D02 - A Quiz Game

Musical games to help develop improvisational skills.
User avatar
CarlWestman
Posts: 574
Joined: Sun Dec 04, 2011 3:36 pm
Location: Chattanooga, TN

D02 - A Quiz Game

Post by CarlWestman » Tue Feb 11, 2014 1:32 am

Inspired by the research of Frederic Brochet into wine tasting, and clever extensions to the work done by Robert Hodgson and Richard Wiseman -

see http://www.theguardian.com/lifeandstyle ... e-analysis

- I want to see if we can, in the words of John Ruskin, "substitute facts for appearances and demonstrations for impressions."

I am curious about a number of things with respect to conventional wisdom we take for granted about classical guitars. For instance, it is well-accepted that solid tops sound better than laminated tops. However, I'm unaware of any double-blinded studies (or even single-blinded) that have specifically tested this. And of course, posing a question in this forum isn't exactly a model of a well-designed protocol. But it might provide a hint of what a really well-designed study might find, something in which we would have greater confidence.

Unfortunately I do not have a solid-top, full-sized classical guitar to compare alongside my Yamaha CG101A. So I would encourage anyone who has one of both types of tops to do such a test. Stick to audio-only, of course, to eliminate the possibility of identifying the model of guitar by sight. And if someone can blindfold you, and you cannot tell by feel which is which, all the better.

So that one will have to wait for another person, or until I get a full-size solid top. However, there's another one I've been thinking about (some would say thinking too much about it, but I enjoy the thought process). If you capo a guitar at the first fret after tuning it down a half step, thus restoring the capo-ed guitar to standard tuning, can you tell a difference in the sound? I did think I could, but I realized I might be fooling myself.

So here are 3 sound samples, presented in an order determined by a random process. The music is from Julio Sagreras' First Lessons (Book 1), lesson 70. It was my final exam when I first took a CG class at a community college a few years ago. I did try to play all equally well (with just enough practice to remember how it went). I'm afraid I'm out of practice, so it's quite imperfect, but I'm not posting this to be judged on my playing of this piece. Instead, I propose a quiz:

Can you identify which one was played on the full 650mm scale, which one was capo-ed at a 613.5mm scale, and which one, with capo at the second fret, had a 579.1mm scale? All should be in standard tuning. I played somewhat softly - perhaps I was subconsciously trying to avoid buzz. In any case, please reply with your guesses, and I will reveal the answers the day before D02 Lesson 8 posts.
You do not have the required permissions to view the files attached to this post.

User avatar
Marko Räsänen
Moderator
Moderator
Posts: 3487
Joined: Thu Feb 16, 2012 11:56 am
Location: Finland

Re: D02 - A Quiz Game

Post by Marko Räsänen » Tue Feb 11, 2014 1:41 pm

This is a wonderful idea Carl!

Referring to the topic of this post, is it your purpose to limit this game for D02 students only? Also, did you intend people posting their guesses into this thread?

I must say that sonically the samples sound surprisingly similar. I would have expected more differences. In fact, if I didn't know that two of the sample were played with reduced scale length (and I am assuming you aren't cheating with the facts you have disclosed here), I would say they are just three different takes of the same piece, played with the same guitar, and by the same player.

But now that I know that I am essentially looking for differences in string tension, I think there are (very) subtle differences between the samples, but it's difficult to be certain, because the samples are quite short. Longer samples would probably better eliminate instances where I think I am hearing evidence of softer string tension, but are in fact anomalies created by your playing. I will refrain from making my guess public at this moment, until you say it's ok for me to do so (as I'm not a D02 student, and am not sure whether the guess should be publicized in the first place).

Regarding your idea of solid top vs. laminated top test, my opinion is that the results wouldn't tell us much. Solid top guitars are generally better built than laminated ones, are louder, and likely to sustain longer. On the other hand, I have two guitars. The first one is a solid spruce top, laminated sides and back, with a price tag of 400-500€. The second one has a solid cedar top, and solid IR sides and back, and it costs 900-1100€. The second one is much better instrument, mostly because its body resonates much stronger, and the notes sustain longer. There is also a considerable weight difference between the two guitars. The second one weighs maybe 1.5 times the weight of the first one.

