Measurements
Table edge
I did say that the treble sounds good and that usually means a good FR yea? It doesn't get much better than this. With 1/24 octave smoothing the room reflections are actually causing bigger zig-zags. (Or, it could be the grille on the tweeter? I don't think so tho.)
1/6 octave smoothing
While S880DB is certainly not the only speaker that achieves this, IMO you need to do this if you want to call yourself a good speaker, and any speaker that can do this is definitely a good speaker. So S880DB goes into the "good speaker" category. (Hear that, PMC twenty.21?)
There is one thing I want to point out, after seeing the measurements of S350DB and R2000DB: I think my measuring setup causes a slight straight dip in treble for like maybe -2dB from 1kHz to 10kHz in a straight line. Could be my mic, could be room reflections, could be my measuring method. In any case, up to you whether you want to apply the correction yourself. It's a good FR in either case. (Slight dip is good.)
S880DB does not have that strange 10kHz peak of S350DB. Fixing that as well as the underlying issues which created that in the first place is probably what fixed the treble from "good but flawed" to "no complaints".
Off-axis is awesome too. Textbook good apart from minor errors in the 4kHz-6kHz region.
Bass can keep up with other speakers until around 90Hz. Then it starts dropping off quickly.
One important thing to note when interpreting measurements is that, most reviewers focus on anechoic measurements. (Or in my case, pseudo-bad-anechoic measurements.) That's why you see most bookshelves' bass stops at around 100Hz. However in an actual room, you can easily get another octave of bass i.e. until 50Hz. So even though the bass above looks almost identical and unusable below 90Hz, in reality the S880DB's bass is still substantially worse than R2000DB. And also we have to consider different-sized speakers have different boundary-loading / "near-field" distances which also change their in-room behavior, resulting in speakers that measure the same anechoically having different performances in room. Maybe we will touch on this more later, or maybe in the next review.
S880DB's treble is better than R2000DB, somehow. Guess that metal tweeter is better than the Eagle Eye tweeter? Or it doesn't matter at this level.
Close mic
Oo the tweeter is smooth. And yes, close mic results may differ from actual results, because, surprise (ok not really), the rest of the speaker also affects the response. But from close mic we know there is no issue here.
Crossover frequency is a bit high tho, and quite not steep, and not much attenuation. But I did not find distortion or directivity issue which is usually the outcome of bad crossover design, so, *shrug*.
Listening position
Yea, where's the bass?
Even if we remove my room boost at 135Hz, the general trend is the bass is getting weaker as the frequency drops. And there's an unexplained dip in 300Hz which I have never seen before. (I have seen a peak there before.)
You still get bass at 60Hz, which agrees with the result of hearing, but it is already down -10dB, which... well my hearing can't quantify loudness.
Notice I show the distortion graph this time. If the volume is reduced by 3dB, the two distortion peaks near 60Hz and 100Hz improve a lot. But this shows that we don't have much headroom.
Equalization
With such a good treble, equalization is easy. None was needed in the treble.
I just needed to smooth out the woofer frequencies a bit.
Removed. See new post below.
The bottom two are greyed-out. These are bass boosts. Let's see how the listening-position measurement looks like with those on:
The frequency response itself looks good right? That is pretty much my "target curve". But the problem is the bass distortion peak is... higher... than the signal.
So in the end I have to do it without the boost. And this is the result. And even then the distortion peak is higher than I like. This is louder than my normal listening levels so it should be ok in normal use, but I really prefer more headroom.
Now it looks ok, and sounds decent, but bass leaves me wanting more. I guess one trick I can do is to add a low shelf filter so I spread out the bass boost over a wider frequency range, and then reduce the bass at 60Hz and 130Hz using peak filters.
So unfortunately, while the performance is otherwise excellent, bass is inadequate and there is no straightforward fix due to lack of headroom.
And now the verdict, I want to call it bad due to the bass, but can I really do that when the treble is this good? And it's not that bad either, because while it does not have the volume it still has decent extension, and it's not too bad for something of this size either. I will call it average bass, good treble. Which is... still not good by my standards. I want to recommend speakers that are good, not average. Because people want to buy products that are good and not average.
Also, the price. For $399............ If overall sound experience is the most important, there are better speakers out there. Sure, these speakers would mostly be bigger, but they will provide better price-performance. I mean... even R2000DB provides a better overall experience, at a lower price.
However it also looks like some shops are selling them at $289. At this lower price, while still not the best choice if just talking about overall sound, I can see the value proposition. I can see people who need aptX. People who need USB. People who need a smaller speaker and do not mind losing a bit of bass, and who cannot tolerate the treble of R1280DBs. It's still not ideal for me, but since I have not thought of an alternative at this price point yet, I will close one eye.
