• 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!

System-Wide SINAD?

MakeMineVinyl

Major Contributor
Joined
Jun 5, 2020
Messages
3,558
Likes
5,871
Location
Santa Fe, NM
If that is accurate, then it would be good to have SINAD for speakers, and in terms of the chained SINAD, the terrible SINAD of speakers completely dominates the net result.

Speakers don't have SINAD since they don't produce noise unless there's something wrong like the voice coil rubbing, and then they could be considered defective. Perhaps one could call reflex port turbulence 'noise' but I would question the validity of that.
 

Bear123

Addicted to Fun and Learning
Joined
Nov 27, 2019
Messages
796
Likes
1,370
In my limited sample, “buying by the (SINAD) numbers” has produced marked audible improvements. Both in theory and in practice. [Have you heard the one about the French Academic, excitedly reporting “this is not only true in practice, its true in theory!”]

Though I agree that there’s one number that can always be measured very accurately, to two decimal points: the price, in dollars and cents!
If you knew which component you were listening to(i.e. the one you spent lots of money on because you are already convinced it would "sound" better), the improvement is as likely imagination as real. It's technically possible it *actually* sounds better, but sighted listening tests are impossible to do without bias affecting what we hear. Given real life conditions, noise floor etc, -80 dB or -120 dB wouldn't be able to be determined in a blind test on real content, despite how much misplaced emphasis is placed on this number for "sound quality".
 

Beershaun

Major Contributor
Forum Donor
Joined
Oct 3, 2019
Messages
1,864
Likes
1,910
I'd rather take the speaker and our ears out of the equation and measure the difference in information going into the speaker input if there is a way to do that? That would tell us, in a speaker and room agnostic way, if our audio chain has measurably different information going to the speaker or not.
 
OP
O

omm0910

Member
Joined
Sep 19, 2020
Messages
26
Likes
19
Still feel slightly lost on what to upgrade next. I guess I’m not going to change the room, or the speakers. I feel like Dirac will give me a lot of bang for the buck and compensate for both the room and the speakers. I’m feeling like upgrading the amp or the DAC won’t give me bang for the buck. I feel like I got a slight improvement from switching to Amazon Music (with often, Ultra HD sampling quality) over USB, from YouTube Music over Bluetooth. I trust Google’s AI to find me new music better than Amazon’s AI, but I wonder if and when Google will step up to higher sampling rates. As far as Qobuz etc go, I would not like to wake up in the morning and find myself competing head on with Apple, Google, and Amazon. Anyway, you can see, I’m back to looking at the full chain.
 

Beershaun

Major Contributor
Forum Donor
Joined
Oct 3, 2019
Messages
1,864
Likes
1,910
Ah. If your question is what to upgrade next to improve your sound then definitely room correction and speaker equalization. Dirac is the best in the business for that. So yes you should add that before spending money on anything else.
 

Beershaun

Major Contributor
Forum Donor
Joined
Oct 3, 2019
Messages
1,864
Likes
1,910
er… good point… https://www.audiosciencereview.com/forum/index.php?pages/SpeakerTestData/

sorry if this is a dumb question but what does a speaker “score” mean as opposed to a “SINAD” for a speaker.

and yeah, if we can look at the speaker part of the chain, it still sounds like by far the weakest link, yes?
I'd recommend you watch Amirs YouTube videos on measuring and rating speakers. It's a great primer on where speaker preference scores come from, and the dimensions used to measure and evaluate a speaker..and why you should care about both the anechoic frequency response and the expected in room frequency response. Along with the other dimensions.

 

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,523
Likes
37,053
Still feel slightly lost on what to upgrade next. I guess I’m not going to change the room, or the speakers. I feel like Dirac will give me a lot of bang for the buck and compensate for both the room and the speakers. I’m feeling like upgrading the amp or the DAC won’t give me bang for the buck. I feel like I got a slight improvement from switching to Amazon Music (with often, Ultra HD sampling quality) over USB, from YouTube Music over Bluetooth. I trust Google’s AI to find me new music better than Amazon’s AI, but I wonder if and when Google will step up to higher sampling rates. As far as Qobuz etc go, I would not like to wake up in the morning and find myself competing head on with Apple, Google, and Amazon. Anyway, you can see, I’m back to looking at the full chain.
The answer will usually be speakers as the place to improve. What speakers do you have?
 
OP
O

omm0910

Member
Joined
Sep 19, 2020
Messages
26
Likes
19

Beershaun

Major Contributor
Forum Donor
Joined
Oct 3, 2019
Messages
1,864
Likes
1,910
Appreciate all the advice. Have Usher BE718 DMD speakers. In a bedroom with imperfect acoustics and placement. SMSL m200 DAC. Audiophonics Hypex 252 amp. I think where I am is, wait patiently for Dirac sound driver for mac. https://www.audiosciencereview.com/forum/index.php?threads/mac-mini-running-dirac.18633/ I’m thinking upgrading the DAC or amp won’t make much difference all things considered.
Thanks for the additional info. Based on your speakers I would say adding a subwoofer would increase your enjoyment the most.
 
