Edifier R2000DB review (WIP)

[wwenze]

Edifier R2000DB review

Oh, right, the review.

There is something about 5-inch speakers. They are like the smallest you can go for decent sound reproduction, they are usually priced above a certain price point, and that price point is also the minimum you can go for decent sound reproduction. Studio monitors seldom go smaller - a few 4.5"-4" exist, a very few even smaller ones exist *cough*iloud*cough*, companies that manufacture studio monitors and also manufacture smaller speakers seldom call their smaller speakers "studio monitors", and they are always lower-priced products while the studio monitors are always at least a certain price.

It is as if manufacturers are intentionally gimping the performance of cheaper, smaller speakers and always using 4-inch or smaller woofers to force you to upgrade to a more expensive 5-inch eventually.

Makes sense, because 5-inch is really barely just enough for bass response. Well, it is not written in stone, and is just that the difficulty of producing sufficient bass becomes harder as the woofer becomes smaller. You can still reproduce bass with a 3-inch... at the cost of heavy DSP, heavy port velocity, and a 80dB maximum SPL.

On the other hand, increasing the size from 5-inch does not guarantee better bass either - keyword being better. More bass, definitely. Better, as in an even bass response, not always, it depends on your room - how close you are to the speakers, how close the speakers are to the walls, and the frequency response of the speakers themselves. I learnt it first hand and hard with the Diamond 8.2 in my room on my desk.

I want to point out that at this point it isn't about the speaker size already, but what the designer wants to achieve, and if the price happens to allow good sound with better drivers, then great. Similarly, if they want to gimp the performance, might as well do it cheaply and consistently. So a small driver is like the manufacturer telling you "screw you, you ain't getting better sound for this money". Hence all the studio monitors at SGD$400-500 and higher. As if there is collusion going on.

To further illustrate, there are cheap 5" or even 6" or even bigger speakers at cheap price points, but their performance is still far behind the entry studio monitors. A $40 6.5" speaker still sounds like a $40 speaker.

And so recommending 2.0 speakers at the $200-300 price point is a difficult task. What do you recommend? Seriously, because I need to consider them in the review. I can think of one direct alternative and one not-so-identical alternative and that's it. I can think of a few 2.1, but even then the number is not many, nor are they direct alternatives.

Fortunately, Edifier reminds us that an option at this price point exists. And also reminds the AMDK brands that if they decide to rest on their laurels for too long, then prepare to be challenged.

 

[wwenze]

Edifier speakers, lots of them

If you go to Edifier's website, they have a speaker for everyone. Heck, I'll just list just the bookshelves below.

S880
R980T
S2000V2
S2000MKII (Kind of like S2000 Pro for the international market, with the two differing in input connections)
R2000DB
R1700BT
S1000
S1000MA
R1000TC (North America version)
R1600T III
R1200TII
R1000BT
R1800T III
R1800BT
R1900TV
And I haven't included the not-so-bookshelf 2.0s. And this is just from the China website, the international website has some models that are not included here (e.g. R2730DB)

Fortunately, the Singapore distributor (Ban Leong) has done some filtering so these are the ones that are available:

I have to say, they did a good job with the selection. At first glance it still looks like a lot of choices, enough to be confusing as to what to buy, but each of this is a clear improvement over the previous, and you have something for which price point you want and which features you want.

With R1280DB you get digital inputs and a remote, with R1850DB you get slightly better tweeter and woofer size and bi-amp, with R2000DB you get the industry-standard tweeter size and woofer size and DSP. With S2000 Pro you get wait what a planar tweeter at this price?

BTW, "D" means digital input (SPDIF), and "B" or "BT" means Bluetooth. Makes remembering the features a bit more convenient.

Unboxing pictures

Normally I don't care about photos, if you can already get the same info from elsewhere. But since there are some things that I don't remember seeing, so here are some photos.

The R2000DB makes Usher S-520 look like a bigger speaker.

Opening up the box, this is how we know this is a unit meant for sale in Singapore. Because the lower left of the styrofoam is a space for the UK plug. In fact I think it is meant for the Europlug to UK 3-pin adapter:

My unit does not have it - it has just the Europlug. But hey, it's a review unit, so whatever. Europlug is also a legit connector in Singapore and nowadays more power strips just directly accepts it.

Versus speakers that come with China plugs which require an awkward adapter which often results in me just using a UK power cable.

Some people say the speaker cable of R2000DB is thick. They have not seen the iLoud's which is even thicker. (The black one is iLoud's.)

I love that they come with protection for the tweeters.

Yea that's a Safety Mark, so this is definitely meant for sale in Singapore.

And I already noticed something interesting about the current rating. 400mA multiply by 100V is what...?

I will decide how much output power it has when I look at the insides. After the sound testing.

 

[wwenze]

Look ma, no DAC

At the heart of R2000DB's amplification system is a TAS5508C 8-Channel Digital Audio PWM Processor that takes in serial audio data (For convenience let's just call it I2S) and outputs PWM which is used to drive a digital amplifier power stage (hidden underneath a heatsink).

