You see @Angolnear , @papydd confirms what I say.
He says that he tried all the rates, and he upsamples to DSD512, because he gets the best sound at this rate.
topping d50s
It is really just a matter of cost and latency and design acumen… We see that in my TEAC UD-501 that DSD is inserted from a dedicated SOIC directly to the PDM output circuitry of the BB DAC… The CHORD DAVE SOIC does quite a lot of number crunching… could it do more? Yes with more CPU cores to do the heavy-lifting… It’s just running software, dedicated to the task it is given… This is why you see third-party devices like the M Scaler and the Canadian company you have alluded to…
This is a subjective observation at best… Try comparing DSD128 to DSD512 and then describe the audible differences to me… Make sure you have eliminated level imbalances and have removed to the best of your ability, the cognitive biases you will apply to the comparison…
évidemment, j’utilise ce dac depuis 3 ans, je me suis même demandé si le problème ne venaint de l’ampli casque à lampe parcequ’il n’a pas fonctionné depuis quelques semaine, mais je le fait fonctionner non stop depuis ce matin mais rien ne change.
why not
So can you paste the debug dump here when you select ASIO and NativeDSD method and try to play.
il faut quand même savoir que quand je débranche le cable usb de l’ordinateur le bruit cesse immédiatement
Maybe it’s the computer, but that would really be a bad luck.
Do you have the noise also while not playing?
The device of the Canadian company does not upsample. It’s a device that allows to connect the USB port of a computer or a streamer to the double BNC ports of Chord DACs. By doing so, it allows Chord users to upsampled 768 KHz PCM sound to their DAC, while avoiding the noisy USB port of their DAC.
Chord does not sell anything similar device, because it offers to its users the pricey MScaler as the only option.
Nevertheless, Chord users who tried both will tell you that you get a better sound with a good computer and the specialized upsampling player, rather than with the MSCaler.
le bruit apparait pendant le changement de piste
Again, and again, there is no audible difference between the sounds.
There’s an audible difference between the quality of the analog conversion that is performed by the DAC, and this difference is audible.
@bitracer
He says that the noise appears when the player passes to the next track.
I already read posts of other users who complained about this issue.
I think you are mistaken…
If the DAC platform has limited digital-audio design support to provide an equal level of data processing or better, than that of a computer + software player platform you have a point… I prefer to let Audirvan up-convert PCMxxx than in my DAC… as you know I modulate all PCMxxx to DSD128 using SoX in Audirvana Studio.
All digital-audio devices operate under some form of OS that is independent of the computer platform that may be feeding signal to it… It’s functional hardware and software design architecture will determine the quality of the final output signal and SQ… there is no given in either scenario that a given platform will sound better or worse, because of how the playback system is implemented… we see it all the time with misconfigured computer platform system designs…
Going to start a go fund me to put you two into couples therapy 
I would agree… so why waste the computing resources?
The PDM signal D/A will only represent the data being converted… If you want to say that a DSD512 waveform is perceived more like a pure analog waveform, than a DSD128 waveform, this would need to be tested in a controlled comparison, and it is not simple to remove cognitive bias and system differentials in the process of evaluation… You will never get me to accept this subjective observation anecdotally.
Why do you say that I’m mistaken if you upsample to the rate for which your DAC was designed to operate the digital to analog conversion?
People with other DAC models upsample to the rate for which their DAC was designed.
That’s all.
I never said that a DSD512 capable DAC is a better DAC than a DSD128 capable DAC. And I never said that with a DSD128 capable DAC, you’ll get a better sound if you upsample to DSD512. Anyway, you can’t do it.
Does additional latency time come into play to stop start of track hash/late start?