User Manual
Hello! This is a user manual for the CHOWTapeModel plugin, currently available as (VST/VST3/AU) for Windows and Mac (and Linux if you compile yourself). CHOWTapeModel is first and foremost a science experiment, with the plugin as a tool for mixing engineers and producers as a secondary benefit. The plugin is based off a 2019 DAFx paper "Real-time Physical Modelling for Analog Tape Machines". While the plugin was originlly designed to model the Sony TC-260 tape machine, it has been extended to model any general reel-to-reel machine.
Binaries can be downloaded from the Releases page.
For experimental features, the most recent builds (potentially unstable),
see the Plugin/Bin folder.
For PC users, download the binary files that you want (VST/VST3/Standalone) and place the plugin files in your plugin folder.
For Mac users, download the MacBuilds.zip folder. This contains VST, VST3, AU, and Standalone versions of the plugin. Unzip, and copy the plugin files to your plugin folder. Note that the Standalone version may not work unless you allow applications from unidentified developers.
Input Gain controls the gain level going into the rest of the plugin. Note that abnormally large levels can cause the plugin to become unstable, so it is recommended that your levels are below unity gain going into the plugin, and any extra gain you wish to add should come from the input gain control.
Output Gain controls the level coming out of the plugin.
Oversampling controls the amount of oversampling being done internally within the plugin. More oversampling will result in a higher quality output with fewer aliasing artifacts and better noise characteristics, but will also be heavier on your CPU. I would recommend using as much oversampling as your CPU will allow.
The hysteresis processor is the most important section of the plugin, and is the most crucial for making the tape model sound like actual analog tape. I've written a blog post about this hysteresis model on Medium, feel free to skip the equations if you like.
Drive controls the level of amplification being done by the hysteresis process. This differs from the input gain in that the input gain amplifies the signal without directly adding distortion, while the drive parameter actually changes the nonlinear characteristic of the hysteresis process.
Saturation controls the amount of saturation taking place in the hysteresis function. Higher levels will result in a more distorted sound.
Bias controls the amount of bias being used by the tape recorder. Internally, less bias makes the hysteresis curve "wider", thus creating a "deadzone" effect at low levels.
Physical tape machines also have a frequency response that is affected by the amount of space between the playhead and the tape, the width of the playhead gap, and the thickness of tape used. The frequency responses of each of these "loss effects" is also dependent on the tape speed.
Spacing controls the amount of space between the playhead and the tape, measured in millimeters.
Thickness controls the thickness of the tape, measured in millimeters.
Gap controls the width of the playhead gap, measured in millimeters.
Speed controls the tape speed as it effects the above loss effects, measured in inches per second (ips). While this control is continuous, it is typical to run tape at 7.5, 15, or 30 ips.
The degradation parameters control a simulation of damaged tape. Depth and Amount control the amount of degradation that is added to the tape, while Variance adds a time-varying randomness to the degradataion.
The chew parameters simulate tape that has been chewed up by a broken tape machine. Depth controls how deep the tape is chewed; Frequency controls how much space there is between bits of tape that have been chewed up.
Tape machines also exhibit timing irregularities, often due to small imperfections in the mechanics of the machine causing the tape to subtly speed up and slow down while being played back. The flutter characteristic in this plugin was measured from my own Sony TC-260 tape machine.
Rate controls the rate of flutter, with higher values causing the flutter to occur faster.
Depth controls the depth of the flutter, with 0 meaning that no flutter is occuring, and higher values making the flutter more noticeable.
"Wow" is similar to flutter but on a much longer time scale, and contains similar controls.
The presets provide a quick way to achieve a certain sound with the plugin. To submit your own preset to be added the the plugin, see the Presets GitHub Issue.
If you notice any bugs or have any questions, feel free to email me directly, or create an Issue ticket in this repository. GitHub issues are preferred since other users can see them as well.
Enjoy!
Jatin Chowdhury