Many speakers with DSP and/or digital crossover just use the DSP/digital crossover for what they are. Convert the incoming audio to digital, do the processing, convert back to analog to feed the amp. There are multiple reasons for this, cost is one, but there are valid performance reasons too. Like for example if the amp stage is a class AB so you would need to convert it back to analog.

I am perfectly okay if the DSP-ed data is eventually fed to a DAC and converted back to analog before feeding an analog-input amp. There are legit technical reasons, and is evident by the many speakers that do use this configuration. (So yea, miniDSP users, you're good.)

What I find not perfectly okay, is if your DSP is digital-input (Like, if you're going to put an ADC between the analog input and the DSP), and you do not provide a digital-input, hence forcing the user to perform an unnecessary digital --DAC--> analog --ADC--> digital conversion, then what the 7th planet of the solar system?

However, I also understand that digital connections are more expensive than analog connections. And analog inputs are still needed since most products don't have digital outputs. So if the speaker is cheaper, I can accept it not having digital inputs even if it uses DSP. Although it still makes me sad. Bose speakers, iLoud, and LSR305 are examples of such. And in the case of studio monitors, many equipment may not have digital out, so a digital in may not be useful for those.

Either way, if you can connect a digital connection to a speaker using DSP, then it is akin to having a perfect DAC feeding a perfect ADC, so why not. It's like saving $15,000 on the perfect DAC, and that is still useless anyway because the ADC inside the speaker is far from perfect.

And R2000DB allows you to do that, and it goes one step further by using a digital-input amp so you can avoid the final DAC too. But this approach has its pros and cons, and the fact that digital-input amplifiers are rare and only started being popular recently shows that it is not easy to implement too. The technical talk of this topic will be long, so maybe later.

tl;dr - You save a DAC by using R2000DB.

 

[wwenze]

First impressions

So I set up the R2000DB by itself normally, like in the first picture above.

Powered on, and... the sound... feels okay... but weird... bass is dynamic and treble is solid... too dynamic and solid in fact.

Then I pressed the "Classic" EQ button on the remote. That dialed things back a bit. Some improvement. Glad I read the manual before playing with the speakers.

Basically, the "Dynamic" is the same feature as what you would find on many speakers by now. It makes the performance inconsistent, and while I can understand some situations and some people may like it, I'm going to leave it off. Just as I leave all of such things off.

But even with the EQ set to "Classic", I felt there is still some DSP going on. May be placebo. Or some of it may be necessary for the performance of the speaker. This is a DSP speaker after all - It uses DSP to compensate for the non-idealities of the speaker.

The sound...

- Treble is clean and clear. There is a lot of separation between instruments, and you can follow an instrument through complicated passages without it disappearing.
- This goes for bass also. Can hear each beat of the fast double pedal bass drum.
- However, sounds sound distant and not-well-defined, with a wide but not-deep soundstage.
- For example, you can distinguish between two sounds at the same time easily. However the piano does not really sound like a piano and strings don't really sound like strings. But you can hear every note hit.
- This may make these speakers well-suited for music professionals, where people want to know if the performers are accurate. But not for mastering though, where how the sound sounds matters.
- Suspect this is a realism/naturalness vs frequency response/accuracy trade off by the DSP. When you try to correct a driver's natural behavior it usually results in higher distortions. And we are not dealing with the most-expensive-tweeter-in-the-world so it comes with its own limits in terms of frequency response and distortion. A better tweeter would start with less distortions and handle more EQ/DSP, and it would also need less EQ/DSP.
- Frequency response... this speaker is... mostly flat I guess? Just a slight recession in the midrange maybe which makes the treble and bass stand out a tad bit.
- Bass is well-controlled down to 55Hz (Based on this test), and drops quickly thereafter. The speaker is not what I would call bass-heavy however. Although the bass extension is good for a bookshelf, it feels like the volume isn't enough. There is not enough weight for those epic moments, and bass-guitar-lovers won't be getting special treatment from these speakers.
- Speaking of epic moments, the sound does not feel dynamic enough. I know it is weird for me to say this when I disabled "Dynamic", but these are different kinds of dynamic I'm talking about here. When all the instruments come in together, things inevitably gets loud and the speaker needs to reproduce that accordingly, but the "Dynamic" setting in most speakers seem to only make certain parts of the sound louder and they do it all the time. In fact many speakers apply dynamic range expansion and compression - expansion to make certain things louder when the rest of the instruments are quiet, and compression to save power and reduce distortion when all the instruments come in.
- Hence the sound is smooth and relaxing with respect to loudness, and makes for a relaxing listen. Not so good for games and movies however. This is where the "Dynamic" button may come in useful. But this is partly also due to not having enough bass volume.
- And to clarify, the amount of bass is actually sufficient and on-par for a 5-inch bookshelf. In fact the extension and control is very good compared to many 5-inch bookshelves out there. I am just spoilt by my bass-boosted S-520.
- The bass knob feels like it affects mainly the lower-bass and just a bit of the upper-bass. Which is good news for those looking at a controlled bass that needs to be adjusted for room acoustics as well those looking to emulate the feeling of having a subwoofer without having the upper bass become a phat mess. But bad news for people who actually want it phat. But for people who are already fine with the volume of piano and bass guitar notes and just want the notes to feel more real, then this is perfect.
- But the way the bass can go too "dynamic" means leaving it at zero or even negative are legit choices too. In fact at negative it sounds like the typical bookshelves out there - not enough bass, but sound is high-quality enough to be what is expected from a 5-inch. This is not those pathetic speakers that sound like smaller speakers once you take away the bass because they are using the bass to hide the pathetic sound.
- Treble seems to be the best at zero. Any higher it sounds harsh, any lower it loses detail (or the subjective perception of). Usually if I can't adjust the treble to become what I like, I usually blame the tweeter. I have the same issue with my Paradigm Atom (modified the crossover by hearing), but the R2000DB's issue is much much more milder.

