• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

SMSL A100 Review (Stereo Amplifier)

Rate this amplifier:

  • 1. Poor (headless panther)

    Votes: 34 18.2%
  • 2. Not terrible (postman panther)

    Votes: 102 54.5%
  • 3. Fine (happy panther)

    Votes: 48 25.7%
  • 4. Great (golfing panther)

    Votes: 3 1.6%

  • Total voters
    187
D

Deleted member 46664

Guest
What?? The dead time distortion of class D amplifiers is a direct equivalent of class B and AB crossover distortion. Why do you think you measured all of that high order distortion mess in the spectrum? And that horrible HF linearity?

The article you cite is 23 years old. A lot of progress has happened since 1999.
For one thing the clock speeds on the outputs has more than doubled, now typically running 400khz. Another major change is post-filter feedback to correct output errors and not a little work on the "dead time" issue bringing it down to just a few nanoseconds.

Some of the best specifying amplifiers currently available are Class D.
 

pma

Major Contributor
Joined
Feb 23, 2019
Messages
4,591
Likes
10,727
Location
Prague
Dead time is a principial issue of class D, sometimes fixed better, sometimes not. It is a general point valid by nowadays. Maybe you are able to decipher it from measurements, maybe not. I would guess not. This forum has not many qualified readers, it is rather visited by laymen public.

BTW, @Douglas Blake , have not seen many measurements made by yourself on amplifiers. You may see mine on class D, class AB and class A amplifiers here on the forum. I do not care much about chat and talk, but I do care about results.

 
Last edited:

pma

Major Contributor
Joined
Feb 23, 2019
Messages
4,591
Likes
10,727
Location
Prague
Yes. That is one advantage of the "chip amps" (even the class AB ones) is that with everything on a common die it is far easier to match and manage transistors, especially the problems with unequal temperatures are reduced.
That’s a plain nonsense. The most important is thermal management and all the chip amps are poor in this. Have you ever designed a higher power amplifier sold in some quantities?
 
D

Deleted member 46664

Guest
BTW, @Douglas Blake , have not seen many measurements made by yourself on amplifiers. You may see mine on class D, class AB and class A amplifiers here on the forum. I do not care much about chat and talk, but I do care about results.

Have you ever designed a higher power amplifier sold in some quantities?


pissed cat.JPG


Nope... sorry... not playing.
 

SmackDaddies

Active Member
Joined
Mar 27, 2019
Messages
173
Likes
353
I'm wondering why you would buy such a low powered mediocre amp,
1) It is not mediocre
2) Sometimes you want a small amp for a tertiary system (I have five running systems and I certainly don't want to spend $350 as a minimum for an amp for my basement or kitchen or bedroom
3) Sometimes what you have is $100. So it something or nothing.
 

Ajax

Active Member
Forum Donor
Joined
Aug 14, 2018
Messages
248
Likes
802
Location
Byron Bay, Australia
1) It is not mediocre
2) Sometimes you want a small amp for a tertiary system (I have five running systems and I certainly don't want to spend $350 as a minimum for an amp for my basement or kitchen or bedroom
3) Sometimes what you have is $100. So it something or nothing.
Yeah, I get it and understand where you are coming from.

My own preference is to use a high quality streamer + high quality amp into a properly designed 4 way speaker selector, which has a direct (straight through) facility if only one pair is playing. I have a pair of speakers on my back patio, front deck and lounge room and simply select all or one at a time. My way obviously prevents different music selection at the same time in each zone, but that is not an issue for me.

No right or wrong, each to their own.
 
Last edited:

TheBatsEar

Major Contributor
Forum Donor
Joined
Jun 16, 2020
Messages
3,084
Likes
4,962
Location
Germany
Why is this amp discontinued tho? I've seen this amp being around for like 1 month and then it's already gone
The chips of the world are getting shorter, but the place on the PCB stays the same.
Such is life during a chip shortage.;)
 
Joined
Apr 5, 2020
Messages
89
Likes
89
Thanks for the updated review including the USB data! Although it looks like they cut corners with the DAC which is too bad because it seemed like a decent all-in-one budget option. Having to add a DAC to get decent performance makes this less appealing.
 

