CS-5501 V2 from TBProAudio
CS-5501 V2 is the most recent version of TBProAudio’s expansive console emulation that I intended to review last year, but pressure of business, places to go and people to see delayed my efforts. At first look I thought “Oh, just another console emulation . . .” but looking at the far right I noticed a column of 17 window boxes with various labels that are different from other “consoles” I have tested and used. Since it has never been reviewed in GearSpace, I’ll start from the beginning.







What is It?
Yes, CS-5501 is a console emulation which you can see above includes an EQ strip on the left, metering near the top centre, a compressor (at least in this view) is below the left meter, and input/output controls below the right meter. If that were all, it would be a simple EQ and compressor emulation, but those labelled boxes on the right edge hint at more. A lot more. Looking at the right column you see down-facing arrowheads below each box which, sure enough, indicates the signal path through all those boxes is from top to bottom. In the screenshot above you see six of the boxes have indicator lights lit, from the top Mt. In (Meter In), EQ1, Comp1, Sat1, Noise2, and Limit1, indicating that these processors are active. You can see there are two of each processor type except the meter, and this leads to some fascinating possibilities since all the modules have series/parallel signal routing possible.

Can you move modules to a different location in the signal chain? Yes, easily done, but not by dragging them with the mouse. Click in any of the module boxes and a drop-down window appears that shows all the modules as seen below. Select any module from the list and it replaces the one already there and the module that was there pops to the old selected module location. The newly chosen module appears in the slot below the left meter (see Gate1 below).



When several modules are active a click on the small left-facing arrowhead under the list module will pop it up under the left meter replacing the previous module displayed there, but the previous module, if active, will continue to operate with the last settings used. While it might seen better to see the full panels of all active modules, that would soon fill your video monitor display!

The Modules: EQ
Starting with the EQ strip which is always shown on the left side of the main view even if you place it down the signal chain somewhere, there are five frequency bands plus a Low Cut and High Cut filter at the bottom. Different than some other EQ sections I’ve used, each of the five bands has the same frequency and Q (bandwidth) ranges: a very wide 10 Hz to 40 kHz and Q from 0.1 to 10. And each can cut or boost up to 18 dB. The highest and lowest bands can be either bell or shelf type filters while the other three are bell only. Each of these five bands can be individually monitored using the small M button below the EQ type label – this is a useful feature that lets you hear what any combination of frequency and Q is passing without the effects of boost or cut. The Low and High Cut filters have six slopes available (6, 12, 18, 24, 36, and 48 dB/oct) and the same wide 10 Hz to 40 kHz cutoff frequency range. In the very upper right corner of the module you can see small S and PM buttons that activate serial/parallel processing (S) and parallel mix control (PM). All modules have these controls which I’ll describe in more detail later.

Common to all the processor modules there is a CH. (Channel) MODE window near the top that allows setting what signal type is being processed, Stereo, Left, Right, Mid, or Sides. And just left of the CH MODE label is a button labelled S – this is a Solo button that will solo the module so that you can quickly hear what that module is contributing to the tonal or dynamic shaping of a sound. And if you hold the shift key (in Windows, at least) when clicking several solo buttons, you can solo multiple modules at once.

I found this EQ very precise if you like, unlike a 50 year old hardware console with noisy pots and component ageing. However, if you turn on “EQ component tolerances” in the Plugin menu (hamburger icon to the left of CS-5501 V2 label at top left) the values of resistors, capacitors and/or inductors in the EQ circuits “get real” and introduce small variations in EQ curves. And every instance of an EQ strip in every instance of CS-5501 using component tolerances ends up a little different, much like real hardware. Noisy pots are not emulated which is a blessing!

The filters are minimum phase designs based on an analogue configuration that minimizes amplitude cramping near the Nyquist frequency – this enables the filter to sound much “softer” than many digital filter designs. BTW, TBProAudio did not “modernize” the EQ with a linear phase option, which is fine with me since I find linear phase can cause more problems than it is purported to solve. And all the great classic rock and pop albums were made using minimum phase designs, so I rest my case.