Overall, it has a sound that is different from the Usher S-520 that I am so used to (well duh, as if that is unexpected), but I don't find any glaring issue that would make it drop to a grade lower. While the R2000DB is not perfect, it also has some things that other speakers don't have. Personal preference plays a key role in determining what speaker to buy in this case, and the R2000DB is certainly a valid choice if you're looking for a speaker with the sound of a proper 5-inch bookshelf. Which is interesting, consider most other valid choices of speakers with the sound of a proper 5-inch bookshelf cost $400 and up. Well this is based on memory anyway. Will the R2000DB stand up to the test when placed beside a $400 and up speaker? Tune in next time on Dragon Ball Z.

 

[wwenze]

Intermission: Let's talk DACs

https://en.wikipedia.org/wiki/Digital-to-analog_converter#Types

The most common types of electronic DACs are:[2]

- The pulse-width modulator, the simplest DAC type.
..........which you will never ever see in audio. Except...... (Hint: Output is PWM)

- Oversampling DACs or interpolating DACs such as the delta-sigma DAC
..........which makes up the vast majority of audio DACs. Unless you're talking really high-end where people seem to hate these for some reason.

- And the rest of the list.

Let's talk history. The earliest DACs are all non-oversampling DACs. Because the logical thought is that oversampling DACs are harder to make. But I'm just going to refer to this documentation from Analog Devices which says sigma-delta DACs came later. (The terms delta-sigma and sigma-delta are used interchangeably in literature)

And so some time after the first chips become available (Usually meaning years or even decades), the market starts seeing these so-called 1-bit DACs. And along with it comes the explanations explaining why 1-bit and oversampling is better than a non-oversampling DAC. (Some like to call those multibit DACs, but I don't like it because it causes confusion with another DAC type coming up later.) A very legit argument is that with non-oversampling DACs, you need to control the manufacturing process of the transistors very precisely. (Let's say this is a current-output DAC) The MSB of a 16-bit DAC switches a current that is 32,768 times of the LSB. So unless you can control the manufacturing that precisely, you are going to get linearity errors. And it gets exponentially more difficult with more bits. John Atkinson talks about this issue here.

Sidetrack, I have tried a thought experiment of the feasibility of non-oversampling DACs today, since with CPUs containing a few hundred millions of transistors costing just a few bucks today, there is no better time to try the brute force method of just using a high quantity of individual elements to compensate for process variations. But it ended up not improving anything, because the final bottleneck still boils down to the fact that you are trying to use one signal to switch a big thing that is 32,768 times of a small thing, and regardless of where this is done you still have to make sure it is accurate down to [1/32,768] / 2. So big transistors, and trimming and calibration, which manufacturers are already doing, are still the better methods.

Why am I talking about 1-bit DACs tho? Because the output stage, the idea of the class-D amp, is that the output stage is a switcher between Vcc and Output that only either turns fully on or fully off.

Now, multibit DACs also utilize switchers. However they switch current, and you can add currents together by connecting two current-outputs together. But speakers are not current-driven; they are voltage driven. You can't add voltage by connecting two voltage-outputs together. Or well you can... if one voltage source is floating - this is the idea behind switched capacitor / charge pump. But this ain't gonna drive a speaker load.

Let's just stop wandering and let's just consider the 1-bit DAC as the closest relative to a class-D amp.

So, yes, problems with precision and accuracy when things become too big, this is what is limiting the scaling of non-oversampling DACs. So oversampling removes this problem, actually no, it shifts some of the problem to the time domain.

But the same thing occurs when you throw everything into the time domain. Although it seems like the general idea/trend is that the time domain has a more consistent performance. (Yea, serial bus over parallel bus, right?) So there was some improvements for a while, until things got stuck again.

So naturally manufacturers combined multiple methods, which gives birth to the multibit delta-sigma DAC. And we're stuck here until another breakthrough happens.

So why am I talking about all this? Well, I'm trying to illustrate why nobody makes a class-D amp that combines the DAC function. Not until recently that is.

If it is already hard to switch a low power 1-bit DAC, imagine switching at the same frequency a big MOSFET driving an 8 ohm load.

So where does the PWM DAC mentioned right at the start of this post come in?

Answer: It doesn't.

It has lots of uses in other electronics, including controlling the speed of your CPU fan, and brightness of LED lighting. But I have never seen one being used for audio.

