• Welcome to ASR. 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!

Do AVR pre amplifier/dac SINAD measurements matter if the power amp Sinad is a lot lower?

SINAD is ratio presented in dB
Yes, but so are distortion and noise. Two very different things. You can have SINAD of -85 db which is noise with much lower distortion. Feed that into a high gain amp and a sensitive set of speakers and you'll hear the noise. You can have SINAD with the same -85 db which is distortion with noise levels far lower and you'll hear nothing into a high gain amp with sensitive speakers. That is why in discussion of how things propagate thru a system of source, preamp, amp and speaker just using db of SINAD to add or subtract can be very misleading.
 
SINAD is ratio presented in dB
sure, but it is important to remember that SINAD is a "synthetic" metric combining two separate and generally unrelated causes. it is best viewed as an "at a glance" figure, but anything beyond that it is best to return to looking at noise and distortion independently.
 
I currently own the Denon 3800H which has sufficient power for my quite transparent speakers. As you seem to know a lot more than me, could I bother you to take a peak at the review of amir and see if you agree that moving up in the range of denon and marantz is not usefull for a SQ piont of view?


Understand if you don't have time, but I would be most thankfull if you do and will send you a big load of karma and thanks! ; )
As I mentioned before, when the gap is that small, you need to dig deeper into the details, instead of just looking at the SINAD number as just one point. So in order to get a more accurate answer to your question, the minimum info required would include the following:

- Your speaker's specs that include sensitivity, impedance/phase angle characteristics.
- Your maximum spl (peak) requirements at your listening distance.
- Listening distance.
That's assuming you will not be using external amp, since you are only considering the SINAD of the AVR-X3800H's 87 dB or so, that is measured at the binding posts.

That way, you will know the likely range of the pre out voltage and will therefore know the SINAD range you can expect.

I know people are concerned more about noise, not so much distortion, still, SINAD is likely enough info because a) it is most likely not all noise, and b) even if the SINAD is completely dominated by noise, unless your room is really quiet, -87 dB is most likely not going to be an audible issue when listening even nearfield, say 2 meters from speakers and even during the quietest passage of music or movie tracks, even -75 dB should be fine, I would worry more if it drops to below 70 dB.

Here's an example of how/why you have to look deeper into the details:

The 87 dB is for about 2.0 V level, if in your application the pre out voltage range is mostly between 0.3 V and 1.25 V, SINAD would be about 90 dB, and if it stays below 0.8 V, it could be as good as 92-93 dB.

I listen from about 11 ft, and my pre out voltage never get close to 1 V, rarely above 0.5 V even during the highest peaks but obviously you have to figure what your maximum requirement is, that's why I would need the info listed above.

As they say, people don't realize how little power they need, or people don't realize how often their amps would clip, in real world use. Both can be true.

Keep in mind that D+M AVRs gain are about 29 dB, so at pre out voltage 1.2 V, the AVR's output would be about 143 W into 8 ohms, or 286 W into 4 ohms, the power amp at that point will be clipping anyway.

So really, regardless of the info requirements I listed above, you should focus on SINAD at pre out voltage below 1.2 V, and SINAD at that point is not 87 dB, but about 91 dB.
That said, people will say things like, that's based on the 1 kHz test frequency, how about at higher frequencies? In that case, you would look carefully on the appropriate curve and pay attention to SINAD when the test frequency is 5 kHz. I wouldn't worry much about test frequency higher than 5 kHz because the harmonics will be at frequencies approaching the inaudible range for humans.

Again, this is just one example of the more detailed analysis required, before your question can be answered with reasonable accuracy and relevancy, and that's because the gap you specified in your original post is small. If it is the SR8015, then I would say yes the power amp section will be the bottleneck, without going through the details, because the gap is just so big, that answer to the same question would be obvious enough.

1739979574258.png
 
sure, but it is important to remember that SINAD is a "synthetic" metric combining two separate and generally unrelated causes. it is best viewed as an "at a glance" figure, but anything beyond that it is best to return to looking at noise and distortion independently.


===========

There is nothing "synthetic" about SINAD.

Look up it's engineering description.

1739981177709.png


I'll leave it at that..
 
