[JediMa70]

I'm just a newbie in the audiophile world but I like to learn and I try to understand, I enjoyed a lot the thread about 16 vs 24 bits, now my question is about DSD what's exactly the difference between DSD music files and FLAC ones? DSD ones are huge so just talking about sound quality, is worth to get them or it's better to stick to FLAC 16?
Thank you

 

[Kammerat Rebekka]

This will be a neverending search on your behalf until you properly find out what your personal thresholds are with regards to sound.
I have personally never seen any proof of people being able to differentiate between mp3 320 and flac...so opting for something like dsd seems like shooting mosquitos with bazookas to me.
...UNLESS the dsd recording is a better/alternate mix...but then it has nothing to do with the encoding but rather the quality of the recording/master/remaster/etc.
Foobar2000 has a free abx program in which you can load your fave tune in flac. Take that same tune and bump it down to mp3 320. Then run it through the abx program (and remember to volumematch the files!!) and see if you can spot the pigeon.
I was absolutely certain I could do this prior to me putting my money where my mouth is...yet ended up failing miserably.

Edith: whoops forgot to arrive at my conclusion which was and is that if you can’t tell the difference between mp3 320 and flac why opt for an even bigger file?

 

[Brooko]

This might help.
https://www.head-fi.org/threads/set...-guide-to-ripping-tagging-transcoding.655879/

 

[castleofargh]

I'm just a newbie in the audiophile world but I like to learn and I try to understand, I enjoyed a lot the thread about 16 vs 24 bits, now my question is about DSD what's exactly the difference between DSD music files and FLAC ones? DSD ones are huge so just talking about sound quality, is worth to get them or it's better to stick to FLAC 16?
Thank you

flac is a compressed format. lossless, but compressed anyway. if you extract the signal to a .wav file, hires music is going to be massive too. but they will still be the same signal with the same fidelity. file size isn't a great way to judge fidelity between formats using different encodings.

for DSD, the encoding is binary, not just converted to a binary code, actually binary. you can imagine it as telling the signal to go up or down at each new sample. the number of samples per second is crazy huge, so you end up with huge files. now because it's a 1bit system, the noise floor without touching anything is at -6dB, which is really bad and cannot be left as is. instead very massive noise shaping is involved to "move" all that noise at higher frequencies and then filter out those ultrasonic frequencies full of noise. so once all is done, despite the crazy high sample rate, the resulting signal of default DSD comes very close to typical hires PCM resolution with similar dynamic range and frequency range. just done differently.

basically, it all works out in the end, but the 1bit encoding of DSD is moronic in my opinion. it was initially an archiving format for a few corporations, and even for that it wasn't anything special. I still don't understand why such a format has reached consumers. we(consumers) have nothing to gain from such an impractical format(again IMO!). if you're looking for a lossless format, I believe that flac and alac are very good options. almost all my library is in flac, at least my archives/backups are in flac.

 

[Kammerat Rebekka]

Nice one Brooko. I actually had a friend set this up for me, because computers and I simply don’t communicate on the same frequency.
Very informal thread. Also (lightly) touches on the subject of what people think they can hear vs what they actually do hear.
I thought a whole new world opened up when I got my first dedicated dap that could play flac files and was furthermore adamant in believing that I’d somehow climbed up the next metaphorical step of the hi fi ladder.
Couple of years later I try a simple blindtest to remove anything but the sound - y’know like closing one’s eyes during Jimi Hendrix’ Pali Gap to divert all the attention one can possibly muster up towards the music?- and it honestly felt horrible afterwards. I felt ‘betrayed’ somehow lead awfully astray by uncontrolled word of mouth and ,granted, great marketing.
...only for a bit though as I suddenly had other things to care about sonically than worrying about ‘loss of sq’.

 

[TYATYA]

Convert pcm to dsd give better separation, stereo imagine and lower noise floor.
I do heard it.

 

[StandsOnFeet]

Convert pcm to dsd give better separation, stereo imagine and lower noise floor.

Now you're just being silly. Transcoding isn't going to add anything to what's already there in the PCM recording.

 

[TYATYA]

Now you're just being silly. Transcoding isn't going to add anything to what's already there in the PCM recording.

You'd better experience the fact before claim base on a theory or base on papers.
On paper any dac sound the same, lossless file obtain by windowmedia is same as the one from jriver.
Chord should not build a scaler....etc

 

[bigshot]

1) If upsampled audio sounds better even if it's the exact same recording, then it would be a good idea to check to see if there is something wrong with your playback chain for the format that doesn't sound right.

2) Of course the other option is sloppy comparisons. The first thing to check would be that you are doing a level matched, direct A/B switched blind comparison.

Check off these two boxes and I'll come back and talk about it with you. I'm not going to bother until you do that because what you are claiming makes no sense. Arguing about it before you check your own errors would be a complete waste of time. Burden of proof is on you. But good luck. If you can prove this, you'll change our understanding of how digital audio works.

 

[castleofargh]

Convert pcm to dsd give better separation, stereo imagine and lower noise floor.
I do heard it.

empty claim.
please share noise floor measurements confirming that at least this isn't totally nonsensical even on your rig.

You'd better experience the fact before claim base on a theory or base on papers.
On paper any dac sound the same, lossless file obtain by windowmedia is same as the one from jriver.
Chord should not build a scaler....etc

advice to trust our guts more than research, not exactly the right section of the forum for that. and the rest are good old strawman arguments, no section of the forum is right for those.

 

[Brooko]

You'd better experience the fact before claim base on a theory or base on papers.
On paper any dac sound the same, lossless file obtain by windowmedia is same as the one from jriver.
Chord should not build a scaler....etc

Have experienced - blind volume matched abx. Question is - what was your methodology for comparison? To me there is no audible difference.

 

[JediMa70]

So in the end whatever is "over" flac 16 is just not worth it if the flac file is correctly coded, this is just good to know cause i'm going to save so much space on my hard disks

 

[Brooko]

Heck - for me, aac256 is enough (blind abx). My archive is in FLAC because I want a lossless copy. All my portables are AAC.

 

[gregorio]

[1] Convert pcm to dsd give better separation, stereo imagine and lower noise floor.
[1a] I do heard it.
[2] You'd better experience the fact before claim base on a theory or base on papers.

1. Converting pcm to dsd gives a way higher noise floor AND much higher distortion and therefore worse separation and stereo imaging!
1a. Apparently you heard EXACTLY THE OPPOSITE of what's actually happening?! In all fairness though, the massively higher noise floor and distortion are largely outside the range of human hearing, so you wouldn't hear how massively inferior dsd actually is.

2. First of all, dsd only exists because of the theory. And secondly, your or anyone else's "experience", "impressions" or marketing induced beliefs are irrelevant, we can easily measure the performance and obtain objective information, which unsurprisingly confirms that dsd operates exactly how it's designed to operate (which is ENTIRELY according to the principles of digital audio)!

G

 

[TYATYA]

1. Converting pcm to dsd gives a way higher noise floor AND much higher distortion and therefore worse separation and stereo imaging!

G

Sorry, higher dynamic, higher SNR, better channel separation, clearer of all notes, much deeper and wide .... seems like a re-master version from studios or even better. That is dsd against pcm.
(I prefer original track than re-mastering bcs I dont like neutralization or eqing, boosting... make it like a HDR imagine.
Above is just comparision)

I cant get a dsd and pcm version of a same track same recording and mastering ... so I listen by this way :
Same track (wav) -> pcm playback -> sp1000 dac mode -> kse1500 (1)

And

Same track wav-> dsd playback(this is capability, pcm to dsd realtime) ->sp1000 dac mode -> kse1500 (2).