Abbey Road Studios are celebrated for many things, among them the unique equipment designed and built by EMI's in–house engineers. Thanks to the studio's partnership with Chandler, some of their most revered processors are available as rackmounting products, and their range also includes plug–in emulations of several key items. Their excellent TG 12413 limiter plug–in is a firm favourite, while the Mastering Pack under review here contains two EQ plug–ins based on EMI's own 'transfer consoles'. These custom–designed mastering mixers were introduced in 1972, and proved so successful that several are still in use today.
Plug–in installation is straightforward, with authorisation to an iLok key, and my only minor quibble concerns the naming of the plug–ins. This is no doubt authentic, but when TG 12413 is joined in your plug–in list by TG 12412 and TG 12414, all of which do different things, there's a certain amount of potential for confusion.
TG 12412 is a four–band equaliser perhaps best described as semi–parametric, although it has some novel features. Each of the four bands is dedicated to its own area of the frequency spectrum — low, low–mid, upper–mid and high — and offers a choice of five stepped corner frequencies. These, unusually, are centred around musical values such as Middle C and its harmonics, rather than neat numbers of Hertz. Up to 10dB of cut or boost can be applied, while the five–position Shape switch offers high and low shelving responses on the outer positions, with 'blunt', 'medium' and 'sharp' bell curves in between. The interesting thing to note here is that any of the bands can be set to low or high shelving, so if you want to, you can apply four different shelving EQs simultaneously, or have a high shelving boost turning over at 128Hz!
TG 12414 is a simpler affair, comprising stepped high– and low–pass filters plus a single band of 'presence' equalisation, with stepped centre frequency and ±10dB gain. The filters can't be switched out of circuit, but have 'Low' and 'High' settings that move the corner frequencies out of the audible range (roughly 13Hz and 38kHz respectively).
It almost goes without saying that both of these EQ plug–ins sound great, but I was surprised at the difference in character. TG 12412 is like the proverbial 'make it sound better' processor, with an effortless sound that invites adjectives such as 'smooth', 'silky' and 'warm'. I was particularly impressed by its ability to add large amounts of boost without any of the usual problems that doing so can bring. It's easy to fatten up the low–mids, with none of the 'cardboard' quality that often plagues attempts to add weight in the 400–600 Hz range. Pushing the low end brings out a fantastic rounded, solid bass, while the high–frequency shelf is quite unlike most digital EQs, with the rare ability to remain fluid and warm even when large amounts of boost are applied. You could argue that it doesn't quite open out the mix in the way that some rivals can, but nor does it ever become harsh or emphasise sibilance. It's a bit like having a well–trained butler apply a gentle polish to the music.
TG 12414, by contrast, makes its presence known in a rather less subtle way. Applying the same gain settings that would result in gentle tonal shaping in TG 12414 here can completely change the character of an instrument or a mix. The manual states that the presence band is useful for "making certain instruments in a recording more or less prominent", and it is certainly a blunter instrument than TG 12412.
If that description makes it sound as though TG 12412 is the more versatile and useful of the two plug–ins, that's probably fair. To my mind, it's probably best to think of 12414 as a useful bonus that you get when you buy its counterpart. I can't recommend TG 12412 highly enough: it's one of the most musical plug–in equalisers I've heard, and its uses go far beyond mastering. Sam Inglis
$499 (TDM) or $249 (native).
$ $560 (TDM) or $335 (native).
Christian Knufinke's SIR VST plug–in has been one of my top freeware tips for a while now, offering, as it does, fuss–free convolution processing that micturates from a great height on most other freebie reverbs. Although the program has been updated frequently since its launch, Christian has now decided to step the development up a notch and deliver SIR2 as a commercial product, albeit still at an attractive UK price.
