CONVERSATION WITH DAVID TUDOR interview by Gustavo Matamoros and Russell Frehling October, 1995 DETAILS AUDIO FILE NAME: FTNK_DT_with_GM_and_RF-excerpt-NR.mp3 DURATION: 14m41s DATE: OCT-1995 LOCATION: Tompkins Cove, NY NOTES: This is a highly edited, noise-reduced version of the (roughly) 90-plus-minute original conversation with David Tudor. This 14m41s version was produced by Gustavo Matamoros to be presented 2016.03.25 during the David Tudor Symposium at Wesleyan University. The original conversation was recorded by Gustavo Matamoros as raw material for SFCA’s Fishtanks Radio Journal. Both are part of SFCA [isaw+subtropics] iSAW Media Archive SUBJECT: Gustavo Matamoros and Russell Frehling converse with DAVID TUDOR about NEURAL NETWORK PLUS, which he performed in a solo concert on 1993.04.01 at Wolfson Auditorium (MDC-Wolfson Campus, Miami) during Subtropics 5 Festival (subtropics.org). Other subjects were discussed CREDITS: © 1995-2016, SFCA [isaw+subtropics] This recording and production are property of SFCA [isaw+subtropics] and contained in its iSAW Media Archive. Publication of this material is permitted for research and educational purposes when acknowledgement is included and notification of its use are provided to us via email to . Thanks! NOTE: SFCA [isaw+subtropics] has plans to compile in book form this and other conversations with artists such as Cage, Earle Brown, Alvin Lucier, Alison Knowles, George E. Lewis, Trimpin, Steve Peters, Kenneth Goldsmith, John Luther Adams, David Dunn, David Behrman, and many more. If you are related to an university press that would like to consider publishing such book, please let us know via email . Thanks! **** TRANSCRIPT: DT: I have a friend who wrote a memoir about my activity. His name is John Holdzapfel. I guess eventually people will be taking the opportunity to take a look at my compositional career. And that’s part of it. Because it took me a long, long time to actually call myself a composer cause I was surrounded by other composers and I did’t want to call attention to myself. But eventually, it catches up with you. EXCERPT OF “NEURAL NETWORK PLUS” - MIAMI, APRIL 1, 1993 WOLFSON AUDITORIUM, MDCC GM: How much of the piece comes from a process of improvisation as you are performing? Or is there some other idea that is more appropriate to describe the process of performing it? DT: Strictly speaking it’s improvisational. What usually happens is that you think in terms of how long a performance that you wish to give, and just insure that there is enough sound source material to give… like sufficient choice so that you are never at a loss to introduce a new element in the composition. And that pleases me because I like it when the composition takes… more exactly… the course of the composition… when it takes an unexpected turn. And that happens easily with the neural network. RF: David, if I may. Can you talk about your concept of these source tapes, and what goes into the preparation of them? DT: It’s all aimed towards exciting the Neural Network synthesizer to perform… [laughs]. I wish I were able to remember the synthesizer more exactly. It’s been a couple of years since I had it under me hands. When I have it in front of me, I know where all the controls are… inexpensive talking clock: It’s 2 o’clock pm… DT: … Thank you, Emma [?] … RF: It seems that your work is so intimately connected with the instrument at hand whether it is the Neural Net at this point, or some of the other systems you developed before that, that the difference between these categories of composer, or performer, or musician, they become irrelevant distinctions. They are one and the same. It seems that those instruments become the structures of the compositions… a physical thing. It’s no longer sort of an abstraction on paper. DT: The setup becomes an instrument itself. I lay the tables out to have access to all the possibilities. When I have a good night, it all gets displayed. But sometimes that doesn’t happen. It doesn’t disturb me at all because if the sound source—the initial sound source—is really coherent, you don’t have to worry about displaying all of its possibilities. RF: Can you describe the sound materials that you found to be effective with the Neural Net? Those kinds of sound materials that excite the system? DT: Well, right now my practice is to limit myself to three source tapes—but actually, that’s just a convenience, cause right now, if you record your initial sound source on disc, it’s much more convenient than having like six machines… [laughs] …which is what I’ve been accustomed to doing for many years. In other words, using everything that is available. But I’ve learned not to do that. And it’s not the idea that, “Less is more”, but, if you set out with plenty, you can’t miss. And the Neural Net circuitry is very responsive to variations like feeding it only low frequencies or only high frequencies. And it just takes five or six different attempts, you get quite familiar with the possibilities. And then you just have to have a compositional plan. GM: Do you device a compositional plan for each time you do it? DT: Yes. I do. GM: If you were to want to perform the piece soon, what would be an interesting plan to follow? DT: First of all I would start out by thinking about how long I would like the piece to go on for. And then you begin to think about what source materials that you do have available. For instance, if you just have three tapes that you are using, you try to equate it with the length of the concert that you wish to produce. Sometimes I think about featuring certain components trying to design a composition which will display what the components can do. You know, how much variety that you can produce with… And there are all kinds of ways to alter the sound of the components, and over the years I’ve become good at it. More and more my tables tend to get smaller rather than larger. Thank god. GM: Still it takes a long time to set up. Is a lot of very intricate connections, it seems. DT: Well… I don’t think of it as a long time. [laughs]. It takes, you know… three hours, just to test it and set it up, and, once it’s done, it can last for quite some time. Occasionally I’ve made installations which go on and on for weeks. In which case you just have to make sure that there is enough variation to satisfy the length that you need to produce. GM: It’s a question of scale, there… DT: Yes! always, and then you also have to think in term of the power that you need. GM: By that do you mean amplification? DT: Yeah. Ahá. GM: That’s also scale. DT: Yeah. That is a related problem, cause some components need a different level at their input to activate the component without overdriving it to distortion. RF: How important does the performance space become to the variations of a given performance? What’s possible? Is the room a big player in the performance? DT: Oh, yes! it’s for me. When I come into a space I always try to sense it, how big or how small, and how much power that I’ll need to drive it with. GM: So what do you look for when you are trying to size a room? DT: I look for anything different. I mean, what makes a space different? Is the space round? Or is it square? Or is it multiple? If I have six rooms to perform in it’s a very different situation. But I enjoy all of the situations and all the differences. That’s what I look for, what makes a space different when it has sound in it… when you put sound into it. RF: In the Neural Net is there an acoustical link in the feedback system where a sound gets put out into the space and brought back in? or does it take place primarily in the electronic realm and then, at the end it gets broadcast into the space? DT: I set up the table so that each component that I use, I can introduce feedback to it. And then, if I think the sound of it is possible when it’s overdriven, then I use it. But feedback is quite important to my process. And recently I had the opportunity to look over some papers from the designer of several of my components that I use and depend upon. What stroke me, what the fact that, because the people that I was working with were able to anticipate how I’d set about using their designs, they dealt with parameters, for instance like heat, knowing that the components would be overdriven, and making sure that the components—the way the devices were made—would be able to withstand that. And I’ve been quite lucky in that, because otherwise it would be, you know, many components that I could use, but I would not call into action because there are other things that gave more possibilities.