3 way shootout: Smabat s-10s (silver) vs. Fiio EM5 vs. Datura Pro
Today we'll take a look at how these earbuds compare to each other in their stock tuning as well as compensated to my v.1.5 earbud target (bass version). For the purposes of this comparison, I’ll refer to the above earbuds by manufacturer name only. I'm going to post this first then edit typos as I find them. I want to get this out as quickly as possible.
A bit on how I build compensation files…
I use the excellent work of Jaakko Pasanen's
AutoEQ project to generate compensation files based on my measurements and then normalized to target curve data of my own design. More details of my work can be found in this thread.
My approach and background
I won't do a detailed review of any of these buds. I'm finding my ability to accurately describe what I'm hearing challenging. I'm still rather new to reviewing and comparing earphones and earbuds. My attraction to the hobby is less of a reviewer and more of a music lover who is constantly enjoying music as I work and read. My personal musical tastes are 90 percent classical music and 10 percent acoustic indie, although for the purposes of this comparison I’ve used a number of well-regarded recordings in the audio engineering community. You can find my test tracks in my Spotify playlist at the bottom of this post. As a baseline for sound preference, my standout favorites in the non-earbud realm are: B&O H6 1st gen (oratory1990 target), AKG k371 (stock), Etymotic er2sr (stock), LZ A7 (oratory target applied to gold filter/pop setting.)
Uncompensated comparison: Faaeal Datura Pro vs. Fiio EM5
No surprise here, the Fiio is much better than the Faaeal overall. What may be surprising to most is that with the exception of the bass response and a couple nuances, the sound profile of both the Faaeal and Fiio are very similar. I spent hours AB testing between the two earbuds to be certain that I can stand behind this conclusion. Here’s a graph of the two earbuds:
The two biggest differences on the FR graph:
- The Faaeal bass falls off a cliff from 200hz and below. Users have noted the bass roll-off here in this thread. It’s pretty dramatic.
- The Fiio begins to roll-off from 125hz and below, dropping -10dB between 125hz to 20hz. The Faaeal suffers nearly -20dB from 200hz to 20hz.
- Post 10k energy. The Fiio carries more acoustical info in this region, notably with more pronounced peaks at 12k and 16k. Note, the 12k peak is also a coupler resonance point and as such is a bit more elevated than in reality.
Non FR differences between the Fiio and Faaeal with stock tuning:
- Soundstage, separation, imaging. In all of these areas, the Fiio is superior to the Faaeal.
- Overall detail retrieval goes to the Fiio. The Fiio has a bass response on the graph that is similar to a large number of buds that suffer from exaggerated mid bass bloom, bleeding into the lower mids. However, Fiio’s large, beryllium coated driver is exceptional in the lower part of the FR (and really across the entire spectrum.) Bass is fast, controlled, and textured. There sub-bass roll-off is still evident but the quality of the mid bass is fantastic.
The Faaeal’s mids are great. They have good presence, depth, and timbre. There is also very little bleed (none, really) into the lower mids from the bass, partly because of the roll-off but I think it’s also an indication of the quality of the driver. This becomes much more apparent when compensated.
Uncompensated comparison: Smabat vs. Faaeal
I’ll keep this simple, most listeners will pick the Smabat over the Faaeal but a detailed comparison is a bit more complicated. I’m going to spend much of this section detailing the shortcomings of Smabat’s stock, additionally as the basis for why the Fiio is much better than Smabat (again no surprise given the cost differential).
The Smabat’s have a 9k trough that is pronounced, so much so it affects the FR detrimentally in a number of ways:
- high hats and cymbals have an unnatural decay and poor timbre. In some cases, the 9k recession will mask or hide aspects of the music completely. Consider these two examples. First, in John Mayer’s Gravity, the opening cymbals between :02 to :07 and 1:00 to 1:06 are completely lifeless and the high hats throughout the entire song are rendered poorly. Second, same issue with the cymbal and high hat wounds in “Hey Nineteen” by Steely Dan.
- live recordings suffer to the point they no longer sound live. This is even more pronounced with classical recordings in concert halls regarded for their fine live acoustics (Amsterdam’s Concertgebouw and Vienna’s Musikverein, for example). In Claudio Arrau’s rendition of the Beethoven Piano Concerto No.5, second movement, there’s a 1st violinist with a squeaking chair. You can hear it throughout the entire performance, or should be able to. The first three times you hear this chair in the opening of the second movement is at :08, :20 and :38. Additionally, between phrases you can hear the hall. The Smabat loses all of this. While on the topic of Arrau’s fantastic performance, the trills from 3:52 forward sound pretty bad. The Smabat is unable to reproduce the resonating qualities of the piano soundboard with any kind of accuracy.
- Soundstage suffers, sounding constrained with very little depth. Separation is still decent, though.
Here's the Smabat graph:
The Smabat does have good bass, so with stock tuning, this quality alone sets it apart from the Faaeal’s stock tuning.
Uncompensated Smabat vs. Fiio
I won’t waste time here. The Fiio is much better than the Smabat when considering stock tuning. I will say that the Smabat bass is great, reaches deeper than the Fiio but has less texture. The bass is also slower on the Smabat. Overall upper mid to treble resolution is much, much better with the Fiio.
Compensated Faaeal vs. Fiio
I’m going to end this comparison with this section but before I do, I want to briefly mention that the correction files for the Smabat help tremendously, though the transformation is nowhere near the Faaeal’s.
With v1.5 bass compensation applied to both the Faaeal and Fiio they both sound fantastic. The improvement on the Faaeal is night and day different. Though the difference on the Fiio is pronounced as well, the improvement is not as striking as the Faaeal's for the simple reason that the Fiio sounds great stock and compensated. The Fiio sounds like two different earbuds, both sounding great. The Faaeal sounds ok stock (I’m being generous) but sounds like it’s in the same price tier as the Fiio when compensated. I’m currently at 8.5 hours invested in this comparison, so I do not say this lightly:
the Datura Pro sounds qualitatively as good as the Fiio when they are compensated.
The similarities in the stock graph between the Faaeal and Fiio become more believable after AB testing these buds with compensation. From this perspective, they really are very, very similar in sound. Here are the few differences when compensated to v1.5 bass target:
1. Faaeal has deeper bass but Fiio’s bass is still more nuanced and textured.
2. Imaging and layering favor Fiio slightly.
3. Stage depth is a bit more obvious in favor of Fiio.
4. Fiio overall sounds a tad more intimate than the Faaeal.
Conclusion: If you can run wavelet on an Android device, for $26 shipped (I'm in the U.S.) you can have access to 99% of the sound quality of the Fiio EM5. If you’re on a pc running peace/apo you’ll have 93% of EM5 sound quality. With my GEQ settings, 92%.
Admittedly, I’m fatigued from nearly 8.5 hours of critical listening. Most of this was due to the fact I had to invest time in listening to these buds at stock tuning (that was a painful experience). Given some rest I might adjust my percentages a bit.
I really do not recommend the Smabat if you listen to any genres with drums, live classical music, piano music, or value stringed instrument timbre. I’m a violinist and every note played on the E string from 3rd position and beyond sounds beyond terrible on the Smabat.
my test tracks