SIR2 is still VST–only, and will run on Windows XP or Vista. Audio arriving at the plug–in's stereo inputs can be pre–delayed for up to two seconds before it is passed on to an M+S matrix that lets you adjust the width of the stereo image fed to the main convolution engine — this can be worth doing to avoid slightly weird results on stereo signal feeds. The convolver can then either process the left and right channels with the left and right channels of a stereo impulse response respectively (as in SIR), or you can use True Stereo mode, where each channel is processed separately in stereo. I have to admit to being sceptical about the merits of this latter mode, especially as it requires you to have a suitable four–channel impulse response on hand and uses more CPU power, but I was surprised at how much of a difference it made to the realism of the stereo image when feeding in a stereo file. It's akin to the difference between the fuzzy stereo of a spaced mic pair and the more accurate stereo of a coincident rig.
After selecting your choice of impulse response from the file browser on the right–hand side of the plug–in window, you can tweak its sound in a number of ways. Separate envelopes are available for amplitude and for a low–pass filter's cutoff frequency, and you can enter as many breakpoints as you like while you zoom in and out of the impulse response's waveform. This is a big step forward from the facilities in the original SIR, and I found that the extra precision was really handy for amplitude control.
A second M+S matrix adjusts the stereo width of the reverb signal, and the effect's tonality can be refined using an improved built–in linear–phase EQ, which is configured with another 'all you can eat' breakpoints system. If you wish, you can quickly reverse the impulse response for special effects, although you could, of course, do the same thing (and more besides) by processing an impulse file itself in an audio editor. Another option is to change the length of the impulse response using sample–rate conversion, as in SIR, and the quality of the conversion has been improved.
While it is possible to save all your plug–in settings within most sequencers, SIR2 also lets you save presets internally. The advantage of this approach is that the presets are all neatly lined up in the right–hand pane under the impulse response that they relate to, which makes it easier to find the correct preset for the job. That said, although the plug–in will load impulse responses in five audio formats including WAV and AIFF, only those in its own '.SIR' format can take advantage of this preset management, although SIR2 will automatically convert the standard file types to the proprietary format whenever you first try to create a preset. Unlike its predecessor, SIR2 can run with zero latency, but also allows you to operate a fixed 1024–sample delay instead if you want, as this can reduce processor load on some systems.
The sound quality of convolution processors relies heavily on the impulse responses used, so Christian has teamed up with Pinguin, a specialist high–end audio company in Hamburg, in order to stock SIR2 with top–notch raw material in the form of special HDIRs (High Definition Impulse Responses). Only four HDIRs are included as standard (Church, Recording Studio, Jazz Chamber and Theatre) but these are available in both Stereo and True Stereo versions and generally sound very slick. A further 12 HDIRs are available as a cost option if you want more to choose from, and while there are now masses of good impulse responses available for free download on the Internet, Pinguin's True Stereo files are a pretty persuasive reason to part with some cash on this occasion.
The HDIRs also have another trick up their sleeve: hit the HDIR Direct button and the unreflected sound of the source signal is also generated by the impulse response, but delayed and balanced to maximise the realism. The value of this provision was a bit lost on these ears, but I'm sure there will be purists out there champing their pipes with glee at the mere thought of it.
The most serious competitor that I can see for SIR2 is Voxengo's Pristine Space, which comes in less expensive, at $119.95, and is capable of surround processing. However, that plug–in doesn't include the kinds of impulse responses included in SIR, and I also can't help feeling that the SIR interface looks much more elegant. Things like Audio Ease Altiverb 6 and Waves IR1/IRL are all well over twice the price, so are unlikely to be in the frame unless you want to get your mitts on their star–studded internal impulse libraries. Overall I think SIR2 deserves to do well: it's easy to use, has a few very good impulse responses built in, and doesn't cost an arm and a leg. Mike Senior
148.75 Euros; HDIR Extras 141.61 Euros; both together 236.81 Euros. Prices include VAT.
$ $189; HDIR Extras $179; both together $299.