So the next question: Why don't we use delta-sigma instead of PWM in this particular situation then?

Quote this article from Analog Devices:

Nonetheless 1-bit sigma-delta modulation is not often used in Class D amplifiers (Further Reading 4) because conventional 1-bit modulators are only stable to 50% modulation. Also, at least 643 oversampling is needed to achieve sufficient audio-band SNR, so typical output data rates are at least 1 MHz and power efficiency is limited.

Or basically the switching frequency needs to be way higher with sigma-delta than PWM. And we're already having enough trouble with PWM as it is.

http://slideplayer.com/slide/5750824/

OTOH if we are not constrained by the limitations of switching transistors, then sigma-delta is better than PWM (This is a general statement), and that's why sigma-delta is used for audio DACs while PWM is used for low-cost stuff. (Referring to 1-bit sigma-delta, because multibit sigma-delta is obviously superior.)

So, in order to make a digital-input analog-output (PWM) power amplifier, the designers need to somehow achieve the performance of a contemporary DAC while dealing with the limitations of a power amplifier, and how they achieve this...

I have completely no idea lol.

And this is apparently not simple because we would be seeing lots of products. Well, the products are slowly starting to come.

But class-D modulation has always been borderlining on magic. It is said that Tripath sounds very good despite having just entry-level specs. (Although Tripath is dead now, RIP.) It is now replaced by TPA3116D2 which beats it everywhere spec-wise, but there are still people who think Tripath sounds better. Similarly, there are lots of class-D amplifier solutions out there, but people like to look at ICEpower, ABLETEC and UCD, because modulation is an art.

 

[wwenze]

Three-way comparison

Yea, this is not an optimal positioning for the R2000DB. I would have wanted the R2000DB at the bottom so that it can take advantage of the table's boundary gain, but then the speaker is shaped in such a way that you can't put anything on top of it. Fortunately, there doesn't seem to be much difference between putting it on the table vs putting it on the... erm... "stands", except for the tweeter direction and the bass loss from being further away from the reflex port, which are both solved by standing up when listening to the R2000DB.

*Although boundary gain is considered bad for speakers designed to be accurate on a stand, it nonetheless boosts bass, and speakers at this size need all the bass they can get. Plus I have always listened to speakers on the table so this would make the sound closer to what I am used to, reduce the bias from listening to a sound I am not used to.

Two Musiland Monitor 01 USD plus the onboard Realtek analog... I'm not going to find another sound card for the third SPDIF out.
Although a single Monitor 01 USD can actually output to 3 SPDIF receivers simultaneously (opt, coax, bnc), keyword is simultaneously and I want to switch speakers using Windows because I'm not going to turn each speaker on and off each time I want to change, takes too much time and is bad for the speakers. Or their volume control if doing that route.

With a third speaker (which costs $400) added into the mix, this confirms something I have been thinking.

That I don't like R2000DB's treble.

At first I couldn't tell whether it is just due to different frequency response and personal preference, but now I dare say that the R2000DB's treble sounds diffused, undefined, a tad like that of a full-range or the tweeter of a cheaper speaker that is trying too hard to reproduce the treble, with the nasalness, the hollowness, the sock-in-the-mouth, the old-radio-ness. All the plus points of being able to identify each instrument clearly and etc remain unchanged however. Now I'm really wondering if this is a design decision. Like, does an Edifier with better tweeter i.e. S2000 sound the same?

Using human speech to test the speakers confirms the suspicion - R2000DB feels more TV-speaker-like compared to the other two. There is just not enough lower frequencies to back things up.

Although, if I didn't have any speaker to directly compare against, the sound is acceptable. And interestingly and perhaps irritatingly, it is okay in some songs, and horrible in others. Just don't try anything sung by Ganaha Hibiki (CV. Numakura Manami) for example.

iLoud's treble sounds similar to S-520's, and iLoud uses a 3/4" tweeter. That said, if you look beyond the frequency response you can hear some of the abovementioned weaknesses in iLoud. But frequency response has the most importance.

Oh and btw, if you try to use increased bass with the knob, the treble sounds even more sock-in-the-mouth/cheapspeaker. So I recommend using this speaker with +0 bass. It has enough bass at that setting already anyway.

Makes me suspect the problem is with the upper frequencies handled by the woofer. So I EQ-ed to give it a boost centered at 500Hz + the adjacent octaves. And true enough, that improves things a lot. Makes things more solid and defined, and the warmer sound that the other two speakers have. But still only fixes half the issue. Further EQ-ing will be needed.

You know, having too much volume in this frequency range is a problem with cheap speakers, but there are just as many cheap speakers that attempt to compensate by giving it the v-shape treatment. The drivers of R2000DB don't look like they can't handle the extra midrange volume, so I'm going to write this down as design decision or personal preference.

If only I have another speaker or at least a tweeter in the $250-500 price range... or do I...? But for now, I'm going to judge it as - If I'm going to get a speaker based on the treble, out of the 3 it is certainly not going to be the R2000DB.

