6. Key terms

The more internal memory you have, the better your samples will sound later, because there will be fewer limitations in sample design. This allows you to use more sub-samples (multisamples) for one instrument sound, assign more length to it, and get the best possible sampling rate. For many devices the RAM space can be expanded later but as already mentioned this requirement almost completely dissapears with software based samplers/players.

The sampling rate determines the upper limit of the frequency response the sampler can reproduce, so the more the better. At a 44.1kHz sampling rate, the playback is about half the sampling rate, ie 22kHz. if possible several sampling rates should be available to give more character to the signals or save to space. A bass drum can for example sound much more interesting at 32kHz and 12 bit resolution than at 44.1 and 16 bits. Professional samplers should have the standard sampling rate of 44,1 and 48kHz in their repetoire, if not only for compatability with CD (44,1) or DAT (old standard: 48kHz) then to avoid problems with dubbing. Each sampler offers at least two sampling rates but 4 or more is preferred.

The maximum sampling time is a result of the RAM memory and sampling rate. This means, in short, that the total sampling time is distributed among the number of samples found in the memory. The longer the total the length of the sample, the more time that can be split up over each sample. This increases the playback quality (not the frequency response) of the sampled sounds, as less material needs to be cut and looped out. However this plays almost no role when using a software sampler/player, since virtually unlimited memory is available.

Modern software samplers/players are only able to load the ''start phase'' of a sound to the internal memory of the computer and play the rest from the hardrive in real time. Streaming the sounds saves RAM by only loading the sounds needed at that moment to the RAM, this also keeps the total loading time down. Streaming is only possible thanks to the high speeds at which modern hard drives can operate. So if you plan to purchase a sampler with streaming possibilities we recommend that hard disc with at least 7,200 rpm and a comparably fast data bus be purchased.

The data structure produced by a sampler is an important part of the decision making process. If it is too complicated and changes to sounds become torture, then your ability to work creatively decreases exponentially. To avoid this, it is particularly important to look at the user-friendliness and musician friendliness of a sampler/player. There are often serious differences between models and manufacturers. Information on these can be found in internet forums, but online test versions are also available for some software as well as downloadable versions, just check the manufacturer website.

At the lowest hierarchical level of a sampler is the single sample, the digitized recording of a sound, after it has been recorded, trimmed (cropped) and normalized (volume optimization). Normally, the sample is placed in a wide range across a keyboard, so that when it is transposed it is playable. There are situations where this isn't desireable though, such as for drum samples, where the individual sounds should be placed one after the other on the keys. From there the opportunities for sound processing and design begin. These include loops as well as reversing a sample, mixing and blending multiple samples into one another, or changing the sampling rate further down the line.

Once again we can use the drum kit as an example: each keyboard zone equals a percussion instrument. With only four zones, a maximum of four instruments can be distributed across the keyboard e.g. bass drum, snare, hi-hat and tom. With 64 zones, on the other hand, it's possible to play a wide range of diverse percussion sounds on the keyboard. The number of keyboard zones therefore equals the number of sounds, making pianos essential for accurately reproducing the sounds of each sample, in other words, making sure they are not transposed incorrectly.

Whether as soft or hardware: The MIDI-multi-mode is the most used operating mode for sound generators like synthesizers or samplers. In this mode the device can receive multiple polyphonic sounds, each on their own MIDI channel, before returning them. The multi-mode, which when using the sampler as a stage keyboard isn't really important, is indispensable as hardware for multi-track operation in conjunction with a sequencer. Only a sampler with multi-mode can simultaneously play drums on channel 16, a piano on channel 1, and strings on channel 3, in this way a single sampler can do the job of several, running mutliple channels at once.

Polyphony is musical term to define a musical ''texture'' consisting of two or more simultaneous lines of independent melody. In terms of samplers, the requirements for polyphony are necessarily tied to the multi-mode function. Ever digital sound generator, sampler or synthesizer is capable of producing a limited number of polyphonic ''voices'' (sounds). In this sense they can be compared with a guitar, which is able to produce a number of simultaneous sounds (notes) in accordance with the number of strings it has. A sampler assigns one voice/sound to each playable sample key. So if you combine two sounds on one key you get two ''voices'' per note. Often the different sounds are heavily used, especially in multi-mode, as multiple instruments need to play at once. Therefore samplers with at least 64 ''voices'' are advised. For software samplers, the number of voices available depends on the computer's processing capabilities.

A sampler has to deal with huge amounts of data, much more than a normal computer does when used in office applications. This data must constantly be loaded in the memory or backed up to it. Because of this every hardware sampler has built in memory drivers (like RAM on your computer), or better yet a SCSI interface leading to an external drive such as a hard drive, MOD, Jaz or Zip drive. With better samplers it is often possible to add a hard drive after the fact and software samplers save their data to an assigned portion of the internal hard drive on the host computer. So make sure you have a large and fast enough hard drive if you plan to use this type of sampler.

While the first samplers could only reproduce sounds in their original form, many modern hardware and software samplers have a large number of useable effects, filters, and modulation options on-board to give each sound an individual flair.

Hardware samplers have separate audio outputs so that the sounds can be allocated to different mixing channels and from there edited with individual effects. For example you could run the bass drum through output 1 to strengthen the bass signal, the snare through output 2 to add a littel reverb, and the bass through output 3 to apply the chorus effect. The number of inputs and outputs available to software samplers depends on the audiocard used.

For hardware devices, where compatible, there are RAM memory cards, SCSI interfaces, different drives, etc. Many older samplers are no longer supported by the manufacturers however. Which means many parts, including spares, are only available used, if at all. Since the trend seems to be heading more and more towards software samplers it is important, when considering a hardware sampler, to purchase the most complete system possible, including all accessories. This goes for the used as well as the new market.