The Virtual Mix Rack could be the most anticipated release in the history of audio plug-in processors, though that might have something to do with the fact that Steven Slate announced the plug-in in October of 2013. Slightly over a year later Slate Digital finally made the plug-in available commercially. Slate Digital products are known to have a lot of hype surrounding them, but the positive is that their products live up to the hype more often than not. Personally I don’t mind waiting for a product that is being perfected rather than use a prematurely released buggy software.
The Slate Digital VMR is a collection of 4+1 analog modeled plug-ins usable in a 500 series modular style virtual mix rack. The +1 refers to Revival, one of the free modules in VMR. Now when it comes to analog modeling, the poster boy for algorithm design is Fabrice Gabriel, the man behind the algorithm designs of all modules in VMR. Fabrice also developed algorithms for the widely used Virtual Console Collection, Virtual Buss Compressors and Virtual Tape Machines all sold by Slate Digital. Clearly, he has experience creating virtual emulations of analog hardware so it's no surprise that his initials are on all the VMR modules.
The Mix Rack
The Virtual Mix Rack with 5 modules loaded. Only 4 are visible.
The VMR is a 500 series style virtual rack that will let you load any combination of the 5 modules in series. You can load a maximum of 8 modules per instance of the mix rack, though in practice this might be overkill, but to each his own. The mix rack will only show you four modules at a time but you can horizontally scroll to see additional modules that are loaded in a chain. Its quite easy to move around modules in the rack and even in between different racks by using mouse drag and drop gestures. You can save presets for individual modules or custom chains of modules. You can also setup A/B snapshots of your entire chains to make comparisons. There is a solo and bypass switch for individual modules, though the bypass is not level compensated and there isn’t a trim control on any of the modules. So it will be quite hard to do true before/after comparisons due to perceived level changes. Steven has mentioned on a popular audio forum that he is addressing the trim issue in a future update.
As of right now there are 5 modules available for use in the mix rack (the fifth being free).
- FG-N - Neve style EQ
- FG-S - SSL 4000 style EQ
- FG-401 - SSL Console style VCA Compressor
- FG-116 - UREI 1176 FET style compressor
- Revival - Harmonic exciter (Free)
Analog Modeling in VMR
There has been a lot of debate on the accuracy of plug-in emulations of analog behavior, but one thing is certain that the algorithms have drastically improved since the early days of digital audio, and computers have become much faster and adept at handling these complex algorithms. Understanding the complexity of these algorithms may seem daunting for the average sound engineer/producer unless he/she has a diploma in computer science or electrical engineering but one thing that can be easily understood and measured is the concept of harmonic distortion with analog processors. At the most basic level, these plug-ins add odd or even harmonics to the signal. Sometimes even both. This kind of distortion sounds musical to our ears and results in that subjective and almost intangible analog warmth that everyone keeps raving about. Purely digital processors do not add this harmonic distortion to the signal being processed which makes it sound cold and sterile, or at least that’s what most people subjectively refer to it as. To make this a bit more objective we can conduct some tests.
The following image show the frequency spectrum of a pure sine wave at 100 Hz. We will send this sine wave through various modules in VMR and observe the resulting frequency spectrum.
A pure Sine Wave at 100 Hz.
Of the two EQs, I found the FG-N to be more aggressive. The predominantly odd harmonics that are introduced with this plug-in is apparent even in its default state with no gain.
Sine Wave processed with the FG-N Module.
The FG-S is a lot more transparent and even with gain on the individual bands the odd harmonics introduced are subtle. If the Low or High shelf is switched to a bell, I noticed some high frequency activity akin to noise on the spectrum analyzer albeit at a very low level.
Sine Wave processed with the FG-S Module.
When the input or output dials are pushed up on the FG-116 compressor there is significant addition of even and odd harmonics as evident in the spectrum analyzer (see image below). If you Shift-click on the Attack dial, the compressor gets internally bypassed but the gain dials still work. So if you want, you can use the module just for its analog goodness this way. The FG-116 also has a noise reduction switch, when turned off introduces high frequency noise as expected but at a very low level.
Sine Wave processed with the FG-116 Module.
The FG-401 has a transformer switch, when switched on introduces significant even and odd harmonics. Switching between the two circuit models doesn’t seem to make much of a difference in the frequency spectrum. I believe these are two different algorithms which affects the reaction times—specifically for low frequency content—rather than the frequency content so the effect will be evident in the time domain rather than the frequency spectrum.
Sine Wave processed with the FG-401 Module’s Transformer switch ON.
Now, I would not want to simplify these modules to just harmonic exciters as they are doing a lot more than just that, but it's a good starting point in understanding the analog experience you get with these modules.
This free sonic enhancer module is very simple and straightforward in use with just two controls, one to add shimmer and the other for thickness. Under the hood there is quite a lot more going on. Steven says the Revival ‘has some of the most sophisticated algorithms we’ve ever developed’. There’s a combination of harmonic distortion, filtering, tube and tape based saturation and also some secret mojo that he doesn't want to reveal.
We can verify the second and third harmonic distortion in the spectrum analyzer. This was brought about by pushing up the Thickness dial.
Sine Wave processed with the Revival module’s thickness dial up.
I particularly like the Shimmer dial on this module and wanted to compare it with the high shelf gain on the FG-N module. At first listen the Shimmer seemed akin to a high shelf, so I thought it might be interesting to compare it with the high shelf on the FG-N. Instead of a sine wave I used a sawtooth wave here. A sawtooth wave has significant even & odd harmonics in the upper register so it was more suitable for this test.