...or so I thought. But I remembered I have heard other more expensive speakers with a similar kind of treble. Which makes it hard for me to conclude.

That's why I hate recommending speakers above $500 - it becomes hard to judge the objective performance and hence the price, and personal preference reigns supreme. That, and the speakers still sound different at this kind of price range. Not as bad as earphones/headphones tho.

With the speakers in an A-B (and -C) setup, some of the observations I had while doing first impressions is proven false or placebo. Like the ability to identify different singers - I originally thought R2000DB is weaker but turns out the other two speakers are the same. I also thought that the sound coming from R2000DB is less dynamic or epic but turns out they are about the same too. Isn't it interesting how we always imagine the sound we're used to to be more perfect than they are? Or that once we listen to a speaker without thinking for a long time we adapt to the sound and stop thinking about the speaker?

But some of the good points were found to be placebo too, like hearing each beat of the fast double pedal bass drum, turns out S-520 can do that too. iLoud can do that too, but it has less bass extension so that doesn't count.

With the speakers in an A-B (and -C) setup, the iLoud also sounds like somebody put a sock in its midrange region. And a bit more midrange volume (which we know from the iLoud review), and bass is all garbled up, and the deep bass is weak in comparison to the other two. But the overall sound is still warmer than the R2000DB.

R2000DB really shows how it is about the bass. At +0, the bass drum volume is bigger than the Usher S-520 with maximum bass from the Marantz PM6005. This is impressive. Although S-520 still goes deeper: Both speakers remain controlled down to 55Hz and start dropping off at ~50Hz, but R2000DB's drop off is much quicker and is gone at ~40Hz, while S-520 with bass boost still has usable volume within this range, although still sharply drops off at ~40Hz, no doubt due to phase cancellation from the bass port.

iLoud? Controlled until 55Hz, obvious reduction at ~50Hz, completely gone below that. Just as frequency response charts predicted.

Interestingly, for all the bass extension it has, for how controlled the bass frequencies it has, the R2000DB does not sound warm. *Feels* slightly v-shaped, hmm that may not be a good description, feels more like somebody used a shelf filter to increase the treble and the bass but not the midrange and upper bass.

To sum up:

- Bass extension and volume is good for a speaker at this price point. Even speakers at the next price point ($400-500+) are not guaranteed to do that. It is possible to do better with a 5-inch, but prepare to spend more and you may still have to tweak. Also, you can still further bass boost to really get that subwoofer experience at a cost of treble quality, but hey at least you have the option.
- Overall frequency response is decent, but you may want to boost the 500-1000Hz range. Experiment further if you have the time.
- If you don't like playing with EQ (or, if the situation prevents you), then there is a good chance you may hate the treble and it may sound unnatural.
- Regardless of previous point, instrument separation is good and sound is clear.
- General sound is slightly v-shaped and not warm. Bass is not overbearing (be it a good thing or a bad thing) but you still get the wub-wub.
- Treble, hard to conclude. Will need to go on to part 3 for this.

 

[abccat]

Creative T3 compare to this Edifier which one is better ?

 

[AZE]

Creative T3 compare to this Edifier which one is better ?

T3 is 2.1, R2000DB is 2.0 .

Not really directly comparable.

 

[wwenze]

Intermission: Inside photos

Soomal already has some photos here, but I tried to see what additional things I can find out.

2200uF 35V capacitors for the main power amps. I'm guessing one for woofer, one for woofer and one for both tweeters. Judging from number of inductors, tweeters seem to be single-ended rather than bridge-driven. Better photos from the FCC reveal bridged-driven tweeters and likely one cap for each amp.

Can't see the power stage clearly, I'm guessing TAS5112?
http://www.savel.org/2008/08/10/class-d-amp/

Interestingly that blog post goes way back, so PurePath (that's what TI calls this digital amp technology) has had a long time on the market. Not a necessarily a bad thing - it means people have had a long time to master its usage and it is time-tested. TDA7294 for example also goes way back and is still used in everything.

The TAS5111 and TAS5112 mentioned in TAS5508 datasheet are rated for 29.5V operation and producing 40-50W (depending on conditions). Edifier claims RMS 24W × 2 + 36 W × 2. The power amp checks out.

Note that 2200uF x 3 may seem like a small amount compared to, say, the 2 x 10000uF inside the M200MKII for example. Or the 8 x 1000uF 50V capacitors of M200MKIII. However note that R2000DB uses a switched-mode power supply so the required capacitance values are smaller; the main purpose of power supply caps is to sustain the supply voltage during the not-charging part of the cycle. And 50Hz power supplies have longer cycles compared to the few kHz or few tens of kHz of SMPS. This is why SMPSes save money.

The power supply.

I originally wanted to determine the output power by looking at the current ratings of the switching elements (i.e. transistors and diodes), but this proved difficult to remove. You win this round, Edifier. But this means losing points on repairability; although the case of M50W has taught us that repairability doesn't matter for most people.

Good to know it is using a 105ºC cap for the PSU.

 

[AZE]

No updates?

