D02 - A Quiz Game

Musical games to help develop improvisational skills.
Laura Staats

Re: D02 - A Quiz Game

Post by Laura Staats » Tue Feb 18, 2014 5:48 am

Neat idea Carl. I don't feel particularly confident in my guesses. :oops:
1: no capo
2: capo 1st fret
3: capo 2nd fret

Set 1

Set 2

User avatar
Posts: 589
Joined: Sun Dec 04, 2011 3:36 pm
Location: Chattanooga, TN

Re: D02 - A Quiz Game

Post by CarlWestman » Fri Feb 21, 2014 2:50 am

Thanks for chiming in, Laura! Beatriz and Satyajit, I think I may only have your guesses for Set 1. I'd encourage you to take a guess at Set 2, which may be easier anyway.

User avatar
Satyajit Kadle
Posts: 203
Joined: Mon Apr 22, 2013 10:30 am

Re: D02 - A Quiz Game

Post by Satyajit Kadle » Fri Feb 21, 2014 5:45 am

Not confident with the second set either. However,
1) 579.1
2) 650
3) 613.5
In the order uploaded.

User avatar
Posts: 589
Joined: Sun Dec 04, 2011 3:36 pm
Location: Chattanooga, TN

Re: D02 - A Quiz Game

Post by CarlWestman » Mon Feb 24, 2014 3:18 am


Don't read further if you don't want to know the answers yet. I've decided to post the answers a little early due to the lack of activity as well as to leave some time for discussion for anyone interested.

I appreciate all those who participated in this listening game; I had hoped that more would chime in, but apparently its appeal is limited. So here are the answers. I hope I interpreted your guesses correctly; feel free to correct my interpretation if I misunderstood you.

For sample set 1, the first recording was capo 2, i.e., 579.1mm scale. Beatriz got this right. The second recording for sample set 1 was full scale 650mm. Marko and Satyajit got this right. The third one was capo 1, i.e., 613.5mm scale. No one got that.

For sample set 2, the first recording was full scale; the second was capo 2 = 579.1mm scale, and the third was of course capo 1 = 613.5mm scale. Everyone got the third one, but not the first two.

Of course, one cannot draw any firm conclusions from this, but it appears that it can be hard to tell the differences in scale (and string tension) through audio recordings. It is probably more obvious to the player, who feels the string tension differences. This also led me to think about sound generally. As players, I suspect the sound qualities we hear vary from any in-person listeners, because of our position above and behind the soundhole. It might be interesting to position the recorder at our own ear level and compare it to what it sounds like in front of the guitar. I bet it differs significantly. If so, this also has implication when we test out guitars in shops and stores - what we hear isn't necessarily what our listeners (or mic) hear(s).

Anyway, thanks for playing along. Cheers,

User avatar
Marko Räsänen
Posts: 3816
Joined: Thu Feb 16, 2012 11:56 am
Location: Finland

Re: D02 - A Quiz Game

Post by Marko Räsänen » Mon Feb 24, 2014 9:16 am

It looks like we might have just taken random guesses. The probability of of guessing one out of three right (but the remaining two wrong) is in any sample set 3/6. The probability of getting none out of three right is 2/6, and the probability of getting all the three right is naturally 1/6. So it's two times more probable to guess at least one right than all 3 wrong.

I'm not really surprised about the result, as I think my own guesses were quite 'optimistic'. I wonder how different the situation would be if there were three identical guitars, tuned the same way Carl did, and the listener was blind-folded? That would be the same test, but without the recorder in-between, and possible differences in sound volume could be heard. I guess the main thing that makes this very hard is the player possibly compensating for the string tension. Though it could be that differences in tension are so minimal that even with a robot arm plucking the strings, no difference could be heard.
Alhambra 4P spruce
Almansa 457 cedar
Cordoba C12 spruce


Re: D02 - A Quiz Game

Post by soufiej » Sun Mar 09, 2014 7:20 pm

I will take the time to listen to the samples a bit later when things are somewhat less noisey around the house. However, if I may, I'd point out a few flaws in your quiz. The wine tests are somewhat well known and used in many debates which center on subjective perceptions vs objective "facts". In other words, the perceived desire of the well heeled to prove they are superior to the rest of us. My career has been mostly involved with high end audio and that has taken me into numerous ... er, "discussions" which involve whether what has been claimed can actually be perceived and repeated. Or, is it all "snake oil" and objective "facts" are all the facts that count and the amplifier with the lower distortion and higher measured wattage will always be the superior product? Due to the dollar amounts asked for certain pieces of equipment and the highly personal situation in which music is perceived by the individual listener, there are grand arguments waged between the various camps who at times have their very existence - if not their reputation and their following of loyal supporters - on the line.

The ability to detect small and at times momentary dissimilarities between components or even something as seemingly inconsequential as the directionality of signal flow in a speaker cable is what "high end" audio is based upon. The basic question is, why would anyone spend the money for any product they did not perceive to be "better" than another? The answer to that question is to resort to double blind tests which are meant to determine whether perceptions are honest and consistent or simply a function of expectation bias and the commonly known placebo effect ... and its near relative, the "no-cebo effect". If you believe, you are 90% of the way to perception or lack there of.

