Alchemy provides a total of 15 multi-mode filter modules. These can be used at the source level or as main filters that process all routed signals. Filters used at the source or main filter stages affect each voice independently. You can also insert filters in the effects section, but they are monophonic when used at this point in the signal path and affect all voices (the overall signal) simultaneously.
Two main filters. See Alchemy main filter controls.
Four sets of three source filters. See Alchemy source filters.
The MMFilter module in the effects section. See Alchemy filter effects.
There are dozens of filter types to choose from in the main filters and the MMFilter effects module.
A lowpass (LP) filter passes the portion of a signal below a specified cutoff frequency and rolls off the portion above that frequency. A bandpass (BP) filter passes the portion of a signal occupying a band surrounding the cutoff frequency and rolls off the portions above and below that band. A highpass (HP) filter passes the portion of a signal above a specified cutoff frequency and rolls off the portion below that frequency.
There are multiple LP, BP, and HP filter designs in Alchemy, each with distinctive characteristics that you may prefer for a given purpose. The available LP, BP, and HP filter designs include:
Clean: Bi-Quad filters. These filters have no Drive option. These were the BQ filters in previous Alchemy versions.
Edgy: Two- or four-pole analog-modeled filters. Four-pole variants provide a steeper roll-off of frequencies beyond the cutoff. These were the XP filters in previous Alchemy versions.
Gritty: Two-pole filters designed to saturate heavily at higher Resonance and Drive settings. These were the Bee filters in previous Alchemy versions.
Rich: Two- or four-pole analog-modeled filters. Four-pole variants provide a steeper roll-off of frequencies beyond the cutoff. These were the MG filters in previous Alchemy versions.
Sharp: Two-pole analog-modeled filters. These were the K20 filters in previous Alchemy versions.
Smooth: Two- or multi-pole state-variable filters. These were the SVF filters in previous Alchemy versions.
The three principal filter controls have standard functions for all filter types.
Cutoff: Controls the filter cutoff frequency.
Resonance: Controls the filter resonance or emphasis. Higher settings boost frequencies in the immediate vicinity of the cutoff frequency.
Drive: Allows the filter to be overdriven; the precise effect varies with each filter design.
A formant filter works like a bandpass filter with a potentially very narrow bandwidth. It is designed to mimic a formant, or fixed resonance, such as those produced by the hollow body of a guitar or by the shape of the vocal cavity (which is adjusted during speech to produce different vowel sounds).
Cutoff: Controls the resonant frequency of the formant.
Amount: Controls the width of the band surrounding the resonant frequency. Low values produce a narrow band, high values a wider one. Low resonance values may allow little or no sound to pass through the filter.
A notch filter cuts a narrow band around a resonant frequency. The remainder of the signal is affected minimally.
Cutoff: Sets the frequency of the cut band.
Amount: Controls the amount of attenuation.
A peaking filter boosts a narrow band around a resonant frequency. The remainder of the signal is affected minimally.
Cutoff: Sets the frequency of the boosted band.
Gain: Controls the amount of boost. Higher values are generally the most effective.
BW: Sets the bandwidth around the cutoff frequency.
A comb filter mixes the original signal with one or more copies of the signal which are delayed by a very short time interval. At some frequencies this mixture causes phase cancelations, while at other frequencies it causes reinforcements. The result is a spiky frequency spectrum with multiple resonant peaks. Graphically, these peaks resemble the teeth of a comb, which gives this filter type its name.
Alchemy offers three comb filter designs, each with its own character. The best choice is very much a question of your preference and the type of sound you are trying to create. That said, there are some distinguishing characteristics that may help guide you.
Comb Pos uses positive feedback on the delay lines, while Comb Neg uses negative feedback to produce less extreme effects, often with a hollow quality. These two are the less powerful combs and offer a much more gradual increase in resonance. They can be useful when you require either a less dramatic effect or you want to hear more of the exciter signal character in your sound. The latter point is noteworthy as this trait can be useful when you want a more naturalistic modeled sound.
