• 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!

Topping A90 Discrete Review (Headphone Amp & Preamp)

Rate this Headphone Amp/Pre-amp

  • 1. Poor (headless panther)

    Votes: 16 3.9%
  • 2. Not terrible (postman panther)

    Votes: 14 3.4%
  • 3. Fine (happy panther)

    Votes: 70 17.2%
  • 4. Great (golfing panther)

    Votes: 306 75.4%

  • Total voters
    406

nagster

Senior Member
Joined
Jan 24, 2021
Messages
367
Likes
598
100 ohm is quite low...if A90D input impedance is more than 1K...this should meet the 1:8 ratio so perhaps it is not a mismatch of impedance. Anyway, A90 input impedance is found here, and it is stated for balanced, it is 2K and 10K for unbalanced. If this is the same too for A90d, then there shouldn't be an issue of impedance mismatching.
 

raif71

Major Contributor
Joined
Sep 7, 2019
Messages
2,342
Likes
2,542
so, it is similar to A90, right?
 

nagster

Senior Member
Joined
Jan 24, 2021
Messages
367
Likes
598
so, it is similar to A90, right?
Assuming his answer is 1kHz, the XLR input is the same as the A90 and the RCA input is about 0.2% lower than the A90.
 

bigLP

Member
Forum Donor
Joined
Feb 8, 2020
Messages
71
Likes
72
Location
Rochester, NY
I finally got around to comparing the A90 D and my schiit freya s. This was as pre amp only, not HP amp though. Listening was not blind, but observations were kept private until end. I level matched(72 db) using a chesky test tone and cell phone app. then used a cd quality chesky recording, node 2i to modius e to pre amps to ATI 542 ncore to ls50.. All balanced. I told my son to listen to tonality of instruments and "image", which I purposely didnt define.
My 21 year old son and I listened to same 30 seconds of music 12 times on the a90d in high gain. It took about 30 seconds to switch cords. Then 12 times on schiit freya s in its active mode. I told him to write down, in short phrases, how the two compared.
we both noted: more bass for a90D, slightly hot treble for Freya, cellos sounded more prominent on a90d, violins on schiit(elevated highs?). schiit wider, bigger image
contradictions: subjectively we preferred different pre amps, I distinctly noticed a smaller more forward image on the a 90d and better and better bass extension. I thought the schiit's image depth was subjectively better, but more diffuse.......my son said he did not really understand my lingo or agree with what i heard. he thought they both sounded good, and not real different in the end. I will not impart that whole vernacular on him. This part belongs in psychoacoustics probably, but Noam Chomsky and others have shown that the vocabulary(and understanding of it) we have can influence/limit our cognition in those areas. I sometimes worry I'm hopelessly biased by 30 years of audio reading. i dont want to pass that on.
On tuesday i will cover rack, so it is blind for my son. Same process, but only 6 plays each. by memory and his own written description i will see if he can tell them apart.

I'm pretty sure i wouldn't sell either. I've had no operational problems with a90D, but have not even listened to the headphone amps..........
 

Arnas

Active Member
Joined
Nov 9, 2020
Messages
169
Likes
229
I finally got around to comparing the A90 D and my schiit freya s. This was as pre amp only, not HP amp though. Listening was not blind, but observations were kept private until end. I level matched(72 db) using a chesky test tone and cell phone app. then used a cd quality chesky recording, node 2i to modius e to pre amps to ATI 542 ncore to ls50.. All balanced. I told my son to listen to tonality of instruments and "image", which I purposely didnt define.
My 21 year old son and I listened to same 30 seconds of music 12 times on the a90d in high gain. It took about 30 seconds to switch cords. Then 12 times on schiit freya s in its active mode. I told him to write down, in short phrases, how the two compared.
we both noted: more bass for a90D, slightly hot treble for Freya, cellos sounded more prominent on a90d, violins on schiit(elevated highs?). schiit wider, bigger image
contradictions: subjectively we preferred different pre amps, I distinctly noticed a smaller more forward image on the a 90d and better and better bass extension. I thought the schiit's image depth was subjectively better, but more diffuse.......my son said he did not really understand my lingo or agree with what i heard. he thought they both sounded good, and not real different in the end. I will not impart that whole vernacular on him. This part belongs in psychoacoustics probably, but Noam Chomsky and others have shown that the vocabulary(and understanding of it) we have can influence/limit our cognition in those areas. I sometimes worry I'm hopelessly biased by 30 years of audio reading. i dont want to pass that on.
On tuesday i will cover rack, so it is blind for my son. Same process, but only 6 plays each. by memory and his own written description i will see if he can tell them apart.

