Improve CD Sound By Shaving?

We always enjoy the odd things that people do to try to get better audio reproduction. Exotic cables, special amplifiers, and higher resolution digitization come to mind.  Most of this is dubious, at best, but [Techmoan] brings up something we must have missed back in the day: shaving CDs with a gadget that was marketed as the “CD Sound Improver.” The theory is that bad CD reproduction comes from light scatter of the laser. The solution, according to the maker of this vintage equipment, is to cut a 36-degree bevel to act as a light trap. You can see the gadget in the video below.

The device claims it reduced vibration, improved audio, and even helped DVDs playback better video. As you might imagine, this has little hope of actually working. The box is essentially a motor-driven turntable, a razor blade, and a port for a vacuum cleaner to suck up the mess. You were told to color the edge with a marker, too.

On the one hand, this seems ridiculous. Of course, it also seems crazy that a marker on a CD would defeat some forms of copy protection, but it did. Apparently, quite a few people plunked down around $500 to trim the plastic off of their CDs. There were plenty of positive reviews, too. But as you might expect, there’s no actual benefit to doing this operation. Nevertheless, [Techmoan] did sacrifice a [Dave Brubeck] CD in the name of science. We don’t think you’ll be surprised at the results.

There’s always debate about how much difference small changes in equipment make to your audio. Unfortunately, most of the easy things probably don’t make a difference you can hear, and to make that kind of difference is much more difficult.

89 thoughts on “Improve CD Sound By Shaving?

  1. coloring the edge with a marker was actually a legitimate thing to defeat some of the stupid copy protection stuff.

    But being digital, improving the accuracy of reading from the CD will only matter for a very marginal CD that is sometimes unable to read the data (it’s 1s and 0s on CDs, no fancy multi-level encoding, remember how old the technology is and that backwards compatibility prevents it from improving in fundamental ways)

    1. Note that that wasn’t the edge as it’s defined here; you actually needed to color the outside ring. The idea was that those CDs had two sessions: the inner and largest one was a normal audio session as read by your CD player, while the outer, smaller session was a corrupted data session. Audio CDs wouldn’t even look at the data session and happily play the audio; PCs would go for the data session and choke on that, not allowing you to play (and thus rip) the audio, in theory at least. Coloring in the outer ring would make the data session so corrupted the PC wouldn’t bother with it anymore, making it act like it was a normal audio CD again.

      1. Interesting. In all of my years of cd collecting, I don’t think that I have ever run into a cd with such anti-copy protection and I have thousands of cd’s but mostly not mainstream modern music.

        1. Blatantly stolen from techmoan channel . But digital is digital the only thing that counts is the quality of the digital to analogue conversion along with integrity of the digital information. Another thing audiophiles need to learn is how electrical signals REALLY travel along conductors. Do the research.

      1. You are correct. This is total rubbish. It is so stupid. Like really I’m aan audio engineer and an electronic technician 66 years old. And these people that think their cd’s sound better after shaving and coloring their discs need to talk to me about my special magic beans i have for sell. ✌️😎🖖👽

      2. Not really. I don’t know what a modern CD player does, but back then it was common for audio to skip, stutter, garble, etc. Most players couldn’t tell good data from bad. Different drives handled troublesome discs differently, and some were sought after because they could defeat certain copy protection schemes or were known to read damaged discs that other drives couldn’t.

  2. On Cds, nothing mechanical makes any difference – ie you either get the complete digital stream or you don’t as there is too many errors for it to fix… What you do with the digital signal is of course a completely different story, as A/D can be done very well or very badly, amplifier can be crap or good, etc etc…

    What surprises me with the above product is a) there are people corrupt enough to want to make money selling a fake product (I know, there a lot of them..), and b) there were people stupid enough to buy it…

    And one simple thing I tell people to do – which actually works – is get a decent pair of cables for your speakers.. They don’t have to be gold plated unobtainium, just some decently thick copper with reasonable connectors..

    1. There might be some legitmately shitty speaker cable out there though. Discount store here was overstocked, had some 50ft packs for 50 cents, grabbed a couple for random hookup wire… stupid stuff is magnetic, thin copper plate over steel cores. Think it was “Welson” on the package, which brand I’ve had okay basic adapters and TV cables from before I think, so I was a little surprised they went as low as that.

      1. Since coathanger wire has been tested as having no audible difference from ‘high end’ speaker cabling (and also works just fine for S/PDIF), a given bit of random wire would have to really exceptionally crap to be of any issue. Worst I could think of would be using too small a conductor diameter for a very high power externally amped driver, but that’s about it – and the failure mode there would be olfactory rather than audible.

