Wednesday, March 21, 2012

Mythbusting with Marlene: The green Marker Tweak



The craze for the green Marker began in the late '80s when some guy wrote about this tweak in a newsgroup post, allegedly he was trying to make an Aprils Fools' Day Joke. According to other people it happened this way: an audio magazine wrote a quite serious article about the effect of a green felt marker on a CD. Whoever it was presenting this idea first started a discussion that has been going on to this day: as usual objectivists claim that no measurment a CD can be exposed to shows significant change while subjectivists claim that there are audible improvments regardless of the measurments. The green felt marker can also be accused of starting a new trend in audio tweaking, the trend of treating the CD itself with several fluids, stabilizing mats, etc. All these tweaks work towards one goal: optimizing the reading process of the disc, usually by improving the readability and therefore reducing the need for error correction inside the player itself. Optimizing the readability of a CD is of course a curious goal when one considers that CD technology already posesses an almost perfectly working error correction, a mechanism that´s built into the CD standardization itself - so it cannot be "turned off" by something external. But  the companies still argue that avoiding error correction reduces jitter, the evil of our digital age. There are roughly three basic tweaks around:

  • Coating the outer and inside edges of a CD with a green felt marker
  • Placing a mat onto the CD before it vanishes inside the player
  • Applying a cleaning fluid to get rid off traces of production wax left on the CD

As usual, tweaks like this tend to be very expensive when some company decides that they want to make a few bucks with it. For example, the CD Stoplight Pen costs $ 25. Stupid really when a "normal", not audiophile-approved pen with the same effect would cost just $ 4 - they both are nothing more than usual felt markers. A CD mat like the Millenium CD mat even costs $ 119. A fluid like the L'Art du Son costs $ 45. Yes, you can spend much money on these tweaks! But why would you buy those? Exactly: you won´t need to of course because everything can be achieved with much less investment. As I´ve said above, just use a green felt marker (water resistant) for $ 4. As for the fluid you will have the same effect when you bathe your CDs in warm water with a bit of liquid dis(c)h soap. Or you choose not to try any of these tweaks because you believe that a CD only contains 0 and 1 - computer data - which either works or not. With an opinion like that you are of course prejudiced, exactly like people like me probably believing the opposite. Why don´t you try it out? Maybe you do hear something - or maybe you won´t.

Don´t worry, it´s just a green felt marker


Anyway, since I first read about the green pen tweak in german AUDIO magazine 18 years ago I have been using it. Believe it or not, I always heard an effect when I played such treated CDs with my old Technics SL-XP 300. Yes, this effect could have been a by-the-book placebo effect of course. But the effect of the green marker disappeared when I started ripping my collection to my HDD eight years ago in order to play all my music with my computer. I´ve often ripped a CD twice, once brandnew without the marker and the next time washed in warm water (with dishwashing liquid) with green paint coating the edges. When I compared these rips I always thought at first that I were hearing something but when I listened a bit closer the effect disappeared or was so small that it could have indeed been a placebo. But here´s my dilemma: I cannot stop painting the edges or washing the CDs, even though I know that there probably isn´t any effect at all. I simply grew accustomed to this process during those ten years prior when I was using normal CD players, it´s an addictive compulsion to treat all my new CDs because if I´m not doing it I´m growing nervous. Still, even then I had the nagging thought if it would be possible to measure the effect of a washed and green coated CD. With my recent interest in vintage portable CD players I became aware that I had the ideal tools at my disposal to do some tests with subsequent measurments; I assumed that portable CD players with their need to conserve power as much as possible (-> an underpowered laser) would be more suceptible to green coated discs compared to big home components. My reason to do the test wasn´t trying to disprove objectivists, I simply wanted to find out if there would be any difference at all.

So I devised a relatively elaborate test routine: I created a disc with five test signals (made by RMAA), five tracks for jitter measurment and two tracks with music. I used five tracks for measurment and jitter because I wanted to rule out any deviations I might encounter, deviations created by the possibly unreliable equipment I have to use: my PC in combination with the E-MU. I also used a brand new disc (The Perfect Storm by James Horner) I had purchased recently (the purpose of the music examples will be explained later). My test setup hasn´t been changed since the last test, I´m still using my E-MU 0202 USB with a recording samplerate of 192 kHz and a bit depth of 24 Bits, recording is done via ASIO with Sound Forge. The RCA cable still is the same for every step: an Audioquest King Cobra. For the tests I used the following three CD players:

  • Technics SL-XP 300 (1991) as an example for an old CD player
  • Sony D-465 (1996) because it reacts unreliable to some CD-Rs
  • Sony DE-J 915 (1999) because it´s the newest of all my players

I then recorded the untreated discs played back by these three players with my E-MU. Then I washed the CDs and painted their edges with the green felt marker you can see above. After the paint had dried I repeated the tests in exactly the same way with the treated discs. I then edited the recordings so that they could be analyzed by RMAA (I could only use four of the five test tracks for the charts below) and to measure the amount of jitter. The final step was resampling the tracks to 44.1 kHz, I kept the bit depth of 24 Bit. Here are the results:

