Beta Testing of Bitwig Studio 5.1

NOTE: Do not use a beta version to work on important projects! Project files created or saved with the beta cannot be opened in previous versions of Bitwig Studio. So if you are opening working projects, save copies of them for beta testing (instead of saving over your original files).

If you think you have found a bug, please drop us a line at beta@bitwig.com. Please be sure to include:

• steps to reproduce the issue
• operating system
• audio interface and any other hardware information
• if this is a new issue in v5.1

And if you get a crash report dialog from the program, please click Send Report. Adding a comment is most helpful, but this can still be useful to us regardless.

System Requirements

• Windows - Windows 7, 8, 10, or 11 (all in 64-bit)
• Mac - macOS 10.14 ("Mojave") or above
• Linux - For installing the Flatpack file, any modern distribution with Flatpak installed (more information here). For installing the DEB file, Ubuntu 20.04 or later.
• a CPU capable of SSE 4.1

What's New in Bitwig Studio 5.1, Beta 1 [released 18 October 2023]

10 New Characters — Modern Filters and Waveshapers, Ready to FX

• Welcome three modern character filters
  • Each use a nested structure, taking a general outer filter and placing a second 'filter' inside the feedback circuit
  • Each algorithm is unique, with a mix of modes for different behavior, tuning controls, inner modulation, and tweak points
  • Drive and Resonance / Feedback are connected, so pulling down Drive a little can often leave more room for resonance to take over, etc. etc.
• Welcome five character waveshapers
  • Each remaps incoming audio to its current shape, transferring each level to the shaper's position
  • Each shaper has its own morphing path, by moving or modulating the single Drive control
  • A little high-level math keeps the shaping tight and smooth via the Anti-aliasing option
• Also along for the ride: a formant filter (Vowels) and a modeled ring Diode shaper

New Filter+/Sweep/Grid module: Fizz (Filter)

• A modern Character filter for spreading harmonic nodes around
• Has a bit of a moving formant sensibility
• Takes place inside a stereo, resonant low-pass filter, with standard options:
  • Drive to affect the incoming signal level
  • Main Cutoff Frequency control
  • Stereo input for cutoff modulation, with Cutoff Modulation Amount set in semitones
  • Key Tracking Amount, for using incoming note pitches to affect the cutoff buss
• For this algorithm, additional controls include:
  • Feedback Gain, which feeds or chokes the nested filter
  • Feedback Cutoff Frequency, for tuning the nested filter
  • A bipolar Color control, which shifts the placement and variation of formant peaks
  • An Alternate Color toggle, for a reorienting and different tuning of the circuit

New Filter+/Sweep/Grid module: Rasp (Filter)

• A modern Character filter that can scream or whisper
• Creates resonant peaks on top of the standard filter
• Takes place inside a stereo, resonant filter, with standard options:
  • Drive to affect the incoming signal level
  • A Filter Type setting, to switch between an outer Low-pass filter or a Band-pass model
  • Cutoff/Center Frequency control
  • Stereo input for cutoff modulation, with Cutoff Modulation Amount set in semitones
  • Key Tracking Amount, for using incoming note pitches to affect the cutoff buss
  • A Resonance Limit Inspector Panel (or Q Limit via right-click context menu) control, for adjusting when the filter model saturates
• For this algorithm, additional controls include:
  • Resonance, which enunciates or chokes the nested filter
  • A Brightness Mode setting, with various options for how resonance peaks move:
    • Shift - Gently moves past the main cutoff, usually emphasizing a central peak
    • Double - A tuned mixture of the Shift and Gravity modes
    • Gravity - Pulls and pushes toward the main cutoff with a bit of magnetism
  • The bipolar Brightness control applies the set Brightness Mode, bending the new resonant nodes thru various harmonic — and inharmonic — positions

New Filter+/Sweep/Grid module: Ripple (Filter)

• A modern Character filter with hyper-resonance
• Three elemental modes provide different levels of fun/wreckage that often anchors to harmonics of the incoming signal
• Takes place inside a stereo, resonant filter, with standard options:
  • Drive to affect the incoming signal level
  • Main Cutoff Frequency control
  • Stereo input for cutoff modulation, with Cutoff Modulation Amount set in semitones
  • Key Tracking Amount, for using incoming note pitches to affect the cutoff buss
• For this algorithm, additional controls include:
  • Bipolar Feedback Gain, which feeds or chokes the nested filter
  • Feedback Cutoff Frequency, for tuning the nested filter
  • A Nature setting, with various models for the filter:
    • Earth - Gently moves past the main cutoff, usually emphasizing a central peak
    • Wind - Focused feedback, ready to blow
    • Fire - Broad feedback, with some motion
  • Two additional toggles, Tweak Feedback and Tweak Feedforward, modify those points in the filter circuit, either dampening or expanding resonance
  • A Low Quality toggle (in the Inspector Panel or via right-click context menu), for adjusting the filter's tuning and reducing the CPU load