OP
O

omm0910

Member
Joined
Sep 19, 2020
Messages
26
Likes
19
Can you compute SINAD for a chain of equipment? Some say yes, some say no. I say yes. Consider. You can feed a signal into a DAC, measure the output, and compute SINAD. Based on measuring "the average power of the signal, noise and distortion". And you can do the same for an amplifier. With me so far? Then, you feed the signal into the DAC, connect the output of the DAC into the amp, and measure the amp output as before. You now have SINAD for the DAC-amp as paired. I'm prepared to go a step further, and say that if you have the SINAD of the DAC, and you have the SINAD of the amp, you don't even have to get out your oscilloscope, you just do the math, as described above, and we even have a little snipped of python code for this.

Is there SINAD for speakers? Again some say yes, some say no. Looking at the definition of SINAD https://en.wikipedia.org/wiki/SINAD my take is that you can still apply the formula even if speakers don't create noise. Just plug in a zero for the noise terms. Maybe that's a misuse of SINAD, let alone the next step of a 3-component long chain of DAC-amp-speaker calculated SINAD. So yeah maybe as was suggested you go back to examining the noise and the distortion separately. (And chain these?) In any event, I'm thinking the key take away, from the math, is that we're in a weakest link in the chain situation, even just looking at the DAC and the amp. But further, the weakest link is the speakers. Given that speakers have a nominal SINAD of 60. This completely dominates improvements in DAC or amp, unless one of those was way down in the SINAD chart.

Reading this https://www.audiosciencereview.com/...er-preference-ratings-for-loudspeakers.11091/ and still trying to understand any definition of "speaker (preference) score". Found this too. https://www.sausalitoaudio.com/wp-content/uploads/2018/07/Interpreting-Spinorama-Charts.pdf Maybe I'll not be getting any "speaker SINAD", but will accept that its the weakest link, and just generally want to improve it based on "score".

We could get out into somehow trying to measure "enjoyment", but I'm prepared to believe that a subwoofer would be fun. As much as I like headphones or IEMs, you listen with your whole body.

And, ok the other weakest link is the room. I suggest there’s a lot of bang for the buck in Dirac room correction.
 

Beershaun

Major Contributor
Forum Donor
Joined
Oct 3, 2019
Messages
1,864
Likes
1,910
Yep I'm with you. Is the system just the sum of the parts or does the upstream performance have an oversized effect on the downstream components? I don't know.

On preference scores and enjoyment, our host has also created a video on this you can watch to learn about the research that the Canadian Gvt and Harman performed. This is the science his tests and results are based on.

 
OP
O

omm0910

Member
Joined
Sep 19, 2020
Messages
26
Likes
19
Is the system just the sum of the parts or does the upstream performance have an oversized effect on the downstream components?
Not to put to fine a point on things… I know you’re using “sum of the parts” metaphorically, but to put it exactly, the formula is not sum, its the logarithm formula cited earlier in this thread and expressed in simple Python.

And further, I’d say that every component has an equal effect, addition being commutative in that formula.

And further, in practice, we find ourselves with the reverse. In practice its the downstream components (speakers) that have the oversized effect! If the stream starts with the DAC, you can get fantastic SINAD for a few hundred dollars. Next downstream is the amp, and even the $3000 Benchmark only manages a SINAD of 113. This 113, relative-to-the-DACs SINADs in the 120, becomes the weak link. If you play with the chained-SINAD formula, there’s little reason to upgrade your DAC. And then if we consider speakers to have SINADs in the 60’s, speakers end up totally dominating (degrading) the overall system fidelity.

Hope I’m not getting to carried away here, but that’s my interpretation of the science, and I don't think it strays too far from, lived experience.
 

DonH56

Master Contributor
Technical Expert
Forum Donor
Joined
Mar 15, 2016
Messages
7,835
Likes
16,497
Location
Monument, CO
If you already knew the answer, why ask the question?

Unless you know the gains, signal levels, and contribution of noise and distortion along the signal chain, you cannot simply "combine" SINAD whether you use a Python script, hand calculator, or a slide rule. You can indeed measure at the output point and watch how it changes as you add additional components to the chain before the output. Since distortion changes with signal level, any change in level will change the reading.

Speaker performance is generally dominated by distortion so you can ignore speaker noise. Of course, unless you have an anechoic chamber, you need to take measurements well above the room's noise floor and determine the noise and distortion of the mic, mic preamp, and so forth that sets the measurement noise floor.
 

pma

Major Contributor
Joined
Feb 23, 2019
Messages
4,591
Likes
10,727
Location
Prague
Output noise of the components connected in series as well as their gain must be known. Otherwise individual SINADs do not tell much.
 
