An interesting point is in the recording we hear and see of Diego, he is using different guitars usually ones of students or foreigners that were in Moron at that time. All were most likely cypress B & S and spruce top.

The similarity between Arabic oud and old school flamenco guitar is interesting and also some traditional flamenco guitar melodies resemble some old Greek melodies I have heard played on the 4 string bouzouki.

I had a 6 double string fretted oud back in the early 70's and tuned it like a guitar instead of the traditional oud tunings. Interesting instrument.

As I mentioned before maple was used in flamencos up until about 1930 when cypress took over because of economics. Maple produces sweeter trebles and crisper basses without have the mellowness of rosewood. In the early 70's the popularity of the negra started and flamenco started leaning toward more sanitary playing to appease the Segovia types that thought the gypsy juerga style flamenco vulgar. The advent of the idea that the ultimate in flamenco guitar was to be a virtuso solo concert stage player helped the more mellow tone negra to grow in popularity. Some started making a bit larger body and longer scales became popular to get more volume. In many cases the only difference between a classical and a flamenco negra is the thinner B & S and top on the flamenco and a lower action.

I think the modern flamenco guitar-player obsession with extremely fast attack times combined with extraordinarily long sustain times is misguided. The easy way to accomplish that is to use steel strings on a heavy guitar. Steel strings provide extremely fast attack times, because steel strings stretch very little when they are picked, causing a guitar top to move almost instantly in sympathy with string vibrations. Heavy guitars have long sustain-times, because vibrational energy is lost more slowly in heavy, less-flexible wood.

Listen to this short clip of Paco Peña playing his Cancion (peteneras) on what sounds to me like a nylon-string flamenco negra guitar with rather slow attack times:
Paco-Peña-Cancion.mp3


Compare that to approximately the same thing played on a steel-string acoustic guitar:
Steel-String-Accoustic-Cancion.mp3


The attack and sustain characteristics of modern flamenco guitars are not that extreme, but some have metallic-sounding characteristics that approach steel-string-guitar tonal qualities.

Now, compare the guitar sound qualities exhibited in both those recordings to the much more beautiful (in my opinion) sound qualities of the 1903 Ramírez guitar that Sam referenced in a post further above: 1903 Ramírez

Which of those three guitars would you prefer to listen to for a long period of time? Which would you least prefer?

-Bob

Bob,

It sound to my ears that the Peña track has half-a-ton of reverb added, either that or he's playing in a very resonant room. I was hard-pressed to make it through all 19 seconds of the steel-string track. The Ramirez is the best of the bunch (no surprise there), but I'd really need to hear all three played by the same person in the same studio to make a fair comparison.

davinort wrote:Bob,
It sound to my ears that the Peña track has half-a-ton of reverb added, either that or he's playing in a very resonant room.

Electronic reverb may have been used, although years ago it was common to record in large halls or recording studios with special wall treatments, place microphones further away from musical instruments, and rely on natural room acoustics to achieve desired reverberation effects. Now that an enormous variety of computer-generated reverb effects can be applied and adjusted easily after recordings have been made, recordings generally are made with close microphone placements to minimize room acoustical effects and reverb is applied later. Regardless of how the Peña recording was made, I agree that it contains reverb.

davinort wrote:I was hard-pressed to make it through all 19 seconds of the steel-string track.

It is very difficult to listen to. I posted it only to demonstrate and emphasize what extremely fast-attack and long-sustain sound like.

davinort wrote:The Ramirez is the best of the bunch (no surprise there), but I'd really need to hear all three played by the same person in the same studio to make a fair comparison.

It certainly isn't fair to compare guitars with different guitarists playing different music recorded in different places with different recording equipment. However, in my opinion it would take quite a guitarist to make the extremely fast-attack and long-sustain steel-string guitar sound better than either of the other guitars playing any type of flamenco.

-Bob

The steel string reminds me of the aluminum wind chimes on the back porch. Gusty winds here today producing a peteneras. Perhaps if Peter had made the steel string http://www.youtube.com/watch?v=KF5ERTQNlJQ

Hey Bob,
Were you trying to torture us with that steel-string recording?
I don't think you can compare the sound of any instruments without eliminating as many external variable as possible, such as the technique of the player, the recording, the room, etc, etc.

I liked the recordings you posted of the oud and flamenco player on page 1.
I've built enough guitars to know that making a guitar lighter in construction does'nt do much to reduce the sustain of the guitar, although you're right in that it does generally reduce the sustain, especially in the low end register.
But the player's technique, both right and left hand, are way more important that a slight reduction in weight or thickness of the guitars' woods.
Action is also a bigger factor in reducing sustain. Take a classical guitar with 4mm action bass side, and 3mm action treble side and play it, and try to get the notes to ring. You'll get good sustain on almost any instrument, unless it's a total piece of crap.
Take the same guitar and drop the action to 3mm bass side, and 2.5 mm treble side, and play it appoyando with the thumb and do picado rest strokes flamenco style and you'll get a totally different sound.
In a recording you could even tell someone that the first one is a "classical guitar" and the second one is a "flamenco guitar." Most people would not argue, but what they're really hearing is different playing technique and a different set-up.