Tritone Digital's entire plug–in range is convolution–based, and all about providing 'analogue' character. They sample hardware EQs with up to several thousand combinations of frequency, gain, Q, and filter settings, and then build their fully interactive EQ plug–ins from this massive sample set. ColorTone–Pro is slightly different, since instead of concentrating on recreating one detailed model of an analogue EQ, it instead offers a host of different analogue flavours captured from analogue consoles, tape machines, hardware EQ, and other desirable hardware gear, so you can apply this 'essence of analogue' to your own sounds.
ColorTone Pro uses a three–pronged approach for maximum control over this character. First, you can load in short Impulse Response files that capture the static frequency/phase response of a large variety of real–world hardware devices. The plug–in is bundled with a collection of such 'fingerprints' captured from desirable analogue hardware, but it also includes instructions on how to sample your own gear.
Enthusiastic users also contribute to an on–line ColorTone IR Library where I was able to download 74 additional IR collections (some comprising a single IR, others a selection captured with differing gear front–panel settings so you can click through them one by one and hear the result of altering their front–panel controls). Some very interesting gear has been sampled here, including hardware from Fairchild, Oram, Rupert Neve, SPL, SSL, Studer, Tube–Tech and the BBC.
Using the Color control, the intensity of this EQ contribution can be smoothly magnified up to five times the captured setting for more extreme curves, or be subtracted from the dry sound for an inverted curve, the resulting frequency response being displayed in a graphic window. Some of the IRs provide more extreme curves that you can vary like single–band hardware EQs, but many offer more subtle tonal variations, with bass humps, mid–range ripples and top–end peaks or roll–off — you'd find it hard to recreate most of these curves even with a bevy of parametric EQ bands.
The second ColorTone–Pro stage is Warmth, which adds a few per cent of odd–harmonic distortion to frequencies below 600Hz, along with a side–helping of low–shelving EQ, and can be added to or subtracted from the original audio to taste. This can be used to warm or cool your sounds, or to balance out more extreme Color settings (coupling a thin–sounding IR with extra warmth, or vice versa).
However, it's the third stage that's most intriguing: in Norm mode you always get 100 percent wet convolution, but if you switch to Blend mode you get a dynamic crossfade (with Soft or Hard knee transition) between the original dry and convoluted wet signals, with the proportion decided by the input signal level. In other words, the hotter your input, the more coloured the output becomes. Even better, you can alter the wet level using the Trim control — if it's lower than the dry level, louder signals effectively become compressed, while with the wet level higher than the dry, ColorTone–Pro becomes an expander. Clever stuff!
You could simply patch ColorTone–Pro in and dial up your favourite gear curve to add that certain 'something' to a track, and there's plenty of scope for adding character to digital synths, enhancing cheap instruments, warming up drums, and so on. However, I found it just as useful in creating tone colours that you could never achieve elsewhere, using the Blend mode to emphasise or tame transients (on drums or picked guitars, for instance), or create subtle backward filter 'thweeps' and other dynamic effects.
I spent many happy hours experimenting with ColorTone Pro, and although its various controls can be initially confusing (especially in combination), it's well worth persevering, since together the three stages offer a huge number of possibilities. There's a ColorTone–Free version available containing a few preset IRs and a Color control to vary them, but with no Warmth or Blend options. However, the extras on the Pro version will add a great deal to your tonal palette, and I think it's good value. Martin Walker
$145 (approx £73).
Cobalt is the first VST Instrument from Leslie Sanford, and is inspired by classic 'hybrid' synths such as the Korg DW8000 or Kawai K3, in that it models digital technology in the oscillators but analogue technology in the filter section. Cobalt offers two oscillators — each offering 26 waveform types, many resynthesized from classic hybrid synths — the output from which is combined before passing through the single filter. Both oscillators include pulse width modulation and two stages of frequency modulation. In all cases, any of Cobalt's two LFOs and two envelopes can be selected as the modulation source.
