Zasilanie - czy zasilacze i kable grają? clickbait

[MichalZZZ]

O! Ciekawe. Tym bardziej, że Rockna mnie też interesuje. 

 

Teraz jestem na wakacjach, ale jak wrócę, to sobie to odsłucham i dam znać. 

 

Dziękuję 👍

Edytowane 29 Stycznia 2022 przez MichalZZZ

[Gość]

 

[Gość]

 

 

Małe podsumowanie:
Przedmiot - To co o nim pisałem - po testach Amira

 

Kable słuchawkowe - nie ma powodu żeby miały wpływ na dźwięk - żadnych różnic
Zasilacz Burson Supercharger - dość kontrowersyjny i marketingowy opis, ale kupiłem, bo była szansa na polepszenie jakości dźwięku i tak się stało - wyraźnie lepsze wyniki pomiarów
Kable USB - nie ma powodu żeby miały wpływ na dźwięk - żadnych różnic

 

Teraz to już w ogóle nie mam powodu do testów kabli

Edytowane 20 Marca 2022 przez Gość

[pabloj]

@jacmiszcz92Zobacz sobie ten tekst, gdzie autor porównuje pliki dźwiękowe, nagrane przy podłączeniu 3 różnych kabli USB.

 

[Suszec]

4 godziny temu, pabloj napisał:

@jacmiszcz92Zobacz sobie ten tekst, gdzie autor porównuje pliki dźwiękowe, nagrane przy podłączeniu 3 różnych kabli USB.

 

Troszkę za daleko idące wnioski gość wyciąga z tej próby przetestowania kabli. Dobrze to tłumaczy jeden z komentarzy:

 

"I appreciate the attempt to take an objective look at the issue. Unfortunately, I think you're making some incorrect assumptions that led to your methodology lacking validity, and you didn't attempt to assess the reliability of the methodology. When you take a signal from DAC to ADC again, you cannot assume that the ADC is sampling the analog waveform at the exact same instances of time as the original digital samples. As far the ADC is concerned, this is just another analog signal to convert. It will make that conversion based on its own clock, which is guaranteed to not be in sync with the clock of the DAC unless both are tracking a single external clock source. That's why all your inverted-and-summed cable tests still have an audible version of the music. All the waveforms are still present after summing the inverted sources, just at varying levels of amplitude based on how "in-phase" the two digital clocks were (plus additional noise due to differences in the consistency of timing for each sample of the clocks, which can vary a fair amount). You get more and more left-over amplitude as the sound frequency rises, because those waves are shorter and more effected by those tiny differences in sample timing. I understand you aligned the samples between the recordings, the problem is that those samples don't actually contain identical information and never will given your methodology, because you aren't actually testing for thing you think you are testing. If you had run the test 10 times using only the generic cable with no "changes," you'd see exactly the same results: that the subsequent tests still produced audible music when inverted against the first test. This has nothing to do with the USB cable and everything to do with the fact that anytime a signal goes through DAC and ADC conversion, it's going to be resampled at different spots on the waveform. Another wrinkle: I bet if you did this exact same test with super-high sample-rate digital sources and/or DAC and ADCs slaved to the same clock, you'd hear a lot less left-over audible differences after inverting (because the differences in sample timing would be so small as to eventually drive the "audibly" effected frequencies outside the range of our hearing). Even if error-correction was 100% on the USB signal, you would get exactly the same results on this test. That's because you're testing the wrong thing. It would be really hard to test and would likely require some programming skills, but the only way to test the "sound" of a USB cable is to have a computer program record and compare the raw digital bitstreams generated by those cables. And those are going to be identical unless there are connection problems or the cables are so long that they can't sustain the necessary voltages (like when HDMI cables get too long and start dropping data). And of course, I guarantee a double-blind scientific test of actual listener preference will show zero difference in these cables."

[hammerh34d]

"And of course I guarantee".
Prove it

[pabloj]

13 hours ago, Suszec said:

If you had run the test 10 times using only the generic cable with no "changes," you'd see exactly the same results: that the subsequent tests still produced audible music when inverted against the first test.

No właśnie koleś nagrał plik jeszcze raz generycznym kablem, aby sprawdzić czy po odwróceniu sygnału będą róznice czy nie, i nie było (dla tego samego kabla). Widocznie powinien zrobić to 10 razy, żeby sie zgadzało z komentarzem. Swoją drogą mógłby też odjąć sygnał nagrany z jednego kabla od innego, aby zobaczyć czy na wyjściu były jakieś różnice czy nie, bez odtwarzania ich ponownie (bo tu jest racja, że zachodzi ponowna konwersja).