Technics SL-XP 300, untreated

Technics SL-XP 300, washed and painted
Sony D-465, untreated
Sony D-465, washed and painted
Sony DE-J 915, untreated
Sony DE-J 915, washed and painted

"GM" means Green Marker. And as you can see, RMAA revealed that there weren´t any differences. Even the graphs showed no differences, in the following pictures all tests for the respective players have been combined into one graph and in order to improve the visibility I changed the colour for all lines to the same green. I will present to you the graphs for Intermodulation distortion of the D-465, these two shall serve as a placeholder for all the other graphs RMAA produced:

IMD, Sony D-465, untreated
IMD, Sony D-465, washed and painted

As you can see there still is nothing to see. All three players produced similar pictures for noise, dynamics, THD, IMD, crosstalk and frequency response, whether the discs had been treated or not apparently didn´t matter. Needless to say that I was surprised, I expected to see differences with the Technics because it was the first CD player with which I´ve heard the effects all those years ago. But the graphs for the Technics were similar all the time. By now you must be thinking that I was a victim of an exemplary placebo effect all those years ago. The jitter tests should therefore provide more insight to the effects because jitter always is proclaimed to be influenced the most by the green marker:

Technics SL-XP 300, untreated
Technics SL-XP 300, washed and painted
Sony D-465, untreated
Technics SL-XP 300, washed and painted

Shall I go on and bore you with basically the same pictures for the Sony DE-J 915? It seems that the proponents of the jitter theory have been wrong all these years. But there´s one test still ahead, I´ve said I wanted to find out if the output produces differences with normal music, differences RMAA was unable to detect because it´s entirely possible that RMAA doesn´t measure everything with the effect that some things go by unnoticed. For that purpose I used a software called Audio DiffMaker. According to the author this software does the following: "Audio DiffMaker is a freeware tool set intended to help determine the absolute difference between two audio recordings, while neglecting differences due to level difference, time synchronization, or simple linear frequency responses." I won´t bore you with technical explanations, I just would like to add that it seems to do its job well enough so that a person attempting to find differences will be presented with evaluable results, results this person can listen to. That´s when the music tracks came into play, I recorded them just for the purpose of extracting their differences with the Audio DiffMaker. For your convenience I made these differences - or lack thereof - visible with iZotope's spectrogram:

Extracted audio differences between treated/untreated discs, Sony D-465
Extracted audio differences between treated/untreated discs, Sony DE-J 915
Extracted audio differences between treated/untreated discs, Technics SL-XP 300

As you can see, both Sonys sounded the same both times, the two pictures above are only two examples because as with the measurments above every musical track looked equally "empty". The picture showing the diffmaker result from the Technics however is the interesting one because it does indeed show differences. Again, I extracted the differences of three musical pieces, the result always looked similar to the one above. When listening to the extracted audio signal I was clearly able to make out the music, it just was stripped of bass and mids, leaving treble only. This is a strong indication that the Technics is suceptible to the effect of the green marker, whatever the effect is and despite the RMAA measurments not producing any results. It also explains why I could hear differences when I used the green marker on my CDs all those years ago. Before you ask: yes, I did a null-test after I reviewed the results from the Technics (no differences).

Conclusions

Judging from my experience and my measurments I think it´s safe to say that the green marker does have a small effect... somehow. But before anyone starts to scream "I was right!" let me remind you that my results should be taken with a grain of salt. First of all, the Audio DiffMaker could work imprecise, it´s even possible that it won´t differentiate every sonic aspect between two files. The software seems to differentiate frequency responses but what about staging? Staging in a musical piece is a complex construct not only influenced by frequency response but by several other factors like perfect impulses, timing etc., does it extract those differences too? Another possible explanation is that - despite my null test - the resolution of the differentiation process isn´t high enough with the result that the program produces differences where none exist (even though I tweaked the process to be more thorough). It could also mean that my Technics is faulty and cannot read a CD properly - although I doubt that since it reads CDs that another player won´t even accept. Another explanation is that older CD players react more strongly to the green felt marker and newer ones don´t. One thing however is crystal clear: the differentiated signal is very low in amplitude, even if it shows on a spectrogram and is audible on its own it still doesn´t mean that it´s audible every time. However, after writing this article I feel as smart as I did before, I´m not very lucky with myself. It would have been better to have no differences with every player or to have differences with all of them. Whatever the reason behind the effect, I will continue coating the edges of my CDs, simply because I am addicted to it without any logical reason. Call me deluded or ill, it doesn´t matter: I cannot be calm without doing it. Why don´t you try it out for yourself and comment about your experiences? I´d love to read about it - but not now because I´m going to undress my stockings to go to bed. Good night, my dear readers.

EDIT 04.02.14: Almost two years have passed since I wrote this article and something has happened... well, I make it simple: I´ve stopped coating the edges of CDs. Yes, I don´t use the green marker anymore. The reason is that the Technics SL-XP 300 is the singular player reacting to it, none of the others is susceptible to the 'effects' of green colour coating the edges of a CD. Really, one player out of 30 or so is not enough to merit the time needed for the procedure. I´d rather spend my days listening to music.


Last update: 04.02.2014
Related Posts Plugin for WordPress, Blogger...

The Socials