The Modules: Metering
While the view at the top of this review shows traditional VU Meters, there are options for this section of CS-5501. In fact, four different analysis views are possible:



The composite view above shows the VU Meter view, the Oscilloscope view, Spectrum Plot and Peak Plot views. The VU Meters can display traditional left/right signal levels, left, right, mid or sides levels. In addition they can be set to monitor the input signal (input at the meter location in the signal chain, usually the first module), Gain Reduction for any dynamics processing, the output signal (at the very end of the processor chain), and an input/output mode that shows input level on the left meter and final output level on the right meter. There is a CAL knob that allows setting the VU zero level (Volume Units for those new to the craft) in terms of dB full scale, from 0 dBFS to -36 dBFS. The MON(itor) button switches what you hear from “center” to “in-place” when you select left, right, mid or sides modes. Center plays the selected source as a mono signal while in-place plays the selected source in its original location, on the left, on the right, a centred mid signal or out of phase sides signal. Very nice feature. There is also Smart Silence Processing (SSP) that will reduce CPU load (to practically nothing) when signal drops below a set value. This is excellent in a large project where, as is almost always the case, many tracks are silent for portions of the song. The meter scales can be set to PEAK (dB), RMS (dB), EBU ML (LU), EBU SL (LU) and the classical VU (dBVU). And this is just for the VU Meters!

The Oscilloscope view can monitor the input (orange), the sidechain signal (green, not shown above), the output (blue), and any two or all three of these at once. It can display summed stereo, left, right, mid or side channel. The time scale can be set either in milliseconds or tempo synced to the DAW as shown in the view above. In this view it displays the effect of six different EQ settings on a repeated section of a loop (using DAW automation).The effect of any CS-5501 processor on signal waveform can be easily viewed this way, but as always, your ears should be the final judge of any processing effects.

The Spectrum and the Peak Level plots can also display either summed stereo, left, right, mid or side channel, and show any one, two or three of the input signal, the side-chain signal and output signal. The Spectrum view has six “tilt” settings from 0 to 6 and includes a handy EQ plot (yellow-gold line in screenshot above) that shows the overall EQ curve of EQ1 and EQ2 if both are engaged. All plots, like the meter view, have the MON and SSP controls as described for the VU Meters. Certainly a comprehensive set of measurement tools. Far more than any classic hardware console.

More Modules: Compression and Friends
One primary processor in a console is the compressor and CS-5501 has two of them available as mentioned earlier. One is shown below on the left along with the Noise1 and Clip1 modules to the right. I’ve animated the compressor view to show the four possible compressor types without always covering the Ratio control – the other two modules have nothing hidden behind their pull-down windows so I saved bandwidth not animating them!



There are four different compressor “circuits” you can use as seen above which provide different actions (with the FET 1 mode modelling the classic 1176 most accurately) and of course there are controls for Threshold, Ratio, Attack, Release (including an auto-release) and a tiny metronome icon button below AUTO REL enables DAW tempo synced release times. This all yields a very comprehensive range of dynamic control.

You can see the compressor has an “active” light as do all modules, a Solo button and CH(annel) MODE window (to process stereo, left, right or mid-side) at the top. The next section down is sidechain control with LC (low cut) and HC (high cut) filters available with six filter slopes as described earlier (6, 12, 18, 24, 36, and 48 dB/oct) and 20 Hz to 20 kHz cutoff frequency range. There is also a window switch for internal or external sidechain signal (external uses audio channels 3 and 4 in DAWs like REAPER – other DAWs like Studio One have dedicated sidechain sends) and you can monitor the sidechain audio using the MON button. In all, a very capable set of compressors.

The Noise module shown above is simply a white-noise generator with four output levels as shown. I didn’t find this module terribly useful, but it’s there if you want to add a little noise within the console itself. It is intended to create electrical thermal noise (Johnson noise for you electronics fans) found in all electronics circuits. BTW, Dither 16 bit produces noise at about -97 dBFS rms and Dither 24 bit is way down at about -145 dBFS rms, which seems unnecessary! A future improvement might be to make the level variable. The Clip(per) module has only the Ceiling control and four Modes as shown. The modes affect how a signal is clipped/distorted when it exceeds the Ceiling level. I found a nice range of subtle to extreme crunchy distortion with this module.

Another Trio of Processor Modules
What console could call itself a console without a gate or two? CS-5501 has two gates, as well as two DeEssers and Two Limiter modules.



The composite view above shows these three processors and you can see the Gate and De-Esser have the CH. MODE and sidechain section like the compressor. A sidechain with a gate can provide some fun pumping as well as useful ducking, and if you look to the left of the Gate THRESHOLD control you see a slide switch with GATE and EXP(ANDER) choices. The difference is shown below:


Transfer function with input signal level on the lower x-axis and output signal level on the y-axis.

