Choosing a new guitar can be a daunting task even with a clear idea of ones cost limitations. While this essay was originally written in reference to choosing a nylon-string guitar, most of what follows would also apply to choosing a steel-string instrument. The overriding guideline should be to find an instrument that meets your needs and expectations. This can only be done by comparative shopping at several sources, trying many guitars - preferably in side by side comparisons. Of course, in a non-specialty shop, in rural areas, or via the Internet, this is not always possible. That’s when things get interesting.

First off- it's helpful to put a cap on what you're willing to spend. You won't do this more than a few times in your life, so don't shortchange yourself. With heavy playing and regular fretboard maintenance, a guitar may last in perfectly playable condition for 15 years or more. Unlike violins and other orchestral string instruments, guitars eventually self-destruct due to string tension that warps the neck and caves the top. I look at it like this; let's say I have $1,000 to spend: $1,000 divided by 15 (years) divided by 12 (months) equals $5.55/month - not a bad rent-to-own fee.

At the same time, one must be realistic and fair to oneself about the price tag. I believe that I should own the guitar my technique and ears deserve. If your guitar doesn’t play in tune due to improper bridge placement, or if your guitar is irreparably damaged to the point that the listener (you or the audience) suffers - then you need a new guitar.

When trying-out guitars I first look the instrument over for the general quality of craftsmanship and materials. I like straight, close, even-grained spruce or cedar for the top and rosewood sides and back although many other woods serve equally well for the back and sides. In Barcelona, I once played the famous Torres guitar built with stiff cardboard back and sides. Sometimes described as “papier-mâché” this material is actually closer to the dry mounting board used by photographers. It is, however, obviously paper. The neck and top of this instrument were of excellent wood, and all parts were normally braced and assembled. Torres built this test model to prove that it was the top or soundboard that matters in tone production. After almost 100 years, the guitar was still playable and sounded remarkably good.

I also played an all-wood Torres guitar that had belonged to Francisco Tárrega. To my ear, it sounded superior to the guitar with cardboard back and sides. So, while the top may be the key element in tone production, the material used in the back and sides also matters. All of my classic guitars but one happened to have had either Brazilian or Indian rosewood back and sides, but I am not opposed to trying guitars made of other tone woods. I just like the typically “dark” sound of rosewood.

It is becoming increasingly difficult to find Brazilian rosewood, but Indian rosewood serves just as well. A local acoustic builder here in Pacific, MO uses Osage orange for the back and sides and the wood is gorgeous. My last three classic guitars have had cedar tops and I have grown to like that sound too. I like to see a slight arch in the top both lengthwise and side to side as structurally, the arch will resist caving between the bridge and sound hole. Aesthetically, all of the wood in the body of the guitar should be book-matched. If you are going to spend a healthy chunk of your life with an instrument, it should be beautiful.

Flamenco guitars are usually made with different back and side material - traditionally cypress. The really big cypress trees are often found in Spanish cemeteries, so grave robbing has taken on a new meaning where Spanish luthiers are concerned. Smaller trees give rise to three and four-piece backs. The flamenco guitar is usually much lighter in weight and characterized as more “percussive” with a tone that is less "dark" or profound then the classic guitar. Tap guards are absolutely necessary to protect the top of a flamenco guitar but I don't think they alter the sound significantly.

Before evaluating the acoustic properties of a guitar, be sure it is accurately tuned. I usually bring along my own tuning fork for this purpose. Mike and Corey Christiansen, father and son Mel Bay authors, alerted me to the concept of singing into the soundhole of an acoustic guitar to determine the built-in pitch of the top and sound box. This is accomplished by singing a slowly swooping or glissando tone in the F to B range directly into the hole. The guitar should respond with a sympathetic resonance - G for flattops, A for classics. You can identify the resonant note by matching it on the tuned guitar. I’m not sure of how critical it is for an instrument to pass this test as my own classic guitar responds with a G and has good sustain throughout. While I have chosen guitars for more than 25 years without including the sympathetic resonance factor in the mix, I will try it in the future.

Again, being sure that the guitar is accurately tuned, I then play a chromatic scale on each string, playing a firm rest stroke on a classical guitar or fairly hard with a pick on an acoustic to disclose any fret buzzing. Any guitar can be made to buzz so don't overdrive the guitar, but do insist on reasonable volume. I listen for evenness of tone and volume in the low, mid, and high registers as well as for precise intonation. Compare the pitch of the note played at the twelfth fret on the E and B strings with the open string and the harmonic that sounds at the twelfth fret. The nylon third string (G) on a classic guitar often has problems passing this test, so I usually discount it. The harmonics and stopped notes should match exactly on the first and second strings.

If the harmonic and stopped string at the 12th fret don't match, it may indicate that the bridge is not placed correctly in relation to the frets and therefore the frets are disproportionately placed along the string length. Untrue or aged strings can also influence the results of this test. If all other aspects of the guitar look promising, change the offensive string(s) and test again. If it flunks again, go on to another guitar.

When you find a promising candidate, try playing a tune from your repertoire that makes use of the entire fretboard. The guitar should play easily throughout the fretboard. Even if the sound is good, you don't want to have to be at odds with your instrument in terms of ease of play. I once owned a 1972 Ramirez 1A- a great sounding instrument but the string length and consequent fret distances seemed inordinately long even for my large hands. Unless I was in absolute top condition in terms of left-hand legato and extension ability, it was a very hard guitar to play. I understand that Ramirez has since offered a 1A concert model with a shorter string length and a radiused fingerboard for greater ease of play.

The last time I helped choose a guitar for a student, we went to a known supplier of quality instruments. After trying at least a dozen guitars by Ramirez and Kohno, we subjected a select few to a "blindfold test." In the test, my student and I taped an A, B, and C on three prime examples from each maker and then played them for each other with the listener’s back to the player, rating them on a scale of 1-5 points each. My student chose the guitars that scored the highest between us; he then bargained with the manager and got a terrific deal on the best Ramirez and the best Kohno. It was guitar-shopping heaven and, fortunately, my student was able to take advantage of the situation.

As an alternative to in-store shopping, many on-line guitar merchants will send out instruments "on spec" for your examination. You would only have to pay return freight on the instruments you reject. The problem is, you can't really compare instruments when they are being sent to you one at a time, and the shipping and insurance fees can mount up quickly. The only recourse here is to take detailed notes on each instrument you try with regard to specific evaluation criteria (tone, sustain, intonation, volume, playability, soundhole singing test, craftsmanship and materials, etc.). If you have the facilities, you might also record the same piece on each guitar and judge them retrospectively.

Overall, it’s a good idea to audition guitars under optimal circumstances arriving at a store early on a Saturday for example, or finding a quiet spot at home after the kids have gone to bed. Secondly, try as many guitars as possible to help establish your own listening criteria. What is it that you especially appreciate about an instrument, even if it’s out of your price range? Providing yourself with such guidelines and parameters will put you one step closer to the guitar that is right for you.

For Mel Bay Publications,
Stephen Rekas
Guitar Sessions Editor