You get a lot of luthiers like Richard Brune and Eugene Clark, for example, who really don't approach the two guitars differently. Eugene was even quoted once as saying that his classical guitars are the same as his flamencos. He just charges "$2000 more to not put on a tap plate!"

I do approach the two types of guitars differently, but I think that way to many luthiers overstate the differences between classical and flamenco guitars for marketing reasons, or maybe because a lot of them don't play flamenco so they think the sound is all wrapped up in the guitar itself.

-Marcus

Marcus Dominelli wrote:Hey Bob,
Were you trying to torture us with that steel-string recording?
I don't think you can compare the sound of any instruments without eliminating as many external variable as possible, such as the technique of the player, the recording, the room, etc, etc.

I agree. That is true. The point I was trying to make with the steel-string guitar recording was simply that steel-string guitars have extremely short attack-times combined with extremely long ring-times and that combination is not suitable for traditional flamenco. No doubt a steel-string guitar could be played and recorded in ways that would make it sound better than in that example. I didn't intended to argue otherwise, but I think all flamenco guitar players know that played and recorded in the best possible ways a steel-string guitar still wouldn't be very suitable for traditional flamenco. If that is true, then it follows that traditional flamenco guitarists shouldn't want guitars with those characteristics.

Furthermore, I wasn't intended to argue that steel-string guitars are not suitable for the types of music they customarily are used for or that they are intrinsically bad. They may be ideal for playing music that is customarily played on them. My point again was simply that they obviously are not suitable for playing traditional flamenco because of their extremely short attack and long ring times and that, consequently, extremely short attack and long sustain times are not guitar characteristics that traditional flamenco guitarists should seek.

Marcus Dominelli wrote:I've built enough guitars to know that making a guitar lighter in construction does'nt do much to reduce the sustain of the guitar, although you're right in that it does generally reduce the sustain, especially in the low end register.

I haven't made a single guitar, let alone the large number you have made, so I am not qualified to comment from the perspective of a maker. However, from the perspective of a player who has played a lot of guitars since first learning to play 65-years ago, it has seemed to me that light-weight guitars typically have less sustain.

Marcus Dominelli wrote:You get a lot of luthiers like Richard Brune and Eugene Clark, for example, who really don't approach the two guitars differently. Eugene was even quoted once as saying that his classical guitars are the same as his flamencos. He just charges "$2000 more to not put on a tap plate!"

I can believe that many, possibly even most, guitars sold as flamenco guitars nowadays are essentially classical guitars with tap-plates, because that is what most of them sound like.

Out of curiosity, I decided to make some comparisons today between my Ramírez classical and Ramírez flamenco guitars. They came from the same shop and they both have Amercian western red cedar tops. Even so, they sound dramatically different. The biggest construction difference is their weight, although the classical also has a thicker sound-box. The classical has heavy Brazilian rosewood back and sides. The flamenco is an unusually light-weight guitar with cypress back and sides.

Of course, you wrote above:

Marcus Dominelli wrote:Take the same guitar and drop the action to 3mm bass side, and 2.5 mm treble side, and play it appoyando with the thumb and do picado rest strokes flamenco style and you'll get a totally different sound.

So, based on your experience, it could be argued that the difference in sound between my Ramírez classical and flamenco guitars is primarily due to the difference in their action. That may be true. However, I also have a heavy Contreras flamenco negra that has heavy Brazilian rosewood back and sides, and an Amercian western red cedar top. As I am sure you know, Contreras worked for Ramírez before starting his own shop and my Contreras flamenco has strong physical construction similarities to my Ramírez classical, except that it has very low action and a tap-plate. Despite having low action, it sounds much more like my Ramírez classical than like my Ramírez flamenco. Even beyond that, its sound qualities are not in between my Ramírez classical and Ramírez flamenco guitars. Its sound is "on the other side," so to speak, in that it sounds even more "classical" and less "flamenco" than my Ramírez classical.

I decided to record and post the same music played on these different guitars to demonstrate these characteristics. However, I have mostly played my new Tsiorba blanca the past several weeks and didn't realize how bad the strings on my other guitars are. I am going to replace those strings today and will post some comparison recordings in a day or two after the new strings have stretched enough to stay in tune.

Despite the worn strings, I made some other interesting comparisons this morning that I thought you might be interested in. I have laboratory test equipment capable of precisely measuring sustain times and other sound characteristics. It wasn't designed for guitar testing. It is just general-purpose electronic laboratory test equipment, but it is well-suited for the purpose. I also have software that can make a subset of the same measurements using a computer. I used the software this morning, so I could capture screen images to post in the forum and not have to take test equipment photographs, but the measured results would have been the same either way.

Guitar sustain times, temporal and frequency domain energy levels, and other sound characteristics are different playing each note up the chromatic music scale on the same guitar due to sound-box resonances, the "Q" of each resonance ("Q" depends on the percentage of remaining energy lost to heat during each vibrational cycle), and various other factors. I compared those characteristics for a large number of different notes played on my two Ramírez guitars and then later played on my Tsiorba. After looking at the resulting data I selected the results from playing "F" on the fourth-string at the third-fret as being fairly representative to post here.