RandomEar

Senior Member
Joined
Feb 14, 2022
Messages
333
Likes
769
Quick comparison to similar-priced contenders (plus "bad" and "good" examples out of the price range):
THD+N Comparison_A100.png

Sources:
  • NAD D3020
  • SMSL A100
    [this thread]
  • Aiyima A07
  • Allo Volt+ D
  • Hypex NC122MP

I'd say it could be worse. But it most definitely could also be better ;)
 

dr_mick51

Active Member
Joined
Mar 12, 2022
Messages
200
Likes
209
Hello guys. A couple of months ago I bought this amplifier. And I did a review with teardown pictures:

SMSL A100 teardown

Amir, I noticed that this amp does not have any form of choke or inductor at the output. This means you cannot measure them directly connecting the analyzer to the outputs.
I noticed that the Loxjie A30 does have tiny chokes on the back on the PCB, therefore the measurements are better than compared to this amp.
The way that the Meris MA12070 chip works is that it uses the speaker coil inductance as part of the filter.
 

Attachments

  • Screenshot_2022-05-13-17-57-55-523_com.google.android.youtube.jpg
    Screenshot_2022-05-13-17-57-55-523_com.google.android.youtube.jpg
    495.4 KB · Views: 182
  • IMG_20220322_081755.jpg
    IMG_20220322_081755.jpg
    165.1 KB · Views: 192

dr_mick51

Active Member
Joined
Mar 12, 2022
Messages
200
Likes
209
Review updated with USB input:

index.php


Please read the brief notes on the testing of this in the review.
Amir, the reason why you see a big difference between this amp and the Loxjie A30 is because the Loxjie has 4 tiny chokes at the output filter which the SMSL lacks.
Some of the amplifier with the Meris chip include output inductors, the SMSL amps do not have these chokes (SA300, A12 and A100). The Merus chips are able to use the speaker coil inductance as part of the output filter.
Archimago on his review of the SA300 used 4 22uH inductors (according to the Merus Datasheet) for accurate measurements. If not , then you have a lot of distortion on the high frequencies because there is no filter.
Archivo SMSL SA300 review

Another article about this topic:
First look Merus
 

dr_mick51

Active Member
Joined
Mar 12, 2022
Messages
200
Likes
209
This is a review and detailed measurements of the SMSL A100 A100 stereo amplifier. It is on kind loan from a member. I can't find it on sale anywhere but seems to have cost US 100.
View attachment 204839
I like the simply but rather elegant/symmetrical design. Same for thoughtful rear connections:

View attachment 204841

Almost every manufacturer puts both speaker outputs together making it so hard to plug in the banana plugs. Here they are spaced apart which essentially solved that problem.

Inclusion of an AC power supply is unheard of in this price class and nicely declutters your desk.

Oddly the USB input is type A. At first I thought this was for external storage but manual talks about using this with a computer. I didn't have a type A to type A cable so didn't test this and stuck with Aux analog input.

The heart of the unit is Infineon MA12070 class D amplifier.

SMSL A100 Measurements
As usual we start with our dashboard:

View attachment 204844

Distortion is actually better than 90 dB but there is rising level of noise and high order spikes which drag it down to 76 dB. This puts is slightly below average:
View attachment 204845

Edit:
Request was made to measure using USB input. I had to use a special breakout box to interface to the unit given its type A connector. Here is the dashboard:
View attachment 205959

As you see, performance is degrade fair bit. My breakout may be causing some clock jitter but then again, if async USB is implemented, this should not be the case.

Noise performance is "OK:"

View attachment 204846

Multitone shows the same problem at high frequencies:

View attachment 204847

I expected frequency response to have load dependency but thankfully, it did not:
View attachment 204848

What ringing it has is above audible band which is good.

Crosstalk is very good, beating some home amplifiers:
View attachment 204849

Power output is naturally modest:

View attachment 204850
View attachment 204851

View attachment 204852

Sensitivity to high frequency input becomes very clear as we do power sweeps:
View attachment 204853

Unit didn't need any warm up.

Conclusions
The message here is a sum of good and not so good. I really like the frequency response for a class D amplifier and inclusion of AC power supply. Distortion on the other hand rises with frequency. I like to see perfection but we don't have such an example below $350 currently. Overall I would say not a bad attempt. Strange that despite having become available late last year, it is discontinued already. Maybe there are part shortages.

I am going to recommend the SMSL A100 as a budget amplifier.

----------
As always, questions, comments, recommendations, etc. are welcome.