My point is this: For anyone who has played a decent solid wood guitar, the difference between that and and a laminate guitar can be heard immediately in most of the cases. There are some very good laminate guitars, but they're only good in their price range. There are also some bad solid top / body guitars, no doubt. For your test to work, you would need to compare several guitars, trying to guess which ones have a solid top, but then you would be essentially comparing cheaper guitars against more expensive guitars, which in my opinion isn't a fair comparison, if you're only interested about the effect of the top construction (including bracing).
Alhambra 4P spruce
Almansa 457 cedar
Cordoba C12 spruce

User avatar
CarlWestman
Posts: 574
Joined: Sun Dec 04, 2011 3:36 pm
Location: Chattanooga, TN

Re: D02 - A Quiz Game

Post by CarlWestman » Tue Feb 11, 2014 3:52 pm

Marko,

Feel free to chime in. I titled it D02 just because all the other threads in here seem to be titled that way. I guess maybe it need not be limited in this instance, but I'd like to keep the title anyway, as I would like to disproportionately attract those going through at my level. To me the D02 group is my primary community here at Delcamp.

If you read the wine studies, yes, you could get the idea that I could be pulling a fast one, trying to lure an excessively self-confident player into declaring that his/her ear is so good, she can tell differences that aren't there. I suppose I can leave that open as a possibility, but the intent is to have listeners post guesses of scale length in one-to-one correspondence with the sound samples. There are 6 permutations of order
1, 2, 3;
1, 3, 2;
2, 1, 3;
2, 3, 1;
3, 1, 2;
3, 2, 1

but If someone wants to guess they're all the same, I suppose that's a 7th possible ordering (someone could also say that 2 are the same and 1 is different, etc). I suppose guess however you like. It is the same guitar, same strings, same player, same recorder and same room.

I may post some additional sound samples as well. Perhaps I will play more aggressively, and/or set the recording level higher. I'll try to keep them in the same order as I posted them (that is, I'll use the same scale for example 1 of both postings). In case I don't get to it, please go ahead and guess now, and feel free to revise your guesses based on a future posting of an additional set of sound samples.

I have some other audio tests in mind for future games as well, not just solid-top vs. laminate.

Speaking of that, I encourage you to read the Guardian article linked above fully, if you have not done so already. You mentioned that you felt that there would be a range of good and bad construction or builds within a given price range. I agree with that, but an analogy to the wine study, the real test is, can a blinded listener tell the difference between guitars of different price ranges. Ideally the player will not even unintentionally play one better than the other (and this is why double-blinded studies are the gold standard). But the wine studies have certainly found that the vast majority of experts cannot tell cheap wine from expensive wine. The few who can appear to have done so by chance, because they generally cannot repeat their success at rates better than chance. So if there is a similar "myth of expertise" in hearing guitars, then success rates should be statistically insignificant from chance, and it should cross price categories as well. The analogy may not hold up ... but I am curious to find out.

User avatar
Marko Räsänen
Moderator
Moderator
Posts: 3487
Joined: Thu Feb 16, 2012 11:56 am
Location: Finland

Re: D02 - A Quiz Game

Post by Marko Räsänen » Tue Feb 11, 2014 5:41 pm

I didn't read the wine tasting studies yet, but I intend to (preferably over a glass of some red wine). It sounds very interesting!

