Sonnox Oxford Limiter Operation Manual Version 1.
1. Introduction The Oxford Limiter plug-in has been developed from decades of professional audio experience to provide a very high degree of quality and capability in programme loudness control and limiting functions.
2. Operation 2.
2.2 Pre-Process Section The Pre-Process section provides a musical programme levelling function. Its primary purpose is to control programme level over a wide range, in order to provide optimum conditions for the following enhancement stage. When the ENHANCE control is disabled in Normal Mode (at 0% with SAFE MODE deselected), the PreProcess section can be used as a conventional levelling section in its own right.
2.3 Enhancement Section The purpose of the enhancement process is to provide sample value limiting and overall programme loudness improvement. The process follows the Pre-Process section in the signal path, and is controlled by a separate ENHANCE fader from 0% (no action) to 125% (maximum action). In Normal Mode, the range from 0% to 100% fades in the effect to full level, at which complete sample value limiting occurs.
Also, since the plug-in has internal headroom, transient levels greater than notional maximum modulation can pass from the limiter section into the enhancement stage.
2.4 Attack Timing The addition of an ATTACK timing control is a significant departure from conventional limiter applications, and requires some explanation for the best results. Because the level detection within the plug-in looks ahead of the gain control, peaks in the programme material are acted upon in advance of the gain reduction process. Therefore, at the fastest setting of the attack control, programme peaks are controlled within a very small margin (less than +0.
In general, very fast attack times will more readily remove extremely fine detail and shortterm events, but will produce greater harmonic disturbance. Slower attack times will progressively allow finer detail to escape the harsh sound of fast limiting, and longer term events will tend to assume a more rounded peak profile. Such settings are usually kinder to the musical programme.
2.5 Release Timing The RELEASE control has a very wide range to accommodate the maximum possible extent of programme and production style. The ability to set very fast release times is provided specifically to allow for the modelling of short-term peaks over restricted gain reduction ranges, up to a maximum of around 4dB. Such settings will result in high levels of distortion for larger gain reduction ranges, and are therefore unsuitable for overall level control situations.
If the AUTO GAIN function is not selected moderate release time settings (above around 0.2 seconds) may produce audible gain ‘pumping’ due to longer and more noticeable recovery periods. If such settings are needed the aim becomes one of ‘fitting’ the release timing to the natural rhythm of events in the programme.
2.6.1 Loudness Maximisation The aim of this procedure is to achieve an overall average increase in the level of the programme by reducing the size of short-term peaks, and applying extra gain to move the programme up into the extra range freed up by the removal of the peaks.
2.6.2 General Gain Management The aim of this procedure is usually to preserve the short-term dynamics of the programme as far as possible, whilst ensuring that no levels surpass maximum peak modulation. This most often entails responding to the peak level of the programme as quickly as possible, and rescaling the gain in the longer term, such that musical dynamics are only minimally affected in the short term.
2.7 Reconstruction Metering Section An important fact, which is often overlooked, is that in any discrete time sampling system, it is possible to create sample values that may not be decoded and reconstructed correctly.
Since the vast majority of metering within workstation environments responds to sample value only, the above example would show a level of around —3dB below clipping. However any further increase in the level of the signal would result in a potentially illegal output level from the system converter. As this error would not be reported on metering within the workstation, in this particular case a possible 3dB overload can result if the signal is increased to a maximum reading on the workstation meters.
The Recon Meter option allows you to correct for inter-sample peaks manually if you want to, by showing how much over unity your inter-sample peaks are. Reduce the output fader to compensate if you do not want to use the Auto-Comp feature. The Recon Meter in DAC simulation mode will not work correctly unless the samples are already limited to digital maximum.
2.7.1 Meter Operation When the RECON METER function is selected (see right), the meter is switched from conventional peak sample value mode into reconstruction mode. In this mode peak reconstruction levels will be displayed on the meter. Levels in the red overload range of the meter represent the presence of potential reconstruction errors, as illustrated below using the previous example: Two methods are provided to correct for this: 2.7.
When the AUTO COMP button is selected, the level of the output is automatically controlled to repair reconstruction errors by the minimum amount required, and only for the duration of the error. In this way the loudness of parts of the programme unaffected by the errors remains as high as possible.
