NAD M23 Stereo Amplifier Review

[Julf]

Sure, here is the link to LTS that performs such tests.

Ljudtekniska Sällskapet

Ljudtekniska Sällskapet är en oberoende förening som verkar för högsta möjliga kvalitet såväl i fråga om hifi-anläggningar som musikproduktioner av alla slag...

www.lts.a.se

Hope you understand Swedish

No problem, it is my native language. Too bad there was no link to the original test (in Swedish).

It was 14b sst model, not 4b.
Found a link in English:

https://www.hififreak.nl/wp-content/uploads/2020/01/Swedish-test-Bryston-14B-SST.pdf

Where do I even start with that test? It wasn't a double blind test, just single, but with everybody allowed to discuss the differences before submitting their opinion, and with an extended sighted listening and discussion too.

Here is an interesting post from ASR member Blumlein 88 who found Spectral amp transparent while other amps not:

Limitations of blind testing procedures

Do you think there is something that can't be electrically measured that distinguishes one amp from another? Its more the definitive interpretation of the data, where comprehensive test data is actually available, and its relation to sound. This is not relevant and continuing down that line...

audiosciencereview.com

That thread also contained the observation from the Swedish tests:

"So my stance is that there certainly exists sonic differences between power amps that drive a reasonably difficult speaker load that is clearly detectable in DBTs, but in reality, they are so small that I do not concider them an issue because there are other issues that impact sonic performance orders of magnitues more."

Anyway, A few non-scientific "tests" won't really determine anything this way other. What we need is proper studies, with peer review, verification and replication.

 

[WaterFlame]

If you wish, but it wouldn't be my first choice. @pkane 's software is far better and more versatile. Transparent high-quality recordings of amplifier outputs are trivially easy to make these days to make a file.

Any clips you wish can be made using any standard editor.

Stop making excuses about why other people aren't chasing your uninformed objections. Go put in the work.

This will not cut it. I want to capture interaction with speakers and how well speakers are driven by the amp.

 

[WaterFlame]

No problem, it is my native language. Too bad there was no link to the original test (in Swedish).

Where do I even start with that test? It wasn't a double blind test, just single, but with everybody allowed to discuss the differences before submitting their opinion, and with an extended sighted listening and discussion too.

That thread also contained the observation from the Swedish tests:

"So my stance is that there certainly exists sonic differences between power amps that drive a reasonably difficult speaker load that is clearly detectable in DBTs, but in reality, they are so small that I do not concider them an issue because there are other issues that impact sonic performance orders of magnitues more."

Anyway, A few non-scientific "tests" won't really determine anything this way other. What we need is proper studies, with peer review, verification and replication.

The flaws in the protocol you highlighted would actually reduce the lekelihood of the transparent result of Bryston. Also if you subscribe to lts and read the article in Swedish, you would see the protocol is actually quite robust there. They first test sighted and then blind with different signals. They had no problem picking a difference with other amps but with Bryston they could not even hear it in the chain when sighted. Not sure i understand your point here. How exactly this protocol led to positive results for all other amps but not bryston ?

 

[WaterFlame]

No problem, it is my native language. Too bad there was no link to the original test (in Swedish).

Where do I even start with that test? It wasn't a double blind test, just single, but with everybody allowed to discuss the differences before submitting their opinion, and with an extended sighted listening and discussion too.

That thread also contained the observation from the Swedish tests:

"So my stance is that there certainly exists sonic differences between power amps that drive a reasonably difficult speaker load that is clearly detectable in DBTs, but in reality, they are so small that I do not concider them an issue because there are other issues that impact sonic performance orders of magnitues more."

Anyway, A few non-scientific "tests" won't really determine anything this way other. What we need is proper studies, with peer review, verification and replication.

Ok, I think I need to clarify what I meant in the previous post.