Using a gate with external sidechain in the expander mode, possibly along with low and/or high cut sidechain filtering, can provide excellent, smooth ducking control.

The De-Esser is truly comprehensive and includes the CH. MODE selector, both internal and external sidechain control, sidechain filtering, along with nine additional variable controls and three switched settings including Listen that let’s you hear the de-essing control signal itself. The Ratio control is something else, literally – it runs from 10:1 compression down to 0.1:1 which yields expansion! There are both frequency and bandwidth controls (Q from 0.1 to 10) of the “de-essing” target frequency, Attack, Release, Gain of control signal level, Range of reduction (or of gain if expansion is used), and Output level control. Although I don’t recall I’ve ever used an external sidechain with a De-Esser before, it’s possible with CS-5501, and it can be used for frequency controlled ducking and even frequency controlled dynamic boosting as well as de-essing vocals. This is about the most comprehensive de-esser I’ve ever seen. Very flexible and effective!

The Lim(iter) module at the right is simple but effective, and as with the other modules can operate in five channel modes, Stereo, Left, Right Mid, Sides. I found it able to crank signals up 10 dB or more with very little distortion added, depending on the original signal level. Note that this limiter uses lookahead processing and creates some latency (2.8 msec at a 48 kHz sample rate). It is the only module in CS-5501 that adds latency.

What About Saturation?
Of course CS-5501 has saturation available with a somewhat different approach than most saturation plug-ins I’ve used, as you can see below.



First there is a dial to select from one to five Stage saturation and to its right a dial labelled Fluctuation. Then, as I show in the middle view, there is a window to choose from four characteristic types, Clean, Warm, Crisp and Classic. There is also the CH. Mode selection with the usual suspects I show in the right panel for completeness. And below the Stages and Fluctuation dials are an ODD dial and EVEN dial, and yes, these, along with the Character chosen, can produce a range of possible harmonics, from 2nd harmonic only (Clean Character, 1 Stage, using Even dial) to 3rd harmonic only (Clean Character, 1 Stage, using ODD dial) to only multiple odd harmonics (Clean Character, 2 Stages, using ODD dial) to primarily even harmonics (Clean Character, 2 Stages, using EVEN dial) to a vast range of even and odd harmonic variations using Clean, Warm or Crisp and two or more Stages. The Classic Character is unique – it deactivates Stages, Fluctuation and Even controls and provides only odd harmonics using the ODD control. All this might seem enough, but, there is more! Increasing the Fluctuation control setting creates increased random (in time) variation of odd and even harmonic levels with any of the other settings (except Classic Character). Talk about a wide range of possible saturation effects! And while I’m at it, saturation in terms of THD could be run from “nothing” (I measured below 0.00001%) to over 20%. Much fun to be had!

What Does Parallel Do?
As I mentioned in the first few paragraphs, each module has a series/parallel switch, and a parallel mix control (PM) as illustrated for Comp2 below.



This works only if you activate both of any module type – you can run two compressors, two EQ, two saturators, etc. in parallel but not an EQ in parallel with a compressor, for example. But the results are fascinating and useful. For example, two EQs in parallel with one band in each set to the same cutoff frequency and same Q, but with the gains in opposite polarities (say one +6 dB and one -6 dB) yields a Pultec type curve. In series two such EQ settings will cancel out and the result will be no peak or dip, but in parallel you get a “dippy-peak” (not sure what to call it) like the classic and famous Pultec (not that it directly emulates a Pultec). With two compressors in parallel you can get a nice, controllable two stage ratio effect – in series the lower threshold unit will often end up controlling all the gain reduction, and although it’s possible to obtain two-stage ratios, it’s tricky to control.

I wondered what happens if a pair of paralleled modules are not adjacent, with other modules in the signal path between them. Turns out that is a case where the signal path shown at the right is not applied – two modules in parallel always act as if they were adjacent and the location in the signal path is where the first module sits. This is logical – I tried to figure how parallel processing can be performed with other processes occurring between the first and second module, and can’t really come up with a concept! At any rate, parallel processing is another useful feature in this feature-rich console emulation.