 

[wwenze]

Next update probably on the weekend of 29th Apr

 

[wwenze]

Part 3

Ok, so, I held off giving my verdict during the previous part, because I wasn't sure whether the midrange performance (or should I say tuning) of R2000DB falls within the range of speakers I have seen so far, and if that's the case it would fall under user (or designer) preference and not performance inadequacies.

So here I have two more speakers.

It's hard to identify the speakers from the photograph. The bigger speaker with a 6.5-inch woofer is Wharfedale Diamond 8.2, the smaller but still boxy speaker is Paradigm Atom v.1, and the one with the red terminator glowing eyeball is the R2000DB that we all know.

And in the middle is something I put just to show what it looks like if somebody decides to put a "for-computer" soundbar in front of a living room TV.

The bass, man
The speaker with by far the most bass, is, yup, R2000DB. The Diamond 8.2 has never been a strong basser - Bose Companion 2 (Series II) has more bass than it, but still, being a 6.5-incher it has more bass than the typical 5 to 5.5-incher that I have heard and owned. And R2000DB just made it look bad in this department without a sweat.

And it's not just volume either, it hits deep. In fact, I think it feels nicer in this living room setup than in the desktop setup. Might be the rear-facing bass port at work.

Some people have commented the bass of R2000DB is sufficient for usage without a subwoofer. I can see where they are coming from. But first and foremost, I need to stress that this is the same as debating whether 80Hz or 50Hz is low enough or if you really need the subwoofer to cover the lower frequencies in both extension and (overboosted) volume. But please also note that there are some subwoofers that don't even go that low. And in the sub-$200 price range or subwoofers intended to work with sub-2" satellites, replace "some" with "many". And lastly, when you try to adjust the subwoofers for a controlled amount of bass to match the satellites, it usually ends up at a lower volume that is determined by the satellites' own bass, at least when you're not going for the "herr derr I'm a gamer I need my table to shake also my kdr is 0.5 that's awesome rite must be my gaming keyboard at work I only lose because I keep getting bad teams".

Point is, the amount of bass coming from R2000DB is comparable to those systems with small subwoofers. Which is not really unexpected at this point. I mean, when you have two woofers that have more bass than a 6.5-incher, you would expect them the match the bass coming from a 5.25-inch sub and be close to the bass coming from a 6.5-inch sub yea?

But, man, the R2000DB with bass at minimum still has more bass than the Paradigm Atom v.1.

Everyone sounds nicer in the living room
The treble of the 3 speakers sound very alike. (The bass is all over the place as we know would happen when the listening area is big.) None of the speaker sound particularly wrong (minus the obvious differences in bass), I can select any speaker and enjoy the sound. That, or maybe all three speakers sound so horrible in a big space that I can't differentiate them anymore. Let's choose to be optimistic.

But in all seriousness, the R2000DB performs very well in the living room scenario, much better than it did on the desktop, at least relative to other speakers. Which is a good thing since it has multiple inputs including optical and BT.

However the speaker that gained the most in the treble department by shifting to the living room is Paradigm Atom v.1. That thing was giving me ear cancer in nearfield situation, and still sounds horrible if I move close to it in the living room. But it performs acceptably when far away. Although bass is totally missing now. Though mind you, I already reduced the tweeter amplitude via crossover mod so your unit may still be cancerous.

The insufficient midrange of the R2000DB is much less intruding, although if I listen carefully, compared to the Paradigm Atom v.1 the frequencies that are lacking are still slightly lacking. On the other hand, the Paradigm Atom v.1 has other midrange frequencies lacking, and actually benefited slightly more when I apply the EQ. Overall R2000DB sounds more normal now, but voices still sound slightly distant and sockinthemouth compared to the other two speakers.

I'm not sure how I should handle my verdict. The reason for doing this part 3 is to determine whether R2000DB's weird midrange falls under the spectrum of normal. Although it is not far from the speakers tested, it is still detectably different, so I have to penalize it for that. Maybe for just half the original intended amount. Also, the midrange is overall at the same level as the Atom v.1 if not better, so I should not penalize it. But then again Atom v.1 is considered a cheap speaker by now so being at the same level is not something laudable. And finally, although the midrange improved in a living room setup, this also means it is still bad in a desktop setup. I think I will just round down or round up, I'll see how it goes. Because I don't want to give a final score like for example 4.0625 stars.

Verdict
0.5 star: Complete garbage not even worth taking home even if it's free
1 star: Crap that you shouldn't be using unless absolutely necessary
2 star: Is worth <=50% its price when compared to other best-in-class products
3 star: Is worth ~75% its price when compared to other best-in-class products
4 star: A decent choice alongside other strong competition; worth 100% of its price when compared to other products
5 star: Unmatched; worth 100% of its price (and more) and you have no choice but to get this

Due to the weird treble (I'm using treble and midrange somewhat interchangeably), the general presentation of R2000DB takes a hit in desktop scenarios. I will actually choose iLoud or at least consider both speakers as comparable options, despite the iLoud having worse bass and slightly more "cheap" in the midrange. Won't be surprised if there are 2.1 that can beat it in this department too. Although for 2.0... we know how the market looks like at this price point for 2.0. And who knows, maybe my faith in 2.1 is misplaced since I haven't heard a good one for years, and the only $300 2.1 I have right now (or more precisely it's a soundbar but wtev) sounds like... erm... you know.

