StandardGATE Manual

StandardGATE is an easy-to-use audio-plugin for gating audio. It offers zero-latency processing with low CPU-consumption.

A gate is most commonly used to remove unwanted noise or spill from signals by reducing the volume of low level sound in the audio signal. However, other creative effects can be created with this gate.

The scrolling audio display shows in real time how the waveform gain is being processed. This gate includes the usual standard controls to help clean up audio recordings.

There is EQ on the sidechain input which can be used for focused gating applications. Also, the external sidechain option allows for creative gating effects. There is MIDI control of the gate too to enable exciting rhythmic effects.

The envelope filtering options can be used to reduce spill with more natural sounding gating. As well as gating, there are ducking and upward expander modes.

Open the dmg-File you have downloaded and double-click on the containing pkg-File. Select "continue" on the welcome screen. If you see the "Select the Destination" screen, you can click "continue" as well. Then, select "Install" and enter your admin password. The actual setup software will start afterwards.

Please click on the "I agree" button if you agree to the "License and Legal Information".

If this is a first-time installation, the license key will be requested at this point. First, copy the license key from the mail you received after purchasing the software. If you can't find your license key, please make use of our License-Key Recovery Tool.

Select the license key and press the key combination "Command" + C, so that you can copy it. To paste the license key into the input field of the installer, press the key combination "Command" + V.

This picture shows an example of what the setup application looks like after the license key has been inserted. The license key consists mainly of letters and numbers. For some products it also contains the email address and special characters. In the product delivery email, the license key is a separate block of monospace characters. (The license key on this screenshot is generated for demonstration purposes only and does not work).

The correct plugin paths for VST, VST3 and AAX plugin formats are pre-configured automatically. If you want, you can change them manually, but that is not recommended. Press "Install Now!" and confirm the installation settings.

Open the zip-File and double-click on the containing setup executable-file.

The unpacking process may take a while. Once it's over, you'll see a Windows notification asking "Do you want to allow this app to make changes to your device?" Please select YES.

Please click on the "I agree" button if you agree to the "License and Legal Information".

If this is a first-time installation, the license key will be requested at this point. First, copy the license key from the mail you received after purchasing the software. If you can't find your license key, please make use of our License-Key Recovery Tool.

Select the license key and press the key combination "Control" + C, so that you can copy it. To paste the license key into the input field of the installer, press the key combination "Control" + V.

Choose the correct plugin paths for VST, VST3 and AAX plugin formats. If you are not sure where the plugin directory of your DAW software is located, please look into the manual, or use the settings dialog of your DAW. Please make sure that the host is using the same architecture (32bit/64bit) as the plugin.

Download the installation package and double-click on the containing PKG File, then click on:

"Continue" -> "Install"

The custom de-installation process starts afterwards.

Accept the license agreement and click on the "Uninstall"-button at the top right corner. To finalize the process, click the "Uninstall Now!" button.

Open the setup-application, accept the license agreement and select "Uninstall" on the top right. To finalize the process, click the "Uninstall Now!" button.

The setup uninstall routine will remove all major files (plugin binaries and content library) from your computer. However, a few very small files, which store configuration properties like the license key, will reside on the machine. If you want to get rid of these files, you can find them in these directories:

/Users/username/Music/Audio Music Apps/SIR Audio Tools

%APPDATA%\SIR Audio Tools

There are three modes in StandardGATE.

Most of the controls are the same in all modes, however in some modes some controls change their purpose or aren't used. In this manual, assume that all controls are universal across the three modes unless otherwise stated.

This is a gate or expander. Sound below the threshold is cut or reduced. This is most commonly used to reduce background noise.

This is a ducker. Duckers would mostly use an external sidechain input so that one sound source ducks the volume of another when it is present. The most common use is on the radio where the DJs voice ducks any music so that the DJs voice can be clearly heard. That's a shame really because usually the music on the radio is more interesting than any histrionic DJ!

Duckers act in the opposite way to a sidechain gate. With a sidechain gate, when the signal is present in the sidechain, the sound passes; with a sidechain ducker, when the signal is present in the sidechain, the sound is attenuated.

Upward is an upward expander. Like a gate this increases dynamic range and has a very similar audible outcome to gate on lower ratios. This is an alternative way to reduce the relative volume of noise or spill. Desired sound above the threshold is amplified, leaving less desirable noise/spill relatively quieter. Amplifying peaks would increase the output, possibly causing distortion, so StandardGATE automatically compensates by reducing the output gain when using upward mode.

