You are not logged in. Please login or register.
RME User Forum → ADI-2 & 2/4 Pro series, ADI-2 DAC series → Native DSD
There is an album I wish to purchase and it is available in multiple formats. The highest resolution is DSD512 FS. Will the RME ADI 2 DAC play at this resolution?
Thank-you.
A music album recorded and produced in 24 bit 44,1kHz is undergoing remastering and resampling and resold as super special DSD.
Sounds to me as a big hoax.
Has anyone yet listen a DSD file and compared to the standard original file an improvement was audible ?
According to the manual the ADI-2 DAC is limited to DSD256, sorry.
I have many 24bit 44.1khz recordings and a few DSD recordings. The DSD recordings are superior sounding on my system. What you say may be true to your ears, but to mine the DSD are a delight.
Thanks for the information about the 256k limitations.
I have many 24bit 44.1khz recordings and a few DSD recordings. The DSD recordings are superior sounding on my system. What you say may be true to your ears, but to mine the DSD are a delight.
Thanks for the information about the 256k limitations.
Your Dsd recordings are better not because dsd is better, but just because they are mastered differently. Converting them to 44.1 pcm will give you identical sounding.
If converting the DSD recording to 24/44 would sound the same, would the 24/44 recording contain the same number of bytes as the DSD recording? DSD recordings I have are much larger than any of the wav files.
If converting the DSD recording to 24/44 would sound the same, would the 24/44 recording contain the same number of bytes as the DSD recording? DSD recordings I have are much larger than any of the wav files.
comparing the bytes is meaningless.
You can convert a 16/44.1 to 32/768. The latter is more than 30 times larger, but it contains zero extra information.
Your DSD512 files are most likely converted from much smaller PCM source files (such as 24/96kHz) as well.
why? because no major DAW can mix DSD files. It's also impractical/meaningless to mix anything beyond 192kHz.
So essentially all those extra bytes in the DSD512 are plain garbage.
In reality, even 24/96kHz files are only useful for mastering purposes.
your ear will be hard to notice anything higher than 16/48kHz, if the conversion is done right*.
You find your DSD version file is better than your redbook version file. That may be true.
But It does not mean high resolution files are better.
They are most likely different mixes.
It's just a business model --- many hifi recording companies simply sell better mixes, only in higher resolution format, for a premium.
It's not that they cannot achieve the same good sound in 16/44.1. That's not a technical limitation at all.
They do it because by this way they earn more money.
Down-converting those DSD512 files to 16/44.1 will sound equally better than the 44.1 version they sell.
In addition, because of the nature of DSD, you will not be able to do any DSP.
To the extent that no volume adjustment is possible.
So in most scenarios, you may believe you are playing a DSD file. You are not.
As long as you use volume control on your DAC, The DAC chip is silently converting your DSD file to PCM format without telling you.
Not to say other useful functionalities that improve sound quality (such as room correction, EQ, loudness and crossfeed).
My suggestion is, if your DSD version sounds better than PCM because they are mixed differently,
just converting it to a PCM file. You don't lose anything. You gain from DSP improvements.
--
footnote:
*Theoretically a trained human ear is possible to notice difference beyond 48khz PCM,
if different filter artifacts are taken into account.
But such difference is very very subtle, and can only be achieved for some specific sound.
Also by doing that he/she is no longer listening to music, but merely looking for artifacts...
You can test if you are able to tell those artifacts by selecting different filter in RME ADI-2.
If you are playing a 44.1kHz file and can reliably tell the difference among
Sharp/SD Sharp/SD Low Dispersion, you may have a trained ear.
Most people just don't listen to those artifacts. They listen to music..
If converting the DSD recording to 24/44 would sound the same, would the 24/44 recording contain the same number of bytes as the DSD recording? ....
Not so easy, here is a short list for 5 minutes 2 channel music
DSD64 (2,8 Mhz) = 201 MB
DSD128 (5,6 MHz) = 403 MB
DSD256 (11,3 MHz) = 806 MB
DSD512 (22,6 MHz) = 1616 MB
PCM 16 bit (44,1 kHz) = 50 MB
PCM 24 bit (44,1 kHz) = 76 MB
PCM 24 bit (48 kHz) = 82 MB
PCM 24 bit (88,2 kHz) = 151 MB
PCM 24 bit (96 kHz) = 164 MB
PCM 24 bit (192 kHz) = 329 MB
Because of the different techniques PCM and DSD will not be exact the same, if both files have the same size. But already DSD64 to PCM 24 bit (96 kHz) shows no significant difference in blind audio tests. IMHO: DSD512 is absolute overkill with no real advantage.
Because of the different techniques PCM and DSD will not be exact the same, if both files have the same size. But already DSD64 to PCM 24 bit (96 kHz) shows no significant difference in blind audio tests. IMHO: DSD512 is absolute overkill with no real advantage.