This is the only thing that is stopping me from giving it a 5 star. Because the rest of the performance is stellar. Bass, for a 5-inch bookshelf at this price, this is as good as it gets, not in just volume but also in extension and quality, it surpassed my expectations. Not forgetting it also has bass controls. Tweeter, I have not mentioned anything regarding the tweeter in this review, perhaps except how clean it is. This is because I didn't notice anything wrong with the frequencies covered by the tweeter. Considering the tweeters of S-520 and Diamond 8.2 aren't weak sauce either, this is very good for an active speaker at this price.

Without listening to the speakers, the only alternative at this price range that I know is Swan D1080-IV, unless you want to count the slightly-higher-priced M200MKII. This puts the R2000DB at the 4.5 stars mark. After listening, the bass performance puts it at a 5.0, but the general presentation drags it down to 4.0, but I need to adjust for the better performance in the living room, so desktop/living room is 50/50 weightage, and the midrange is less problematic and also comparable to other speakers, so 4.375 stars. And the living room performance stomped all over Diamond 8.2 - the active version of which until recently still sold for $600+ per pair - so it deserves some bonus points. And because I don't want to deal with a score like 4.375 stars. So 4.5 stars it is.

And there are other things too, mostly not related to the R2000DB's sound but help to solidify the 4.5-star score. For one, it has a DAC, or rather, it throws the conventional concept of DAC out of the window. I will not discuss whether an expensive DAC is necessary in the audio chain, but for those of you who believe in DACs and are planning to buy one, good news, you just saved your money. And even for those who don't believe in DACs, the optical input is good to have since it avoids common-mode noise (or some of you call it ground loop) and for me it avoids bad contact which results in channel imbalance. And you can send two channels perfectly over a thin cable. How nice.

But anyway, multiple inputs, check. Remote control, check. Treble/bass controls, well D1080-IV has that too, but compared to other speakers, check.

1-inch Ru-Fe-B silk tweeter vs 0.8-inch metal tweeter of D1080-IV? Ah-hah...
Although iLoud's tweeter is 0.75-inch but the overall sound doesn't lose to R2000DB. But then iLoud's tweeter is silk...
D1080-IV has a slightly bigger 5.25-inch woofer, but woofer bass is the last thing that R2000DB is missing.

Class D amp with DSP (TAS5508 + output chip) vs class AB amp (TDA7296), this is... debatable...? A lot of the magic is only possible because the former is used. FWIW, TAS5112 is rated at 0.03% THD+N, while TDA7296 is rated at 0.005% THD.

So... yea... still hard to tell which is better. So the 4.5 star rating stays. But at least now I know how it sounds like.

Summary
4.5 stars - Marred by a perhaps major perhaps minor imperfection in the midrange, but hits beyond its price point in the other departments, with a shortage of competition. Bundled features provide bonus value proposition.

 

[wwenze]

Also I realized even if I change the title of the first post, it doesn't change the title of the thread. What am I supposed to do here

 

[wwenze]

I'm unsure whether I want to post this part. But I don't feel like I explained the midrange issue well enough. Yet it may be due to personal preferences skewing the perception so it may be inaccurate anyway and only makes things more confusing. So, you have been warned.

But/hence/so, my personal opinion, with the R2000DB on my table, right now, comparing it to the other speakers I have, it really sucks for vocals. Echoing what has been mentioned in the earlier parts of the review. And this opinion has been strengthened further with the Leviathan in a side-by-side comparison.

So, how did it reach a final score of 4.5? Or where did the initial 4.0 come from?

My honest personal opinion, for the vocals, based on the scoring system it would be a 2.5. The current price of R2000DB is SGD $250 to $300, so the vocal performance can be reached by speakers in the $125 to $200 range. As much as I want to give it a 2.0 (i.e. worth 50% or less), I do not have enough experience with speakers in the $100+ price range, and I'm not sure if speakers in the ~$200 range (you know, the usual names) actually do any better. And it is on-par with the Atom v.1, which definitely would cost way more than $125 if talking speaker + amp. OTOH Atom v.1 may or may not be considered a best-in-class product, depending on how much I value it. As in monetary value.

The problem comes from the handling of the criteria "best-in-class products". Defined as the popular/famous models at their various price points. The problem is, yes these best-in-class products are best-in-class based on overall performance at that price point, but are they best-in-class in a specific performance? (In this case vocals.) Or if I take another best-in-class product from the same price point, will that product perform better or worse than the R2000DB in vocals? I'm not sure. For the first scenario, we have for example z263 when sold at $150-200, and for the second scenario, we have Promedia 2.1 (at its lower selling price). I find it hard to believe R2000DB losing to either.

OTOH Swan M10 exists, and I am hopeful it might perform better. And I *think* M50W *might* have sounded better tho I can't really remember now but I don't remember this issue sticking out like a sore thumb. Tho M50W is also slightly more expensive and has its own pros/cons. It gets even more confusing when the Razer Leviathan also does vocals better (well, in some ways), or at the very least matches the performance. The problem is how much should I valuate Leviathan to be? $150? $200?