The main display shows the audio signal in real time as a scrolling waveform. The input signal is the darker shade. The output signal is the brighter shade.

On the left-hand side is a graph of input (x-axis) against output (y-axis). The current level is illuminated as a circle on the ratio line.

Three of the controls of the gate are controllable on the display by dragging the sliders: threshold, tolerance, and range.

"Det." or detection shows the sidechain detection signal as an extra line draw on the display. This could be useful to see how EQ on the sidechain is affecting the gate triggering, or if an external sidechain is being used to check that it's working as intended.

"Red." or reduction shows the amount of gain reduction as an extra line drawn on the display. In upward mode, it actually shows amplification because gain is increased.

The scroll speed can be altered between "slow", "medium" (default), "fast" and "fast+".

The display defaults to logarithmic ("Log"), reflecting the decibel scale which is most useful for most applications. However, if a linear display is required "lin" can be selected.

Clicking on "Zoom" allows you to adjust the range of the display can be selected. The default is -60dB which would cover most recordings. There are two sliders: the bottom slider moves the bottom of the range; the top slider moves the top of the range. Click away from the slider to remove it when finished.

When the threshold level is exceeded, the gate opens, letting sound through unchanged. Signal that falls below the threshold is reduced in volume by the gate. If the threshold is set to maximum, 0dB, no sound will be audible; all sound will be cut out. Reduce the threshold to let more sound through ungated yielding a more natural sound. Increase the threshold to cut out more of the signal to remove more background noise.

When signal exceeds the threshold in the sidechain, the ducker reduces the volume of the sound being processed. Lower thresholds give more sensitive ducking, i.e. quieter sidechain signals would trigger ducking.

When the signal exceeds the threshold, gain is increased. Signal below the threshold remains the same volume. This control is used to find the desired signal that needs to be amplified.

Usually, the gate open and close threshold would be the same. With some recordings, e.g. some slowly decaying sounds like a cymbal or reverb tail, it's possible that the signal could temporarily fall below the threshold and is cut, then momentarily rise again opening the gate again causing an unmusical chattering gate on the tail. The tolerance control is a hysteresis control which enables the gate to close at a lower volume than it opens. By increasing the tolerance, unmusical chattering gating can be eliminated.

There is no tolerance control for the ducker or upward modes.

The range controls how much quieter the output sound is than the input sound when the gate is fully shut. Setting the range to -120dB essentially makes the gate a true noise gate and gives silence when the gate is shut. Having the range set closer to 0dB makes the gating have less effect so would yield a less "choppy" output sound.

The range controls how much quieter the output sound is than the input sound during ducking. Setting the range to -120dB essentially mutes the signal when the ducking sidechain signal is present. Having the range set closer to 0dB makes the ducker have less effect so the sound can still be heard when the sidechain signal is causing ducking.

The range controls how much louder the output sound is than the input sound when full activated. High ranges yield more amplification so more expansion of dynamic range. When the range control is moved, the output gain is automatically reduced to compensate for the overall increase in peak volume to avoid distortion and so that peak volume is maintained for a fair comparison when adjusting the range control. When adjusting the range, you won't hear the peaks get louder; you'll hear the sound below the threshold get quieter.

The sidechain is the signal that controls the gate.

For most gating applications, e.g. noise removal, the sidechain signal would be the same as the audio signal that is being processed.

It's possible to use an external signal to trigger the gate. For example, the gate could be on a synth pad sound; then if a kick drum was the trigger for the gate, the synth pad would only play when the kick drum was played.

When using ducking mode, the sidechain must always be set to external.

The sidechain signal, either external or internal can be EQed. This would most commonly be done if the internal signal is being used to isolate the instrument that needs gating, e.g. if a kick drum mic was being gated, EQ could be used to remove mid and high frequencies so that only the low frequency content of the kick drum triggered the gate instead of the snare and other drums.

By clicking on the headphones signal, the sidechain signal can be monitored. This would be useful so that you can hear how the EQ is affecting the sidechain signal and whether you are correctly isolating the desired instruments with the EQ settings.