The first graph shows the damaged spectral result of passing a low-level 1KHz sine wave through a 16-bit undithered truncation: The next graph shows the exact same signal and truncation to 16-bits, but with the HP TPDF dither applied: As it can be seen, all the harmonic errors have been removed. Also, since the FFT analysis method provides an enhanced view of the signal below the noise floor, it can also be seen that there is effectively no low level floor below which a signal will fail to pass.
The graph below shows a 1KHz signal at —120dBr passing through a dithered 16-bit system. This corresponds to a signal 24dB below the level of the least significant bit; the effective channel SNR (Signal to Noise Ratio) is added in blue for illustration purposes. This illustrates that dither turns a quantised numerical signal conduit into the equivalent of a naturally continuous (un-quantised) system, which exhibits a finite signal to noise ratio with no practical limit to harmonic signal resolution.
2.8.2 Noise Shaping Dither If for some reason SNR figures of 93dB at 16bits (or 143dB at 24bits) prove insufficient, noise shaping can provide an apparent increase in SNR, but there are some potentially hidden costs. Noise shaped dithering is a mechanism that aims to reduce the perceived loudness of the noise of a dithered signal by either forcing the spectrum of the noise out of the audible range or placing it into frequency ranges to which we are less sensitive.
This means the noise level necessarily increases in some ranges of the spectrum. A level increase anywhere in the spectrum must be accommodated by an increase in total peak noise level. The following graphs illustrate this in action, under the previous test conditions. The first graph above shows a sample value plot of the conventional TPDF dithered signal. The second graph shows the increase in values caused by the application of the TYPE 1 noise shaping at 100%.
The Oxford Limiter includes four noise-shaping curves: Types 1 and 3 are fifth order and Types 2 and 4 are third order designs, representing a varied set of trade-offs to suit most programme types, as illustrated below: Whilst it is understood that the selection of noise shaping type is largely a matter of user preference, generally speaking Types 1 and 2 produce the most dramatic reduction in overall noise loudness, with Type 1 being the most effective of all.
Generally speaking, high levels of noise shaping render a signal that is more fragile. Almost any change to the produced audio file after noise shaping could potentially result in unwanted effects. For these reasons the DEPTH selector is provided to put you in charge of the degree to which noise shaping is applied. When any of the noise shape curves are selected, the depth selector varies the degree of noise shaping from 0% to 100% in 10% steps.
Another important factor is that the effectiveness of psycho-acoustic noise shaping relies heavily on our sensitivity to noise spectra at the threshold of hearing. Therefore, if noise shaped dither actually gets to be heard directly, it will often sound quite strange and intrusive and may detract from the listener’s experience of the programme.
3. Description of Controls The plug-in’s user interface is divided into three main areas: an Input section, the Pre-Process section, and an Output section: Note that the parameter fields provide continuous feedback of settings values, and have typein fields so you can change values directly from your keyboard. Input Section INPUT GAIN fader — controls the input gain to the limiter processing from —18dB to +18dB.
INPUT GAIN display — displays effective gain at all times and allows settings to be typed in manually Pre-Process Section ATTACK fader — provides setting of attack timing from 0.05mS to 1mS (ref. 10dB gain change). RELEASE fader — provides adjustment of release timing from 0.05mS to 1.038 seconds (ref. 10dB gain change). SOFT KNEE fader — adjusts the soft limiting threshold from 0dB (hard limiting) to 10dB (maximum soft limiting).
DITHER TYPE button — selects the output dither type, which includes conventional TPDF and four different types of noise shaping options (see description in the Operation section of this manual for details). DEPTH button — selects the effective degree of action for any selected noise shaping option from 0% to 100%. OUTPUT meter — displays the plug-in’s final output level.
Options Menu Clicking the Sonnox button produces a drop-down options menu (see right). Clip Lights — these options determine the approximate time that an overload indicator will stay on for when the plug-in has detected a full-level sample at either its input or output. Enable Sonnox Toolbar — displays or hides the Sonnox Preset Manager Toolbar (see Section 4). Show Preset Name Path — shows the hierarchical directory path for Presets that are stored in sub-folders of the default Preset folder.
4. Additional information for Pro Tools AAX/HDX In preparing the Oxford Limiter for the Pro Tools AAX plug-in format, we have striven to provide the closest possible integration within the new Pro Tools framework, and where possible have made improvements to the original RTAS/TDM design. Dithering: Unlike previous TDM and RTAS versions, the AAX version has the new option of not dithering the output. This should be used whenever the plug-in is not inserted last in the master insert chain.