Any donations are much appreciated using: https://www.audiosciencereview.com/forum/index.php?threads/how-to-support-audio-science-review.8150/
Amir, you might believe I'm biased, but this amp really sounds better than the Aiyima A04/A07 I own.
If you compare this amp to the Loxjie( which includes the four tiny SMD output chokes and makes measuring easier because there is nothing that's needs to be added) and remove all the high frequency distortion you end up with more than 90db THD, which is higher than the Loxjie.
Why SMSL does not include the choke inductors? Worldwide shortage of SMD choke? Cutting corners? I don't really know. But the lack of those chokes does not affect the listening experience because of what I already mentioned, the inductance of the speaker coil is generally greater than 22uH and are used as part of the filter. But if you add those chokes in the test the measurements are completely different.
Also, the A100 uses one OPA1678 as preamp which is better than other similarly priced Merus amps use.
@amirm , I'm afraid you don't have those chokes around, do you? It would be nice and fair to see the results with those chokes. It would pretty much measure the same or better than the Loxjie A30.
 

Attachments

  • IMG_20220514_073714.jpg
    IMG_20220514_073714.jpg
    419.4 KB · Views: 176
  • IMG_20220514_071026.jpg
    IMG_20220514_071026.jpg
    121.5 KB · Views: 178
  • IMG_20220514_070936.jpg
    IMG_20220514_070936.jpg
    125.1 KB · Views: 175
Last edited:

RandomEar

Senior Member
Joined
Feb 14, 2022
Messages
333
Likes
769
How to evaluate the Crosstalk? Why is it very good?
The rating "very good" is certainly subjective. The A100 is about 30 dB better than the crappiest amp Amir could find, but also still about 22 dB away from the best of the best. Also, more than 80 dB of separation around 3 kHz is very, very unlikely to ever cause any audible crosstalk, seeing the typical application scenario of this amp (that is: desktop usage, not PA-level blasting). I'd say that's at least "good", but not quite "excellent".
 

dr_mick51

Active Member
Joined
Mar 12, 2022
Messages
200
Likes
209
Hello guys, I want you to realize about the bargain this amp is.
I'm going to use measurements from other amps to let you know of the capabilities of this amp.

First, we have the Loxjie A30 which uses the same chip but worse opamps. But it has 4 small chokes on the output filter.
This is how it measured under the same conditions as the A100 (almost 29db Gain)
1652969701763.png


After reducing the gain to merely 8dB, this are the results:
1652969651661.png


Now we have the SMSL A100 which does not have the 4 chokes at the output filter:
1652969864699.png

Now, how this amp wouls measure if: One, we reduce the gain to 8dB as well. Two, if we add 4 chokes at the output.
Also note the absense of Third Order Harmonic distortion, and the dominance of the Seconf order harmonic distortion:
1652970103221.png


Another thread by @RandomEar Help with odd problem: SMSL AO200 emits hissing sound only when connected to a tweeter
shows the effect of using a similar Merus amp connected directly to a Tweeter with not enough inductance to act as a filter.

Now, let's see at the Topping PA3s with a lower Gain than the A100 in balanced mode:
index.php

You can see a dominance of Third order harmonic distortion.
I have not seen full pictures of the inside both front and back of this amp but for me its has some kind of inductors or chokes at the output.

Just think about this a little.
 

dr_mick51

Active Member
Joined
Mar 12, 2022
Messages
200
Likes
209
Review updated with USB input:

index.php


Please read the brief notes on the testing of this in the review.
Amir, have you considered the possibility that the absence of chokes/inductors at the output filter is affecting the performance of this amp specially when using the RCA inputs? It could be much better, I know that the USB and BT inputs are crap, i don't care about them..

Maybe SMSL realized this and that's why the amp was out of stock shortly after launch. Maybe they will be adding those choke and voila!.
 

TheBatsEar

Major Contributor
Forum Donor
Joined
Jun 16, 2020
Messages
3,084
Likes
4,962
Location
Germany
Amir, have you considered the possibility that the absence of chokes/inductors at the output filter is affecting the performance of this amp specially when using the RCA inputs? It could be much better, I know that the USB and BT inputs are crap, i don't care about them..
Well, nobody (except you maybe, you seem to be knowing what you are doing) will change, remove or add SMD parts on a cheap amp to get less distortion.

If the measurements are bad, they are bad because of the manufacturing. No value in changing things around until they measure better if there are other manufacturers that do a better job. You can't make a nun out of a pig. :p

USB and BT seem to be indeed an afterthought. But it doesn't have to be like that, we have seen others having acceptable USB/BT performance in low cost devices.
 
Top Bottom