There is a four band EQ, the default filter types are described below and will cover most applications:

This removes low frequencies of the sidechain signal. When clicking on the red dot, the controls for the HPF appear. The cutoff frequency can be adjusted by dragging the red dot or rotating the knob. The steepness of the slope can be selected between 6dB/oct, 12dB/oct, 18dB/oct, 24dB/oct, or "more" options. The higher the dB/oct, the steeper the slope so this increases the ability to isolate frequencies. The HPF can be disabled by selecting "off".

There are two parametric EQs which would mostly be used for controlling the mid-frequencies of the sidechain signal. Clicking on the yellow or green dots reveals the controls for these parametric EQs. The centre frequency and gain can be adjusted by dragging the dots or rotating the two knobs. The Q is the bandwidth of the peak. Higher Q values narrow the bandwidth so provide more focussed isolation of frequencies. Having the Q set to maximum can isolate particular frequencies.

This controls the amplitude of high frequencies of the sidechain signal. When clicking on the blue dot, the controls for the high shelf EQ appear. The cutoff frequency and gain can be adjusted by dragging the dots or rotating the two knobs. The steepness of the slope can be selected between 6dB/oct, 12dB/oct, 18dB/oct, 24dB/oct, or "more" options. The higher the dB/oct, the steeper the slope so this increases the ability to isolate frequencies. The high shelf EQ can be disabled by selecting "off".

For each of the four EQs described above, by clicking on the filter type icon (visible with the controls when hovering over the coloured dot) a menu will pop up and it is possible to change the filter type from a selection of low shelf, band, high shelf, band stop, high pass, low pass, band pass and all pass.

The real-time spectrum analyser shows the frequency content of the sidechain signal. Command clicking or right clicking accesses a menu that lets you select either two modes, FFT or bandpass. It can also be disabled.

The input gain can be controlled using the "In Gain" slider on the left-hand side. The default position is 0dB. This would rarely need adjusting as long as the signal entering the plug-in is at a sensible level. The processed signal gain can be controlled using the "Wet Gain" slider on the right-hand side. The default position is 0dB. This would not normally need adjusting. However, when using the StandardGATE as an upward expander, this signal is automatically reduced to avoid the overall amplitude increasing.

The attack time is how long it takes for the gate to open after the sound exceeds the threshold. Long attack times cause the sound to fade in slowly after the sound has exceeded the threshold. By having the attack time too long, vital transients could be removed from percussive instruments, so the attack time should be reduced. If the attack is too short, i.e. less than 1ms, then undesirable clicks could occur because the waveform is cut mid-cycle without a short fade. This especially can be a problem when you are using an external sidechain. Therefore, StandardGATE uses an internal minimum attack time of 0.2ms when using an external sidechain.

Technically, for this plug-in, the attack is the time taken for the signal to reach 63% of the possible range.

The attack time is how long it takes for the ducker to reduce the volume of the sound after the sidechain signal exceeds the threshold. Longer attack times cause the sound to fade away more slowly after the sidechain signal has exceeded the threshold giving a smoother transition.

The attack time is how long it takes for the gate to apply gain after the sound exceeds the threshold. Long attack times cause the amplification to fade in slowly after the sound has exceeded the threshold. By having the attack time too long, vital transients would not be amplified sufficiently so percussive instruments could lack bite, so the attack time should be reduced. If the attack is too short, i.e. less than 1ms, then undesirable clicks could occur because the waveform is amplified mid-cycle without a short fade (though these would be less noticeable than when gating because the ratio is lower).

The hold time is how long the gate stays open with no change in volume when the sound falls below the threshold. After the hold time has elapsed, the release phase begins. Longer hold times can reduce chattering gating on tails. More commonly, hold times are short with the closing of the gate controlled by the release.

The hold time is how long the ducker stays ducked when the sidechain signal falls back below the threshold. Longer hold times will leave the sound ducked for longer.

"Always hold while detection higher than threshold"-Option:

When this switch is on, as long as the detection value is higher than the threshold, the reduction is increased or held but not decreased. In this case, the hold value describes how long the reduction is held even if the detection value is lower than the threshold.

When the switch is off, the reduction is held regardless of threshold for the specified time period.

If the algorithm should work more like a traditional limiter or compressor, then this switch must be turned off.

If this feature is not visible, check if you have installed the latest version.