New Filter+/Sweep/Grid module: Vowels (Filter)

• An Inspired filter that produces vowel sounds
• Can be used several ways, including:
  • Setting (or hard modulating/automating) a single vowel
  • Setting and morphing anywhere between two and five vowels
  • Any combination, all with different configurations and vowel models
• Standard filter controls include:
  • Drive to affect the incoming signal level
  • A Resonance Limit Inspector Panel (or Q Limit via right-click context menu) control, for adjusting when the filter model saturates
• Five Vowel Position choosers are available, located around a central, bipolar Vowel Blend knob:
  • Vowel Blend at -100 % uses only the nearby Vowel Position 1
  • Vowel Blend at 0 % uses only the nearby Vowel Position 3
  • Vowel Blend at +100 % uses only the nearby Vowel Position 5
  • Vowel Blend corresponds to Vowel Position 2 at -50 % and Vowel Position 4 at +50 %; if set to vowel sounds, only those values will be heard; if set to None (the default), the surrounding vowels will be blended perfectly there
  • Each position offers 27 different vowel sounds to choose from:
    • i - As in “see” or “eat”
    • y - With a rounded w-, like “we”
    • ɪ - As in “sit” or “hit”
    • ʏ - A medium oo, like “ooze”
    • ɨ - An exagerated oo, like “eww” (gross)
    • ʉ - A slow oo, like "ooh!" (surprise)
    • ɯ - As in “hook” or “book”
    • u - As in “pool” or “cool”
    • e - As in “say” or “rain”
    • ø - With a closed -l, like “ool”
    • ɘ - Partly closed, as in “eh”
    • ɵ - As in “foot” or “would”
    • ɣ - Partly closed, as in “uh”
    • o - First sound in “coat” or “bold”
    • ə - As in “run” or “ton”
    • ɛ - As in “get” or “rent”
    • œ - With a round -l, like “ole”
    • ɜ - Partly closed, as in “ah”
    • ɞ - Partly closed, as in “aw”
    • ʌ - As in “fun” or “come”
    • ɔ - As in “more” or “floor”
    • æ - As in “cat” or “hat”
    • ɐ - With a subtle -r, like “are”
    • a - First sound in “hi” or “fight”
    • Œ - With an open -l, like “all”
    • ɑ - As in “far” or “star”
    • ɒ - As in “want” or “job”
  • Each Vowel Position can be set in two ways:
    • Clicking on any position opens a pop-up menu of all available vowels sounds and description texts
    • Clicking and dragging on any position starts moving thru the vowels sounds, for a quick workflow with audible results (if audio is passing)
  • In The Grid, a stereo in port (Vowel In) is available for adding to the Vowel Blend value
• Profile selects which set of vowel data to use, with choices including:
  • Women 1 - an older data set, with average values from several women
  • Women 2 - a modern data set, with average values from several women
  • Female - values from one individual female
  • Men 1 - an older data set, with average values from several men
  • Men 2 - a modern data set, with average values from several men
  • Male - values from one individual male
  • Kids - average values from several children
• The Topology chooser (on the right edge of the module) sets the structure of the filter, from three choices:
  • Cascade - Serial low-pass filters; a bit better for traditional text-to-speech sounds
  • LP/BP - Low-pass and band-pass filters, processed in parallel; a synthier, Autobahn-friendly vibe
  • LP/BP/HP - Low-pass, band-pass, and high-pass filters in parallel; adds a slight bit more highs
• Several parameters influence the tuning of the internal filters in use:
  • Cutoff Pitch Offset moves the internal filters by semitones
  • The Cutoff In port and its associated Cutoff Modulation Amount attenuator allows stereo manipulation of the Cutoff Pitch Offset
    • Note: While this is like moving the cutoff of a normal filter up and down, the result is different and you might want to start by disabling pitch modulation
  • The Cutoff Frequency Offset (in the Inspector Panel or via right-click context menu) allowing linear frequency manipulation, which can be interesting for formants
  • Resonance adjusts the relative sharpness of the internal filters