Anyhow, my guess is, in descending order of scale length: 2, 1, 3
I'm not too confident about it being the correct permutation, but based on what I heard, I find it to be the most likely one. Again, it is interesting that there is so little tonal variation between the samples. I think I can hear the metallic tone of my capo much clearer than that, although it could be that your capo's construction allows for more neutral sound. That is one of the reasons I suspected that there might be some foul play involved :D

My guess is somewhat based on my playing experience with lute tuning (3rd string pitch is lowered half a step into f#), and drop D tuning (the 6th string is lowered by a whole step), and how that changes the feel and the tone of the strings, and extrapolating from that. I don't think for a second that the actual scale length could be heard, at best the changes in string tension, which makes the actual plucking sound, and the time it takes for the note to settle down to its pitch, slightly different.
Alhambra 4P spruce
Almansa 457 cedar
Cordoba C12 spruce

Beatriz Martin

Re: D02 - A Quiz Game

Post by Beatriz Martin » Tue Feb 11, 2014 11:05 pm

Carl, what a great idea! Thank you for starting this fun game.
I am going to take a guess, since I don't see much difference.
3, 2, 1.

User avatar
CarlWestman
Posts: 574
Joined: Sun Dec 04, 2011 3:36 pm
Location: Chattanooga, TN

Re: D02 - A Quiz Game

Post by CarlWestman » Wed Feb 12, 2014 2:58 pm

OK, here's another set of soundclips. What I did here was to set my recorder on auto-level, which tends to be much higher than my typical manual setting. It picks up more background noise in the silences. That said I don't know how much detail may be lost in the conversion of wav to mp3.

Again, this is a piece (well, the first half of it) I learned years ago - Sagreras Second Lessons no. 10. I play it periodically just because I like it, but I don't try to keep it up at performing/recording level. And that's why I only attempted half of it - not meant to be a performance recording, so I just wanted something quick for this purpose. 2 of these were first take, the other was second take. I play it a bit more boldly than the previous one, so you may notice something with the string tension, sympathetic notes, etc.

Please note, however, that I made one change. I randomized the order again. Thus, there is a 5/6 chance that it's a different order from before ... but a 1/6 chance it is the same. So just listen and see what you can tell. Make your guesses independent of your guesses for the previous set of recordings for S1L70. Again, the scale lengths used are 650, 613.5, and 579.1.

I'm also interested in hearing why you guess what you guess. There may be different clues.
You do not have the required permissions to view the files attached to this post.

User avatar
Marko Räsänen
Moderator
Moderator
Posts: 3487
Joined: Thu Feb 16, 2012 11:56 am
Location: Finland

Re: D02 - A Quiz Game

Post by Marko Räsänen » Thu Feb 13, 2014 7:58 am

For the second set my guess is 2, 3, 1 in descending order of scale length. This set was perhaps more difficult than the previous one, because some of the samples sounded a bit out of tune. The guess was based on bass response, the softness of playing the treble strings, 'jitter' of the pitch and what I thought was some fret rattle. I can only hope that I was able to guess at least one of the scale length extremes right.
Alhambra 4P spruce
Almansa 457 cedar
Cordoba C12 spruce

User avatar
CarlWestman
Posts: 574
Joined: Sun Dec 04, 2011 3:36 pm
Location: Chattanooga, TN

Re: D02 - A Quiz Game

Post by CarlWestman » Thu Feb 13, 2014 1:25 pm

Thanks for submitting your guesses, Marko! Although I checked my tuning carefully with my digital tuner before playing, I wonder if what sounded like out-of-tune playing was a result of the auditory side-effects of using shorter scales, or using a capo to effect that. For instance, you guessed that the third one was the 650mm scale ... did that one sound the least out of tune to you? And the middle one the most? Just curious.

User avatar
Marko Räsänen
Moderator
Moderator
Posts: 3487
Joined: Thu Feb 16, 2012 11:56 am
Location: Finland

Re: D02 - A Quiz Game

Post by Marko Räsänen » Thu Feb 13, 2014 6:11 pm

No problem Carl! It's a fun game as long as I don't take it too seriously :)
CarlWestman wrote:For instance, you guessed that the third one was the 650mm scale ... did that one sound the least out of tune to you? And the middle one the most?