The hold time is how long the upward expander stays with the additional gain when the sound falls back below the threshold. After the hold time has elapsed, the release phase begins. More commonly, hold times are short with the returning to unity gain controlled by the release.

The release time is how long it takes for the gate to shut after the sound falls below the threshold. Long release times cause the sound to fade out more slowly when the sound has fallen below the threshold. By having long release times, longer tails of instruments such as cymbals and piano wouldn't be cut abruptly. However, long release times would leave more low-level noise or spill present. If the release is too short, i.e. 1ms, then undesirable clicks could occur because the waveform is cut mid-cycle without a short fade.

Technically, for this plug-in, the release is the time taken for the signal to fall to 37% of the possible range.

The release time is how long it takes for the ducked sound to return to the original volume after the sidechain signal falls back below the threshold. Longer release times will allow the ducked sound to fade back in more smoothly.

The release time is how long it takes for the amplified sound to return to the original volume after sound falls back below the threshold. Longer release times would leave peaks amplified for longer, letting them fade more naturally.

Detector is the knob just to the right of release; it's not actually labelled "detector". This gives the method used to detect the level of the input signal used by the gate threshold. The default setting is "Peak" which is most commonly used.

With peak detection, the gate threshold looks at the signal at a given instant. Even brief signals would trigger the gate to open. For most applications, using peak would suffice. This peak detector has a special adaptation so signals with low frequency content are detected reliably.

Using RMS instead of peak causes the input volume to be averaged over the length of time specified by the knob; the gate threshold is looking at an average volume instead of the volume at that instant.

Peak dB values are greater than average RMS dB values of the same signal. Therefore, when using RMS settings, a lower threshold is needed to give an equivalent gate threshold, i.e. reduce the threshold if using RMS settings for similar removal of low level signals.

Using RMS slows the response time of the gate so the gate could delay opening, cutting the attack portions of an instrument; transients could be lost.

There are two RMS settings:

RMS Window treats all signal within the length specified by the knob with the same importance.

RMS Decay applies different importance to the signal according to its age; older signal has less importance. RMS Decay provides a smoother sounding gate with slower response times than RMS Window.

This is a clever detector just for detecting transients. This can be used to extract transients from the signal. This detector is able to find the transients even if all of the transients are different volumes. For example, with an acoustic guitar signal, even if the strums are different volumes, the gate will still be triggered only by the transient of each strum. Also, the gate won't stay open too long on louder peaks.

This detector could be used to ensure that any quieter peaks are still caught by the gate.

If it's just the transients that you want isolated, use short attack, hold and release times. These could then be mixed with the dry signal, to create a very punchy signal.

This detector could be used for other creative purposes such as thinning out a rhythmic part if longer release times are used.

This detector is a blend of peak and transient for quick and reliable detection of percussive signals.

This is a precise peak detector, similar to "peak" but with the special bass adaptation not present.

For normal gating applications, the ratio would be set to the maximum of 100:1 for cutting out low level noise completely. Any sound below the threshold is removed. By decreasing the ratio, StandardGATE becomes an expander, so sound below the threshold is reduced in gain rather than eliminated completely; the quieter sounds are now quieter. Low ratios will have less effect on the sound below the threshold; higher ratios will have a greater effect and will make quieter sounds more quiet. This increases dynamic range.

The ratio is the ratio between input and output below the threshold. Above the threshold the input = output so that is a ratio of 1:1. Below the threshold, if the ratio is 2:1 then the output signal will have half of its original gain, so for every -1dB of input below the threshold, the output would be -2dB.

The ratio governs how much quieter the ducked signal will become. With a ratio of 1:1, there will be no ducking. With higher ratios, the amount of ducking will increase.

The ratio governs how much louder the amplified signal will become for sound above the threshold. With a ratio of 1:1, there will be no change in gain. With higher ratios, the amount of upward expansion will increase. Higher ratios increase dynamic range.

The knee control smooths the transition around the threshold between gating and not gating. For knee settings greater than 0dB, there is some gain reduction above the threshold and gain reduction immediately below the threshold would be softened giving a smoother transition between gated sound and ungated sound.

This is the parameter that makes the most subtle audible difference to the processor. For most gating applications, the knee could be set to 0dB; the ratio, attack, hold and release make more of an audible difference.

