Speed Reading DAC Audio Measurements

[amirm]

Thank you, very useful video. You know better how many new users and visitors read the reviews, it may be useful for them to have an (unobtrusive) link on each review to these kind of videos for each category, assuming you make one for each category of the products you test (ninja style of saying please do so ).

That is exactly my plan.

 

[amirm]

1kHz SINAD / THD+N is not a useful predictor, especially not for IMD.
One can design two almost identical amps with the same superb 1kHz THD (THD, not SINAD or THD+N). Let one of them have 20dB less loop gain for correction at 20kHz and it will have much worse IMD.

Best universal test signal which allows much more insight than 1kHz single tone is a 7kHz+13kHz twin tone when measured with a 22kHz bandwidth, in an IMD (not IMD+N) analysis.

Lumping noise and distortion into one single value is a bad idea and always has been.

IMHO and IME, there are no shortcuts to a complete technical and mathematical understanding of what's going on when interpreting measurements. Speed reading and jumping to quick conclusions based on a single number is bound to fail.

Did you watch my video? It seems you missed the explanation I provided.

 

[Sokel]

Best universal test signal which allows much more insight than 1kHz single tone is a 7kHz+13kHz twin tone when measured with a 22kHz bandwidth, in an IMD (not IMD+N) analysis.

Like this or I need to include any other parameters?

 

[voodooless]

Speed reading and jumping to quick conclusions based on a single number is bound to fail.

You can come really close though. A multitone at 44.1 kHz sample rate and visualize up to 100 kHz would reveal almost everything you'd need to know in a single graph. it shows a fair approximation of frequency response, gives a good measure of IMD, shows filter attenuation and possible issues, as well as distortion and noise products, all in a single graph. Yes, you can dig deeper, but I'm pretty sure this will be a good predictor of the other measurements 99% of the time. The current 192 kHz multitone fails to show a great many issues simply because they are shifted to the right of the visual area.

 

[xaviescacs]

1kHz SINAD / THD+N is not a useful predictor, especially not for IMD.
One can design two almost identical amps with the same superb 1kHz THD (THD, not SINAD or THD+N). Let one of them have 20dB less loop gain for correction at 20kHz and it will have much worse IMD.

Best universal test signal which allows much more insight than 1kHz single tone is a 7kHz+13kHz twin tone when measured with a 22kHz bandwidth, in an IMD (not IMD+N) analysis.

Lumping noise and distortion into one single value is a bad idea and always has been.

IMHO and IME, there are no shortcuts to a complete technical and mathematical understanding of what's going on when interpreting measurements. Speed reading and jumping to quick conclusions based on a single number is bound to fail.

If such numerical representation of other measurements like IMD and linearity could be build, one could in fact check if the models to predict those numbers from SINAD are good or not. I'm far from understanding all technical details here, but if some variables are highly correlated there is some relation between them and one can use one of the two to predict the other. I'm sure there will be cases in which this is not possible because the device has some flaw or whatever, but then thanks to the model one could assess the degree of deviation with respect to it.

I'm not proposing anything, just following that reasoning of having a predictor variable.

 

[Boxermotor]

Ah, good word! I am going to steal it for future discussions.

Would this not be a predictor variable, which is used to forecast rather than having a direct correlation to the outcome (as opposed to an independent variable which can be tested)? A good or bad SINAD doesn't influence corresponding results in other areas, but may infer another variable that does (e.g. quality control process, attention to detail, testing, etc.).

 

[Multicore]

I watched the video all the way through now and I still don't understand the jitter test.

 

[xaviescacs]

Would this not be a predictor variable, which is used to forecast rather than having a direct correlation to the outcome? A good or bad SINAD doesn't influence corresponding results in other areas, but may infer another variable that does (e.g. quality control process, attention to detail, testing, etc.).

If two variables have a strong correlation (linear or non-linear) one can predict the value of one given the value of the other. This is always true as long as you can build a model.

Of course the root causes of something, the causality chain, the causal graph, may look different, but in the end what matters is what you want to achieve. What amir says and more or less can be verified reading some reviews is that good SINAD usually implies good other measurements, so there is a correlation between them, and hence one can define the SINAD to be a predictor of the others. What is dependent and independent is set by design.

 

[AdamG]

Ah, good word! I am going to steal it for future discussions.

We use the term “Barometer” to quickly describe the very complex weather patter and prediction of what’s to come. Choices…Speakers next. Like others have said you use lots of different measurements and graphics to do this that are confusing without context. My as well be ink blots…

 

[Dennis_FL]

I had a couple of DACs that didn't "handshake" well. For instance, I would switch from say Tidal to Amazon Music or PCM to DSD and I would get loud white noise. Is that a measurable anomoly or was I doing something wrong?

 

[JktHifi]

Emotiva vs Zidoo, and money lies again

 

[Ken Tajalli]

I am baaaaaaack!!! A new video in a series to become proficient in reading measurements, in this case, DAC measurements. The video itself is rather long but hopefully once you watch it, you can scan a text review in a minute or two.