OP
O

omm0910

Member
Joined
Sep 19, 2020
Messages
26
Likes
19
I wonder Amir would consider this a worthwhile experiment. He has the test equipment, and a stack of DACs and a stack of amps on loan. It would be straightforward, he already has SINAD for each DAC and for each amp. The calculation is simple. My “null hypothesis” is that the calculated result is close to the measured result from the chain.

I like to think that the DAC can “drive” the impedance of the amp without any interaction. (Though I wouldn’t say that about the amp driving the speakers). You predict interaction, how does it come about?
 

DonH56

Master Contributor
Technical Expert
Forum Donor
Joined
Mar 15, 2016
Messages
7,835
Likes
16,497
Location
Monument, CO
One example:

DAC output is 110 dB SINAD at 4 V output
Amp output is 100 dB SINAD with 1 V input

If you just use the script or otherwise calculate the resultant SINAD, you'll find the answer is about 100 dB. But if you do the experiment at those levels, you'll find the amplifier has higher distortion due to the higher input level and the SINAD is reduced, because the amplifier's input level (from the DAC) is much higher than the original SINAD used. There are a myriad of potential combinations.
 

pma

Major Contributor
Joined
Feb 23, 2019
Messages
4,591
Likes
10,727
Location
Prague
One example:

DAC output is 110 dB SINAD at 4 V output
Amp output is 100 dB SINAD with 1 V input

If you just use the script or otherwise calculate the resultant SINAD, you'll find the answer is about 100 dB. But if you do the experiment at those levels, you'll find the amplifier has higher distortion due to the higher input level and the SINAD is reduced, because the amplifier's input level (from the DAC) is much higher than the original SINAD used. There are a myriad of potential combinations.

Yes. Let's consider DAC SINAD is mostly defined by its output noise. Than we have -98dBV noise at DAC output. Always, in case of digital output volume control of the DAC.
Let's consider that the amp has gain of 28dB. This makes -70dBV noise at amp speaker terminals just by amplifying the DAC output noise. And let's consider Amir's 5W/4ohm test, which equals to 4.472V amplifier output voltage, or +13dBV. We get resulting SINAD of 83dB at 5W from the components with 110 resp. 100dB SINAD and the signal source would be the main contributor.
 

Spkrdctr

Major Contributor
Joined
Apr 22, 2021
Messages
2,212
Likes
2,934
After all is said and done, the speakers and the room dominate and swamp so much of the "other" things people focus on that we need a sea change of thinking in audio. As stated many times DIRAC is awesome and gets you started, the speakers themselves contribute a huge effect to the sound and then the room finishes it off by butchering the sound. Dirac tries and does stop a lot of the butchering. So, as we have mentioned many times. The biggest and most important issues are the room and the speakers, assuming you have DIRAC as a base requirement. Once you get your set up sounding and testing pretty good, your brain will take over and make it sound amazing. It is an amazing sound cleaner upper in its own right. In my mind focusing on the equipment that will not make much of a difference is the wrong thing to do. BUT and it is a BIG BUT, buying new equipment IS FAR EASIER than room treatments and measuring/tweaking the room. This is from aesthetics and cost. So, most everyone thinks, I will buy better speakers and that will give me a big sound quality boost. This may not be accurate at all it will probably sound equally as bad. A properly treated room to just the the basics, plus DIRAC, plus speaker location etc. will allow a $1000 a pair speakers blow away a $8000 a pair set of speakers that are just plopped into a room and not setup. So, that $7000 would have easily made the system sound so much better than just buying expensive speakers. But, then as most know on this site, I have been preaching this for quite awhile. :)
 

luft262

Senior Member
Joined
Mar 25, 2021
Messages
465
Likes
236
Location
Phoenix
Well you don't sum them no. The part of your chain with the least SINAD is the bottle neck which will set your system SINAD. Usually that will be the speaker and room noise.

For an extreme example you have all SOTA components, and live in an apartment next to a construction site where a jack hammer is running non-stop. What limits the level of detail you can hear in your music? The high noise of the jackhammer.

So in your example of a 114 db DAC and 94 db amp, the amp will set the result of those two at 94 db or slightly less.

Using SINAD this way is really a misuse of it. In general few speakers will do better than 60 db SINAD because distortion will be at least that high. As I say often, you really get further along to split up SINAD in to THD and noise separately. Doing that noise is likely to be limited by room noise and THD is likely limited to that of the speaker.
If speakers are the bottleneck and they can only achieve a SINAD of about 60dB then why do we even care if our amps and dacs can do much better than 60dB?
 
Top Bottom