New Filter+/Sweep/Grid module: Push (Shaper)

• A Character soft clipper with a detailed curve
• One Drive parameter for going thru the module's unique path, with optional high-order Anti-aliasing (AA)

New Filter+/Sweep/Grid module: Heat (Shaper)

• A Character S-shaped clipper that starts soft but can drive hard
• One Drive parameter for going thru the module's unique path, with optional high-order Anti-aliasing (AA)

New Filter+/Sweep/Grid module: Soar (Shaper)

• A Character soft wave folder that makes the quietest parts loud
• One Drive parameter for going thru the module's unique path, with optional high-order Anti-aliasing (AA)

New Filter+/Sweep/Grid module: Howl (Shaper)

• A Character wave folder puts different parts of the signal into loud focus
• One Drive parameter for going thru the module's unique path, with optional high-order Anti-aliasing (AA)

New Filter+/Sweep/Grid module: Shred (Shaper)

• A Character non-linear wave folder for subtle cancellation or big-time artifacts
• One Drive parameter for going thru the module's unique path, with optional high-order Anti-aliasing (AA)

New Filter+/Sweep/Grid module: Diode (Shaper)

• A Parametric shaper modeling the classic circuit in a modern way
• Offset parameter for biasing the signal to be asymmetric
• Drive parameter for pushing the signal to bend
• Low-pass Cutoff Frequency control for rounding it off and restoring some order
• One for going thru the module's unique path, with optional high-order Anti-aliasing (AA)

New device: Filter+ (Filter)

• A dead-simple FX box, letting deploy any waveshaper and filter from The Grid directly onto a track
• Pick one of ten filters from three categories:
  • Structural choices for classic circuits:
    • Low-pass LD - A ladder filter, with variable slope and nonlinear option
    • Sallen-Key - 16 various low-, high-, and band-pass configurations
    • SVF - Highly resonant multimode (high-, low-, band-pass & notch) filter
    • Comb - A comb filter with timed feedback & dampening
  • Inspired options that speak:
    • Low-pass MG - A Moog-style low-pass filter, including drive character
    • XP - An Oberheim-style multimode filter, with 15 configurations)
    • Vowels - A morphing vowel filter, with various models, pitch and frequency offsets
  • Character ideas for something new:
    • Fizz - A nested filter circuit that can sparkle, shimmer like a phaser, or bump
    • Rasp - A filter that adds brightness around the cutoff, so it can scream or whimper
    • Ripple - A hyper-resonant circuit for playful feedback, subharmonics, or even distortion
• Pick one of 14 waveshapers, sorted for you:
  • One Knob classics with a singular control:
    • Chebyshev - Nonlinear shaper that can target harmonics
    • Distortion - Gentle distortion
    • Hard Clip - Simple, hard clipper
    • Quantizer - Signal resolution reducer
    • Wavefolder - Reflects each cycle back on itself
  • Parametric options that offer more control:
    • Diode - Classic circuit model, used for biasing and clipping
    • Rectifier - Independent positive and negative attenuators
    • Saturator - Waveshaper with loud/quiet settings + bipolar skews
    • Transfer - A freely drawable, segmented waveshaper, with BWCURVE-file support
  • Character ideas, for unique paths and simple control:
    • Push - Soft clipper with a detailed curve
    • Heat - S-shaped clipper that starts soft but can drive hard
    • Soar - Soft wave folder that makes the quietest parts loud
    • Howl - Wave folder that puts different parts of the signal into loud focus
    • Shred - Non-linear wave folder for subtle cancellation or big-time artifacts
• Signal flow is simple: audio input → waveshaper → filter
• Pre FX and Post FX chains are also available, for nested other devices or plug-ins
• A modulation section offers two built-in sources:
  • A stereo LFO module gives four waveshapes with sync-able Rate and Timebase controls
  • The incoming audio itself provides a second modulation source, with optional low-pass filtering and rectification (to make the modulation go in only one direction)
  • Both LFO and Audio Mod sources are normalled to the filter's cutoff buss, with attenuators on the filter
  • These sources are also available as modulators for free control elsewhere, including shaper Drive controls, other filter controls, or control of any nested devices in the Pre FX and Post FX slots
• Additional Inspector controls for Stereo Spread and Wet Gain
• Other parameters are available in the device's Expanded Device View, which exposes the embedded Grid patch. These include:
  • LFO Skew (to bend the shape), Phase, Phase Offset (R) (for the right channel, controlling the stereo effect), Bipolar, and Sync to Global Transport toggle (on by default)
  • Pitch Buss toggle (with = icon) to not attenuate the audio mod source, giving it a ±10 octave range
  • A simple Pan module, for placing the signal
• Being a Grid-powered device, polyphony and voice stacking are uniquely available in this audio FX device
• By right-clicking the device header, functions are available to:
  • Convert to Sweep, for bringing all settings into that device
  • Convert to FX Grid, for full patching control

