Audio Sampling the sampled instruments will become more detailed. Samplers will, however, stay fundamentally the same. 9.1 Modeling and convolution meet sampling Modeling is the new wave of the future. Modeling is not a form of sampling but accomplishes similar end results. When creating a model of an instrument, many measurements are taken. These measurements are used to create a software instrument that simulates a sound source using mathematical modeling. Modeled instruments can reproduce these real instruments very accurately. In the future, it will be possible to create indistinguishable replicas. One difference between sampling and modeling is that, with modeling, the end user's creativity is limited by the sounds that are included with the modeled instrument. There are parameters that can be tweaked by the end user, but the fundamental sounds and programming cannot be altered. The amount of software development and programming that goes into a modeled instrument's creation is enormous. This precludes most musicians from creating their own modeled instrument from scratch. A part of modeling is the convolution process. Convolution, in relation to sampling, is the ability to transfer the charac- teristics of one sonic item to another. It is possible to capture the essence of a concert hall's reverb and use it in a recording studio on a mix (this process is explained in more detail in Section 5.4). Convolution techniques are a part of the future of sampling. The best thing about convolution is its accessi- bility: almost anyone can create convolution samples. Not only is it relatively simple to sample the reverb of a given space, but it is not much more complex to capture the char- acteristics of other pieces of equipment. An example of this is sending a sweeping sine wave through the filters of an analog synthesizer such as the classic Mini Moog synthesizer (Figure 9.4). Once the resulting wave is 202