Below I will post three things related to playing that "F" note on each of those three guitars (I should have included the Contreras, but didn't take time to do that today).

• First, will be a seven-second audio recording of the note being played.
  
  
• That will be followed by a plot of air-pressure fluctuations twelve-inches in front of the guitar face vs. time. Up on the plot is increasing air pressure. Down is decreasing air pressure. The air-pressure and time scales are the same for all three guitars, so they can be compared directly. The air pressure changes were sensed by a laboratory-quality electrostatic transducer designed for these types of measurements.
  
  
• The third image is a plot of spectral vs. temporal domain energy in the radiating sound waves. The vertical axis is the spectral domain (the domain containing the fundamental sound frequency and its harmonics). The bottom is the fundamental frequency with harmonics of the fundamental extending upward. The horizontal axis is the temporal domain (seven-seconds in the time domain). Third-dimension sound energy levels are depicted by different color hues, with violet being the lowest energy level (the background noise level in the room), red being an intermediate level, and yellow being the highest (loudest) sound level.

Results for the Heavy Ramírez Classical Guitar
Ramírez Classical.mp3 Sound Recording

Sound Amplitude vs. Time after the Note was Picked

Sound Spectral Energy vs Time

Results for the Light Ramírez Flamenco Guitar
Ramírez Flamenco.mp3 Sound Recording

Sound Amplitude vs. Time after the Note was Picked

Sound Spectral Energy vs Time

Results for the Light Tsiorba Flamenco Guitar
Tsiorba Flamenco.mp3 Sound Recording

Sound Amplitude vs. Time after the Note was Picked

Sound Spectral Energy vs Time

-Bob

Hey Bob,

Your empirical findings are to be expected.

Up until about 1950 or 1960, or maybe later, most musicians and luthiers did not even make a distinction between "classical" versus "flamenco" guitars. This is pretty well documented. The big difference between the two types of guitars was in the string set up, and the technique of the players. IMO, even to this day, these are more important to getting the sound (fom either style) than is the factor of guitar design.

While it's true that guitar design has an effect on sound (as your recordings suggest) The thing we hear in a musical performance is mostly the technique of the player. Listen to the difference between a free stroke and a rest stroke.
For example, I heard Pepe Romero in concert a couple of years ago. He played his usual repertoir of both classical and flamenco dance pieces. He could get whatever sound he wanted out of one guitar. Was it a classical guitar or a flamenco guitar?
If someone from the audience asked him "Pepe, is that a classical or flamenco guitar" he probably would have chuckled to himself.
Before about 1950, I would say that the only difference between the flamenco and classical guitar was a tap plate, a set up, and playing style. Many people like Brune and Eugene Clark still believe this to be true.

But maybe they're just Old Timers......

Bob, Very interesting the rate of decay on the 3 guitars. Are the amplitude spikes we see after the original decay of fundamental note, responding harmonics? I noticed the rate of decay of the Ramirez flamenco and the Tsiorba to be similar, but the following spikes are more pronounced in the Tsiorba indicating to me the responding harmonics would be more noticeable in Peters guitar.

I think there are four important factors in getting the old sound as the original poster is seeking in his guitars.

1. Wood choice for back and sides.
2. Thickness or lack thereof of the top and B & S.
3. Set up ... mainly distance of strings from top.
4. Body size, mainly depth.

How much effect the golpeador has on the decay would be an interesting study. I am guessing it would be very little. When I converted my rosewood and redwood to flamenco, the plate made no noticeable difference. Lowering the strings made a noticeable difference.

Marcus makes a good point about the playing style and resulting sound, however one can play a classical piece on a cypress flamenco, then on a rosewood flamenco of the same model and make and a difference can be heard. Some times recordings cannot capture the difference and one has to be there to hear it.

I think the rate of decay is most affected by the thickness of the woods.

Marcus Dominelli wrote:Up until about 1950 or 1960, or maybe later, most musicians and luthiers did not even make a distinction between "classical" versus "flamenco" guitars. This is pretty well documented. The big difference between the two types of guitars was in the string set up, and the technique of the players. IMO, even to this day, these are more important to getting the sound (fom either style) than is the factor of guitar design.

It may be true that there wasn't the distinction between classical and flamenco guitars in the distant past that there is today. However, some 19th century makers labeled certain models of their guitars sevillanas or fandango guitars and others "guitarras tribunal música" (court music guitars), so a distinction existed long before 1950 or 1960.

Rather than flamenco guitars having been derived from classical, it seems more likely that the reverse is true. The following is quoted from http://www.fernandezmusic.com/Flamencovsclassical.html

In his article, "Cultural Origins of the Modern Guitar," (Soundboard, Fall 1997) Richard Bruné has argued that the modern flamenco guitar is closer to the 19th century guitar developed by Torres than modern 20th century classical guitars. In his very interesting and persuasive article he turns conventional thinking on its head and suggests that the flamenco guitar is not a specialized version of the classical guitar but vice versa, that the modern classical guitar is derived from an earlier flamenco-type instrument. Read this article it is very good.

-Bob