The AAX Limiter continues to exhibit exceptional noise shaping performance for the dithering and truncation, resulting in a noise floor that is in some places nearly 30dB below that of standard (TPDF) dithered/truncated methods. One improvement in this new version is the elimination of the tiny dc offset that was introduced by the truncation.
In True Peak mode, the true inter-sample peak value of any sample sequence is calculated. This means that if you use slower attack times and do not use Safe Mode, the Auto-Comp section when enabled will now catch and compress any over-range peaks escaping the main limiter section. For this specific case, this gives a different sound to the hard clipping that occurs in the DAC Simulation (RTAS-TDM) version. When Auto-Comp is on in True Peak mode, the clip LEDs will not light.
for 16-bit dither), and so the Pro Tools clip LEDs will not necessarily indicate an agreement with the clips on the Oxford Limiter plug-in. As for the 16-bit mode, since the 24-bit integer value has hit digital max, and the samples are destined to be written to 24-bit fixed precision media, the clip LEDs are drawn with an elevated danger level of Red with a ‘24’ indication in them to remind the user they have a particular meaning. 5. Suggested Workflows 5.
8. If you changed the Input level in stage 7, you must repeat stage 6 and 7 to check where the peak sample turns out to be. This is because changing the input level will have non-linear consequences to the output level. 9. Set dither mode to 16-bits (for CD) or 24-Bits (for a good DAC feed by a digital output), enable Auto-Comp, and rerun through your mix to check that the output does not clip. If it does, reduce the Output fader by a fraction, say 0.0008dB. 10.
3. Set dither mode to 16-bits (for CD) or 24-Bits (for a good DAC feed by a digital output). 4. Set Peak Hold On and Auto-Comp On. 5. You can now experiment with pushing the input level while listening to the sound. The more input you have, the more ‘compressed’ the perceived result, up to a point. You can also experiment with pushing the enhance fader to fatten your sound.
6. Run through your entire material. At the end, note how much over unity your peak value is. 7. You now have four methods to eliminate the clips, which can be used individually or in combination: Choice 1 - Change the output gain: Reduce the Output Fader by the same amount, enable Dithering with noise shaping (if needed), enable Auto-Comp in DAC Simulation mode, and replay your mix to check if you reduced enough, repeating until there are no clips.
6. Preset Manager Toolbar The Oxford Limiter plug-in comes equipped with its own onboard Preset Manager, which is displayed as a toolbar at the top of the plug-in window, just as if the host created it (see right). The reasoning behind this is to allow increased portability of your presets across all the host applications, while also providing a consistent and versatile interface.
7. Specifications 7.1 Instantiations per DSP Chip Pro Tools 48kHz 96kHz Mono Stereo Mono Stereo Mono Stereo 6 5 3 2 1 1 AAX 13 (no Recon) 9 6 4 3 2 Accel 4 2 2 1 1* 1* HD 1 1 1* 1* DSP TYPE AAX 192kHz * Due to processing load restrictions, the Reconstruction Meter and Auto Compensation options are not available. Native processing overhead will be dependent on CPU speed, the number of CPUs, and how much of the DSP algorithms are engaged by the settings used. 7.
Similarly, note that at 192kHz, the full mono version will not longer fit into a whole DSP chip, even at 150MHz; so the Auto-Comp and reconstruction meter functionality is disabled for the mono version at this sample rate. For Pro Tools AAX DSP, AAX Native and LE, VST and AU Native versions, there is no need for any restriction. 8. Copyright and Acknowledgements Trademarks and content copyright © 2007-Present Sonnox Ltd. All rights reserved. This Product is manufactured and supplied by Sonnox Ltd.
Platform Specific Supplement S1. Supported Platforms Avid Pro Tools (LE, RTAS, M-Powered, Pro Tools HD) and Pro Tools 10 (HDX & Native) VST Native Audio Units Native S2. System Requirements These requirements are current at this revision of manual. For latest system requirements, please see the website www.sonnox.com Pro Tools Pro Tools 7, 8, 9 & 10 Approved Pro Tools CPU, OS and hardware configuration (see www.avid.com) Mac OSX 10.4 or higher; OSX 10.6.
