TC Electronic Finalizer

The processor that revolutionised home mastering is at last available in software form.

TC Electronic first introduced digital multiband compression in their M5000 hardware back in 1994, but the concept really took off with the launch of their Finalizer in 1996: the music industry would never sound the same again! This was a powerful tool, but its relatively accessible pricing, combined with the automatic loudness compensation and the lack of an overview of what really was happening to the music, made it all too easy to overdo the processing. So it’s fair to say it’s a processor that divided opinion.

Nearly a quarter of a century, and a DAW revolution, later, TC have released a standalone software version of the Finalizer. But while the algorithms have been ported from the company’s System 6000 hardware into this software, this is not a simple clone of the old hardware. Indeed, the manufacturers suggest the algorithms have been ‘enhanced’, and the software’s new one‑window interface is a lot more useful, not least because it does so much to address the aforementioned lack of overview.

On the left of the GUI, you’ll find TC’s brand new Spectro Lab metering section, with full track analysis in the form of a Spectral Dynamic Contour and Average Spectral Curve, as well as Real Time Spectrum metering; more on this later. The middle shows the chain of processing modules you’ve selected, along with their settings. To the right is a module library, and the right‑most section hosts the monitoring section and output metering. The bottom of the screen shows the waveform of the processed audio or chosen reference track. One window, many possibilities.

Preparation Is Everything

Today, the workflow for many musicians and laptop producers involves having some mastering plug‑ins on the stereo bus, to help render a ‘finished’ sound. Having to move mixes into the Finalizer software could be seen as a limitation, but I’d argue that it’s a better approach: switching software and being limited to performing processing on the stereo mix forces you to take off your mixing cap and look through the mastering monocle. It shifts your focus from doing small changes in the mix to considering the whole song as a single entity, while comparing the processing result with reference tracks.

The curves show the spectrum and dynamics of the unprocessed audio. The green wave fields show the processed audio.When importing a stereo mix, the software automatically analyses the loudness, True Peak maximum level, and peak‑to‑loudness ratio. It then performs loudness normalisation. The reference loudness level can be set to ‑12, ‑16 or ‑20 LUFS, so no matter how loud or soft the mix measures, it will be set to approximately the same level as all other tracks imported into the software. It’s also possible to load up to 20 reference tracks that will automatically be loudness matched with the processed audio, making A/B comparison a breeze. Finalizer will perform sample‑rate conversion if the file sample rate differs from the project setting; Finalizer supports sample rates up to 192kHz, but the internal processing sample rates are 44.1 or 48 kHz, with 64‑bit floating‑point precision.

Spectro Lab

TC have developed a new metering system called Spectral Dynamic Contour (SDC). This shows the spectral content and energy distribution of the whole track, with a 1/6‑octave resolution, by displaying six curves. The top‑most curve shows the peak level of the whole track, at different frequencies. The third curve down shows when the level reaches 83 percent, the fourth shows 50 percent of the level and so forth. The closer together the curves are in a specific frequency range, the less dynamic the content is. Depending on the style of music and level of compression, the ‘perceived’ spectral balance is somewhere between the third and fourth curve.

The average energy for each frequency, calculated across a full track. The blue curves shows the energy of the unprocessed audio, while the green one shows the processed audio. The third, purple line shows a reference track for comparison.What makes the SDC different from a standard spectrum analysis is that it’s offline, so the curves are static, which means the energy content is displayed in a more comprehensible way. It’s easy to switch between metering of the unprocessed audio, the processed audio, and an automatically loudness‑matched reference track, and to see the differences in spectral balance. An interesting aspect of this SDC metering is that you can easily see the effect of any dynamics processing, such as limiting or multiband compression; that’s not so easily achieved with traditional spectrum analysers.

There are two other metering options. First, a Real Time Spectrum (RTS) meter displays a real‑time spectrum analysis behind the SDC curves. The two moving curves shows a peak and an RMS spectrum analysis during playback, which is a handy visual aid when trying to identify instantaneous peaks and bursts of energy in the source material. The third meter, called Average Spectral Curve (AVG), shows the average energy for each frequency, calculated across the full track. The audio is loudness‑normalised during the analysis, making it easy to compare the spectral balance of different tracks.

I like the new metering. Indeed, during the test period I found it highly informative and especially useful for comparing the low‑end content of my in‑progress masters with reference tracks. I also found myself experimenting with broad shelving filters to match both the low‑end and the midrange/treble spectrum balance more closely with my reference tracks. The final decision should always be based on what you hear, of course, but, that said, the metering can be helpful when deciding on the spectral balance and when identifying frequency areas that require more attention.

It’s easy to switch between metering of the unprocessed audio, the processed audio, and an automatically loudness‑matched reference track.