High quality blind testing is a very difficult thing to accomplish. Small errors which would not present themselves in another situation are heightened in their effects once a subject is asked to determine which is which and which is superior. An almost ancient article written by the man who created subjective listening audio reviews back in the 1960's is included here;

One point which stands out in any such test is the approach taken by the test subject once they have been put on notice their responses are being graded. In such a situation, the listener mentally switches gears and is no longer listening to music for pleasure or even as a critic of the performance of the player. Another portion of the listener's mind is switched on and now the listener is being asked to determine which is which amongst what have been established as unequal values. In other words, you've already established in our mind there are differences between the samples and we now focus on the differences and not on the music. We are literally testing ourself as to whether we can determine which sample is which and no longer are we listening as if we were not under such a test. This variation in skills is not how we normally listen to music. What we come to is a test of the test subject rather than a test of the equipment. It is no longer about whether subjective differences exist and it becomes very much about whether any one listener has the ability to perceive those differences we have been told exist. This is likely the single most difficult value to overcome in designing a non-biased test and where most test designers fail to think ahead.

Secondly, yes, by its very design, there is significant information lost in any of the numerous MP3 formats. If someone were to come into my shop and want a demonstration of a high quality sound system, they wouldn't be given that demonstration using MP3 material. If they told me they were intending to use MP3's as their primary music source, I would never suggest they invest heavily in quality audio equipment since the subtle improvements found in superior audio equipment will be lost in the source player and format. This doesn't even get into the variables presented by offering such samples on a computer sound card system. Which playback format will be used? Will the samples be audition through headphones or speakers? What is the quality of the system being used? And, will the lower quality of an average computer system played back through less than transparent equipment favor one sample over another?

This is not a discussion of the possible losses found in the transfer from one digital format to another but simply a statement of how MP3 was created to supplant quality with quantity. The MP3 format was created through the use of audio masking technology which throws away information when it would (possibly) be obscured by louder sounds in various frequency bands. Attack and decay are inferior to both WAV format and the real thing since there are less bits carrying the same amount of information from the original music source. In short, MP3 is a compromised format and should not be the choice of decisions based on somewhat critical impressions of quality.

Please don't take this as a criticism of your quiz but rather just another view of the testing procedure you have chosen. It is extremely difficult to design a test of any one's subjective perceptions and not run afoul of the many pitfalls which are inherent in such tests. In the end, while some results might be interesting to discuss, if your objective is truly impartial and meant only for your own enjoyment, then that is where your test must end. No real world conclusions can be made from such tests.

I still own my laminated Yamaha classical from the late 1960's. I also own what I would consider several rather nice solid wood instruments. When playing the Yamaha I am still taken by its overall tone which is warm and mostly unobjectionable. Its sins of omission are largely a matter f it does not possess sins of commission. It is, though, a very difficult guitar to play when it comes to projection, attack, sustain and (not so) subtle variation in emotion. It is a far more difficult guitar to play effectively than are my solid wood instruments. If a test subject's priorities are geared mainly towards tone and tone alone or the player is not quite as skilled at bringing forth those varied expressions of musical performance and the listener is not aware of their existence or the music example selected does not lend itself to those elements of performance, there could easily be a conclusion reached which would not favor the higher quality instrument. I would liken this to those clients who desire a high end audio system yet are completely unfamiliar with the sounds and perceptions involved in a live music experience. If your reference is your car stereo, what you will find in a high quality home system will not be very satisfying. This is even more the point with a hand built instrument which can involve virtually every variable imaginable from the quality of materials and construction to the humidity on the day of the recording.

Posts: 338
Joined: Mon Dec 26, 2011 10:12 pm
Location: Norway, Hole

Re: D02 - A Quiz Game

Post by Håvard.Bergene » Wed Mar 26, 2014 8:59 pm

My guess in the first sample set (not looking ahead) is: 2,3,1 (full, 1st, 2nd)
And in the second set: 1,2,3 (My guesses are based on which sample I think sounds more "sloppy". 1=thightest, 3=sloppiest)

Fun game. Sorry I didn't find time to to the quiz earlier.
Alhambra 11P

User avatar
Haris Karachristianidis
Posts: 459
Joined: Sat Dec 14, 2013 8:50 am
Location: Kavala, Greece

Re: D02 - A Quiz Game

Post by Haris Karachristianidis » Thu Apr 10, 2014 4:13 pm

First sample set:
650mm scale: sample #1
613.5mm scale: sample #2
579.1mm scale: sample #3

User avatar
Duang Turongratanachai
Posts: 938
Joined: Thu Jul 12, 2007 11:56 pm

Re: D02 - A Quiz Game

Post by Duang Turongratanachai » Sun Mar 08, 2015 4:10 am

It's really hard to hear the difference. I think I will go for...
The first set...
650mm # 3
613.5mm # 1
579.1mm # 2

Return to “Musical Games”