Right. Even with filter artifacts taken into consideration, 96kHz is already way too perfect.
It takes me a lot of efforts to tell the difference between 320kbps MP3 and 16/44.1kHz.
Whereas the distortions and noises introduced during recording/mixing are a million times higher than those small artifacts.
Sadly some companies mislead their customers by selling better mixes of the same recording only in hi-res format for a much higher price.
Those poor customers...
rickd wrote:If converting the DSD recording to 24/44 would sound the same, would the 24/44 recording contain the same number of bytes as the DSD recording? DSD recordings I have are much larger than any of the wav files.
comparing the bytes is meaningless.
You can convert a 16/44.1 to 32/768. The latter is more than 30 times larger, but it contains zero extra information.
Your DSD512 files are most likely converted from much smaller PCM source files (such as 24/96kHz) as well.
why? because no major DAW can mix DSD files. It's also impractical/meaningless to mix anything beyond 192kHz.
So essentially all those extra bytes in the DSD512 are plain garbage.In reality, even 24/96kHz files are only useful for mastering purposes.
your ear will be hard to notice anything higher than 16/48kHz, if the conversion is done right*.You find your DSD version file is better than your redbook version file. That may be true.
But It does not mean high resolution files are better.
They are most likely different mixes.
It's just a business model --- many hifi recording companies simply sell better mixes, only in higher resolution format, for a premium.
It's not that they cannot achieve the same good sound in 16/44.1. That's not a technical limitation at all.
They do it because by this way they earn more money.
Down-converting those DSD512 files to 16/44.1 will sound equally better than the 44.1 version they sell.In addition, because of the nature of DSD, you will not be able to do any DSP.
To the extent that no volume adjustment is possible.
So in most scenarios, you may believe you are playing a DSD file. You are not.
As long as you use volume control on your DAC, The DAC chip is silently converting your DSD file to PCM format without telling you.
Not to say other useful functionalities that improve sound quality (such as room correction, EQ, loudness and crossfeed).My suggestion is, if your DSD version sounds better than PCM because they are mixed differently,
just converting it to a PCM file. You don't lose anything. You gain from DSP improvements.--
footnote:
*Theoretically a trained human ear is possible to notice difference beyond 48khz PCM,
if different filter artifacts are taken into account.
But such difference is very very subtle, and can only be achieved for some specific sound.
Also by doing that he/she is no longer listening to music, but merely looking for artifacts...
You can test if you are able to tell those artifacts by selecting different filter in RME ADI-2.
If you are playing a 44.1kHz file and can reliably tell the difference among
Sharp/SD Sharp/SD Low Dispersion, you may have a trained ear.
Most people just don't listen to those artifacts. They listen to music..
rawac wrote:Because of the different techniques PCM and DSD will not be exact the same, if both files have the same size. But already DSD64 to PCM 24 bit (96 kHz) shows no significant difference in blind audio tests. IMHO: DSD512 is absolute overkill with no real advantage.
Right. Even with filter artifacts taken into consideration, 96kHz is already way too perfect.
It takes me a lot of efforts to tell the difference between 320kbps MP3 and 16/44.1kHz.
Whereas the distortions and noises introduced during recording/mixing are a million times higher than those small artifacts.Sadly some companies mislead their customers by selling better mixes of the same recording only in hi-res format for a much higher price.
Those poor customers...
Thanks ning for these two very interesting posts. I totally agree... I've made the blind comparison between 16/44.1kHz and the muddy/shitty 320 MP3, it is very very hard to tell... And as you said, even if there is a tiny difference, you are no longer listening to music, but looking for artifact...
When I see the complicated process of listening to DSD, for no audible gain, no thanks... But as long as some people buy it...
The invention of digital audio was not done overnight. Sony and Philips spent about 20 years research and science into this and the standard sample rate and bit resolution is not just as is.
There is given frame by nature, the human ear and the frequency range a healthy young person can hear.
Then comes a lot of mathematic and at the other border of the frame the technology the computer chips etc.
The computer chips have changed and made the frame wider, but the ears are the same.
Would be interesting to have a name of such a DSD album...?
Guys, I believe that in this (sort of heated?) discussion you are mixing several subjects, such as (i) technical superiority of various formats/codecs/bitrates, (ii) original material recording quality (please note that there are some labels natively producing in DSD), or even (iii) pricing strategies of music labels... If you wished to focus on just the first aspect = what sound best, you might visit 2L.no ( http://www.2l.no/hires/index.html? ) where they actually put dozens of files, decoded from same original DXD to various lossless or lossy formats. In my view - there is no doubt that high-res formats sound SIGNIFICANTLY better than "CD quality", altough at certain point it becomes difficult to hear any meaningful difference between the formats. And this also refers to MQA, that many of you seem to hate for some principle reasons - MQA also sound much better than CD.
Please take my apologize if I was rude.