The Processing Modules

Just like the hardware equivalent, the processing is based on modules that can be arranged in any order in the signal‑path. Your main options are: five‑band EQ, Dynamic EQ, Stereo Width, Compressor, three‑band Legacy Compressor, five‑band Modern Compressor, Master Limiter and Loudness Limiter. There are also one‑band filters, which make it easy to try out different filter settings by just adding or bypassing one module. Clicking on an active module will reveal its parameters. The interface is simple and unfussy, with just sliders, numbers and compression metering; those who prefer to rest their eyes on fancy front plates of vintage compressors or EQs will have to look elsewhere!

The multiband compression is impressive, which is no surprise given that the algorithms’ origins lie in the System 6000. The three‑band Legacy Compressor is based on TC’s MD3 algorithm, but the five‑band Modern Compressor is new; according to TC it’s based on the MD4 but with refined dynamic precision behaviour and linear‑phase crossover filters. Both perform well and sound excellent, taking control of the music without putting a lid on things or changing the spectral balance. It’s basically impossible to make them sound bad, as the automatic gain control, dynamic behaviour and pass‑band filters work so well together.

The Dynamic EQ is a one‑band dynamic equaliser, and it is excellent for taming a loose and boomy kick drum in a narrow bass frequency range, or providing sibilance control at 6‑10 kHz. Actually, it can be a good idea to have two of these dynamic EQs standing ready in your default chain, precisely for those two applications.

The static five‑band EQ and the one‑band filters are just what you expect of TC Electronic: super clean. They won’t add any colour to the sound, just change the spectral balance and notch out problematic frequencies, without affecting the transient response. I was able to add both low‑end weight and high‑end sheen in a couple of mastering projects in a very satisfying way.

Limiting & Soft Clipping

Two limiters are included: Master Limiter and Loudness Limiter, and both are based on the True‑Peak BW2 algorithm from the System 6000. The Master Limiter works as a traditional brick‑wall limiter, with limiter gain and maximum output level controls, and there are three different limiter profiles intended for different source material: Electric, Acoustic and Classical. The Loudness Limiter is rather nifty: the user can set a target loudness level (say, ‑14 LUFS for publishing on YouTube, Tidal or Amazon Music), and the limiter will then analyse the whole track and apply the appropriate amount of limiting to take it to the target loudness.

Just like the multiband compressors, it’s practically impossible to get a bad result out of these limiters, even with hard brick‑wall limiting: they handle the low end without clipping it and shave off midrange transients without smashing them. If you are not aiming for a dynamic range of less than DR6, the Finalizer software will be able to handle the job brilliantly.

Both limiters have a soft clipper before the limiting takes place. I haven’t previously been a big fan of TC’s SoftClipper, because it has just never sounded very musical to me. Happily, then, something must have been updated in the new Finalizer software because now it sounds really good! Adding just a touch of smooth soft clipping holds the sound stage together a little better, and allows you to ease off the master limiting a bit too. It’s well worth experimenting with.

Final Words

Really, the intended users of this software are musicians and laptop producers in need of easy‑to‑use and great‑sounding software to help them put the final touch on their own music. With that market in mind, Finalizer makes perfect sense: it’s easy to use, the automatic loudness compensation helps the user stay focused on mastering the music without being seduced by the change in loudness, and, most importantly, the multiband compression, EQ and limiters all sound great.

Mastering engineers, on the other hand, will soon come up against its limitations: the processing is limited to 44.1 or 48 kHz; there are currently no Mid‑Sides processing options except for the Stereo Width module; there’s no support for external hardware or third‑party plug‑ins; it doesn’t allow automation; and there are no real editing possibilities except for trimming the start and end point of the file. All of which might make you wonder where the plug‑in version of these great‑sounding modules is. Well, there aren’t any. Or rather, not of all of them. TC Electronic have two separate mastering plug‑ins that are available for purchase: the Brickwall HD Native and the Master X HD. The latter is a three‑band dynamics processor much like the Finalizer’s three‑band compressor module, but with a large metering window. There are optional hardware controllers for both plug‑ins too.

While no software can replace the craftsmanship of a skilled mastering engineer, if your budget is limited there is software available with AI functionality, and automated online mastering services that provide a ‘one‑click’ solution for mastering. So should the final touch of your music be done by an automated process, or should you spend another half hour doing the mastering yourself? With the Finalizer software and a couple of good reference tracks, I recommend that you try doing it yourself. Not only does the sound quality of the processors and the automatic loudness compensation make it possible to get decent results but, just as important, you’ll evaluate and improve your mixing skills while doing it. I highly recommend downloading the 14‑day demo version; all that is needed is an iLok account.