-
WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!
Does DSD sound better than PCM?
- Thread starter dragonspit4
- Start date
- Status
It is not so much about the format itself, but instead how converters operate. There are quite a bunch of side-effects from SDM-PCM-SDM back and forth rate conversions. If you look for example how 192k PCM output from most ADCs look like it is just like it would have been converted to 192k PCM from 6.1 MHz DSD128. Because that's pretty much how it is. And if the ADC chip happens to be for example TI's PCM4202/PCM4204 then that is exactly what has happened.
Using DSD allows simplifying the digital path, but at the cost of extra work at production time.
andymok
Addicted to Fun and Learning
Ah now I see why there's no response. Here is the whole discussion, which I find it extremely interesting, very technical (to me), explains a lot. Strongly recommend
https://www.gearslutz.com/board/rem...nd-location-recording/30990-dsd-wdsd-dxd.html
https://www.gearslutz.com/board/rem...nd-location-recording/30990-dsd-wdsd-dxd.html
What percentage of recording engineers prefer to work in DSD?
The ones that get paid to, would be my guess.
Not using NAA. But, HQPlayer accessing files from NAS over WiFi couldn't handle DSD256, but had no problems with 192k PCM.
No idea.
Well I have taken the time to speak to people in the industry who do this for their livelihood and have many years experience of working with analog, PCM and DSD recordings and the overall impression I have is what I've already stated.
I'm totally well aware of all the factors you list believe me but my personal experience and therefore preference at this time is that the DSD recordings I have - even if the music is not my favourite genre - simply sound.....better. Can't say more.
Thanks for that - worth wading through. The comments on mixing DSD are particularly interesting....
- Joined
- Jun 19, 2018
- Messages
- 6,652
- Likes
- 9,406
No argument with your preference, just not at all convinced your preference comes down to to any real technical difference between the formats
Cookie Marenco.
That may be related to something else than just bitrate. 192k PCM was FLAC or some uncompressed (WAV/AIFF)? We are not talking about any huge amount of bandwidth here...
You can take a look at nativedsd.com for example. But one could also ask what percentage of mastering engineers don't participate in loudness wars and push PCM productions to heavy compression and digital clipping...
11.28m * 2 = 21.5Mb/s for DSD256
192k * 24 * 2 = 8.8Mb/s for PCM192
So, almost 3x the sustained bandwidth required, even before compression in FLAC (most of my files are compressed FLAC).
21.5 Mbps is very little for anything remotely modern... You'd have trouble watching 4K videos too, for example Netflix is sending about the same as your DSD256 for heavily compressed 4K video over the internet! Use something like GoPro in pro-video 4K mode and you'll have much more. For comparison, this video gives me around 38 Mbps when there is full area movement (camera pans and such):
For HQPlayer NAA, PCM is always sent out as 32-bit regardless how many bits of it are used. So it is 192k * 32 * 2 = 12.288 Mbps.
PCM is sent as file bytes when accessed from NAS, so definitely not 32 bits per sample. And 4K video streaming works fine over exactly the same wireless repeater as the one used for audio duties.
AFAIK, Cookie records and works in analog and converts to DSD for distribution. Has that changed?
Well maybe, I find a lot of pop music has excreable SQ only suited to listening on earbuds or in cars, and even then not nice.
I was genuinely interested in whether there really is a significant number of recording engineers genuinely using this technique because it actually sounds better.
Nobody seems to know, and I suppose it is unknowable for sure.
I, personally, don't hear any difference between 2 files, one original 96/24 and the other the same file downsized to 44/16 then re-calculated to 96/24 so the DAC treats both files in the same way. Since tagging ripped files bores me I am largely back on CDs.
The fact is that the difference in the recording quality of music releases is massively bigger than the difference between different bits of hardware and file sizes and types and is sort of independent of the technology IME.
Yesterday I listened to a new (to me) CD of Elgar's "The Dream of Gerontius" recorded in 1965, digitally mastered in 1999. The sound quality is absolutely fabulous.
I have yet to see any plausible evidence that the recording technology is the limiting factor in SQ of music releases, I wait with bated breath.
As I referred earlier, the process of making a recording also matters, for example using DSD enforces certain ways or at least promotes certain ways.
So possibly in many cases RedBook made from a DSD recording may sound better than one made with typical PCM hardware and ProTools. Different converters also sound different too, Grimm AD1 sounds different from Merging Hapi. Same goes for different "PCM" ADCs (that are all SDM devices).
I still have no problem hearing difference between RedBook conversion of DSD recording vs DSD recording played as DSD.
After the Pacific Microsonics Model One and Two I have not seen any actual PCM ADCs being sold for pro audio. So most of todays "PCM recordings" originate from SDM anyway.
I must be lucky. The first time I heard a recorder where the output was indistinguishable to my ears from the microphone feed on the sort of music I record(ed) was the StellaDAT.
I easily hear the differences between different microphones, and the even more marked differences between microphone locations, but to my ears even early ADCs were transparent enough for me - simple classical music recordings of orchestras and choirs.
- Status
Similar threads