Usually, the blind tests in audio are not very sensitive for picking up small differences. It is very rare that somebody can identify a difference in which case your argument about consulting with other prior to submitting the answer would be ok since it can make you believe that you actually heard the difference. In case of the test their are doing, they manage to identify amplifier in the chain almost always. So if there was any chance that Bryston was not transparent, it would have biased other participants towards the positive result. Since they could not identify Bryston in the chain, the results is even stronger because of it.

I have reread the article. They consulted only during the sighted part of it. Not during the blind phase when they submitted the answers. The pattern of their answers also is totally random which means it was a pure guess.

Also, why do you insist on double blind? It is not for publication in the journal, single blind is more then good enough for me to begin to doubt consensus in ASR regarding amps.

 

[HarmonicTHD]

This will not cut it. I want to capture interaction with speakers and how well speakers are driven by the amp.

Then record the output with the speakers attached. Do it for two amps and run it through Deltawave. If you can’t hear the result you won’t hear a difference in an ABX test either. Not so difficult. Actually much easier than a blind test and much more reliable and less prone to mistakes.

 

[Julf]

In case of the test their are doing, they manage to identify amplifier in the chain almost always.

I would like to see what the "almost" means, and the statistical analysis. And just to clarify, "identify" in this case means getting an A/B choice right, not going "ah, yes, I can hear that this is another Quad" or something.

Also, why do you insist on double blind? It is not for publication in the journal, single blind is more then good enough for me to begin to doubt consensus in ASR regarding amps.

I insist on double blind, because it is the only reliable way, as recognized in the audio science community for more than 50 years. If single blind is more than good for you, that is your choice, but it makes me think you would probably doubt ASR consensus anyway.

For proper, professional ways to do tests, I tend to recommend the industry standard, ITU BS.1116 : Methods for the subjective assessment of small impairments in audio systems.

 

[WaterFlame]

Then record the output with the speakers attached. Do it for two amps and run it through Deltawave. If you can’t hear the result you won’t hear a difference in an ABX test either. Not so difficult. Actually much easier than a blind test.

Thanks for the tip. Will this capture how speaker drivers react to amplifier and how well they are being driven or just how amplifier reacts to the reactive load connected to it?

 

[SIY]

This will not cut it. I want to capture interaction with speakers and how well speakers are driven by the amp.

OK, why is that hard to do? Record the signal at the speaker terminals.

Excuses, excuses, excuses.

 

[IAtaman]

Thanks for the tip. Will this capture how speaker drivers react to amplifier and how well they are being driven or just how amplifier reacts to the reactive load connected to it?

 

[WaterFlame]

Thanks for the discussion. Going back to M23, I am happy to answer questions about it. Has been a great product for me.

 

[WaterFlame]

View attachment 348719

M23 has been stable into difficult load yes. Do not see the graps for hypex.
However, this does not measure distortion at 1w, 5w, it measures only max output into those loads (is it 1% thd used as a cutoff or a knee?)

 

[WaterFlame]

I would like to see what the "almost" means, and the statistical analysis. And just to clarify, "identify" in this case means getting an A/B choice right, not going "ah, yes, I can hear that this is another Quad" or something.

I insist on double blind, because it is the only reliable way, as recognized in the audio science community for more than 50 years. If single blind is more than good for you, that is your choice, but it makes me think you would probably doubt ASR consensus anyway.

For proper, professional ways to do tests, I tend to recommend the industry standard, ITU BS.1116 : Methods for the subjective assessment of small impairments in audio systems.

So far, it has been only Bryston that was shown to be fully transparent, hence almost always. For all other amps, they have managed to guess with stat significance that amp was in the chain.

 

[CleanSound]

So far, it has been only Bryston that was shown to be fully transparent

What? Where in the world are you getting this from? You mean the most transparent for high power output? Or in general?

 

[WaterFlame]

What? Where in the world are you getting this from? You mean the most transparent for high power output? Or in general?

I meant in the the tests of LTS. I cannot find any recent tests they did, especially for class d amps. Of course, there are many transparent amps. I brought up lts just to illustrate the point that not every competently designed amp with sinad of 80 is transparent.