Excellent and thank you.
Just one observation regarding the 32 tone test.
I always thought a 32 tone test but with pink-noise equivalent levels would be much more informative.
In real life, no DAC amp or speaker would ever encounter anything resembling the 32 tone, all at the same level.
Thank you for all your efforts.

 

[neoheresy]

Very, very nicely described to an average audio playing hardware user. It made a ton of things much more clear to me. Extra appreciation note goes for your proactivity and answering the "unasked questions" - like, how do some of the graphs correlate with each other. Great job, Amir!

 

[Ken Tajalli]

Thank you so much everyone. What would you like the next speed run video be about? I am thinking of either headphone amplifiers or speakers.

Headphone measurements.
Specially on how difficult and approximate the measurements can be.

 

[IAtaman]

IMHO and IME, there are no shortcuts to a complete technical and mathematical understanding of what's going on when interpreting measurements. Speed reading and jumping to quick conclusions based on a single number is bound to fail.

I agree completely if we are talking in isolation. If we take into consideration where we came from and who is the audience of the video however, I think you might also agree that it is a good general indication of an audio electronic product's performance, if the audience is looking for an easy to grasp metric to compare products.

For where we come from, let us not forget before ASR popularized SINAD the only other popular metric was a) price, and b) what did the guy with lots of speakers and oscilloscopes in the background said about this product. In absolute terms SINAD might not be great but compared to that SINAD is a very good metric.

And for the matter of audience, I don't think the target audience of these videos and reviews are people like you, who know and understand the topic in detail. It is consumers and hobbyists who just want to have high level understanding of the topic without getting into all the nitty gritty details. Again for that general purpose, the electronic measurements provided are great.

Having said all that, I think measurements have reached to a certain level of popularity now, people do speak about THD+N and IMD and linearity etc, and believe it would be good to create a more multidimensional performance comparison, not unlike the ones Julian Krause does for Audio Interfaces, to tell a bit of a detailed story of the products rather than a one dimensional best case SINAD one.

 

[Robbo99999]

Nice, I think these kind of videos are crucial for newbies that visit this site, especially if they are in an easy to find prominent position when they first come to the site. I recommended this site to someone recently, and when I asked them a week later what they thought to it they answered that they thought it "was over their head" and "they didn't know who to believe" amongst all the opinions. So videos like this are useful for newbies, but I think they need to be put in a prominent position so anyone who lands here can quickly & easily see where they need to go to find explanations on how to understand the various elements of reviews, whether they be DACS / amps / headphones / speakers.

EDIT: in fact maybe it would be good if there was a seperate tab right at the top of the page that was entitled as something like one of the following: "Instructional Videos" / "Tutorials" / "How to understand this site" / "How to understand the technicalities" / "Confused? Look here" / "Confused about the Technical Elements, look here" / "Technical Explanations" / "Technical Fundamentals" / "I don't understand this site? Click Here".....whatever, it could be described as whatever makes the most sense to the most people, but you could put in an extra tab at the top of the site so newbies can instantly see where they need to go for explanations:

 

[Timcognito]

@amirm Just finished the tutorial and can't wait to showoff my specmanship to ne'er-do-well "audiophile" friends. Keep it up Amir. Thanks again.

I too vote for more on speakers but maybe few tutorials on those like one on spin-o-rama, one on EQ how to, battling room effects and such. I think speakers are a mile wide and mile deep because it's electronics and acoustics. As you said in this tutorial, one can almost tell everything about a DAC with few measurements, not so much with speakers.

 

[FreaK]

I watched the video all the way through now and I still don't understand the jitter test.

Amir has a more detailed video on this, look it up in his older videos (sorry, don't have the time to search it right now).

 

[amirm]

Nice, I think these kind of videos are crucial for newbies that visit this site, especially if they are in an easy to find prominent position when they first come to the site. I recommended this site to someone recently, and when I asked them a week later what they thought to it they answered that they thought it "was over their head" and "they didn't know who to believe" amongst all the opinions. So videos like this are useful for newbies, but I think they need to be put in a prominent position so anyone who lands here can quickly & easily see where they need to go to find explanations on how to understand the various elements of reviews, whether they be DACS / amps / headphones / speakers.

EDIT: in fact maybe it would be good if there was a seperate tab right at the top of the page that was entitled as something like one of the following: "Instructional Videos" / "Tutorials" / "How to understand this site" / "How to understand the technicalities" / "Confused? Look here" / "Confused about the Technical Elements, look here" / "Technical Explanations" / "Technical Fundamentals" / "I don't understand this site? Click Here".....whatever, it could be described as whatever makes the most sense to the most people, but you could put in an extra tab at the top of the site so newbies can instantly see where they need to go for explanations:
View attachment 273583

Excellent suggestion. Let me see if I can make that happen.

 

[amirm]

I watched the video all the way through now and I still don't understand the jitter test.

The jitter test is explained in this text: https://www.audiosciencereview.com/...derstanding-digital-audio-measurements.10523/

And in some other videos as noted.