Announcing (fanfare): I’ve now started the process of creating a set of sample libraries featuring instruments from the Hopkin Instrumentarium (which is to say, featuring instruments that I’ve made). The collection is to be produced and released by Soundiron, one of the leading companies distributing such libraries. The audio engineers at Soundiron, led by Mike Peaslee and Gregg Stephens, recently brought their considerable experience and expertise to my studio for the first of what looks to be long series of recording dates. The plan is for a set of libraries to be released over the coming months and years, each featuring a different instrument or category of instruments. The first ones will be lamellaphones (instruments in which the initial vibrating body is a prong or tine).
*(If you’re not familiar with what samples libraries are about, see the footnote at the end of this article.)
While I’ve been building instruments these many decades, I’ve not previously gotten into sampling or sample libraries myself, either as producer or as end-user. So it’s been quite interesting working with Mike and Gregg to learn something about how things are done in this world, and what’s expected of a good library by people who use these libraries every day in their composing and recording work . I’ve learned that it’s a huge amount of work to build a well made library! The number of individual recordings to be made is enormous: individual recordings are required for each of the instrument’s notes, each taken at several different dynamic levels, and often each taken in multiple articulations as well. The total can easily run to hundreds of individual samples per instrument. Following the recording process, the samples must each be processed and edited as needed. Furthermore, for each library a dedicated user interface has to be built reflecting the controls and manipulations most suitable for the instrument type represented. The quantity of recordings is further increased by a prevailing philosophy which says that it’s best to give the end-user more options rather than less. Whenever the question arises, “should we include this sound in the collection?” the answer always seems to be yes. I’m still a bit uncomfortable with this: wouldn’t it be better to let people see only the best we have to offer rather than throwing a whole lot of stuff at them? But the prevailing more-is-better approach reflects the idea that you never know what the user may want. A sound that I wasn’t thrilled with may happen to be just the sound that someone needs in some context I could not have foreseen. And apparently the interface is designed in ways that make it very easy for the user to sift through available sounds to find just the right one. So the audio engineers and I have frequently had conversations like this: Me: “We probably don’t need to include this instrument in the collection; it’s quite similar to that other one did but not quite as nice.” Mike and Gregg: “On the contrary, we need to include both because the sounds this one offers offers are slightly different and someone may find them useful.”
With regard to the whole idea of sampling musical instruments, this question arises: should it bother me that my instrument sounds will be made available for decontextualized use; that a set of digital files will seemingly claim to replace the real physicality of the instrument; that in these decontextualized digitizations the unique gestural qualities and playing-feel of the physical instruments will be discounted? All of these attributes, based in the corporeality of the instrument, mean a lot to me. Over a lifetime of instrument making I’ve come to appreciate the physical and gestural qualities in musical instruments almost as much as the sounds themselves. If the player doesn’t even touch the instrument, if the listener isn’t even in the same room as the instrument, much is lost.
But, to answer the original question: the de-physicalizing inherent in the use of samples does not bother me enough to make me reject the idea of sampling. Of course I’ll always love the physicality of each instrument and value the opportunity to see it realized in human interaction. But even without that I’m very happy to see the sounds extended further out into the world, made available in a most convenient and useful fashion to people who’ll be able to make the most of them. I’m excited by the idea that if more people get their hands on the sounds of my instruments, many of them will likely come up with ways of using them that I never would have thought of.
Here’s another reason I’m happy to be sampling these instruments. Some of them, however interesting their sounds may be, have inherent limitations of various sorts. Most obviously, many have baked-in tunings; too bad for you if you want to produce pitches that aren’t in the available note set. And in some cases, for practical or acoustical reasons, those note sets are fairly limited: many of the instruments are diatonic or even pentatonic, or else are relatively small in compass. With sampling it’s not difficult to fill in missing pitches or (within limits) extend the range.
Also, for reasons having to do with the playing techniques involved, some of my instruments cannot be played with complete fluidity. Others have the practical limitation of being being too big or too fragile to transport easily to wherever they may be wanted for live playing. Once again, sampling gives the user an easy end-run around the limitations.
An example: A while back I made an instrument called Apres Baschet Air. It has a set of glass friction rods coupled to tuned plastic diaphragms covering the ends of tuned air resonator tubes. Its sound is one of my favorites, slightly sad and moany like a sick cow, but very full and rich, with some of the awkward beauty of, say, a bassoon. For practical purposes, the instrument has two serious problems. One is that it’s awkward in form and fairly big, and the glass rods of course are fragile, making it difficult to move to wherever it might be needed. The other is that, although I diligently tuned the diaphragms when I made the instrument, the tuning turns out to be is undependable, and you can never be quite sure that you’ll get the intended note on any given stroke. For these reasons, I’ve never done much with the instrument despite the immense appeal (to my ears) of its sound. But this is a situation custom-made for sampling. I’m much looking forward the day when Mike and Gregg aim their microphones at it, and then to the possibility that someday someone will be able to compose something really lovely for Apres Baschet Air by means of the resulting samples.
*Here’s a brief description of the idea behind sample libraries. Skip this note if you’re already familiar with this information. A sample library is a collection of digital audio recordings of individual notes or phrases played on one or more musical instruments or sound sources. With an electronic keyboard or a computer interface designed for the purpose, a player or composer can play the sampled sounds in an improvisation or composition. Thus, if you’ve written a part for violin but aren’t ready to hire a violinist to play it, you can use a set of violin samples running through the keyboard or computer to play the part. Further, you can create entire orchestral compositions this way. In the early days of sampling, as you’d expect, the results tended to sound stiff and unnatural. Over the years the quality has improved. At the same time, some people have explored the idea that sampling systems can be used in ways that aren’t necessarily intended to replicate natural sound, but instead explore possibilities that might not be otherwise feasible. Many excellent libraries of orchestral and popular instruments are commercially available. In the realm of new and experimental instruments, there’s less on offer so far, and this new collection aims to help fill that space.