 

[HarmonicTHD]

Thanks for the tip. Will this capture how speaker drivers react to amplifier and how well they are being driven or just how amplifier reacts to the reactive load connected to it?

You are comparing amps. This will tell you all you need to know.

You seem to be quite reluctant doing the one test which can clear it all up. Seems like you are clutching for straws to prove your erroneous perception.

 

[Julf]

I meant in the the tests of LTS. I cannot find any recent tests they did, especially for class d amps.

Maybe there is a reason they don't do the tests anymore?

Of course, there are many transparent amps. I brought up lts just to illustrate the point that not every competently designed amp with sinad of 80 is transparent.

And how do we, in your opinion, know which ones are competently designed?

 

[WaterFlame]

You are comparing amps. This will tell you all you need to know.

You seem to be quite reluctant doing the one test which can clear it all up. Seems like you are clutching for straws to prove your erroneous perception.

So you want me to buy another amp in addition to my m23 just to make you take my arguments more seriously? No offense but no thanks. Lts tests i was referring to are much better designed than most of the blind tests published by members here. If this was not motivating you to try to understand amplifier coloration a bit better then i do not think it is worth for me investing my time in tests that will make you happy. Especially considering assymetry here between how negative results and positive results are being scrutinized here. I know, burden of proof you would say.

I have done single blind tests (preference-based) between Bryston 3b3 vs Hypex Nilai, between Benchmark ahb2 + topping dx7 pro + and hegel h190+ dx7 pro +, nad m33 and m10, lyngdorf tdai 1120 vs hegel h95. Those were informative to me but will not pass scrutiny here although volume was matched with a voltmeter.

From now on, I will reserve my replies to practical questions anout products I own or have owned in the past. I just hope more open-minded members, such as pma, KSTR will continue to post on asr. They are true researchers driven by curiosity.

 

[WaterFlame]

Maybe there is a reason they don't do the tests anymore?

And how do we, in your opinion, know which ones are competently designed?

Pma posts about amplifier design and testing can give you some hints. Also there was a topic about audibility thresholds which are much higher than sinad for just middle of the road class ab amps. Nad m23 is a very good amp and clearly competently designed

 

[HarmonicTHD]

So you want me to buy another amp in addition to my m23 just to make you take my arguments more seriously? No offense but no thanks. Lts tests i was referring to are much better designed than most of the blind tests published by members here. If this was not motivating you to try to understand amplifier coloration a bit better then i do not think it is worth for me investing my time in tests that will make you happy. Especially considering assymetry here between how negative results and positive results are being scrutinized here. I know, burden of proof you would say.

I have done single blind tests (preference-based) between Bryston 3b3 vs Hypex Nilai, between Benchmark ahb2 + topping dx7 pro + and hegel h190+ dx7 pro +, nad m33 and m10, lyngdorf tdai 1120 vs hegel h95. Those were informative to me but will not pass scrutiny here although volume was matched with a voltmeter.

From now on, I will reserve my replies to practical questions anout products I own or have owned in the past. I just hope more open-minded members, such as pma, KSTR will continue to post on asr. They are true researchers driven by curiosity.

Where did I say to buy another amp? You claim, that you compared several ones.

How else do you want to compare amps unless you own them or have access to?


Again recording the output of each amp while attached to a speaker and running the result through Deltawave will tell you with much more certainty about audible differences than an error prone blind test. And if you don’t have the other amps anymore than record the one you have and obviously don’t run it through Delawave but simply post the FR and we can see if there is an audible load dependence. Plus you will have facts to post here and not anecdotes.

As it is very very doubtful that you heard differences btw all these amps you described but more likely have an error in your blind test (which once again is not even needed if you would have done what now several of us proposed multiple times). Do it right if you want to be taken seriously.

All seems like very weak excuses.

But I am wasting my time. I am out.

 

[pogo]

This could also be interesting for M23 users: Link