Review: Soundtheory Gullfoss - EMusician

Review: Soundtheory Gullfoss

Taking intelligent EQ to the next level
Publish date:
Social count:

Soundtheory bills its Gullfoss plug-in as an “intelligent automatic equalizer.” While that might bring to mind other smart processors, such as iZotope Neutron 2 and Sonible Smart EQ, it works quite differently.

Gullfoss is an adaptive EQ that makes at least 100 adjustments per second based on the incoming audio.

Gullfoss is an adaptive EQ that makes at least 100 adjustments per second based on the incoming audio.

Gullfoss employs a proprietary “computational auditory perception technology” process, which adjusts your audio in real time. The goal, according to the developer, is to improve clarity, detail, spatiality, and balance.

Soundtheory’s Andreas Tell, a physicist who’s been researching auditory perception for many years, developed Gullfoss, and it is the first commercial product he has released based on his work. The fundamental way in which this plug-in differs from other smart EQs is that it makes its processing decisions by comparing the audio in real time to an “auditory model” that Tell developed based on his research on how humans perceive sound and, in particular, what makes a sound pleasing to the ear. Gullfoss accesses that model as a reference, at least 1,000 times per second, and can make at least 100 frequency adjustments per second.

This 64-bit AU, VST/VST3 and AAX Native plug-in is Mac-only, at the moment: According to Tell, a Windows version is in the works.


If you’re wondering what Gullfoss means, it is the name of a waterfall in Iceland. The sound of a waterfall is considered to be a pleasing one to the human ear, which is presumably why Soundtheory chose the name.

You can use Gullfoss on individual tracks, subbuses, the master bus, or in a mastering situation. Its interface features a large frequency graph, and with Gullfoss set at its default, you won’t see any action on the graph except for the horizontal Input and Output meters at the bottom. Once you start adjusting parameters, however, you’ll see a frequency display that moves to reflect boosts and cuts that the plug-in applies in real time.

The parameters on Gullfoss are decidedly nonstandard and, therefore, a bit confusing at first. But once you read the online overview document (the closest thing there is to a manual), you will begin to get a sense of how to set them.

Gullfoss provides five adjustable parameters—Recover, Tame, Bias, Brighten and Boost. Recover and Tame are the key ones because they define the type of processing the plug-in will apply. Recover increases the intensity of frequencies that Gullfoss has determined are being masked, whereas Tame reduces the intensity of the masking frequencies. So primarily, Recover boosts and Tame cuts, although there is boosting and cutting going on in both processes.

The Bias control tells Gullfoss how much to emphasize the Recover or Tame processes. Negative percentages tilt it toward Recover and positive toward Tame. If you leave the Bias at zero, then the ratio of Recover to Tame processes are governed solely by their respective settings.

The Brighten parameter also impacts the Recover and Tame processes, doing so in a way that pushes the signal to be brighter on positive percentages and darker on negative. The Boost control, though expressed in dB, changes the frequency response to match what a listener would hear when the audio is louder or softer.

To compare how a particular parameter is affecting the process, you can Alt-Click on it to toggle between its current setting and zero. That same key command also resets the adjustable gain control, a green line on the right side of the interface that you can drag up and down (see Figure 1).

Fig. 1. The one standard parameter you can adjust on Gullfoss is the output gain, on the far right.

Fig. 1. The one standard parameter you can adjust on Gullfoss is the output gain, on the far right.

Small vertical meters to the left of the display give you a visual sense of the relative amount of Recovery or Tame processes as they are applied. Similar-sized horizontal meters under the frequency display reflect how the processing you’ve dialed in is affecting perceived brightness.

With so much going on inside it, Gullfoss is necessarily CPU-intensive. If you don’t have a fast Mac, you will have to be careful with how many instances you use at a time.


Sonically, Gullfoss is impressive. I tried it on individual tracks of many types, on buses, and on the master bus. It gave me positive results on almost every source I put it on, though its impact was the most dramatic when placed on the master bus because it noticeably added to the clarity of the mix.

One of the other applications in which Gullfoss shines is on multitrack drums. I put separate instances of it on kick, snare, stereo rooms, and overheads, and the difference in the entire drum sound was quite noticeable; the processed version was clearer and brighter.

Gullfoss seems to work best with its parameters set relatively low. Like on any EQ, extreme settings yield extreme results.


I can’t claim to fully understand the underlying science behind Gullfoss, but the results are unquestionable. While it’s not a panacea that will make poorly recorded tracks instantly sound great—nothing can do that—it will add clarity and balance to virtually anything you apply it to. Its ability to adapt to a changing sound in real time is unlike any EQ I’ve ever used.

Overall, Gullfoss is a groundbreaking product that is sure to be coming soon to a mix near you.

Adapts to audio in real time. Uses model based on how humans perceive audio. Welldesigned metering.

Manual needs work. Windows version in development.


Mike Levine is a composer, producer, and multiinstrumentalist from the New York area, and is the Technical Editor—Studio for Mix. Check out his website at