Converting DSD Files to FLAC

QuadraphonicQuad

Help Support QuadraphonicQuad:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
I would convert to PCM and then normalize to a sensible level on a per-album basis (to preserve inter-track dynamic relationships). I find -1dBFS to be a fairly safe normalization target.
I have converted the DSD files to FLAC. Should I just amplify in 24 bit directly or move to 32 bit float first and render in 24 bit again? Thanks
 
As always the best way to decide is to compare the various files (WAV and FLAC).
Along with the different software and hardware approaches to converting the files from their original format to FLAC.
And of course your comparisons must be double blind or equivalent (e.g. ABX testing). If you go to the effort, you'll find that you can't tell the difference between the sound of various flac encodings and wav, because they're identical by the time they hit the DAC. So much snake oil.

P.S. flac is flac; I want my baby back.
 
I have converted the DSD files to FLAC. Should I just amplify in 24 bit directly or move to 32 bit float first and render in 24 bit again? Thanks
Just leave it in 24-bit. Depending on the software you use, the processing will be done in up to 64-bit floating point. And if not, any difference in result won't be audible, anyway. Save the time, CPU, and drive space.
 
First, I use Sonore iso2dsd (a free Javascript app) to rip a SACD iso image to DSD files.
(I feel like there should be a better way to do this... A Java app? ... But it appears to work and it's simple.)

I'll use XLD to convert to PCM and I save that in FLAC in the end.
Settings here are important!
You want 32:1 decimation to convert the DSD losslessly to PCM. This results in a sample rate of 88.2k.
Anything else gives a lossy conversion and the lossiness goes way beyond normal PCM to PCM sample rate conversion!

The levels on DSD program can end up anywhere from 6db low to 6db over zero after conversion to PCM.
If you have a hot example and convert it straight to FLAC (or any format in 24 bit), you will clip! Conversely, a lower level example will lose precision (even if only 1 bit). So it is critical to convert to floating point first!

Convert the DSD to 32 bit floating point wav using 32:1 decimation.
Open the wav files in your favorite DAW app.
Normalize the levels to just below zero.
Now render those files to 24 bit fixed and convert to FLAC.

Re: 176.4k sample rate sounding better than 88.2k.
You may have hardware converters that perform better as ultra HD sample rates vs. HD sample rates. (In the same way that most consumer converters found in AVRs run better at HD than SD.)
You could convert some program in 88.2k to 176.4k and A/B listen to try to confirm that. There's no actual meaningful audio content captured above 20kHz. HD sample rates are all about the machines we build (AD & DA converters in this case) running better. I'll assume you were being critical and this wasn't a red hearing chasing a different volume level. (You must match levels within 0.1db or you will pick the louder program as sounding better.)

Jim,

I just want to be sure that I'm getting this procedure correct. So I've shown the following screen shots of what I believe you are recommending. Please let us know if this is correct or needs adjustments. Thanks.

1. Convert the DSD to 32 bit floating point wav using 32:1 decimation:

1(a) XLD DSD Importer Settings set to 88200Hz and 32-bit Floating Point:

1 XLD DSD Importer Settings.png


1(b) XLD Output Format set to WAV:

2 XLD Output format.png


2. Execute the conversion to WAV in XLD.

3. Import resulting WAV file into Audacity (or other DAW App).

4. Normalize the levels to just below zero (-1.0 dB appears to be Audacity default):

4 Normalize in Audacity.png


5. Render to 24 bit fixed and convert to FLAC in Audacity:

5 Export 24 bit flac.png
 
@ar surround

Step 1a is not correct. Select 32:1 Decimation instead of sample rate conversion.

I'm not as familiar with Audacity features...
FYI, make sure you select normalize (common gain) or whatever Audacity calls that. Normalizing the entire album of songs equally to the loudest peak for the whole album as opposed to normalizing each song separately which would alter the volume levels song to song. I'll usually normalize to -0.01 or -0.02.

I'll add that I haven't seen one of the 6db over transcodes in a while. The up to 6db low variety appears most common. That leads to: 9 times out of 10 you'd probably be none the wiser going straight to 24 bit fixed from DSD. The result would be a 23 bit recording. Match the level and A/B against the original and I'm pretty sure no one is going to pick that out of a shootout!

So a little bit academic. But DSD -> 32 bit -> normalize -> 24 bit is the path for a critical transcode and you'll avoid clipping the odd over zero ones that come up. That's what I think I know.
 
I read your trying to normalize the volume levels in the 32bit floating point domain in WAV format and then converting to 24 bit flac.

If you are not normalizing the volume of the PCM due to having no clipping, there is no point in converting to 32 bit wav for editing. Go straight to 24bit 88.2/44/.1 KHZ. That's what i've gathered from reading on this.

I have noticed that the dsd file has a greater dynamic range than the 24bit 88.2khz pcm render. What is the reason for this? They are volume matched.
 
They are normalizing the PCM because raw converted DSD-->PCM peak level can be as much as -6dBFS. (The subject of the first 20 or so posts of this thread). There is no way to know what the actual peak is until you do the conversion.

I agree that it's unnecessary to do normalization in a 32 bit realm, if that's what you're saying. 24bit is fine.

Where and how are you calculating the DR of a DSD file?
 
Jim,

I just want to be sure that I'm getting this procedure correct. So I've shown the following screen shots of what I believe you are recommending. Please let us know if this is correct or needs adjustments. Thanks.

1. Convert the DSD to 32 bit floating point wav using 32:1 decimation:

1(a) XLD DSD Importer Settings set to 88200Hz and 32-bit Floating Point:

View attachment 61818

1(b) XLD Output Format set to WAV:

View attachment 61820

2. Execute the conversion to WAV in XLD.

3. Import resulting WAV file into Audacity (or other DAW App).

4. Normalize the levels to just below zero (-1.0 dB appears to be Audacity default):

View attachment 61821

5. Render to 24 bit fixed and convert to FLAC in Audacity:

View attachment 61822
Thanks for these instructions, this really helped me as I worked through my rip from SACD through to editing and saving as FLAC.
 
