Floyd has/had nothing to do with marketing & selling. Most of his time has been spent as a researcher at the National Research Council Canada facility where the big anechoic chamber lives.
Let me flip your query over: can you prove measurements track sonics.
Toole was the lead in a lot of work in loudspeaker measurements correlation with sonics, and it seems they have gotten somewhere, althou in a limited set of hifi. Geddes has shown that the loudspeaker distortions we measure ar emostly meaningless.
Neither of these results has been verified by a second researcher (largely due to it requiring lots of $$$)
dave
Such is called "self-fi", fidelity to each own taste.
How low is very low distortion? Numbers please.
According to you? OK.
Please keep in mind that planet10 sells stuff.
You could do the research yourself. Search for phase noise in digital audio. There are plenty of youtube videos, some are good, most do not explain anything, and some explain in detail how to measure the phase noise. There are also scholarly articles available, for a small price. If you are still a student, you may be able to get them for free.
Once done with that part, there's a decision to be made... do you believe the phase noise is important, or not.
Then, if you conclude that it is important, the next logical step would be to prove to yourself, empirically, that the stuff works.... replace a standard crystal oscillator with a low phase noise crystal oscillator and see if you can tell the difference.
Or, you can engage here in debates about why you think the stuff works/doesn't work.
Guys, let's not waste any more time with measurements.
To achieve a perfect musical reproduction, that is, exactly the same as what we can hear live but emulating it with scientific means, it will be by applying quantum physics.
The older ones here may remember a movie - The Fly 1958 - in which a scientist achieves teleportation - a perfect replica of objects from one place to another - and, encouraged by his success, decided to try it with living beings. Who better than him? So he got into the cabin, but, as fate would have it, he was not alone, a fly had entered! The result was almost perfect, but although the bodies were perfect, there was an exchange of heads! Just a little oversight, the day will come, don't lose hope.
This wine is very good, huh ! hic !
😊
To achieve a perfect musical reproduction, that is, exactly the same as what we can hear live but emulating it with scientific means, it will be by applying quantum physics.
The older ones here may remember a movie - The Fly 1958 - in which a scientist achieves teleportation - a perfect replica of objects from one place to another - and, encouraged by his success, decided to try it with living beings. Who better than him? So he got into the cabin, but, as fate would have it, he was not alone, a fly had entered! The result was almost perfect, but although the bodies were perfect, there was an exchange of heads! Just a little oversight, the day will come, don't lose hope.
This wine is very good, huh ! hic !
😊
Which one of those did you study and passed?
How would you set up the test rig to ensure that both are heard at matched level, aural memory is optimized and personal bias is controlled? Details please.
Look up live vs. recorded sound demonstrations. They go back many years.
Have you ever played with these low phase noise crystal oscillators?
... or you fall in this category:
??
... or you fall in this category:
??
It seems that you are quite interested in the topic. Maybe you can't hear the difference??
This could be because the rest of your system is not revealing.
What amplifier and speakers are you using?
What is your source, and how is it powered up?
This could be because the rest of your system is not revealing.
What amplifier and speakers are you using?
What is your source, and how is it powered up?
Depends on what you look at:
The audio is cut off abruptly at 20 kHz for CD, anything above 22.05 kHz is rubbish, while vinyl records can easily have some audio content well beyond 30 kHz.
Record players don't go into hard clipping at musical peaks, like most DACs do.
Records have no quantization errors like CDs do; you can dither the quantization, but it remains an error that is statistically dependent on the signal, unlike additive noise.
Just about everything else is worse on vinyl records than on CDs.
Of course what's more important is how much fun it is to listen to records or to CDs. I don't know any way to quantify that.
Last edited:
Not necessarily. It is looking like at least with ESS it does result in better measured performance and is at least partially the cause of IMD issues. cant state that categorically as investigation is still ongoing, but certainly the dacs that show it the worst appear to be the ones running 50M clocks. i'm not sure what is stone age about it.
I have not seen much improvement in measurements using synchronous operation with ES9038Q2M. IME lowering MCK has bigger impact on performance (lower high-order distortions). ASRC is useful for cleaning up SPDIF inputs.
hey i'm not saying the change is likely to be audible, or that significant. Just looks like they are there. Good chance it has more to do with their particular brand of ASRC in combination with other gotchas. I have no issues with ASRC in general and I would agree that its useful for spdif (which I dont tend to use for anything but audio-visual systems) It may, as you say, end up being more a matter of the 2 coinciding, rather than being the root cause. for spdif, i'll likely end up using ASRC on the receiver (not the ess internal) and feed that to the DAC in sync. resampling algorithms (software based) are very good these days too. dont really have a dog in this fight, but I do still like to use the sample-rate unchanged, if I can.
that's probably distortion as many cutting heads won't go above 15kHz except with half speed mastering.
Is mistracking not a pretty 'hard' clip?
What about all those records from digital masters or digital mastering computers?
Yes, the possible audible differences with sync mode may well be caused by using a lower MCK with sync. Many ES9038 boards have 90M-100M on-board clocks (IMO way too high) and I would guess that most who have tried sync mode have used clocks well below 90M.
ES9038 boards often have 100M clocks because people think they should be able to playback 768kHz content. Nobody tells them there is any tradeoff involved in having that capability.
Regarding ESS ASRC, IME if DPLL_Bandwidth is quite stable at a setting of "1" then ASRC sound can be pretty close to synchronous. Not quite though. Heard a couple of different ES9038Q2M implementations that be could run either way. Sound seems to clean up some in sync mode. Maybe ASRC PPLL tracking jitter had some effect (PPLL = Poly-Phase Locked Loop, it estimates the incoming sample rate in order to select the correct ASRC filter coefficients on the fly).
Regarding ESS ASRC, IME if DPLL_Bandwidth is quite stable at a setting of "1" then ASRC sound can be pretty close to synchronous. Not quite though. Heard a couple of different ES9038Q2M implementations that be could run either way. Sound seems to clean up some in sync mode. Maybe ASRC PPLL tracking jitter had some effect (PPLL = Poly-Phase Locked Loop, it estimates the incoming sample rate in order to select the correct ASRC filter coefficients on the fly).
Last edited:
Yep, in order to cater to the ever increasingly stupid samplerates that these chips say they cater for. The only sync clocks I have tried at the higher 90.xx/98.xx required for 384/352.xx have been clock generators, not crystals
lol, should have hit send when I typed this an hour ago
lol, should have hit send when I typed this an hour ago
Many implementations of ESS would struggle being stable with lowest DPLL bandwidth. You need to have your ducks in a row for that to be stable (particularly power supply). I found that my fridge starting was enough to disturb that and the noise that comes from that being upset can be VERY jarring and potentially damage speakers
So you have no problem with the fact that the whole audio stream is recalculated when using an ASRC?
//
- Status
- Not open for further replies.