Cobalt uses a 12dB–per–octave filter and offers the usual cutoff and resonance controls. Four filter types are provided: low–pass, high–pass, band–pass and notch. Two frequency modulation sources can be selected, and the filter can also respond to aftertouch if your keyboard supports it. The two envelopes are standard ADSR types, while the LFOs feature six different shapes (sine, triangle, down ramp, up ramp, square and sample & hold), as well as the ability to engage Mod Wheel control for the LFO amplitude. The feature set is rounded off by the effects section which includes overdrive, noise, panner, chorus and delay.
Cobalt is supplied with an impressive preset collection and these do a good job of demonstrating what the synth is capable of. There are some excellent lead sounds (for example, 'Cognitive Distortion'), pads (such as 'Dawn'), strings (the aptly named 'Strings') and a goodly range of keyboard, bass and bell–like patches. Cobalt is also capable of the dark and daft — try the 'Haunted Halls' and 'Green Jelly' patches for some examples.
The user interface, while a little on the dark side, is extremely easy to find your way around. Usefully, two otherwise identical versions of the plug–in are supplied, set up to occupy different amounts of screen space; I found the text a touch small on the compact version, but it will be more manageable for laptop users. In all other respects, Cobalt is a breeze to use and, because it offers a fairly straightforward control set, is very easy to program. New users also have the benefit of a very useful programming tutorial within the PDF documentation — a nice touch.
Cobalt might not be the most advanced software synth available, but it is certainly a very capable instrument and, because it is easy to get to grips with, those without a PhD in synthesis ought to find it very accessible. I experienced absolutely no problems during testing within Cubase 4, so it also seems to be fairly robust. For those with a well-stocked virtual synth locker, I suspect Cobalt might not offer anything you don't already have, but it certainly ought to appeal to those still building their VSTi collection. And at $25, it is a bargain that almost anyone can afford, with a fully functional demo version available for download so you can try before you buy. John Walden
X–EQ is the latest optional plug–in for SSL's Duende platform and is designed as a flexible, multi–band parametric equaliser for use in both mixing and mastering situations. There are 10 EQ bands, of which the first and the last are dedicated high– and low–cut filters, each with a choice of five different filter curve types. Of the remaining eight bands, two and nine are shelving EQs with a variable Q of up to 10.3. As you might imagine, these filters take on a resonant character at higher Q settings, making them useful for sound design as well as for corrective processing. The middle six bands are all fully parametric proportional–Q band–pass or bell equalisers; every band can be tuned anywhere between 20Hz and 20kHz, with a cut/boost range of ±20dB. Additionally, the six band–pass filters can be switched from the usual series operation to a parallel mode.
The interface not only includes the now–familiar EQ curve, so that you can visualise the effect of any adjustments, but also an FFT display that shows the spectrum of the EQ'd waveform in real time. EQ settings can be adjusted by dragging the curve points or by using a single set of conventional parametric controls, which act on the selected band. There's an A/B function for comparing settings and X–EQ's window includes stereo input and output metering with peak hold, a separate RMS meter and a dynamic range display, all of which are vitally important when you have such a wide cut/boost range to play with.
This all adds up to a very powerful equalisation tool accessed via a single window, but how does it sound? What impressed me about this equaliser is that it doesn't have an obvious sound: it just does what it is supposed to do, and usually accomplishes it with relatively modest amounts of cut or boost. It has the capability to smooth out the sound of a mix without making it dull, and there are plenty of bands for more surgical tweaking. It displays no obvious vices when used within normal parameters, and it is certainly extremely versatile.
Switching to parallel mode produces very different results for the same EQ settings, making it easier to tackle separate problems in different parts of the frequency spectrum. With very small adjustments I soon turned a lacklustre mix into something with all the sparkle and punch I needed — most impressive. The analyser display also works well, but how important this is depends on how much you rely on your eyes to locate potential mix problems. I sometimes find it helpful in setting the Q of filters to ensure that I'm only adjusting the part of the spectrum that needs it, once I've used my ears to find the problem frequency.
If you're a Duende owner you'll find X–EQ a powerful and useful tool both for tracking and mastering. I'm anticipating using mine quite a lot in the future. Paul White
527.58 Euros including VAT.
$ 449 Euros.