A filter can be applied to the attack, hold and release phase of the gate, though it's most noticeable and useable in the release phase (as described in the individual controls below). This is most useful for further removal of spill, allowing more musical lower thresholds or longer release times, so the desired signal is less affected by gain reduction. There are different filter types for different types of spill that need eliminating. The low-pass filter or de-bleed are probably the most commonly used.

The filter is applied to the ducked sound. If a very low ratio, e.g. 1.2:1, is used then this means that the amplitude isn't reduced that much at all, then the filtering can be the predominantly audible effect. The ducked sound appears filtered rather than attenuated. High % values are necessary to achieve the filter having more of an effect than the attenuation, over 100%. The low pass filter would be most useful for setting the ducked sound back in the mix (without affecting it's volume too much) when a lead instrument needs to be more prominent.

There is no filter control for the upward expander. The individual controls below apply for gate only, but the same principals apply for the ducker.

There is no filtering applied to the release phase. The % control has no effect.

This is a high shelf EQ where the gain and cutoff frequency of the shelf decrease through the release phase. This could be useful on a snare microphone to help filter out the hi-hat.

This is a low pass filter where the cutoff frequency decreases through the release phase. In practice, this sounds very similar to "de-bleed" and would be used in similar applications to remove high frequency spill/noise.

There is a resonant option where the cutoff frequency is emphasised. This is useful to the engineer so that they can clearly hear the effect of the filtering so that the release and threshold can be fine-tuned more easily, then the engineer can revert to the non-resonant low pass filter. Alternatively, the resonant option could be used creatively for rhythmic filtering effects.

This is a high pass filter where the cutoff frequency increases through the release phase. This would be useful to remove low frequency spill/noise, for example a kick drum on a snare microphone.

There is a resonant option where the cutoff frequency is emphasised. This is useful to the engineer so that they can clearly hear the effect of the filtering so that the release and threshold can be fine-tuned more easily, then the engineer can revert to the non-resonant high pass filter. Alternatively, the resonant option could be used creatively for filtering effects.

This enables the gate to act before the signal passes above the threshold, allowing some sound to be let through before any peaks occur. This helps preserve transients when using longer attack times. Longer attack times are sometimes useful to prevent clicks and other artefacts, especially with low frequency sound sources like bass guitar.

Some DAWs take some time to get back in sync when the latency of a plugin has changed. For others, sound playback must be stopped and restarted. With some DAWs the Plugin-Delay-Compensation must be activated first.

Although DAWs should compensate for plug-in latency, lookahead should be used with care in live situations because when switched on, the latency cause a musically noticeable delay.

The maximum look-ahead time can be chosen between 0..10ms/100ms/1000ms (*), which also defines the used plugin-latency.

* Please note, the maximum look-ahead time of 1000ms is not supported when using the AAX plugin format for compatibility reasons.

When using stereo signals, the gate can be triggered by different aspects of the stereo signal. The most commonly used setting would be "All" which is the default. If you decide that you need to use specialist settings other than "all" then it would be sensible to set the detection and process settings the same, or for most signals leaving detection on "all" would be most straightforward for predictable results.

For most gating applications, "All" would suffice. This uses the maximum of the left and right channel to trigger the gate. This would be the most commonly used setting.

Mid uses the mono-information of a stereo signal (left+right), ignoring anything in the sides (left-right) to trigger the gate.

Side uses the stereo information (left-right) of the signal, ignoring the mono information to trigger the gate.

Left and right use only one of either the left or right channels to trigger the gate.

When using stereo signals, the gate can be assigned to process one aspect of the stereo signal. The most commonly used setting would be "All" which is the default.

If you decide that you need to use specialist settings other than "all" then it would be sensible to set the detection and process settings the same, or for most signals leaving detection on "all" would be most straightforward for predictable results.

For most gating applications, "All" would suffice. The signal is treated as one stereo signal and gating is applied to both left and right channels identically.

Anything present in the mono-part of the signal (left+right) and not in the sides (left-right) will be gated. When signal is below the threshold, this effectively removes any mono-information from the signal, just leaving the stereo-information (with inverse polarity on each opposite channel).

Side signal below the threshold is gated, yielding a mono output (just the mid remaining) when the sound falls below the threshold. For example, in a stereo drum overhead, loud snare drums triggering the gate will appear big and stereo, and all other elements would appear mono.

This setting could be used creatively to make louder sounds sound bigger and wider.

Left and right use only one of either the left or right channels to trigger the gate.