        1. Kinda of like the gold-plated TOSlink connectors? (aka optical digital output) Or the $1000/meter twisted pair digital cables (which is really just a very overpriced cat5e cable, and I can make those in my sleep)

          Granted, due to some hearing loss in my irresponsible youth, most audiophile grade stuff is useless on me anyway. :)

        2. Can’t believe some moron thinks coat hanger wire for speaker connections will sound as good as a decent guage/quality ofc copper cable.More likely equipment connected to it is substandard and no amount of good cables will make it sound better.There is a lot of snake oil in the hifi game but you need to look at quality of metals used.Silver ,gold and occ copper materials in a well constructed cable will most definitely make a noticable difference in a quality system,it’s that it wont make your crappy av surround sound anp from harvey norman sound much better!

          1. with large enough guage copper wire (of any type), or probably even copper-plated something (isn’t it all about surface area, or “skin effect”?), the only worries you have are corrosion and connections (besides isolation, of course). Almost any wire that isn’t sufficient for an audio cable can be doubled as many times as necessary to get to get high enough cross-section (surface area?) and sound equal to your gold or silver. Are the speed of electrons faster in gold than in copper (or in fact steel)?

          2. Have you ever simply tried lamp cord? That’s what I use with a Sansui AU-9500 and ESS AMT-1s. Granted, my stereo’s like 6 or so years older than I am, but the amp has been serviced.

            It sounds brilliant. I don’t have “golden ears”, but I highly doubt I could ever tell the difference between this and audiophile investment grade (😆) cable. And yeah, my gear is vintage, but I don’t often hear a stereo that sounds way better +, though I don’t hang out with any audiophiles).

            You don’t need precious metals for @#$ speaker wire or connectors. Well, some people might, but those people would have been able to explain in detail why. If you value the gold ones, it’s not my place to dispute it.

            If I,’m wrong, please explain why. I would like to know.

    2. >you either get the complete digital stream or you don’t

      As someone who worked in the industry, this isn’t really true. CDs are read analogue, and that signal is converted to digital. You’ll always get a digital stream, but it may be junk. A PC drive will usually reject this because a single flipped bit in an app is unacceptable, triggering a re-read. And the OS will then report the whole file or CD unreadable, giving you the impression that you either get a complete stream or don’t.

      Not the case with audio.

      Audio CDs in particular have less ECC, so they are more prone to errors, and audio players would often play through any errors they read rather than go back and re-read because they had to send something to the DAC, and audio players typically weren’t 32x drives able to re-read multiple times before they got something that passes ECC. Also audio players were often portable, which brought its own selection of failure scenarios and large ram buffers.

      So yes, audio players often did send damaged data to the DAC. Typically this resulted in pops or noise in the audio, which an analogue or digital filter might attempt to remove, or momentary skips.

      This device is total junk though.

      1. > You’ll always get a digital stream, but it may be junk.

        True, but the quality reduction will be very different from what people know from analogue media. There is no gradual quality reduction. Either the digital error correction works and is able to reconstruct the exact bit stream, then there is no loss at all. Or the error correction fails, in which case you get essentially random values, since there is no reason why lower-valued bits are more likely to get flipped than the higher-valued ones. The result will be very loud noise.

        Moreover, if the CD is even damaged more severely, the player loses its track completely, in which case there is a 50% chance that it jumps ahead or backwards by one turn (roughly 1/10s) – in the latter case sometimes as an infinite loop.

        Most of these errors are very noticeable. I would call a CD which suffers from such errors broken. So if you do not have such a broken CD, improving the analogue signal quality during readout has literally no effect on the audio quality, because all errors are mathematically perfectly corrected.

      2. “CDs are read analogue, and that signal is converted to digital”

        Only in the optical regime between the laser and the photodiode amp, and only in as much and any digital transmission medium is ‘analogue’ before thresholding.

        The signal stored is digital (pit or land, no in-between), the output of the laser system is digital. It is no more ‘analogue;’ than the eye-pattern viewed on a scape for any high speed signal signalling method.

        1. The output of the laser is high or low, digital, and coded to a voltage with a rise and fall time, analogue, then the voltages go to the error correction in the servo and are converted to a digital stream with all the added, none audio, data parsed out and this goes to the dac, through receiver, reclocker, buffer etc.

          Truth is, a cd player uses both analogue and digital transmission of data at various points, but up until the ass end of the dac chip, it’s all encoded digitally.

      3. Cd players can’t re-read it wasn’t part of the red book spec. Any cd playing box that does this isn’t a cd player by definition, it’s some other class of cd reading device.

        Cd has two types of data integrity protection, the data isn’t continuous, it’s spread over multiple data blocks, none sequentially. Secondly only about 80% of the data is music, the other 20% is error correction, parity bits and other stuff. So even if it can’t read some data it can reconstruct it with 100% accuracy. If the data loss is so great that error correction can’t cope then error concealment takes place, which basically guesses the missing samples, and lastly if the data loss is so gross it exceeds the ability to guess the data then it just glitches or stutters.

