We are trying to determine how these four sets of violin strings differ in sound:

For the first trial, we strung the electric violin with four A strings, one of each type. Sam plucked each one, and then bowed each one (keeping the bowing style and pressure as constant as possible).

The listeners seemed to be able to distinguish the one non-D'Addario string from the others, but could not tell the difference between any of the D'Addario strings. When bowed, the sound of the D'Addario strings lasted longer, and the sound swelled within a second or so after the attack, before dying away. Oddly, the reverse was true when the strings were plucked - the D'Addario strings died out more quickly then the other.

We think this difference in sound was due to resonance, because it disappeared when we damped the other three strings.

For the second trial, we used a real violin, strung with all the D strings. We changed the arrangement so that the strings which had been on the outsides of the bridge on the electric violin (the E and G positions) were now in the center (the A and D positions). This was to correct for any effect of the strings' positions on their sound.

As Sam bowed each string in turn, he damped the other three with his fingers. This eliminated the resonance effect and helped us concentrate on the sounds of the strings alone.

The result was that no listener could reliably tell any of the four strings apart.

We're not sure, but these strings probably should not sound completely identical. Because their cores are different, they should act very differently under the same bowing conditions, producing a different sound.

A violin string can be imagined as an ideal string with infinitely small radius and perfectly homogenous along its length. The bow pulls the string at a point and then releases it, resulting in a sawtooth wave. (This is accomplished through the sticky rosin on the bow, and is called the "stick-slip" effect.)

With an ideal string, the wave would be perfectly horizontal. Of course, this is not the case in real life. Real strings have imperfections along their length that can very slightly deflect the wave in places. And strings with multi-stranded cores have an additional twist factor in their waves. As the bow grips the string, the strands start to twist around each other at that point, and when the bow releases, the strands un-twist and re-twist rapidly as the string vibrates.

The following table shows the spectrum of string types, from most to least "ideal". It's important to remember that the physically "ideal" string is actually a very poor violin string. The more perfectly horizontal the string's vibration, the more it sounds like a synthesizer, and not like a real violin.

		String	Core	Characteristics
Most Ideal		Prelude	solid steel	single strand, acts the most like an "ideal" string
		Helicore	multi-stranded steel	7 strands, acts less ideal than the single strand
		Zyex, nylon strings	multi-stranded polymer	hundreds of strands, more irregular than metal
Least Ideal		gut strings	animal intestine	hundreds (thousands?) of strands, organic, more irregular than synthetics

If the strings really are distinguishable, then why couldn't we hear the difference? We think the following issues may have played a role:

Not enough information. We played the strings in the simplest way possible, which conveys the least possible amount of information about the character of the strings. Differences between the strings are much more apparent during transitions between different notes and different bowings than they are when the string is simply vibrating in a basic wave.

Feedback loop. The violin player naturally corrects for minor variations between strings, so most kinds of strings sound similar when played by a good violinist.

Distance from strings. Most of the audiable differences between two strings are in the high-frequency "noise" (scratching or squeaking sounds). While the violinist himself can hear this noise, it decreases exponentially with distance and is not usually audiable to the audience. (This may not apply here, since in these experiments the audience was sitting next to the player.)

Psychoacoustics. Just as the violin player is trained to produce the same sound with different violins, bows, and strings, the human mind is trained to hear all violin strings as sounding more or less the same. Even when a listener is close enough to hear the "noise" of the strings, they generally fail to notice it unless someone specifically points it out. (Professional violinists are probably more sensitive to the noise, but none of the listeners in this experiment were experienced violinists.)

We have two ideas that should help us find the differences in the strings (assuming that there are any):

Interviewing violinists. Someone more experienced with violin strings may be able to notice differences between the strings that we cannot. As a future experiment, we plan to ask some more experienced violinists to review the strings for us and explain any differences that they notice.

Experimenting with the most wildly different strings. Fan Tao suggested three kinds of strings that should sound very different, even to non-violinists. Gut strings, he says, sound very different from the steel core strings. Strings called "Dominants" have a noticable buzz when they are new, and a brand called "Supersensitive Red Label" has about the most high-frequency noise (squeaking and whistling) of any brand. Also, Fan Tao is sending us two A strings which are basically the same make, but have different levels of damping. He believes we should be able to hear the difference between these strings, and it may help us recognize the differences between the original sets of strings.

Thank you to Fan Tao of D'Addario for donating the strings for this experiment.