Not that one can hear these differences...but the company already currently makes processors with 100+ SINAD at 4V RMS output (in fact Audioholics just got 108dB SINAD on their latest test). Please break out your wallet and we can quiet things down here.
That's the thing, Amir changed the methodology, so you cannot directly compare the two graphs. Also, take any read carefully, as I see it, it is more like 87.5 or slightly higher SINAD, than 86 dB at 0.3 V, but you are right, even then the difference is almost 6 dB, still a lot. But again, quite a bit of that difference is due to the different method used for this sweep.
I posted the following graph in one of mine earlier posts to try and help lessen people's concerns of this very point:
If you look at the one for the PCM5102A, you can see the SINAD for the DAC at 0.3 V should be higher than 90 dB, probably about 91 dB, so in that case the gap due to the DAC chip itself is not significant at all, compared to the 4700's. I hope that helps settling our differences.
Better yes, but not more representative. And I also have a problem with that, that the measurement methodology was changed in favor of 4800. Only in this one case! It is then poorly comparable. I'm very much looking forward to what method Amir will use to measure the nearest non-Harman AVR...Thank you...I stand corrected, that does indeed show about 87.5 dB @300mV. God that's so much less intuitive to read
As for the methodology could anyone comment on which would produce the better performance, altering the volume in the digital domain and keeping the preamp gain higher or having full digital range output and lower the pre-amp gain?
I think my intuition (edit: and the above graphs) tell me that altering the gain in the analog domain would produce better results. I'm thinking if the DAC output is at full range you are keeping the levels as far away from the (likely somewhat constant) noise floor as possible and that at preamp stage you are basically attenuating the signal and noise together so the ratio would be the same.
If I'm not mistaken, the old methodology was altering the output in the digital domain thus making the signal vs the noise lower. Shouldn't the new methodology then produce better (and more reprensentative) results all else being equal??
I meant representative in the sense that if I want to change the output volume of my AVR, I reach for the volume knob which I understand is connected to the analog pre-amp and does not affect the output level of the DAC.
) and say the previous measurements using the old method actually were under-reading SINAD in which case the older 4700 model might measure even higher than the 101dB it achievedI meant representative in the sense that if I want to change the output volume of my AVR, I reach for the volume knob which I understand is connected to the analog pre-amp and does not affect the output level of the DAC.
If this is correct then I would assume that Amir's change is for the better and will produce more accurate results. I would also make the jump (somewhat out on a limb here though
If I'm wrong in any of these assumptions please do let me know and at the very least I'll be learning
Exactly, I exchanged a ESS dac Pioneer with the 4800h and i like the Denons sound better. But there will always numbers people, no matter how meaningless they may be. Each their own. Some wait and wait and wait for the next best thing and when they have it they get annoyed something else comes along . Until Amir tested the (older) Marantz they where up there for music lovers and received all the praise for their sound. They measured bad. Got me laughing.
But... If they are both trasparent, you should not ear any differences between the Pioneer and the Denon
Difference is coming from audyssey vs mcacc measurements. Which can probably be tweaked to sound mostly the same.
I meant representative in the sense that if I want to change the output volume of my AVR, I reach for the volume knob which I understand is connected to the analog pre-amp and does not affect the output level of the DAC.
If this is correct then I would assume that Amir's change is for the better and will produce more accurate results. I would also make the jump (somewhat out on a limb here though
If I'm wrong in any of these assumptions please do let me know and at the very least I'll be learning
You are not wrong.I meant representative in the sense that if I want to change the output volume of my AVR, I reach for the volume knob which I understand is connected to the analog pre-amp and does not affect the output level of the DAC.
If this is correct then I would assume that Amir's change is for the better and will produce more accurate results. I would also make the jump (somewhat out on a limb here though
If I'm wrong in any of these assumptions please do let me know and at the very least I'll be learning
Yes. I understand what you are saying. I just don't know why it was necessary to change the measurement method for this particular model so that it is not possible to compare the old measurements of the 3800. I am waiting to see how it will be measured in the future.
Hmm, still trying to make sense of this tbh but I read it differently.
That is an unfortunate side effect of correcting the method...I wouldn't be overly hard on Amir though as he does a huge amount of work to bring us the facts of how products perform. Maybe a period of time giving both measurements to aid comparison would have been a good idea
I agree, but unfortunately measurements of old AVRs with the new method are probably no longer possible repeatedly.
I actually meant the new ones being tested both ways for a time so that we can compare old with new and also have the more accurate measurement...this could also be confusing though not to mention extra work
HDMI handshake is definitely a thing regardless of the AVR - let us know if it continues to be a problem after you get it back. Sometimes it's software sometimes it a bad chip, we just don't know.
