Next to a synthesizer, the sampler is probably the most common instrument found in any modern recording studio. At the core a sampler is an electronic instrument that emulates physical instruments by recording notes from real instruments and assign them to be played back through a midi keyboard, drum pads, etc. Obviously you can record and trigger other audio material like foley and sound effects as well.
Early samplers were hardware devices, and include many famous products like the Akai MPC series, Akai S950 and the Emu SP1200. Hardware samplers are still produced (eg. the SP404, the Elektron Octratrack and MPC to name a few), but today modern sample based instruments are often virtual instruments running as plugins in your DAW. The industry leader here is certainly Native Instrument Kontakt, but each DAW has a more or less famous native sampler, like Logic’s EXS as well as Ableton’s Sampler and Simpler devices.
The big upside on with software based samplers is that modern computers have large RAM and hard drive space, which allow you to store a sampled instrument where each note is recorded in multiple velocities or articulations, which give you impressive realism and expressiveness.
While the core principles and workflow are straight forward, if you dive deeper into samplers (both hardware and software), you’ll encounter terms that might not be instantly clear to you. To save you from the menu diving and google searches, I’ve compiled a list of explanations below.
Rompler — In the days of hardware samplers, you found samplers which came with an internal sound library. While technically a sampler, they were more like sample playback devices. The Emu Planet Phatt for example which gave you a specific sound selection pre-installed into the ROM memory. A sampler that doesn’t let you sample your own sounds, is referred to as Rompler.
Multi timbral — The ability of a single electronic instrument to simultaneously provide multiple instrument sounds for recording & playback. Back in the days it meant on each one of the 16 different possible MIDI channels a different instrument voice could be played back. Nowadays it is expected that you can open different instruments in your DAW. Some virtual instruments still offer the multitimbral functionality to be able to layer sounds, like different sampled pianos in Kontakt. With multitimbral functionality you can achieve richer, complex sounds.
Root Key — Since the sampler is based on recordings of real instruments and lets you play that instrument on a standard midi controller, samplers have to transpose the sample. That means they play the sample from all keys raising or lowering the pitch. For this to work properly a root key has to be defined for the original sample. If you recorded your guitar at C#, you will need to set the root key in your sampler to C# to have it mapped properly.
Transpose — After you mapped the root key, the sampler will transpose the sound across the keyboard. This works well within a few semitones, but transpose it too much and your instrument will sound unrealistic. It’s like playing a vinyl record, you dont only change the pitch, but also the the attack and decay times – the overall tone/timbre of sound.
Multisampling — This is why sample libraries get larger and feature multisampling. You record several notes of different notes from the same instrument and assigning them to play back from the appropriate keys on a midi keyboard.
Key Zone / Key Map — In multisampling, each sample has its own root key that describes its original pitch and a key zone, which determines how many other notes on the keyboard will also play that sample automatically transpoing it up and down. This is only relevant if not all possible notes have been sampled from the original instrument. In the most detailed multisampled instruemnts, each note of the original was sampled, so each sample will be played from one key (its key map is one single key), the root key. When you defined all the root keys and key zones, they are combined into a key map – the sound source from that sampled instruemnt. SImilar to an oscilator in a synth, this is the starting point of the patch.
Velocity Zones / Layers — Each sample is just a snapshot of the tonal variations you can get from the real thing. A piano key sounds different if you hit it hard or soft. When you turn down the velocity of a hard hit in your software sampler, it will not sound soft, but like a hard hit with the volume turned down. You are not getting the softer attack and less agressive tone. You can get close by using the ADSR and filter settings on your sampler, but most sample libraries now use multiple velocity zones / layers to achieve realism. That means they sample the instrument at multiple levels, and assigned a velocity zone. Each zone is assigned a midi veloctiy value from 0-127.
Keyswitching / Articulations — A string instrument has different playing styles, like legato, staccato and pizzicato, etc. To make these available it would require seperate keymaps / sampled instrument. But this is sometimes inconveinent if you want to switch articulations frequently within a musical phrase. With keyswitching you get several instrument articulations layered on top of each other and with a specific set of keys you can pick which articulation you want play. SO you can start playing a melody in legato in the C3 range, iht a key switch at C1 and instantly continue playing staccato. switch in real time
Looping — Some instrument might need to provide an indefinite sustain, like on pads. Also looped rhyhtmic drum and percussion elements should run indefinitely. Repeat automatically as the key is held down. IF you import your own samples, the task of looping the sound correctly falls on you. You have to go to the edit page in your sampler of choice. Especially on drum phrases the looping is pretty straighfrward as all the elements are visually easy to identify. Looping a sustaining sound is a lot more diffcult to get right. You need to find the loop start and end point which don’t produce a click or glitch each time the wave moves around. In general you want to find two zero crossing to get the loop sounding smoothly. When that proves difficult, most samplers have a crossfade option, which easier to get right, but often the sound will be phasy. Just be prepared to spend some time getting the loops right. Commercial libriaries like … come with looped sampler patches already
Sample Rate / Bit Depth — Just like you have a project in your DAW set to a specific sample rate like 48khz or bit depth like 24bit, a sampler also has this. In early days of sampling, ram memory was at a premium. CDs were pressed at 16bit 44khz, but most samplers were at a much lower resoultion, initially 8 bits, later sampler like the sp1200 defined the golden era boom bap sound with its crunchy 12 bits sampling. Combined with the limited sampling time, these machines were a far cry from being realistic representations of insturments like the ones we get with a 70GB Kontakt library these days. Nowadays the edgy lo-fi quality of these machines is their charme and many of them fetch far higher prices than what they used to sell for 20 years ago. If you can’T afford them, there’s of course virtual instruments in your DAW that mimic the sound of these machines. The funny thing is that a machine like the EMU Emulator II had on-board memory of …MB while its equivilant Kontakt Library comes in at a whopping 16.000MB. Modern, realistic sample libraries at 16bit 44.1khz minimum, often 24bit 48khz. Especially for sound design and film you even find higher sample rates like 24bit 96khz.
Time Shifting — Almost all DAWS let you change speed while maintaining pitch and vice verca. Hardware and virtual samplers often have this function built in, althought their name might change, like Kontakts ‘Timemachine Pro’ or ‘Beatmaschine’.
Hope this little excursion into sampler knowledge helped you a bit. Let us know if you have any specific things we’ve missed.