New device: Sweep (Filter)

• A performable effect devices, combining and blending waveshaper and filter from The Grid
• Everything said about Filter+ is true of Sweep, except Sweep has a second filter slot and generalized controls for this setup
• Joint Frequency Control provides control of both filters in a range ±3 octaves
  • An Invert option flips the direction that the Joint Frequency Control applies to filter B, allowing you to move their cutoffs in opposite directions
• The Routing Blend control smoothly moves thru various device configurations:
  • At 0 %, only filter A → waveshaper is heard
  • 50 % is a parallel routing, with device audio input going straight to both filter A (then out) and to waveshaper → filter B (then out)
  • 100 % is fully serial, with device audio input going to filter A → waveshaper → filter B
  • Positions in between blend these routings, for a continuous range
• Stereo Pan is similar to the Joint Frequency Control, except it applies the same stereo adjustment to both filters
  • Positive (rightward) settings move right channel cutoffs up and left cutoffs down, and negative (leftward) settings move left channel cutoffs up and right cutoffs down
  • Good, quick stereo-ization control

Voice Stacking Expansion

• Voice Stacking remains a unique feature of Bitwig Studio, allowing any polyphonic device — including compatible plug-ins — to create multiple layers of sound
• 16 is now the maximum, allowing each note played to trigger up to 16 independent voices
• When voice stacking is active, any single voice in the stack can be soloed
  • This temporary gesture makes sound programming easier and fast
  • This option is available in the device Inspector Panel, along with the Voice Stacking setting
• The old Voice Stack (Note-driven) modulator has been replaced by two modulators in the new Voice Stacking category:
  • Stack Spread offers the original four spread modes, along with eight new ones:
    • Flipped - Provide simple reciprocals [1, 1/2, 1/3, ... 1/n]
    • Straight - Gives harmonic relationships [1, 1/2, 1/4, ... 1/2ⁿ], whether for pitch or rhythms
    • Primes - Inverted prime numbers [1, 1/2, 1/3, 1/5, 1/7, 1/11...]
    • Golden - The Fibonacci sequence [1, 1/2, 1/3, 1/5, 1/8, 1/13...]
    • Rand+ - Creates a unipolar random value for each voice (at note on), with all values adding up to 1
    • Rand+ ↕︎ - A scaled, stronger version of Rand+, tending towards more large values
    • Rand± - Creates a bipolar random value for each voice (at note on), with all values adding up to 0 (zero)
    • Rand+ ↕︎ - A scaled, stronger version of Rand±, tending towards more large values
  • Voice Control provides 16 individual modulator sources for programming any of the 16 voices uniquely
  • With more spread modes, you may want to insert more than one Stack Spread modulator
  • Previous presets or projects using the Voice Stack modulator are automatically converted to use the new modulator(s)
• Three new Grid modules offer a new level of control when voice stacking is active:
  • Voice Stack Info (I/O) provides two signals:
    • Voice Stack Index outputs a polyphonic signal of the current voice's index within the stack
    • Normalized by default, the signal is ideal for creating your own spread modes (maybe try running it thru a Bend [Level] module, or any other shaper)
    • Voice Stack Size outputs a mono signal reflecting the current voice stacking setting, which could be useful for dynamic gain adjustments, etc.
  • Voice Stack Mix (Mix) is a processor that provides a mixer interface for each voice in the stack
    • This provides a Level, Pan, Solo, and Enable control for each voice in the Inspector Panel
    • Each control can be mapped, modulated, or automated
    • The module can be inserted anywhere in a patch — and as many times as you like. This isn't just for audio.
  • Voice Stack Tog (Mix) is a processor the provides simple Enable controls voice each voice
    • Like Voice Stack Mix, this is a processor that can be inserted anywhere in a patch and is free to be mapped or modulated
    • This module could be useful in many cases, including with Note Grid (before the Gate Out, for example)
  • Since the Voice Stack Mix and Voice Stack Tog modules share an Enable control for each voice, swapping between the two modules will keep these particular data
• Since Filter+ and Sweep are based on FX Grid, there are now three audio FX devices that can use polyphony and voice stacking
  • Simple idea: try a band-pass filter in Filter+, enable Voice Stacking, and spread the cutoff