And there is More!
There are more features to describe than I have time to cover, but a few are worth a mention. CS-5501 includes a built-in ABLM Lite loudness matching system that can automatically match the output level to the input level of the full processor chain to a very close degree. This helps compare the effects of different processor settings without loudness bias. There is DAW automation possible for every control of every module – a huge number of controls! There are separate left/right input faders and output faders that can be linked or separately operated. And you can invert (flip phase as people erroneously say) the left and right signals independently. There is a GUI colour control (tiny paintbrush icon above the EQ section) that provides nine colour schemes – I used the Black scheme in these examples which I find pleasantly neutral. And the GUI can be freely sized or use one of 11 fixed scales from 50% to 300%. There is a Preset browser with 17 factory examples and, as always, you can create and save your own. And every module has a tiny hamburger icon in its upper left corner that opens a Module Preset menu enabling you to copy, paste, save or load all the settings of the module – very handy when you want several instances of CS-5501 to have the same settings in a particular module. You can also separately turn on/off tooltips and knob overlays (that show settings digitally). And if aliasing bothers you, there is 2X and 4X oversampling available at a CPU resource cost, of course.

How Does It Sound?
There is so much possible that all I can say is it sounds excellent, with any set of modules in action. The EQ section is smooth as I mentioned, with the Pultec-like cut/boost ability. The compressors can act like a range of different dynamic processors and the de-esser is a fabulous, very flexible dynamics processor that I’ll be using on many source signals, not just vocals. All the modules do what they claim, in excellent fashion. From subtle warmth to screaming harsh distortion, it’s all here. If one wanted a stripped-down studio in a laptop, the CS-5501 alone (well, a dozen or two instances) could cover almost every audio processing task you need.

Technical
Requires Windows 7, or later, OpenGL 2 GFX card/Mac OS X 10.11 or later, Metal GFX card. Plugins provided for Windows: 32/64 Bit VST, 32/64 Bit VST3, 32/64 Bit AAX. Formats for OS X: 64 Bit VST, 64 Bit VST3, 64 Bit AU, 64 Bit AAX.

I tested CS-5501 V2 using a PC Audio Labs Rok Box with Intel Core i7-4770K CPU @ 3.5 GHz, 16 MB RAM running 64 bit Windows. CPU load varies with the number and type of processors being used, as well as use of oversampling. With no oversampling, running at 48 kHz in REAPER, and running CS-5501 on multiple tracks with four to six modules active in each track, the average CPU FX load was about 0.4% per track. Using 2X oversampling doubles that and 4X quadruples the CPU load. However, I found no reason for oversampling. Latency is zero for all modules except the Limiter (needed for lookahead) – the Limiter creates a low 133 samples of latency at any sample rate I tested (2.8 msec at 48 kHz sample rate or 1.4 msec at 96 kHz).

Good news on CPU load – the 0.4% per track average was measured with multiple modules such as compressor, saturation, and de-essers active and audio being processed. I modified my main mix template to use a CS-5501 on every input track with the SSP mode turned on in every instance (BTW, there is global switch to turn on SSP in every CS-5501 in a project). Running a project with 32 audio tracks and a heavy mix (often 12 to 24 tracks playing audio at once), total CPU load ran 8% to 10%, Real-Time CPU (the actual critical measure for audio processing) ran 5.7%, and the longest Real-Time Block ran 1.64 msec/21.33 msec. In case all this tech jargon isn’t familiar to you, this means a ‘heavy’ 32 track project is barely ticking the CPU load. A hundred track project with a CS-5501 on every track would be no problem. The SSP feature is great – REAPER’s Performance Meter indicates CPU load drops to 0.01% on a track when audio stops and SSP takes over. In fact, I was so impressed that I have now adopted a new mix template using a CS-5501 on every input track (and removing from each track the previous separate EQ, compressor, and console emulation which saves me several percent CPU resource right off the top).

Pro:
An excellent, comprehensive console emulation with some useful unique features.

Clean clear GUI with fast access to any module.

Great range of signal monitoring with VU, oscilloscope, spectrum and peaks views – the Spectrum view is especially valuable for adjusting EQ and dynamics filters.

The De-Esser is the most comprehensive dynamics controller I’ve seen, and useful for far more than de-essing vocals.

Full DAW automation support

Low CPU load and no latency unless the Limiter is used (and it adds only 133 samples). The SSP mode is efficient, dropping CPU load to nearly zero when audio is not being processed.

Demo version available

A very fair price for such a powerful set of tools! Excellent value for money

Cons:
Nothing really – a recommendation would be to have the Noise module levels adjustable rather than only four settings, one of which is at the 24 bit noise floor level. But that is being rather picky!

https://www.tbproaudio.de/products/cs-5501