User Experience

If it was not for the orange button on each bud, the more deliberate design of the EVA2020 x final earphones would not be easy to make out once in your ears. Seen above are some different tip options used on an artificial ear mold that is similar enough to my own average-sized ears. The first one is with the pre-installed size M final Type-E tips placed at the ear canal entrance, as recommended by the company. It feels secure but not snug, and isolation could be better, too. You can push them into the ear canal as with most ear tips for more, but then it does not feel comfortable. It does seem like final has gone with a balance of comfort and isolation, with the balance tipping towards the former at the expense of the latter.

The SpinFit CP100 is probably my go-to here, and I will mention again that you need to push these in slightly more than you think for the pivot point to rest at the ear canal entrance, which has the tip pivot into the direction of your specific ear canal. This may feel uncomfortable for the first 10-15 minutes, after which it ends up quite comfortable, but that may not be the case for you or your earphones and is why the nozzle size and shape of your earphones matter, as well as the core diameter and design of the tips themselves. Take the CP145 at the right, which looks like any other tip but larger and ends up the least comfortable fit of the lot, albeit still very secure and isolating.

Battery life is a key metric for TWS earbuds, and these promise nine hours of use when fully charged. I hit 8-9 hours regularly at ~80% volume on my phone, and the included charging case provides for another 56 hours (6 charge cycles) even though there is theoretically enough capacity for nine charge cycles. Charging them with the case takes about 90 minutes, and charging the case itself takes about two hours over USB 3.0 (0.9 A maximum). The indicator LEDs on the case show charging progress of the case in 25% increments, which can take much longer if on USB 2.0 and its 0.5 A limit. The case does not support any quick charging in itself that I saw for those wondering about USB PD or Qualcomm QC protocols. Regardless, this makes for very good total battery—I got a week on average while commuting or just using these as a headset paired to my phone.

Audio Performance

Audio Hardware and Microphones

As is the case with most TWS earphones, final uses a single dynamic driver per ear bud. The company claims precision of the driver unit and assembly, so much so that the smaller diameter driver was chosen for "exceptionally high precision in this price range." I also found it funny that final effectively calls out most of the competition for relying on software-based EQ for balancing sound in the high frequency range, which according to it causes listening fatigue. There is more shade thrown out by saying bass is overemphasized by others at the cost of a balanced response. Now, I may not agree with the lack of EQ to give consumers the option to customize things, but then final goes on to say it has a technology to help correct the "irritating high range sound caused by Bluetooth," which achieves a "balance between high-resolution, clear vocals and powerful bass".

The EVA2020 x final earphones use the Qualcomm QCC3020 SoC designed on a low-power consumption architecture intended for TWS stereo earphones. The chipset utilizes triple-core processing delivered by two dedicated 32-bit application processor subsystems and a single Qualcomm Kalimba DSP audio subsystem.

Microphone quality is about average only, with the position right under the vent susceptible to picking up noise from outside but not directed towards your mouth. There is no mention of the hardware used for the microphones, and they end up with mediocre fidelity and the compression associated with Bluetooth earphones that are designed more with listening in mind. They will work in an emergency if you have nothing else, but if two-way communication is an important metric to you, there are far better implementations out there.

Frequency Measurement and Listening

Before we proceed, I will mention that I have a general preference for a V-shaped sound signature emphasizing elevated bass and treble with recessed mids. I also generally prefer instrumental music over vocals, which complements the V-shaped profile.


It took some time to identify all the necessary pieces, order them with a language barrier in place, obtain the necessary software, and finally set up a reproducible testing methodology. This begins with an IEC711 audio coupler earbuds can feed into enough to where you have decent isolation similar to real ears. The audio coupler feeds into a USB sound card, which in turn goes to a laptop that has ARTA and REW running and the TWS earphones paired to it. I begin with an impulse measurement to test for signal fidelity, calibrate the sound card and channel output, and finally test the frequency response of each earbud separately by also keeping the distance from the source to the earbud as consistent as possible. Octave smoothing is at the 1/6th setting, which nets a good balance of detail and noise not being identified as useful data. Also, the default tuning was used for testing, and no app-based settings were chosen unless specifically mentioned. The raw data was exported from REW and plotted in OriginPro for easier comparison. I am avoiding the use of the ear mold for wireless solutions because the base of the mold disrupts the connection somewhat.


Now, after calibrating the sound card and accounting for the base floor noise levels, I am a touch more confident about the SPL label, but there is likely still a significant offset across the board owing to the overall resistance and impedance in the system. I should really also normalize the curves to a set frequency, but the IEC711 is such that you can't really compare these results with most other test setups anyway, just within our own library of measurements. What is really useful information is how the left and right channels work across the rated frequency response in the EVA2020 x final TWS earphones, or at least the useful part of it. The left earbud was separately tested from the right one, and colored differently for contrast. I did my best to ensure an identical fit for both inside the IEC711 orifice, so note how the right channel ends up having a slightly boosted sub-bass response followed by a dip and then is consistently under the left channel. Both generally have the same profile shape across the 20 Hz to 20 kHz test range, which is good to see. I will also mention that there was no discernible break-in period or effect.

This is one of the better-sounding TWS earphones out there to me, and the curve is also more of a V-shape (really more an M-shape) than neutral, which plays a role, too. There is the characteristic elevated bass courtesy the tuning of the single dynamic driver, with enough of a sub-bass response to even surprise some who go in expecting less from such TWS solutions. The brief spike around 180 Hz is barely felt, with a smooth transition to the mids. The mid-bass is the strong point here, which makes sense given the physical and power restrictions. There are no balanced armature drivers for an active crossover as with an IEM in the price range this operates in, but the positioning of the driver speaker right up against the nozzle helps retain clarity across the board in combination with that filter in the driver system. This lends warmth to vocals, with good tonal separation and balance from the background.

The best feature, to me at least, is with the detail imparted to string instruments. Classical music sounds wonderful compared to pretty much any other TWS earphones I have tried, and smooth jazz gets favored, too. Going to the mid-treble, we start to hear the M-shaped curve with its characteristic peaks and valleys. There is enough energy here to excite pop music fans, and I have to say that the filtering was well done across the highs—not only to counter the compression and hiss from Bluetooth pairing, but also for streaming audio over a subscription service such as Spotify, YouTube Music, or Amazon Music. Here too the staging is clearly biased towards vocals and instruments. Orchestral playback will suffer slightly if there are too many instruments playing at the same time because of muddy instrumental separation. Support for aptX is nice, but nothing to distinguish the EVA2020 x final from just about anything else on the Android side of things, where there are plenty of other aptX and even aptX HD TWS earphones. Overall, however, I would say that this is the best-sounding set of TWS earphones I have tested objectively to date, beating out the ATH-CK3TW, ATH-CKS5TW, and ATH-ANC300TW from Audio-Technica, the Lypertek PurePlay Z3/Z3 2.0, and the Creative Outlier Gold.

Testing the SpinFits

So how do the SpinFits fare? Keep in mind that all these do here is provide a different fit inside the audio coupler. There is more isolation and a more custom fit with the larger orifice compared to the ear mold or my own ears, so the CP145 tests near identical to the CP100. I suspect it would have been different on the ear mold, however, so keep that in mind. We get a tighter response, with a smoother bass to upper mids region that has no bumps along the road, and a shift in the treble regions towards the lower frequencies I am not a 100% sure of. These were all reproducible across multiple mounts too, and in practice, I have to say that the SpinFits sounded better, which I attribute mostly to the more custom fit I got with the CP100. The CP145, on the other hand, is going back into the packaging for someone else to use since it's not for me!