This should not be a war against DSD, just keep in mind when you say DSD is better than CD, that means to some people, you CD listeners you have been listening to bad sounding low-quality stuff all the time !
Mac or PC ? Babyface blue or Babyface silver ? Top notch of the beauty is unbeatable the wonderful colored Ladyface !
Guys, I believe that in this (sort of heated?) discussion you are mixing several subjects, such as (i) technical superiority of various formats/codecs/bitrates, (ii) original material recording quality (please note that there are some labels natively producing in DSD), or even (iii) pricing strategies of music labels... If you wished to focus on just the first aspect = what sound best, you might visit 2L.no ( http://www.2l.no/hires/index.html? ) where they actually put dozens of files, decoded from same original DXD to various lossless or lossy formats. In my view - there is no doubt that high-res formats sound SIGNIFICANTLY better than "CD quality", altough at certain point it becomes difficult to hear any meaningful difference between the formats. And this also refers to MQA, that many of you seem to hate for some principle reasons - MQA also sound much better than CD.
I am not starting a war against DSD. If the recording you want to listen to is only in DSD format, or the label only offers a better mix in DSD, go for it. You can always convert it to PCM---because for people in this forum, the ADI-2 Pro/DAC users, there's no question PCM is the preferred format. One of the most important reason people choose ADI-2 Pro/DAC is because of its rich feature set, and using DSD will disable most of its useful features. Moreover, if people are playing DSD with volume adjustments, the DAC chip is actually converting that to PCM on the fly anyway. So people are not losing anything by first converting the DSD to PCM and then playing the PCM on their ADI-2 at all. In contrary they gain all the useful features back.
In previous posts I also conceded cd format may suffer from filter artifacts and original recording/master format in this regards may avoid such issue. but 1) such artifact is so minor that most people can barely tell. 2) doesn't necessary mean DSD is better. a little bit higher frequency such as 96kHz PCM can solve this problem. compared to the cons that DSD cannot do DSP such as room correction or loudness properly, which normal people can very easily tell the difference because they are NIGHT and DAY, imo focusing on those artifacts is pointless.
your second point belongs to the third points. for those label who produce dsd natively in your second point --- this is just a business strategy. they do that only to cater to that small niche population who believes DSD is better. Moving PCM out of the chain means no mixing at all. they can not even do digital volume adjustments not to say EQ during recording, and they are suggesting their customers to follow the same rule when playing. This certainly will sound suboptimal, in most people's (people other than those who want to get rid of PCM in the pipeline and enjoy a pure DSD experience) view.
... Moving PCM out of the chain means no mixing at all. they can not even do digital volume adjustments not to say EQ during recording, and they are suggesting their customers to follow the same rule when playing. This certainly will sound suboptimal, in most people's (people other than those who want to get rid of PCM in the pipeline and enjoy a pure DSD experience) view.
If in the production chain, (multitrack-) mixing and mastering, is done in the analog domain nothing is lost when DSD is used as recording format.
This combined digital/analog way of producing is still quite common these day, no matter if PCM or DSD is used.
If now the consumer decides to listen to a pure unchanged signal (without EQ or other DSP processing) this approach is again quite common and I don’t see anything wrong with that.
Guys, just to clarify: @waedi: I personally do not use DSD (other than per last paragraph) and mostly my sources are of CD or sub-CD resolution (I mostly use Tidal in Flac, but also more and more Apple Music - for purely convenience reasons). And those sound real good on my system and there is no reason to discard CD as a substandard format (btw: it is as good as it always was = very good, and apperance of new "denser" - and yes: better - formats is not anyhow crippling old good CDs:)).
Moreover, having choice of DSD vs. hi-res Flac via RME Dac I would rather use Flac, exactly for the reasons outlined by @ning: (i) for me both maxed-out DSD and even "leanest" Flac 24/96 sound perfect, (ii) Flac is by far more convenient to use (and yes - DSP-ed Flac sounds to me actually better than flat DSD). And this is also very hyphotetical choice as simply there is very little choice of music you can buy in DSD (which is also true for other hi-res formats).
BUT: you sort of implied that hi-res files (of any sort?) are not worthwhile and it is difficult to hear a real difference vs. CD quality, which I personally do not believe to be true. Thus I suggested that you may try those 2L sample files in various formats to see if you really do not hear the difference?
And one more comment - I also have Ayon CD-35 in my system - this gear actually makes a sort of PCM to DSD conversion/upsampling. And the difference in sound signature of CDs played with such upsampling is positive and very real (and obviously there is no real difference when you upsample SACDs, as expected) - which I guess shows that technically DSD is stronger format.
Hi
just to note:
DSD is an old format from Sony from the 90's.
The resolution is around 20Bit with high frequency noise and not anything of bit efficient.
I just avoid DSD.
Peter
RME User Forum → ADI-2 & 2/4 Pro series, ADI-2 DAC series → Native DSD
Powered by PunBB, supported by Informer Technologies, Inc.