I'm pretty sure i wouldn't sell either. I've had no operational problems with a90D, but have not even listened to the headphone amps..........
In my opinion: Dacs and amps sound wider when there is more distortion... The cleaner the signal the stronger the attack, punch, transients or whatever people gonna call it...
Low thd = everything is sharper and POPS a lot more. Low thd dont have that pop and thus sound more distant, more laidback or artificialy wider...
More bass can be explained by lower noise of a90 and hot treble from Freya can be explained by Noise power supply.
 

tomelex

Addicted to Fun and Learning
Forum Donor
Joined
Feb 29, 2016
Messages
990
Likes
572
Location
So called Midwest, USA
BLIND TEST

Check out this simple Blind test, this young gentleman has good enough ears to hear the differences between the A90 and A90D. And he picks up on the differences in seconds not hours or days or weeks, bla bla , and I have always said if you can not hear a difference within a minute or so then it is just a guess after that if you do not admit you do not hear a difference.

Appreciate his work very much, and his observations.

 

tomelex

Addicted to Fun and Learning
Forum Donor
Joined
Feb 29, 2016
Messages
990
Likes
572
Location
So called Midwest, USA
He doesn't. His methodology is flawed in multiple ways which defeats the whole purpose of blind listening tests.
Here's how to do proper blind tests:

Edit: his video was discussed here

Are you claiming he did not hear a difference?

Look, I am totally aware of "proper" DBT, but, his test is about as good as you could expect from at home not using more advanced test gear than an iphone.

Let us assume he is not "cheating", if you can go that far. If we can assume he is not cheating, what technical reason(s) can you produce that says it is impossible for him to hear a difference? And to make this more interesting, can we assume for the purposes of discussion that the level match was 0.1dB.

If you are assuming he is cheating or it was simply a level match issue (he did "weakly" level match twice) then that would be the end of our discussion so if thats the case then at least thanks for pointing out the link to the other thread discussing this test, I am a little late to the table on this one! These threads can be so long sometimes it is hard to keep fully up with them.
 

Veri

Master Contributor
Joined
Feb 6, 2018
Messages
9,597
Likes
12,039
Are you claiming he did not hear a difference?

Look, I am totally aware of "proper" DBT, but, his test is about as good as you could expect from at home not using more advanced test gear than an iphone.

Let us assume he is not "cheating", if you can go that far. If we can assume he is not cheating, what technical reason(s) can you produce that says it is impossible for him to hear a difference? And to make this more interesting, can we assume for the purposes of discussion that the level match was 0.1dB.
But it wasn't within 0.1dB so if he could reliably hear a difference it could very easily be just volume making things appear different.
Also the YouTube comments in the video point out the flaws too. The video's unlisted... it's obviously not perfect.
 

staticV3

Master Contributor
Joined
Aug 29, 2019
Messages
7,844
Likes
12,526
Are you claiming he did not hear a difference?
I'm claiming he wouldn't have been able to hear a difference, were the experiment conducted properly.
his test is about as good as you could expect from at home not using more advanced test gear than an iphone.
That's an incredibly low bar.
Let us assume he is not "cheating"
But he is, though not necessarily on purpose/knowingly
what technical reason(s) can you produce that says it is impossible for him to hear a difference
The known audibility thresholds of noise and distortion of DACs and Amps, which are explored and discussed here
And to make this more interesting, can we assume for the purposes of discussion that the level match was 0.1dB.
No, why would we? Mr Valour has given us no reason of believing it was.
 