A pure Sawtooth Wave at 100 Hz.
The Shimmer is clearly doing a lot more than just boosting high frequencies but it is interesting to see the shape of this boost. The following is a frequency spectrum comparison of the pure sawtooth wave being processed with the FG-N’s high shelf and then the Revival’s Shimmer.
Sawtooth Wave processed with the FG-N module’s high shelf boosted.
Sawtooth Wave processed with the Revival module’s Shimmer dial pushed up.
The FG-N high shelf has a more gradual boost while the Shimmer dial seems to be focusing specifically on 5 kHz and above.
I added some Shimmer and Thickness to a snare sample and you can clearly see in the following two images, the definition being added to the shape of the waveform. The first one is with Revival bypassed and the next, with it on.
Snare Sample waveform unprocessed.
Snare sample waveform with Revival processing.
We’ve looked at a bunch of waveforms & spectrum analysis graphs but probably the best way to analyze these modules is with some sound tests. Let's start with the EQs. I’m using female vocals, recorded with a Sterling Audio ST69 Tube condenser mic going through an Apogee Duet 2. As I mentioned earlier, I found the FG-N to be quite aggressive so while processing this vocal buss I made sure to be very subtle with the settings. The exact same vocals were used with the FG-S but while trying to get a similar tone to the FG-N processed vocals, I noticed I had to push the individual bands gain on the FG-S significantly more. The analog warmth definitely comes out easily in the Neve model but the SSL emulation sounds a bit more subtle and transparent.
Vocal snippet Dry:
Vocal snippet with FG-N:
Vocal snippet with FG-S:
Check out the image below for the exact parameter settings I used for each module. The two modules were used separately, but I put them in the same mix rack here for ease of comparison.
FG-N & FG-S settings for Vocal processing.
For the compressors I used electronic drums as the source material. I also used pretty extreme settings so you can really hear the compression. The FG-116 when pushed hard can really transform the sound. I used a ratio of 4:1 but cranked up the input and output gain to about halfway with fast attack and release settings. Notice how different the kick sounds when compared to the dry version.
Drum Loop Dry:
Drum Loop with FG-116:
I really like that the FG-116 has a mix dial so if you are not a big fan of this type of heavy compression, you can bring down the mix and have that parallel compression sound which to me sounds way better for this drum loop.
Drum Loop with FG-116, mix at 55%:
The FG-401 doesn’t tonally change the drum loop like the FG-116 does. I went for similar settings on the FG-401, with 4:1 ratio, fast attack & release and threshold low enough so the gain reduction was similar to the FG-116. Even with the transformer ON, the tone doesn’t drastically change though I did use the circuit 2 model which gives more space to the bass.
Drum Loop with FG-401:
Again, thanks to the mix dial, we can have the best of both worlds with some heavy compression mixed in with the original dry signal.
Drum Loop with FG401, mix at 50%:
Both these compressors are excellent at what they do and definitely bring something new to the plug-in world. What I like is that both of them have contrasting characters which totally makes sense since one is modeling a VCA compressor while the other a FET compressor/limiter. Always good to have different colors in the palette box. The FG-116 being extremely true to the original hardware it's modeling, which if you have ever used, is a good color to have in the box.
The only drawbacks for me is that there is no auto release switch or side-chain filtering on either of these compressors. When using these compressors on complex material like a drum buss or even the master buss—even though they were not intended for that—the auto release switch would have been very handy. The side-chain filtering could be used to filter out the lows in the detection circuit to get less pumping with high compression settings whenever the kick hits. I can understand the FG-116 was meant to be a faithful model of the original hardware but the FG-401 was tweaked to be different from the SSL compressor being modeled and even has a 2nd circuit option so maybe adding an auto release option and side-chain filter could have been additional tweaks.
Finally, let's have a listen to the Revival on a complete mix. One thing I’ve noticed with this module is that there is a tendency of overdoing the processing. Just because it sounds that good. I noticed some ear fatigue after listening to a mix repeatedly. Clearly this was due to setting parameters too high on the Revival module, especially the Shimmer. In the example below I have made very subtle boosts to the Shimmer & Thickness dials. Pay attention to what this is doing to the kick and snare. Even at just 25% Revival is lifting the overall sound and adding so much definition to the snare and fattening up that kick just enough to sound more appealing than the dry version.
Music Mix Dry:
Music Mix with Revival:
The Virtual Mix Rack with its 5 modules is an achievement that is commendable to say the least. With additional modules coming out in the future, VMR may cover more ground other than just EQ & compression. With Native processing being as powerful as it is today VMR may eventually be the death of hardware-based DSP accelerator systems for plug-in processing. As with most Slate Digital plugins, I noticed very minimal CPU usage on my mid-2012 MacBook Pro. Regarding analog emulation in the box, you can’t go wrong with these plug-ins. Considering the introductory price, it's pretty much a no-brainer purchase. I suppose the more pertinent question is, do you really need this if you have every other analog modeled plug-in in the market? Probably not but considering the price, you will definitely be tempted. If you are starting off in the analog emulation world and can’t afford DSP accelerated systems, VMR would be perfect for you. When it comes to analog modeled plug-ins, VMR is at par with the best in the market, if not even better.
Price: $150 with 2nd gen. ilok. (introductory offer)
Pros: Accurate Analog emulation of classic hardware. Routing flexibility with the Virtual Mix Rack. Compressors have mix dials for parallel compression. Free Revival module.
Cons: Needs 2nd gen. ilok. No trim dials. No auto-release or Side-Chain filter on compressors. Modules can’t be used outside the rack.