And furthermore, as much as I currently hate the sound, it is similar to Atom v.1, which is also similar to many 2.0 bookshelves at around their price ranges. And I lived with the Atom v.1 for years. This is the kind of sound that I won't complain too much about if I have to live with it, and probably become more and more accustomed to it over time. And if you came from cheaper computer speakers, you probably came from the same type of sound too. And if you don't have anything better to compare against, you may not notice the deficiencies.
And then but of course, if there are many speakers that sound like this, how am I to say that this sound is wrong? That is the reason why I did part 3 of the review.

In short, I don't like R2000DB's vocals and general presentation, but I can't find justification for penalizing it. But I am giving it 2.5 stars in this department anyway. And I consider this as 1/3 of the total performance. This is an otherwise 5 star product, so 2.5 stars penalty / 3 = 0.833 stars = round down to 4 stars. Then the further calculations from this point I have covered in the previous part of the review.

In other words, there is a chance that you may hate the vocals, there is a chance that you may hate it a lot. To the point that choosing other speakers that are inferior in other departments but better overall performance may be just as valid a choice. But there is also a chance that you may find nothing wrong with the vocals, and there is a chance that maybe I am just wrongly hating half of all speakers in existence. If you are buying the R2000DB, most likely the odds would be in your favor, but I feel I need to mention the chance of it going all wrong. A score of 4.0 to 4.5 - so a 10% to 20% chance of it going wrong I guess? Some consider that low, some consider that high, but definitely worth mentioning.

Add: Or, to yet again put in other words, some of you may feel the performance is 2.5/5/5 like me (where 2.5 is vocal/midrange/general presentation and the two 5s are the rest), some of you may feel the performance is 4/5/5. And for people who feel the performance is 2.5/5/5, some of you may be the type that would prefer to go for 4/4/4 or even 4/3.5/3.5 instead.

 

[Dhitto]

Do note that for those speakers with rear bass ports, they need a certain amount of backwall clearance to sound correct and get the best out of their bass output.

Usually for these kind of packaged speakers, they will have some recommendations in their instruction booklet. If not, it would be suggestible to experiment with the amount of that to get the best sound from the speakers.

Room setup/layout and acoustics will have very dramatic effects on the sound of various speakers. Do experiment a bit and see if the same conclusions apply though. Personal experience is that they can be quite a dramatic difference.

 

[ExcuseRMJ]

the finger print magnet gloss black plastic is a major turn off though

 

[wwenze]

Found a better EQ solution for me:

Give it a hearty boost in the 100Hz to 1kHz range. Below that isn't necessary because that is where the R2000DB actually has too much of (the supa-bass). There is however a half-octave between that supa-bass and 100Hz which you may want to tweak if you are good at it. (I'm not)
2kHz is probably optional depending on how clarity / nasal you want. It is also close to the crossover frequency already.

3dB boost in 100 to 1kHz makes the sound much better, now it is seriously approaching 5 star overall, or at least 4 star in midrange. But this kind of wide-band boost is what I am wary of - This might as well be a driver volume-matching issue, which means it is a personal preference, or it could be a room acoustics issue. Either way this means it is intended and not a parts quality issue. So it feels harder to score against now.

But, the score stays. Because as long as you're using it on a computer table like me, you have a chance of hating the sound, and the scoring included that in the consideration. And the scoring does not include the use of third-party EQ to improve the performance (I would be taking forever to review everything), though it does allow the use of any EQ the speaker comes with (which didn't help in this case).

But in any case, if you're finding the sound tinny and metallic, try the EQ mentioned above.

Update: Alternatively reduce at 4kHz and 8kHz instead.

 

[PsyClick]

How will the Presonus Eris 4.5 fare against this?

Thanks for the detailed review.

 

[wwenze]

Well don't expect the same accuracy of review if I don't have the two speakers beside each other

Eris E4.5 used to be my default recommendation for $250+ price range. The default sound is, well, is worth its price and size. And would be considered bright. But it comes with midrange volume adjustment so you can adjust it if you don't like it. And I value that a lot.

But the bass cannot keep up with 5-inchers, and worse is it doesn't even give you any way to boost it. So it just feels incomplete compared to the 5-inch studio monitors at $500. (Or nowadays there are some at $400.)

So later, when 5-inch speakers became available at this price range, I removed E4.5 from the default recommendation list.

R2000DB's problem is if your acoustics and your preferences make you hate the sound, you have to resort to external EQ. But I feel it is overall the more proper-sounding of the two, because it is hard to make up for E4.5's lack of bass. And it has more potential too - If you find the right EQ, or if it sounds ok for you as it is, it can challenge $500 products.

And this is without considering the "DAC" and BT and remote.

 

[finclk82]

Ts thx for this. I'm really considering this speaker of the Audioengine A5+. I know it is a significantly different price point but wondering if you have any comments on this? I'm sticking to 2.0 format due to save of space but am willing to pay more if the quality is significantly better.