No matter how noise/distortion were calculated for given SINAD value once SINAD is calculated the level of noise/distortion power level is fixed relative to the overall signal.

If SINAD is calculated = 80dB, then (noise/distortion) power level is 80dB below overall signal level.

It's not a matter of interpretation, discussion, etc.

It's engineering term, math.
 
No matter how noise/distortion were calculated for given SINAD value once SINAD is calculated the level of noise/distortion power level is fixed relative to the overall signal.

If SINAD is calculated = 80dB, then (noise/distortion) power level is 80dB below overall signal level.

It's not a matter of interpretation, discussion, etc.

It's engineering term, math.
No it is not a matter of interpretation. It does require some understanding like much engineering. In the case of SINAD which is a combination of two metrics you can have the same measurement number and different results. Not understanding that causes one to make faulty assumptions about how things will work out. Oh, and btw, SINAD varies with signal level or power level.
 
SINAD of 83dB
system playing at 100dB sound level
combined noise/distortion 17 dB sound level
noise/distortion level quieter than Recording Studio

Not an issue for me
:cool:
 
SINAD of 83dB
system playing at 100dB sound level
combined noise/distortion 17 dB sound level
noise/distortion level quieter than Recording Studio

Not an issue for me
:cool:
Now you are combining the mistake of broadband spl levels vs how the ear works. As in you can hear a 10 db signal on a room at 35 dbspl.
 
Now you are combining the mistake of broadband spl levels vs how the ear works. As in you can hear a 10 db signal on a room at 35 dbspl.

I don't think he's talking about whether he could hear a 10 dB signal, he's just saying:

combined noise/distortion 17 dB sound level
noise/distortion level quieter than Recording Studio

I wonder how quiet a recording studio is, presumably there is some sort of acceptable and/or typical range, pure tone, broad band, or whatever band.
 
Last edited:
No matter how noise/distortion were calculated for given SINAD value once SINAD is calculated the level of noise/distortion power level is fixed relative to the overall signal.

If SINAD is calculated = 80dB, then (noise/distortion) power level is 80dB below overall signal level.

It's not a matter of interpretation, discussion, etc.

It's engineering term, math.
SINAD does not inherently scale linearly with signal level.
That is one of the important reasons the test signal levels are quantified.

When evaluating gain staging one may need to look at the noise floors separately from THD or IM.
 
SINAD of 83dB
system playing at 100dB sound level
combined noise/distortion 17 dB sound level
noise/distortion level quieter than Recording Studio

Not an issue for me
:cool:
A broadband noise floor of -100 dB from DAC into preamplifier with a gain of 20dB and an amplifier with a gain of 30dB.

Approximately what would be the noise floor of the total system?
 
===============

Do you know the noise floor levels of the pre-amplifier and amplifier?
If you don't know the noise floor then the THD+N or SINAD values would give you some ball park values to start with.

===============
 
Last edited:
===============

Do you know the noise floor levels of the pre-amplifier and amplifier?
If you don't know the noise floor then the THD+N or SINAD values would give you some ball park values to start with.

===============
No it won't. Depends upon the details.
 
Isn't the weakest link the bottleneck in the chain?
It would be a mistake to categorically say "the amp SINAD is 90 dB so as long as the pre/DAC is >90 it doesn't matter."
- The noise spectra may not be the same, nor the distortion spectrum.
- It's ONE measurement, there may be transient behaviors not captured by typical measurements. Like noise modulation.
 
The thread title says "a lot" lower, so it also depends on what the OP meant by a lot. For example, would it be 5 dB, or 20 dB? The answer to the question would be different between 5 and 20 dB, all else being equal.
 
As in you can hear a 10 db signal on a room at 35 dbspl.
Yes, possibly, that is indeed the misunderstood point. I've posted elsewhere how a friend has been fighting aircraft overflight noise in National Parks, and the FAA (which PROMOTES aviation, not REGULATES, in real life) doesn't want to hear about any details. They just want to look at numbers. But with different noises (or distortions) the spectra can make a difference, the direction of the sound, and the periodicity as well. (Periodicity = for example you can be by a river and pick out a helicopter more easily from the rotor whup whup whup rhythm than an overhead jet who noise is more similar to the river)
 
Back
Top Bottom