tomelex

Addicted to Fun and Learning
Forum Donor
Joined
Feb 29, 2016
Messages
990
Likes
572
Location
So called Midwest, USA
But it wasn't within 0.1dB so if he could reliably hear a difference it could very easily be just volume making things appear different.
Also the YouTube comments in the video point out the flaws too. The video's unlisted... it's obviously not perfec

Yes, we do not know how close he came to 0.1dB or even 1dB which is hard to hear a difference. I was happy to see someone trying to do some science. It made me giddy with joy as he was displaying some courage and putting his reputation on the line. Volume difference is most likely what happened given his iphone test equipment. I do appreciate the difference was enough for quick and confident outcomes.
 

tomelex

Addicted to Fun and Learning
Forum Donor
Joined
Feb 29, 2016
Messages
990
Likes
572
Location
So called Midwest, USA
I'm claiming he wouldn't have been able to hear a difference, were the experiment conducted properly.

That's an incredibly low bar.

But he is, though not necessarily on purpose/knowingly

The known audibility thresholds of noise and distortion of DACs and Amps, which are explored and discussed here

No, why would we? Mr Valour has given us no reason of believing it was.

OK, we both agree it is not a very scientific test. But can we learn something from it....

But now, can you tell me based on your claim if the test was done properly, considering everything was proper blind, even ABX blind, by say Amir, how would you ensure that the levels as per the volume control were the same so we could rule that out as the last reason that he would not hear any differences?
 

staticV3

Master Contributor
Joined
Aug 29, 2019
Messages
7,844
Likes
12,526
how would you ensure that the levels as per the volume control were the same
I'd play an identical 60Hz or 1kHz sine wave on both Amps and measure the voltage on the output of both Amps across the load, either with my Aneng AN870 multimeter, or with my E1DA Cosmos ADC.
 

iamsms

Member
Joined
Dec 14, 2020
Messages
35
Likes
126

staticV3

Master Contributor
Joined
Aug 29, 2019
Messages
7,844
Likes
12,526
Other than creating an opening in the headphone cable, is there any way to measure the voltage in a loaded condition
This kind of splitter would do the job:
-910988892.jpg
I'd use my Cosmos load board, which has sensing points that are easy to probe
 

tomelex

Addicted to Fun and Learning
Forum Donor
Joined
Feb 29, 2016
Messages
990
Likes
572
Location
So called Midwest, USA
I'd play an identical 60Hz or 1kHz sine wave on both Amps and measure the voltage on the output of both Amps across the load, either with my Aneng AN870 multimeter, or with my E1DA Cosmos ADC.

I would say that is almost no test at all. What about the rest of the spectrum? What about measuring the currents as well? Your meter has no specs on the FR of its AC measuring, although in this test since you are just comparing it would not matter...but... Are the headphones on the guys head or just sitting on the table for these tests? What power levels are the tests being done at, especially if you are comparing music? Is the music monoed or stereo? What temps are the amplifiers at? Chip amps and discrete amps behave differently via temps as feedback has to work harder and component spacing.

How does the output impedance vary over FR between the two amps and what is the sustained reserve power of these amps and on and on. This kind of testing is not easy, as Amir said, and so I was appreciative of the guy who tried to at least do his version of a blind test. As is always stated, testing, quality testing, knowledgeable testing is hard, especially when amplifiers are trying to control complex loads. When you do each amp test do you start out with the headphones at the same temperature and not having just been tested with the other amp?

Finally, when you did all that (and more complex tests), how much of the measured differences, when taken in total will be enough to be heard, thats where the ear comes into play.

Until you understand by lots of measurements (most of above and more) how these two entirely different amps output circuitry react into a particular complex load, you just can not look at measurements and make a definite pronouncement other than same comparisons of tests with other units when it comes to driving complex loads. Amp and transducer pairings are always particular to each unique set up. Inferences can be made though, for example an amp putting out 1 watt into a resistor should be more capable of one putting out 0.1 watt into a resistor all other things being equal..

Insufficient measurements can lead to wrong assumptions and opinions and get objectivists into trouble, the same as for subjectivists that refuse to listen to even the least rigorous "blind" test arrangements.

I am ordering ordered a Topping A90 discrete as it is certainly value for money and can also do duty as test equipment!
 
Last edited:
Top Bottom