New Polymer/Grid module: Bite (Oscillator)

• A Techniques-driven oscillator, offering exponential FM, hard sync, PWM, and ring mod from dual oscillator feedback
• Anti-aliasing and internal feedback allow for some very crisp, fresh, and/or modular analog sounds
• Internal Oscillator A & B are identical, each providing seven waveshapes with Pulse Width controls, as well as fixed Saw and Sine options
  • Like the Union oscillator, the oscillators exhibit some smooth analog drift when Pulse Width, for example, is moved
• Oscillator B can pulse-width modulate (PWM) Oscillator A
• Oscillator A can do exponential frequency modulation (xFM) on Oscillator B
• Oscillator A can also hard SYNC Oscillator B:
  • SYNC is a useful way to use exponential FM without breaking the pitch of Oscillator B
  • Oscillator B also has its own Pitch Offset control, for setting (or modulating) more interesting hard sync waveshapes
• A trio of mix controls set the output level for oscillator A, oscillator B, and a ring-modulated mix (RM) of the two
• The Grid module version has a special Independent Mono Mode toggle in the Inspector Panel
  • This flattens the module to a mono output
  • This also allows individual oscillator targeting via the in ports, routing left channel inputs only to Oscillator A and right channel inputs to Oscillator B

Onset Power and Audio Quantize

• Our new, higher-resolution onset detector from v5.0 provides better data and allows for new functions
• Audio clips/events now have an Onset Threshold setting
  • This value affects playback, setting which onsets are preserved by stretching
  • Onsets below the threshold will appear more dim, and if affected clips/events are onscreen with onsets showing, this will be updated while dragging the Onset Threshold setting
  • This value is in the Audio Event > Stretching area of the Inspector Panel
  • The setting is available for audio using either the Stretch, Stretch HD, or Slice when the algorithm has its Transient Rate set to at Onsets
• A new Audio Quantize… function is now available for audio clips and events
  • Set the desired beat grid in the dialog
  • Adjust the Onset Threshold within the dialog to set which onsets should get quantized (via stretch markers)
    • Again, if the clips/events are visible when the dialog is open, they and their onsets will be brightly previewed in the background
  • The Amount, Shuffle, and Humanize settings determine the movement and position of quantized
  • Click OK to make it so. Boom.
  • An additional Audio Quantize (no ellipsis […]) function is available to skip the dialog and directly apply your last values
• Other functions that can use onsets (Slice in Place, Slice to Drum Machine, Slice to Multisample) now have the Onset Threshold control, interface, and visual preview mechanism

Some Mixer Love

• The various panels in the Mix view and Mixer Panel have been reordered, made scrollable, and show/hide is better ways (when available space changes)
• A permanent left column (below scenes, when visible) has been added for the mixer view toggles and section labels (Devices, Sends, I/O, etc.)
• When multiple tracks are selected, adjusting one track will adjust them all, whether it is volume, mute, output routing, etc.
• The left column also has a horizontal divider line that can be dragged ups and down to resize the volume faders and their level meters
  • The peak hold level for each track is now visible at most heights, just above the level meters
  • Hovering over the peak hold level shows updated stereo information about the track level
  • Clicking the peak hold level clears it for all tracks, as well as the small red peak indicators at the top of each meter
• Tracks can also be made much narrower now
• The mini device view of the mixer now shows clickable icons for native Bitwig devices that have an Expanded Device View
  • Clicking the icon scroll the device (in the Device Panel) onto screen and opens the Expanded Device View above
  • The same is true in the Inspector Panel, except the icons are always visible (give or take EQ+ and EQ-5)
• Holding [SHIFT] when resizing on track resizes all tracks in that section to the same size
  • Holding [SHIFT] now performs the same function when resizing Arranger track lanes
• Track headers have been redesigned to give more space to track names, and to make selections and relationships more obvious and clear