Comb PM uses bipolar feedback on the delay lines. The resonance control is bipolar, allowing you to freely shift from negative (hollow sound) on the left to positive (bright and peaky) on the right. This comb is useful for classic bright Karplus-Strong style sounds, where the exciter impulse is not easily heard and the comb is more prominent. Take care with your resonance level because it is capable of quickly going to extremes, which can lead to feedback. Start with a resonance level of zero and increase (or decrease) slowly to find a suitable effect strength.
For instrument modeling, it is best to assign combs as source filters, rather than in the master filter or effect sections. Although comb filters can be used in the master filter section, it is better to use the main filters for additional processing. Comb filters are also available in the effects section but are not polyphonic, so you will achieve better results if they are placed earlier in the signal path.
Cutoff: Controls the delay time in the comb circuit. Lower cutoff values equate to a longer delay.
F-Back: Controls the amount of feedback in the comb circuit.
Damp: Softens the frequency peaks in the comb-filtered spectrum.
Note: Sending a percussive sound into a highly resonant comb filter causes it to ring at a frequency determined by the delay time you have set with the Cutoff knob.
Ring modulation is a process in which a modulator and a carrier signal are multiplied. Each frequency component of the modulator interacts with each frequency component of the carrier to produce two sidebands: a sum and a difference (carrier – modulator). When a filter is set to the RingMod type, the signal entering the filter acts as the modulator, while the carrier is supplied internally by the filter.
Frequency: Controls the carrier frequency.
Mix: Applies a constant offset to the carrier.
At 0%, the carrier wave varies between –1 and +1, resulting in classic ring modulation.
At 100%, the carrier wave varies between 0 and 1, resulting in classic amplitude modulation. In this case, the carrier signal itself is present alongside the sum and difference sidebands.
Weird: Controls the carrier waveform.
Set to 0% for a pure sine wave carrier. Use this setting to produce characteristic bell-like timbres.
Set above 0% to progressively truncate the carrier waveform. Higher settings produce additional sidebands, typically resulting in noisy timbres.
Alchemy filters offer several distortion effects.
Bitcrusher is a harsher-sounding digital effect.
Downsample is a lo-fi digital effect (similar to Bitcrusher but is frequency dependent).
Mech is a more metallic-sounding distortion.
Tube is a classic tube distortion effect.
Note: Distortion effects created in the source and main filter sections are polyphonic, with each voice distorted independently. This avoids unpleasant intermodulation effects when you play chords. In contrast, the Distortion module in the effects section at the end of the signal path processes a mix of all voices.
The filter controls work as follows when the filter type is set to a distortion effect:
Amount: Controls the intensity of the distortion effect.
Mix: Controls the mix between clean and distorted signals. A value of 0% results in the clean signal only. A 50% value results in an equal mix of clean and distorted signal. A 100% value results in the distorted signal only.
PreGain/Freq: Boosts the input gain. Sets the frequency for the Downsample distortion type.
These tools provide creative options at the filter stage of the signal path. Unlike dedicated FM synths, Alchemy does not have preconfigured algorithms or a mod matrix setup specifically for FM. It does, however, offer the option to run a huge number of FM filters in various series and parallel configurations. FM in Alchemy also has a different character to classic FM synths. FM in Alchemy is more like working with FM on analog synths where you are modulating oscillator frequency rather than phase. As a result, Alchemy’s FM is great for adding different kinds of effects to your sound, as well as “organic” distorted textures.
The filter controls work as follows when the filter type is set to Compressor:
Thresh: Set the threshold level—signals above this threshold value are reduced in level.
Ratio: Set the compression ratio—the ratio of signal reduction when the threshold is exceeded.
Gain: Set the amount of gain applied to the compressed signal.
The filter controls work as follows when the filter type is set to FM:
Freq: Set the carrier oscillator frequency.
Tip: For classic FM sounds, double-click the Freq knob to set it to the center position, then Control-click the knob and choose Add Modulation > NoteProperty > KeyFollow from the shortcut menu.
Mod: Set the degree to which the modulator (source audio) can modulate the frequency of the carrier.
F-Back: Set the amount that the output of the carrier is fed back into itself.