0.009 % distortion is a respectable figure for such a powerful amplifier, but I think some people can hear it.
-
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!
No love for IcePower?
- Thread starter typericey
- Start date
More likely some people think they can hear it.
BillG
Major Contributor
Yeah, that one is ICEpower based from what I can tell.
1200AS is the only ICE Edge module in their line. Maybe the new chip is responsible for this improvement. But I don't understand why didn't they trickle down this tech to the lower powered modules yet... It's not like they don't have competition going on...
Well, I am far to be an expert on the subject of human hearing, but my reasoning is:
- at 1 kHz the ATH from humans is between 0 to 15 dB SPL.
- 0.009% is about -80 dB
=> at 15 + 80 = 95 dB SPL, somebody trained to detect distortion should hear it.
If I'm wrong, please explain without shouting too loud, I had a very bad night
.Edited min ATH
Well, I am far to be an expert on the subject of human hearing, but my reasoning is:
- at 1 kHz the ATH from humans is between 0 to 15 dB SPL.
- 0.009% is about -80 dB
=> at 15 + 80 = 95 dB SPL, somebody trained to detect distortion should hear it.
If I'm wrong, please explain without shouting too loud, I had a very bad night
- 0.009% is about -80 dB, yes, so that would be 80 dB down from the loudest you play - let's assume 105 dBSPL or so. That would put the distortion at 25 dBSPL. On it's own it would possibly be audible, but the noise floor of your room is probably aroun 35 dBSPL. Much more importantly, you never listen to distortion products in isolation. That distortion is totally masked by the fundamental tone at 80 dB louder than the distortion.
According to @amirm testing methodology: The THD+N of the AS1200 is at 5W / 4 ohms, 0.004% maximum. How this translate in SINAD?
You can use this Decibel calculator. You just have to convert the % into a pure number by dividing it by 100 (so insert 0.00004 in the lower field, select "factor" and go "calculate". In this case you get (-) 88 dB.
Ok for for the noise floor, a bit less for the masking. Distortion products can be outside of the masking band.
These links are converting % to dB, not THD+N to SINAD.
To be pedantic, THD+N is a good approximation of SINAD, but it's missing the spurious tones.
SINAD = -10 log (10^(-SNR/10) + 10^(-THD/10) ) (note that it's THD, not THD+N)
The distortion products are still harmonics. Masking might not be the perfectly accurate term, but our ear is much less sensitive to additional harmonics as compared to random uncorrelated noise.
That is why I pointed to the basic decibel converter - the OP started out from THD+N.
Then why dither? It's exactly adding random uncorrelated noise in order to get correlated quantization errors disappear.
In which case they will be part of the noise component.
Its really quite accurate
example
"Correlated quantization errors" is not the same as harmonic distortion.
I agree it's quite accurate.
In fact they are equal if the bandwidth is the Nyquist one. For audio measurements, it's quite often not the case.
I specifically mentioned I would be pedantic
.Some reading: Understand SINAD, ENOB, SNR, THD, THD + N, and SFDR so You Don't Get Lost in the Noise Floor
So using this link to calculate ICEpower 125ASX2 stereo SINAD into ASR standards (1 kHz, 4 ohms, 5 watts has THD+N 0.002% as per the green line on graph) gives SINAD = 94. Right?
If so, this would be right up there in the middle of the blue sector of the master SINAD chart for power amplifiers.
Not bad for a small, all-in-one (buffer+power stage+power supply), stereo $125 module!
Why is it called correlated if it doesn't affect the harmonics?
Anyhow, we are polluting the tread, I will stop the ping pong here.
Ahem, March Audio
Right
Would be SINAD = 70 if 6.67 kHz, 4 ohms, 5 watts.If so, this would be right up there in the middle of the blue sector of the master SINAD chart for power amplifiers.
Not bad for a small, all-in-one (buffer+power stage+power supply), stereo $125 module!
That's why I am personally not a fan of the SINAD at 1 kHz as one and only quality indicator. From engineering standpoint, getting clean signal at 1 kHz is much easier than at 20kHz.
Agreed. That is why the competing Hypex Ncore NC122MP would give SINAD = 100 at 1kHz (SINAD =104 at 6.67kHz!, so in theory "free of added sibilance"), in a small, all-in-one, stereo, slightly more expensive $190 module.
Again, imo, ICEpower should trickle down their ICE Edge tech to these lower powered modules ASAP.
Again, imo, ICEpower should trickle down their ICE Edge tech to these lower powered modules ASAP.
