Yeah. People argue about what they can or can't hear. If not that, they argue about what it would take to convince people the old research missed something. The only solution I know of would be for members of the two sides to do a collaborative research project. That way they both have skin in the game, and both would be hard put not to accept the results.
Yeah, right, one side contributes with their education, labs, development and testing expenses, the other side with their ears and sweet talk. Would work as a business model (Engineering/Marketing/Sales), but not here.
I did the tests, and it says my threshold is -27dB on my computer speakers at my desk. They're decent (Kanto YU4), but it's not a reference system. I don't know how -27 dB relates to %THD, but I'm guessing the +9 dB level is around 10% THD. That would put my THD threshold around 0.15% in this test.
I was wishing for a cleaner music recording, and stereo. In the past, stereo imaging and depth have been my clues on lower distortion levels. I'm not as trained to simply listen for depth. But that's what I had to do here. Tracy's voice already has a lot of harmonic content, too. I was finding it easiest to hear artifacts in the initial drum attack right at the beginning, and then in the way the bass line separated itself from the wall of sound, as the distortion became gradually less.
Without the "solid starting point" story about 3 camps sounds like story of 3 cults/religions.
Oh, definitely.
It was just a plain question to clarify something. So it turns out that your assertion of camp 1 wasn't based on reality. Got it. Thanks.
Now I'm angry with you. You are calling me a liar. I told you that I have worked with real people (whole design groups in large companies) who were definitely in camp 1. That is my real experience with real people in reality.
The idea that only posts on diyaudio are reality is laughable. So I'm laughing at you. And you are now added to my ignore list.
IME listening for IMD on cleanly recorded vocal or choral harmonies is about the easiest way to detect low levels of it. This assumes the rest of the system does not mask DUT IMD. The texture of chordal sounds changes in the presence of IMD. Instead of hearing separate voices, harmonies become increasingly blurred together into one new sound.
Some recordings already have higher IMD levels than you may want to test for. In that case it may be necessary to find a cleaner recording. Also, if playing music from Windows, its important to use ASIO drivers and configure Windows not to mess with the sound. The problem to avoid is having Windows resample your audio in real time and not do it very well. Distortion is added by it that isn't too hard to hear. Its also important to make sure no ground loop issues and so forth as they will dirty up reproduction quality.
@Markw4, re DAC noise floor modulation vs (undithered) DC, may we assume the noise is periodic?
If so, one could find and isolate it even when buried in other residual noise by finding the right period for heavy block-averaging to reduce lesser correlated noise (true random analog noise, especially).
If so, one could find and isolate it even when buried in other residual noise by finding the right period for heavy block-averaging to reduce lesser correlated noise (true random analog noise, especially).
It's unfortunate that presenting evidence is not included in your list of solution.
I doubt any sane person with basic knowledge in audio electronics would say "All electronics sound the same". There are plenty of tube amps sounding different from transistor amps thus the term "tube sound". I also doubt you heard "real people (whole design groups in large companies)" actually uttered such claim. If they really did, those are the people you want to put on your ignore list.
Not sure where you are going with that idea?
Off the top of my head, I'm not sure why try to separate the two effects in that way. For sigma-delta dacs, noise floor modulation can be evaluated as discussed in another thread: ESS Sabre Reference DAC (8-channel) ...starting at Post #2643
Perceptually speaking, quantizing distortion sounds to me like a type of low level distortion riding on the audio signal. It doesn't sound to me like what I think of as a noise.
IIRC, ESS said something similar about the perception of noise floor modulation. That it is not perceived as a noise, but some audible yet vaguely defined difference in sound depending on noise modulation amplitude. At least for people who can hear it. Again IIRC, they say training can help develop ability to notice the effect.
Last edited:
Thank you. I've seen those DNR vs. DC slides before, and I'm asking myself if anybody else ever showed such results. I'm eager to try... though it's unclear what level of DC step size would be needed to expose the "noisy" DC levels, I hope it's not down to full 24 bits.
Then, once I could recreate such a plot for the AK4490 or 4493 I could use a short and low level music signal on top of very slowly moving (or stepped) DC and do subtractive tests. There would be a consistent noise floor difference when choosing proper "operating points", wouldn't it?
Hopefully, it would be not buried in analog and other noise which otherwise could only be reduced by sync'ed averaging which would require a known (found) period size to avoid the averaging taking out the modulation noise as well.
Last edited:
Really? You came up with the $10,000 price by yourself and it miraculously was the same amount?
Yes.
Actually, Earl's estimate was a range, IIRC from several thousand to a few or several tens of thousands of dollars for a preliminary study. The end product of that would some papers with letters and numbers on them, or some equivalent computer file.
My estimate is less vague and results in a collection of SOA or near SOA hardware to constitute an 'listening, training, and testing station' for conducting listening tests more or less the same as I would use in today's world for judging the sound of low level distortion and other audible effects.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.