How close do these modelling plug‑ins sound to the real thing?
Silo SoundLabs are a new plug‑in company with a new technology. SiloDNA is a novel way of recreating the behaviour of hardware processors, and is claimed to offer not only extremely accurate modelling, but also far lower CPU overheads than existing alternatives. Their launch offerings recreate two hardware units from celebrated mixer manufacturers Trident Audio: the current 80B 500-series EQ, and the vintage CB9066 rackmounting parametric EQ. Both are available in all the major native formats on Mac OS and Windows, and are authorised to an iLok account.
Launched in 1980, Trident Audio's Series 80 consoles built on the reputation they had forged with the original A and B Range desks. The channel EQ was arguably the star of the show in the Series 80, hence its availability today as a 500-series module. The rackmounting CB9066 is perhaps less well known: dating from 1974, it was the first Trident product designed by John Oram. For its time, it was extremely well specified, offering three fully parametric EQ bands plus versatile high- and low-pass filters.
Neither a single-width 500-series module nor a 1U rackmounting unit really lends itself to slavish recreation in a plug‑in GUI so, sensibly, Silo have rearranged the front panels in a logical way that preserves the general look and feel. Oddly, though, they've chosen completely different legending schemes. On the original CB9066, for instance, the high band covers 3.5-14 kHz, with 4, 5, 6, 8, 10 and 12 all marked off at irregular intervals. On the plug‑in version, this control spans the same range, but the only intermediate marks are at 6.5 and 10.5 kHz, and it's not clear whether it has the same uneven taper.
In a sense this doesn't matter, because all the controls are continuous rather than switched. But, as luck would have it, we happen to have an 80B 500-series module at the studio where I work. It proved surprisingly difficult to match the sound, and I can't say how much of that is down to the variant legending and how much to intrinsic differences. My general impression was that the high end was actually slightly nicer on the plug‑in, while the hardware offered more 'bite' and authority in the mid-range. At the bottom end, the combination of 50Hz HPF and -15dB of low shelf that almost completely took out a vocal plosive on the hardware was noticeably less effective on the software version. In most cases, though, the differences were small, and the plug‑in definitely shares the distinctive, and very usable, character of the original. You can find a few audio examples from these comparisons in the Zip file (see 'Audio Examples' box below).
The high end was actually slightly nicer on the plug‑in, while the hardware offered more 'bite' and authority in the mid-range.
I've never had the pleasure of using an original CB9066, but the plug‑in version has a lot to commend it. The three parametric bands cover a huge range and can be narrowly focused, expansively broad or anything in-between. The filters not only have fully variable frequency settings but fully variable slopes, which must have been highly unusual in 1974, and is still quite impressive today. Having said that, if I had to choose only one of these plug‑ins, it'd be the 80B. It's not parametric and has fewer bands, but there's just something intrinsically 'right' about the setup, and it's quite a bit cheaper.
What impressed me most about both plug‑ins, though, was that they really are astonishingly CPU-efficient. Even on my very modest MacBook Air, I could happily contemplate using either as channel EQ on every track in a huge mix. In terms of system resources, they're more comparable with the basic EQs that come with a DAW than with any vintage emulation I know of — which means you can treat Silo's Trident equalisers as a virtual console, rather than as special processors that have to be rationed for use only on key tracks.
I compared the Silo SoundLabs Trident 80B 500-series plug‑in with the real thing on these excerpts from a multitrack sent to me by SOS reader Jimmy Kostelidis. In each case, I auditioned the dry sound in isolation, adjusted the controls on the hardware to a setting that I thought worked, and then tried to recreate that setting on the plug‑in.
They were then level-matched. As mentioned in the main text, the process of matching them was made harder by the fact that the legending on the plug‑in dials is different from that on the hardware.
For more critical listening and auditioning in your own DAW, download the Zip file of hi‑res WAVs.
- They sound good, if not completely identical to the hardware.
- They don't burden your CPU.
- Attractive GUIs.
- The legending is not consistent with the hardware.
Refreshingly light on CPU resources, these new modelling plug‑ins sound good, and could be used to deliver Trident-like EQ on every track in a large mix.
80B $89.99, CB9066 $149.99
80B $89.99; CB9066 $149.99