        An average cd player is specced to play ten hours of music without encountering a single bit read error, assuming a clean cd is used. Think about that, anything worse than a single bit of 100% recoverable error in ten hours means your cd player wasn’t good enough to get a licence from Sony/philips back in the day.

        Aiui Audiolab are the only hifi company selling a cd player that can re-read data for additional correction, like EAC does.

      4. Wanted to state, there is a sound quality to the type of material the CD player Laser is reading the 1’s and 0’s from , that will affect distortion, especially at higher volumes. The CD player laser x times sampling plus buffering memory to interpet and produce sound signal, the more reflective the CD material is, the CD player does not need as much sampling, which reduces buffering and “more” of signal gets through without errors to produce a cleaner sound signal. I am referring to CD that are specifically released with gold play surface, gold is highly reflective along with eroding of surface, other than by damage. The gold released CD play crisper, cleaner for lows, highs and mids during playback, especially at higher volumes in the 100db + range.

    3. Spot on. Like your cable comments too. “They” talk about skin effect and other utterly irrelevant specs at audio frequencies…but as we all know, stripping back the bs, we have R,C & L!! In most cases it’s “R” …and decents amount of copper doesn’t have much of that.

    4. > you either get the complete digital stream or you don’t as there is too many errors for it to fix

      Not true. There’s redundancy built into the data on the CD, so that small read errors are corrected on the fly.

      A nice explanation: https://www.irishtimes.com/news/science/high-fidelity-how-the-sound-of-cds-stays-error-free-1.2362987

      I also believe that some CD processing chips can “regenerate” the audio lost to some bigger glitches in a way that’s inaudible to the average listener.

      (and who’s kidding who? Y’all listen to streaming files and MP3s and that’s below CD quality ;-) )

  3. Oh, I was wondering if my beard was absorbing high frequency sounds before they reached my ears!
    Hopefully a disposable razor will work as well as a straight edge!
    Is there a particular shaving cream preferred by audiophiles?
    B^)

        1. That’s nice. Leaving aside the Blu-Ray players with only one laser (guess what color!) as the name suggests red like blue is not infrared. One could perhaps call it “infra-blue”.

      1. CD-R’s are much more fragile. I found one that my brother had left in the sun at the top of a CD rack for about three years. You could see the ‘shadow’ left by the rack, and it wouldn’t play at all.

        1. All of my 25 year old cd-r’s play fine as they were recorded on high quality media – old Taiyo Yuden, Mitsui, old Kodak, Falcon, etc and stored in jewel cases. Most of them still scan like new media with low error rates also. Cd’s and cd-r’s both have a lot of longevity if properly taken care of.

    1. After looking at this shaver, it seems like this would be the only valid use for it. You would still have the problem of shortening the life of the disk, but it would certainly balance it. The only concern I have in that use case is whether or not the stamped data is properly centered on the disk. If it isn’t this will only make it spin smoothly, which while a plus, won’t do much for the readability of the disk at those speeds.

  4. In the beginning I thought it’d take a lazerdisc sized platter to do this thing which should be 24 bit but I fell for the Nyquist thing back by ’78. I hated to see too little used to do audio when I learned the disc had to fit into the nose-piece and 2 knobs DIN VW type radio and still hold Beethoven’s 9th. Too few bits, I never considered CD’s any way near perfection.

    From what I remember when CD’s came out 2 kinds of robust data scattering and error detection are used or they would be full of hicks and ticks and never live up to what was promised. That and the focusing below the surface of little scratches on the surface. All in all it was pretty good till smudges and bigger scratches happened then it crashes. Most perfectionists would have spotless CD’s as well as vinyl maybe a record cleaner or CD polisher.

  5. Any CD player that exposes the error rate on read will also show if there’s something to this or not.

    I’d say that keeping the plastic free from damage is the best way to assure high fidelity. But then again, I went from CDs to FLAC just about 20 years ago.

        1. Technically the FLACs are not a copyright violation as long as one holds onto the CDs. Disposing of the media opens one to liability, at least in theory. While it’s pretty much unheard of these days, one never knows what the future might bring. That’s why I still have boxes of CDs somewhere in the garage.

    1. I suggested this in the comments of the video when I watched it. Just pick up a decent Plextor drive that does C1/C2 scanning and run a new disc through a scan to look at the error rates, then perform this process on it and rescan. The resulting data should bear out changes the process makes. I would suspect you’ll see an increase in error rate just from the dust pickup you’re going to get from clamping the CD in thoese rubberized pads.

    1. As Sir Sean said after he was hit on the head by several books: I have no one to blame but my shelf.

      I cut my meeting with him unexpectedly short. Only later did I realize he had asked me if I wanted to sit in the kitchen.

