DSD Vs. 1536kHz

[Rialta]

I am very sure my ears play tricks on me when I hear what I hear, but it's all good.

"High-end audio is all fake in the end, so just buy the fake that sounds best to you." Not sure who said this.

 

[Blumlein 88]

Still how can we have an opinion from listening with regards to DSD vs 1536 khz when there is no ADC to record at 1536 khz?

 

[Rialta]

You hear those differences when your brain knows what's playing. Do a level matched double blind test to find out what the brain says when only the ears deliver information.

MP3 is lossless and depending on bit rate and your ears may sound different compared to FLAC. However:

• FLAC is lossless and hence the truth even if you prefer lossy MP3.
• If any change in treble is audible I would expect MP3 to have less treble and air, at least for lower bitrates, where all content above 16 kHz is thrown away,

I will be 55 in October, and I can't hear above 13.5 or 14 kHz. If the MP3 algorithm throws away information I can't even hear, (at 16 K), but I still tell the difference between lossless and MP3, then it has to be more than that. To me, MP3 320 SOUNDS like something is 'missing', not smooth at all.

 

[Rialta]

Still how can we have an opinion from listening with regards to DSD vs 1536 khz when there is no ADC to record at 1536 khz?

Does upsampling to those frequencies count? I have no idea, honestly.

 

[Blumlein 88]

Does upsampling to those frequencies count? I have no idea, honestly.

IMO, no it does not count. I can upsample CD to 1536 khz, but why?

 

[Apesbrain]

I think that would be a great blindfold test - "can you hear the difference between 1536kHz or DSD 256/512."

This is hardly worth the bother of testing. PCM 24/1536 is 73.7 Mbit/sec (compared to measly CD at 1.4 Mbit/sec) so that's way more than DSD64 (2.8 Mbit/sec) or even DSD512 (22.6 Mbit/sec). PCM 24/1536 must sound at least 3x better than DSD512; it's practically approaching vinyl!

IMO, no it does not count. I can upsample CD to 1536 khz, but why?

Oh c'mon, which one is better:

Sex

Or,

Sex Sex Sex Sex Sex Sex Sex Sex Sex Sex
Sex Sex Sex Sex Sex Sex Sex Sex Sex Sex
Sex Sex Sex Sex Sex Sex Sex Sex Sex Sex
Sex Sex Sex Sex Sex

 

[Blumlein 88]

This is hardly worth the bother of testing. PCM 24/1536 is 73.7 Mbit/sec (compared to measly CD at 1.4 Mbit/sec) so that's way more than DSD64 (2.8 Mbit/sec) or even DSD512 (22.6 Mbit/sec). PCM 24/1536 must sound at least 3x better than DSD512; it's practically approaching vinyl!


Oh c'mon, which one is better:

Sex

Or,

Sex Sex Sex Sex Sex Sex Sex Sex Sex Sex
Sex Sex Sex Sex Sex Sex Sex Sex Sex Sex
Sex Sex Sex Sex Sex Sex Sex Sex Sex Sex
Sex Sex Sex Sex Sex

I don't think it works that way.

 

[Katji]

Actually, that's too high even for a bat, since they top out at 212kHz.

Thanks for the link, I'd thought of it, been meaning to look it up. I was thinking they - some, like the big fruit bats I know - must have a much wider range than humans, because I can hear them, loud, even though my hearing only goes up to 11-12 kHz. Although that was 15 years ago, maybe I could hear up to about 15 kHz then.

 

[Katji]

I'm getting confused here, high frequency sound versus sampling rates.

 

[Blumlein 88]

I'm getting confused here, high fequency sound versus sampling rates.

Higher sample rate giving wider bandwidth or higher frequencies. That's really all it does.

 

[TurbulentCalm]

Higher sample rate giving wider bandwidth or higher frequencies. That's really all it does.

There is another benefit you missed, vastly increased file sizes that help you fill you multi-terabyte storage devices faster.

 

[Robin L]

Higher sample rate giving wider bandwidth or higher frequencies. That's really all it does.

A whole lot of bits being wasted for sounds that can't be recorded or heard. Your data storage gets all filled up sooner with zero sonic advantage. 100% virgin snake oil.

 

[Lupin]

I know people say 'it's bit-for-bit', but to my ears brains, a CD has a more 'liquid' sound - hard to explain. It's almost as if some of the imaging was taking away. Just slightly, nothing dramatic. And as I said before, MP3 320 does seem to have more of a treble response.

Fixed it so we can be in agreement.

 

[dfuller]

I will be 55 in October, and I can't hear above 13.5 or 14 kHz. If the MP3 algorithm throws away information I can't even hear, (at 16 K), but I still tell the difference between lossless and MP3, then it has to be more than that. To me, MP3 320 SOUNDS like something is 'missing', not smooth at all.

That's because the MP3 algorithm throws away more than just top end - it employs a psychoacoustic model to throw away stuff we (usually) won't notice is missing.

Higher sample rate giving wider bandwidth or higher frequencies. That's really all it does.

There is a point to higher sample rate PCM (but I'm talking 96k, not 16 times that) on the recording side, if you're doing pitch correction or time stretching the algorithm has more datapoints to work with, so it works with less artifacting. DSD holds literally zero advantages and I find it to be probably up there with $20,000 cables as the worst scam in audio.

 

[Tks]

IMO, no it does not count. I can upsample CD to 1536 khz, but why?

Thank goodness you didn't ask this on the first page. Such bait, it's like flies on shit for those who buy into upsampling for audibilities' sake on the consumption end.

 

[Katji]

LOL!



Just struck me - I should've thought of it before - another term for audiophile/fools - batty.

 

[Katji]

That's because the MP3 algorithm throws away more than just top end - it employs a psychoacoustic model to throw away stuff we (usually) won't notice is missing.

Exactly. Usually, or sometimes, and it depends on the music, on the production. And maybe even on the neurology at any particular time.
Same goes for MP3 128 versus 320.

 

[AnalogSteph]

1536 kHz (8x 192 kHz) is intended for connection of external oversampling filters; you will find this feature e.g. in higher-end ESS DAC datasheets. It is not meant to be user-accessible. The delta-sigma modulators of audio DACs do not exactly lend themselves to truly wideband use either, given that they are specifically designed to push noise out of the audible range but not too much further.

DSD as a format was obsolete almost the minute it was introduced. While high performance 1-bit delta-sigma ADCs continued to be made (culminating in the tour de force that was the CS5394 ca. 1997 - 7th-order modulator, 117 dB(A) DR), it was evident that multilevel was the future by 1996-1997. The first such DAC I could find was the ca. 1994 Burr-Brown PCM1710U (5-level, still a rather modest 98 dB(A) DR).
In any case, whatever DSD will do, chances are PCM will store better audio quality in less data.

 

[muslhead]

"High-end audio is all fake in the end, so just buy the fake that sounds best to you." Not sure who said this.

I think it was a well-known music producer, (name escapes me),

 

[Katji]

hmm, and then the fakery is embellished and distortion is added as it goes. Sounds familiar. [So to speak.]