No, I guessed the 2nd one was 650mm scale, and the 1st one 579.1mm, 3rd in the between. If I remember correctly, I chose the shortest scale based on 'out-of-tune playing', because common sense and experimentation with various string tensions says to me that the less there is string tension, the more pitch will change from fretting, accidental string bending, left and right hands handling the string generally. The two longer scale samples were harder to tell apart, because I couldn't hear any definite pitch bending. Another kind of mental imaging that I found useful was to compare the sound to lute and baroque guitar as I remember them, because those instruments use gut strings with much lower tension, which causes a particular tone (that's my theory at least).

I couldn't be exactly sure if the out-of-tune sound was really caused by lower string tension, or just shifted tuning, but that felt like the best strategy to use for identification, so I used it.
Alhambra 4P spruce
Almansa 457 cedar
Cordoba C12 spruce

Beatriz Martin

Re: D02 - A Quiz Game

Post by Beatriz Martin » Fri Feb 14, 2014 2:30 am

ok, just to clarify, 1 means you start playing in fret 1 which is the longest scale, 2 you start playing in fret 2 because there is a capo on 1, 3 you start playing in fret 3. Should we be confused 1, 2, 3, fret 1st, 2nd, 3rd, is this correct?

User avatar
Marko Räsänen
Moderator
Moderator
Posts: 3487
Joined: Thu Feb 16, 2012 11:56 am
Location: Finland

Re: D02 - A Quiz Game

Post by Marko Räsänen » Fri Feb 14, 2014 5:14 am

Hi Bea,

I guess it depends on how you'll want to phrase your answer. In my answers the numbers refer to sound sample numbers, and ordering of numbers to scale length: my first number means the sample number for the longest scale (without capo), and the last number the shortest scale (capo at 2nd fret).

To make my answers so that there can be no confusion about how they should be interpreted:

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

Second sample set:
650mm scale: sample #2
613.5mm scale: sample #3
579.1mm scale: sample #1
Alhambra 4P spruce
Almansa 457 cedar
Cordoba C12 spruce

User avatar
Satyajit Kadle
Student of the online lessons
Posts: 202
Joined: Mon Apr 22, 2013 10:30 am

Re: D02 - A Quiz Game

Post by Satyajit Kadle » Fri Feb 14, 2014 11:59 am

@ Carl, My guess for the first set would be 2,1,3. Assuming 1=650, 2=613.5 and 3=579.1
I could be wrong, but I guess when I heard more of the shifting sound between frets, I assumed a shorter fretboard length :) Good fun!

User avatar
Marko Räsänen
Moderator
Moderator
Posts: 3487
Joined: Thu Feb 16, 2012 11:56 am
Location: Finland

Re: D02 - A Quiz Game

Post by Marko Räsänen » Fri Feb 14, 2014 12:58 pm

I find it interesting that when having to interpret the number sequences, Bea and Satyajit choose to think the numbers as indices to the ordered sequence of scale length, and the position of the index as the number of the sound sample. Whereas I think the number values as sound sample numbers, and the position of the number in the sequence as index to the ordered sequence of scale length. I wonder if that's what Carl's game really is about, and everything else just smoke screen :D

It's also funny that sequence 2,1,3 means the same interpreted either way, as does half of the 6 permutations. With the other half the meaning changes depending on the interpretation of the sequence.
Alhambra 4P spruce
Almansa 457 cedar
Cordoba C12 spruce

User avatar
Satyajit Kadle
Student of the online lessons
Posts: 202
Joined: Mon Apr 22, 2013 10:30 am

Re: D02 - A Quiz Game

Post by Satyajit Kadle » Fri Feb 14, 2014 1:13 pm

:lol: Ok.Super confused now

User avatar
CarlWestman
Posts: 574
Joined: Sun Dec 04, 2011 3:36 pm
Location: Chattanooga, TN

Re: D02 - A Quiz Game

Post by CarlWestman » Fri Feb 14, 2014 7:28 pm

However you choose to respond, just pair up the sound sample with one of those string scale lengths. You can express it the way Marko did, going in descending string scale length and noting the sample that goes with each, or list the sound samples in ascending order (1, 2, 3), and noting the string scale length for each.

I'd recommend against using fret numbers for the purpose of this exercise. It can get confusing enough!

Return to “Musical Games”