New Features

• A non-destructive Normalize function is now available on audio clips/events, inserting a Gain expression value for the clip (or for each audio event, when multiples are selected)
• New Set Clip Start function is available in the Detail Editor Panel, sliding the clip contents to the current time selection
  • Available in the Time menu or in the right-click menu
• Container devices with three or more chains (Multiband-3, XY FX, and XY Instrument) now have mute / solo controls for each chain
• Container devices with two chains (Multiband-2, Mid-Side Split, and Stereo Split) now a solo button for each chain
• FX Grid (The Grid) device: Now has an Auto-gate from Notes parameter
  • If you enable an FX Grid patch to be polyphonic, now you don't have to change your patch at all as some simple enveloping is applied for each note played
  • The Auto-gate toggle is in the Note Source section of the Inspector, with an Auto-gate Release Time setting right beside it
  • Auto-gate is on by default and has no effect if your patch is using True Mono
  • Both Filter+ and Sweep also have this Auto-gate option
• New Toggle In (Mix) Grid module: Switch between two incoming signals, with a button directly on the module
• New Toggle Out (Mix) Grid module: Switch between two outgoing paths, with a button directly on the module
• New Pitch Buss (Pitch) Grid module: Pitch summing buss with attenuators for up to six inputs
  • Attenuators are set in a range of ±36 semitones
  • Inputs two to six also have a Thru option (a clickable = icon) that adds that incoming signal without attenuation, good in the case of actual pitch signals, etc.
• New Invert (Math) Grid module: Gives a button to reverse polarity (× -1) of the incoming signal, with Stereo-ness option
• New Reciprocal (Math) Grid module: Gives a button to flip (1/x) the incoming signal, with Stereo-ness option
• LFO, Curve, and Wavetable LFO (LFO) Grid modules: Now have a Sync to Global Transport option
  • This mode positions (and keeps) the module's clock relative to current global transport
  • This disables the Retrigger In port and the Retrigger on Notes pre-cord options
  • This setting defaults off for the module, and defaults on within Filter+ and Sweep
• Chebyshev (Shaper) Filter+/Sweep/Grid module: Now has a Remove DC option to subtract the function's inherent DC
• Support for Softube Console One devices [33559]
• DAWproject files can now be exported (from the File menu) and imported by Bitwig Studio
  • This new format allows transferring project data from one audio sequencer to another
  • More information on the format is here
• Controller script added for Derivative's TouchDesigner
  • This script from Derivative provides an integration between Bitwig Studio and TouchDesigner, for dynamic live visuals and more
  • More info here

Improvements

• Zooming behavior with shortcuts now centers around the active selection if present, or otherwise around the mouse location if present over the editor
• Various editing improvements:
  • Make Legato function now treats chords together
  • After using the Knife tool to slice, the second clip/event is now selected
  • Time selection tool can now move beat markers when hovering marker head
  • Eraser tool can now be used to delete beat markers and onsets
  • The Arranger loop length can now be extended by clicking outside of the loop region
  • The Arranger loop length can now be shortened by using [CTL]-click ([CMD]-click on Mac) inside the current Arranger loop
  • Dragging in new beat markers from onset locations has moved to top edge of the Detail Editor Panel, where the onset locations are shown
  • [ALT]-dragging now slides the waveform under a beat marker
  • Dragging automation events now feels smoother
  • Better beat marker visibility
• Instrument Selector and FX Selector devices: May now process chains in parallel
• EQ+ (EQ) device: Is now latency compensated
• Test Tone (Utility) device: Now has cleaner Saw and Square waveforms
• Transfer (Shaper) Filter+/Sweep/Grid module: The Reset Curve function (available on right-click) makes a three-point identity function for better custom editing
• Saturator (Shaper) Filter+/Sweep module: Has a Pop-out Editor, making all parameters available even when outside of The Grid
• ≥ and ≤ Grid modules (Math): Now have an Exact Matches Only setting, which can be disabled from the Inspector or the context menu [33705]
• Bounce only processes the track's prerequisites (and not the whole project)
• Support for sample-accurate automation recording from controllers and plug-in windows
• New function to Delete All Project Automation
• New function to Delete All Automation Owned by Selection
• VST3 and CLAP: We now honor any plug-in's realtime requirement
• CLAP: Uses MIDI MPE if the plug-in asks for it

Fixes

• Instrument Selector, FX Selector, and Note FX Selector devices: Index automation is now latency compensated correctly [33766]
• Library clips containing Arpeggiator were previewing twice as fast as they should have [33764]
• Now correctly update the project's PDC, when activating a plug-in for offline rendering results in a latency change
• Touch automation mode now works when plug-in parameters are changed from within Bitwig Studio's GUI [33768]
• Fixed a rare engine crash [32296,32104]