SometimesWarrior

    • When playing back the (digital) recordings of the analog outputs, there was no audible difference, although a difference could be heard when the recording was taking place, before the sound was digitized by the soundcard.


    Well, I'd rather say that my attention was more or less sharp. You know, when listening several times to a given piece of music, we can hear more and more things each time. It's very easy for the hifi shops to make people listen first to the cheap gear, then switch to the best.
    Usually, listening two times to the same sound, it seems better the second time.

    The test simply proves that the differences were in my hearing. If the difference I percieved during the recording was objective, I would hear it in the recorded files.

    Quote
    Originally posted by SometimesWarrior

    • The recordings of the analog output do not match up accurately enough to make quantitative comparisons of the recordings possible.


    Yes, at least with this method that consist in inverting one file in order to paste it, so that the two files cancels themselves, and only differences remains.
    All I could see is that all recording sessions captured a different sound, and the difference between bush/no brush was not significantly bigger than the difference between two recordings made in the same conditions (brush/brush or no brush/no brush), according to this analysis method.

    On "24-bit digital remasters" and remastering in g

    Reply #94 –
    Quote
    Originally posted by Annuka


    The error correction can only recover ONE bit error. 
    If there are two incorrect bits, the error correction can only detect the error but not do anything about it.
    You can even produce enough bit error to make make the error undetectable. I don't remember the math.

    That is very little and that is why audio extraction is such a huge problem on computers.


    sorry, you're thinking of ECC memory in computers, and that has absolutely nothing to do with the error correction data present on an audio CD.

    On "24-bit digital remasters" and remastering in g

    Reply #95 –
    Quote
    Originally posted by Pio2001

    I wasn't kidding. A lens is made of plastic, as is the CD, and it is very sensitive and very close to the CD.
    You know that different signums attracts each other, while same signums push themselves aside.
    Therefore if the electric charge is not evenly distibuted on the CD, the lens will be alternatively attracted and pushed. Alledgedly, the carbon brush solves this problem.


    except that there is a relatively powerful voice coil mechanism in place to position the lens and any static charge on a CD isn't going to be anywhere near powerful enough to have any effect whatsoever on the lens.

    On "24-bit digital remasters" and remastering in g

    Reply #96 –
    Quote
    Originally posted by _Shorty


    sorry, you're thinking of ECC memory in computers, and that has absolutely nothing to do with the error correction data present on an audio CD.


    Enlighten me by telling exactly how the error correction works.

    On "24-bit digital remasters" and remastering in g

    Reply #97 –
    http://www.siam.org/siamnews/mtc/mtc193.htm might be a good start.

    "Current implementations of Reed-Solomon codes in CD technology are able to cope with error bursts as long as 4000 consecutive bits."

    if you google "Cross Interleaved Reed Solomon Code" you should find a great deal of useful and/or enlightening information.

    <edit> there's some more here too http://www.roxio.com/en/support/cdr/cderrors.html

    On "24-bit digital remasters" and remastering in g

    Reply #98 –
    Quote
    Originally posted by _Shorty
    except that there is a relatively powerful voice coil mechanism in place to position the lens and any static charge on a CD isn't going to be anywhere near powerful enough to have any effect whatsoever on the lens.


    It was just an hypothesis, if the coil is powerful enough not to be affected by static charges, then I'm wrong. Anyway, the recording test gave no audible result.

    On "24-bit digital remasters" and remastering in g

    Reply #99 –
    Quote
    Originally posted by Annuka

    Then we applied green paint to one of each cd. One of us played the disc jockey and played the three discs (2 ordinary and one green) in random order on the Denon DCD-580. Then we repeated the tests with another disc jockey. Test results were first discussed after completed testing. The green discs were identified with 100% accuracy. It was really easy, and the DCD-580 generally sounds horrific.


    So, it was a single-blind test (SBT). I wonder if the result would have been the same in a controlled double-blind test (DBT) with more trials. A single blind test can be not very reliable. I don't know of any DBT that has shown differences with green pens.

    Quote
    Me and two other guys tried every tweak in the book. This includes sorbothane(rubber) and spikes under cd player and other units, several different expensive cables, green pen, wax for bottom side, metal plate stabilizer on top, cd-lens cleaner, cable-plug cleaner, garden tile under speakers and other hifi-units, foam rubber on the walls, wadding on the walls and ceiling, various carpets and more. Actually much much more. 


    And, what was the result for these other tweaks? It would be interesting to know.