  6. Clever how he did an PROPER A/B comparison showing how imaginary all of those old review articles were which praised this quack device, all swearing it had a positive effect: he used the Audacity app to combine the inverted waveform from a treated CD with one from an untreated CD and it resulted in a zero audio result.

    Something to balance DVDs and Blu-ray disks would actually be useful if it could prevent the buzz I sometimes hear from my player.

      1. It is. Especially in this case where all he’s doing is comparing two recordings to show that they are identical. It doesn’t need to be sophisticated to do A-B=0.

        Now, *if* there actually was a difference in the ‘sound stage’ (or anything else) between the shaved and unshaved CDs, then all Audacity would be useful for would be to show that there was some kind of difference, but not what that difference actually was.

  7. There’s a lot of snake-oil out here for audio cables, be it for sources, pre to amp, or speaker cables. But I’ve been blown away more than a few times with a huge difference between some of them. Emphasis on the “some”.

    One was DIY speaker cables from dual runs of a coax antenna wire per speaker, cross-connected. With one cable hooked up while I worked on the second, with a TV as source into the system, the difference was staggering; I had no idea that source had such quality. Completed, I heard all sorts of detail on pieces that I’d never heard before. Since then, there are same or better speaker cables for less $ and no hassle. And, cheapish purpose made speaker cable I got a spool on clearance from a home improvement store, had a sound quality near identical to the DIY antenna wire sets described above.

    The other was DIY external speaker cross-overs with the metal shells removed from the large film-wrap caps, and the XO components in a cork lined wood box, then filled with beeswax: goal was to limit movement and damp vibration. Some reduction in noise, huge increase in clarity and a warm natural sound. Meant I stopped looking for a new amp, loving what I was getting from my old Bryston.

    1. Effective resistance of wire at audio frequencies is within a pretty narrow range, so the fat line effect of using coax cable will be pretty minimal. Go ahead and use solid or stranded wire of suitable chunkiness. As for reactive response, the fundamental resonant frequency of most zip cord is in the mhz range — far above the realm of any audio work.

      Higher quality audio cable might look nicer or stand up better to handling, but that has nothing to do with audio fidelity and really doesn’t require anything exotic.

  8. Its fairly obvious why the experiment failed. I own the level V model of this device and it lives up to the full expectations. I have PLL speed control with direct drive. The upgraded tooling to diamond tip also made a noticeable difference in the ambient texture of the sound reverbreations.

  9. As others have commented, it could be something that comes into play once you have a really worn disk. I think you should be able to read out the errors through software. I would get a really scratched disk, test the errors multiples times and plot the averages – then “tune” it, then re-test it.

    1. Tries, and FAILS, to explain exactly what is going on. Nothing conclusive is ever documented.

      From the Conclusion section:

      “From my measurements, it is apparent that none of these CD tweaks have any effect on a player’s error-correction ability or on the amount of jitter in the HF signal.”

      Gotcha. So, article over? Oh, no, he keeps going…

      “However, it is beyond doubt that they increase the musicality of CDs.”

      [citation needed]

      “Just as in analog audio, there are things going on in digital audio that have not been identified, but influence sonic characteristics. There is a real need to explore these questions through empirical measurement and by listening. I am convinced that undiscovered optical phenomena in CD playback affect sound quality. Only by combining critical listening with the scientific method can these mysteries be solved.”

      Why not just hold a séance to contact the dead bits lased off of the disc substrate? Maybe they have some ideas.

  10. I’ve actually found it on some old desktop PC optical drive units, that it can be helpful to block the light from entering from the edges of the bay. I would occasionally use this trick when I had semi sketchy discs that seemed to be having problems reading the day that easily.

  11. most of the confusion around this is from the reconstruction of corrupt data. people seem to imply when there’s a corrupt bit that can’t be read properly it is reconstructed in an imperfect manner from the crc. what actually happens is it either gets perfectly reconstructed to the original data, or the track skips because it couldn’t produce valid data.

  12. Every time I start to think people are finally beginning to grasp the concept of “digital”, I’m reminded that they never, ever will.

    Another “neat trick” sure to imbue your music with richer, deeper ones and fuller, clearer zeroes! The benefits can’t be “explained” with “facts” or “logic”, you just have to imagine–er, HEAR–them for yourself!

  13. Best audio CD readers are more about continuity with minimal HW buffering (tipically 1 word), synchronous and silent 1x reading (no re-read) at expense of not getting rid of e.g. C2 errors which are interpolated (loss of details… transparency). No golden hears are required, if no differencies are noted is likely because your reader cannot expose such details… which can be a very good thing for you: just enjoy your music and not become mad following the rabbit in expensive audiophile circles. :-)

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.