MIDI can be used to trigger StandardGATE. In your DAW, configure the plug-in so that it accepts a MIDI signal. This would be different in different DAWs. For example:

• In Cubase, you put StandardGATE on an audio track, then route the output of the MIDI track-out to the plugin.
• In Logic, create an instrument track and insert Standard Gate. Then use the sidechain to route the audio from a different track to StandardGATE.

When using MIDI to trigger the gate, interesting creative rhythms with a high level of MIDI control can be created on sustained signals, adding layers of rhythm to a mix.

The below controls are described for the gate and then upward. In either gate or upward mode, the audible effect is similar. MIDI can be used to control the ducker in the same way as the gate.

For gate, this is the most intuitive setting, so would be most commonly used. With this setting, if a MIDI note is present then the gate is open.

For upward, again this is the most intuitive setting. If a MIDI note is present, then the amplitude is boosted by the range value, sounding similar to the gate.

If MIDI triggering only is required (ignoring the threshold triggering), then select "MIDI only" (see below).

For gate, this setting has the opposite effect to "note on opens". With this setting, if a MIDI note is present then the gate is closed.

For upward, the amplitude is boosted when there is not a MIDI note present yielding a similar audible effect to the gate.

For both gate and upward, this setting works in an identical way to "note on opens" but with the added bonus that MIDI note velocity affects the gain. The higher the MIDI note velocity, the louder the gain of the sound when the MIDI note is present. MIDI note velocity of 127 would give the same results as "note on opens".

For gate, this setting works in an identical way to "note on closes" but with MIDI note velocity affecting the gain. The higher the MIDI note velocity, the lower the gain. MIDI note velocity of 127 would give the same results as "note on closes gate". Lower MIDI note velocities would result in less gain reduction when the gate is closing, letting more sound through.

Upward, has a similar audible effect to the gate with lower note velocities resulting in more sound present when a note is present.

The audio detector is disabled and only MIDI will trigger the gate. This setting would probably be the most useful for full MIDI control.

MIDI works together with the threshold control to trigger the gate. For example, in "note on opens" mode, if there is a MIDI note or signal is above the threshold, then the gate will open.

This controls the balance between the input and output signal. For most applications, this would be set to 100% wet. If more of the input signal is required, it would be better to use the ratio or range control to introduce less gain reduction.

Next to the dry/wet fader, is a polarity button. This button inverts the polarity of the wet signal. When this mode is active, the dry signal is not affected by the Dry/Wet slider; only the level of the polarity inverted wet signal is adjusted, which is added to the dry signal. This can create some interesting "reverse" effects: the gate could be used to separate out the attack portions of the sound, then these can be subtracted from the dry signal yielding a fading in of each instrumental hit.

To open the advanced settings, click on the oversampling-button -> Expert settings, or use the Preferences button on the top -> Oversampling Configuration

The filter is an important part of the up and down-sampling process. You can choose between two filter-types:

• linear-phase (adds latency to the signal, but doesn’t induce phase shifting to the signal. This type is preferred for most applications (including mastering)
• use minimum-phase filter (adds no latency to the signal, but induces phase-shifting to the signal) for live usage

The lower the filter-cut off the less harmonics above Nyquist-frequency will cause aliasing, but the more effect on the signal the filter has.

Generally speaking the steeper the filter is, the better the oversampling result will be. But keep in mind that bigger filter-kernels also cause more latency, CPU consumption and potential pre-ringing artifacts in the transitional region. On the other hand, bigger filter-kernels have a smaller transition region.

The higher the order, the steeper the oversampling filter, but also the higher the CPU-consumption. Also there will be more phase-shifting in the transitional region.

This button resets all oversampling related settings to their default value.

This tool allows you to compare the effect of different settings quickly with just one button press. When you have the settings you like, press A=B. Then make some changes to see if you can improve on these settings. Then you can flip between the original settings and the new settings by pressing A/B. When you have decided which is better, press A=B again. You can repeat this process as many times as you like to fine tune parameter settings.

This is a useful tool to check if any changes you've made actually improve the sound and not made it worse! Remember that endless tinkering usually leads to tired ears and an awful mix!

Adjust the whole plugin-GUI size.

Special thanks to all beta-testers, to Lloyd Russell for having helped me with the manual and to Alexander Alexeev for additional support.