Sound Quality

The SteelSeries Arctis Nova 3 is equipped with a pair of 40-millimeter dynamic drivers with a specified frequency response of 20-22,000 Hz. SteelSeries doesn't reveal anything about the integrated electronics, so I can only tell you that the built-in DAC can handle 24-bit/96 kHz PCM files, and that the amplifier is powerful enough to make the Arctis Nova 3 as loud as you could possibly want. If you end up finding the headset too quiet in its default state (which I didn't), you can add up to +12 dB of output gain by activating Sonar in the SteelSeries GG suite and scrolling all the way to the bottom of the Game, Media or Aux tabs, where you'll find a Gain slider. Increasing gain this way doesn't introduce any audible distortion.

Recently, I've been thoroughly testing and experimenting with the miniDSP EARS headphone test fixture, probably the least expensive such device on the market. I got to a point where I feel comfortable using it for frequency response measurements to present the results publicly. The miniDSP EARS consists of two miniDSP UMIK-1 microphones placed inside a pair of silicone ears and affixed to a steel headphone stand. It's a USB-powered device, perfectly integrated with the fantastic Room EQ Wizard (REW) software, which is used for headphone and microphone calibration, as well as doing the measurements themselves. Every single miniDSP EARS unit comes with its own microphone calibration files, which are loaded into REW before doing the measurements. The miniDSP EARS certainly has its quirks. The artificial ear canals aren't particularly deep, so I wouldn't necessarily consider it a viable option for IEM testing. The integrated microphones are exceptionally sensitive to headphone positioning, partially because of the screws affixing the silicone ears to the stand, which aren't perfectly flush with the silicone surface. Some users also claim that the fixture is too narrow for very large hi-fi headphones, so they have to use rubber bands to tighten the clamping force. The latter isn't something I had an issue with, though; all my headsets and headphones fitted perfectly.

Because of all that, the miniDSP EARS is prone to user error. Whoever is using it has to know how to interpret the results and, even more importantly, how to spot those that are clearly flawed. After getting rid of the bad data, assuming the initial configuration and calibration were done properly, the resulting frequency response measurements are usable and in line with what one would hear from the headphones (or headset) in question. Once I got to where the frequency response plots corresponded to what I heard from the headphones and headsets I measured, I knew it was time to bring the results into my reviews. My testing methodology mostly follows the best practices of using the miniDSP EARS as covered in great detail in this article on Headphonesty. The most significant difference is that I'm using the updated HEQ compensation, where a subjectively neutral headphone would measure approximately flat. The audio source for my analog headphone and headset tests is the fantastic EVGA NU Audio Pro PCI Express sound card. For the SteelSeries Arctis Nova Pro, I used the supplied GameDAC Gen 2 as the source, as that best represents the experience the future owners of the headset are going to have when listening to it. To compensate for minor seal variances, I'm measuring each ear until I get 10 usable results and re-seat the headphones, or headset, between runs. After I have my results for one ear, I apply a 1/12 octave smoothing and average the measured responses. I then repeat the process for the other ear for an idea of whether any channel imbalance exists. It's important to keep in mind that my results cannot be directly compared to anyone else's, as only an identical methodology conducted on my exact sample of the miniDSP EARS would yield directly comparable results. Still, we now have a relevant visual representation of the sound characteristics of the headphones or headset under review, so I hope you find it useful.

I only just started building my gaming headset frequency measurement database, but you can explore it and follow its progress on my personal Squiglink. Let's take a look at the frequency response measurement of the SteelSeries Arctis Nova 3.



The frequency response you're looking at was recorded at default settings without any equalization applied. The overall sonic character of the Arctis Nova 3 isn't that similar to the Arctis Nova Pro or Nova Pro Wireless. In its default, non-equalized state, the Arctis Nova 3 sounds bassier and warmer, but also audibly darker and more veiled. This is apparent when looking at the substantial dip in the 2-5 kHz range. While that dip makes the headset sound completely unfatiguing, you're also losing some of the treble energy and sparkle, as well as finer details in the presence region. Luckily, these issues can be easily fixed by using the parametric equalizer offered within the SteelSeries GG software suite. After gradually boosting the entire 1-5 kHz region, with a +6 dB peak around 4 kHz, I was able to lift the veil and achieve a snappier, more lively presentation. I wasn't able to achieve such good results by using the treble slider, as the slider is set up to boost the 8 kHz range, where the Arctis Nova 3 isn't lacking.


The bass is pleasantly full and thick, with no overhang to speak of. There isn't a ton of content to be heard in the sub-bass region, but the Arctis Nova 3 sounds surprisingly rich in mid-bass and upper-bass regions. This is what gives the headset its warm sound signature and makes it quite pleasant to listen to, especially if you're not that bothered by being able to hear every possible detail in the music you're listening to. While the treble region is veiled, I still found the Arctis Nova 3 excellent for all kinds of gaming, including multiplayer shooters. The midrange is fairly neutral and detailed enough to make enemy footsteps easily audible. A good gaming headset gives you a sense of awareness and control; the Arctis Nova 3 ticks both of those boxes. Of course, for a less colored, livelier sound, it's still worth boosting the presence and treble regions, so definitely consider taking some time to play with the parametric equalizer in the SteelSeries GG software suite.

In terms of seal and overall fit, the Arctis Nova 3 is definitely more finicky than the Nova Pro and Nova Pro Wireless. The cloth earpads are more sensitive to micropositioning, so sometimes you might feel you're not getting the expected sound quality, especially in the bass department. You're not crazy; you just have to readjust the headset slightly until a better seal is formed. That was the impression I got while using the Arctis Nova 3, and it was confirmed when I did the frequency response measurements – it took me a ton of tries to get consistent results and prevent a substantial channel imbalance due to an uneven fit.

Virtual Surround Sound

On the virtual surround sound front, the SteelSeries Arctis Nova 3 offers a Spatial Audio section with a ton of ways to tune the experience to your liking. First of all, you can pick between the headphone and speaker modes, with each offering a slightly different tonality of the surround sound effect. Then you can play around with individual virtual speaker placement and distance and adjust a slider that has Performance (better sound localization) and Immersion (more ambiance) on its ends. While I didn't find a spatial audio setting that would offer me a more accurate in-game target retrieval or a more engaging music-listening or movie-watching experience than traditional stereo mode, your mileage may vary. The sheer number of possibilities is staggering, and experimentation is highly advised.