I read your trying to normalize the volume levels in the 32bit floating point domain in WAV format and then converting to 24 bit flac.

If you are not normalizing the volume of the PCM due to having no clipping, there is no point in converting to 32 bit wav for editing. Go straight to 24bit 88.2/44/.1 KHZ. That's what i've gathered from reading on this.
You missed the part where the output from the DSD to PCM transcode can exceed zero. (Up to +6db) If you went directly to 24 bit fixed instead of 32 bit floating point, you would have clipped distorted files because fixed point cannot exceed zero.

Any lower level (as much as -6db) program and you would be none the wiser. No one is going to be able to hear anything change from going from 24 bit to 23 bit resolution IMHO.

I have noticed that the dsd file has a greater dynamic range than the 24bit 88.2khz pcm render. What is the reason for this? They are volume matched.
It should be exactly the same unless it's been altered. I'd look into the metering I was using and the signal path for DSD to PCM conversions. Is this measurement being done natively for both the DSD copy and the PCM copy? If not, then one of them is getting converted to the other before the meter sees it. That process needs to be vetted carefully or all meter readings are moot.

I've had the opportunity to compare a few titles released on both DSD and PCM formats. (Water's Amused to Death was one of them. I only have a PCM system, so I convert everything to PCM. Transcoding the DSD copy to PCM and then comparing to the native PCM released copy, the two nulled down to -90db or so. They sounded the same in an A/B test. The difference signal (below -90db) was not audible. None of that can be coincidence. The two copies came from the same source. Proper handling results in a still intact copy that still has the full audio spectrum null against the other copy.

That tells me two things.
1. You can transcode these formats back and forth with virtually no loss at the highest audio standards. (I like this answer!)
2. It's possible to verify when the same recording is put to DSD or PCM.

DSD may have been a cheap cash grab attempt by creating a different "digital language" to lead to incompatibility and competition. But it turns out that it is just as capible as HD PCM as a container for audio.
 
I would strongly urge everyone NOT to normalize anything at all.
Doing this adds another layer of DSP to the process and also at the same time the quantization errors and distortions that will inevitably come with this, and not only that you are also raising the noise floor at the same time.
There is nothing wrong with 24-bit files peaking at -6dBFS - going up to 0dBFS or even to -1dBFS will leave no headroom for transients & ISP's (inter-sample peaks). Yes, ripped SACD are quiet because of the way DSD actually works - bear in mind it was never intended as a commercial consumer format but an archive one & only Sony's greed when they did not get as many DVD patents as they wanted to made it happen in the first place.
 
I would strongly urge everyone NOT to normalize anything at all.
Doing this adds another layer of DSP to the process and also at the same time the quantization errors and distortions that will inevitably come with this, and not only that you are also raising the noise floor at the same time.
There is nothing wrong with 24-bit files peaking at -6dBFS - going up to 0dBFS or even to -1dBFS will leave no headroom for transients & ISP's (inter-sample peaks).
And if it really bothers, wait until after you've converted to PCM, check the peak (true peak, if possible) levels, and then add gain.
 
I would strongly urge everyone NOT to normalize anything at all.
Doing this adds another layer of DSP to the process and also at the same time the quantization errors and distortions that will inevitably come with this, and not only that you are also raising the noise floor at the same time.
There is nothing wrong with 24-bit files peaking at -6dBFS - going up to 0dBFS or even to -1dBFS will leave no headroom for transients & ISP's (inter-sample peaks). Yes, ripped SACD are quiet because of the way DSD actually works - bear in mind it was never intended as a commercial consumer format but an archive one & only Sony's greed when they did not get as many DVD patents as they wanted to made it happen in the first place.

Yes about the history of DSD and archiving.

No about normalizing. DSD was meant to be rendered for CD (or another multiple of 44.1). It was not meant to be rendered to CD at -6dBFS peak.

Just because DSP is used doesn't mean you hear it. Really!

The only 'danger' from normalizing to ~ 0dbFS is if you are running old gear that can't deal with intersample overs. Not that they are super common anyway. If that's a worry , normalize to -1 dBFS. (But normalize the whole album, not track by track..that way you retain the original intent of the producers). And even then *the chance that you 'hear' an intersample over is small*. Audibility distortion from digital clipping depends on how often it occurs within a time span.

I've checked for instersample overs after normalizing DSD-->PCM, using Audition 3. Nothing so far.

Or if you insist that this matters, use something like replaygain to 'nornalize' on the fly (without chaning the audio file)
 
I need to get my SACD ISO files converted to FLAC and I've been putting it off for ages and ages because it seems more critical - and trickier - to get right for issues outlined in this thread, as well as outlined in lots of threads all over the web. Fwiw, I need to convert SACD ISO to FLAC because my OPPO 103D supports gapless playback of stereo and multichannel FLAC via USB; I also use my OPPO for ripping SACDs along with the Sonore ISO2DSD GUI, but it's the next step I need help with.

I rip and manage all my media in Windows 7 so that's the platform I'll be using. I use a Macbook Air for day to day browsing and stuff and so although I have the option of using mac OS, I'd prefer to stick with Windows if possible because all my music is on my Windows computer.

Ideally, it would be great to have a one-programme solution that just needs a few clicks and it's good to go. Things like normalising and replaygain are well outside my comfort zone. I need a simple solution that pretty much does everything for me. Oh, and as well as having the ISO2DSD GUI on my computer, I've also got DVD Audio Extractor, dbpoweramp and JRiver MC19.

So where and how does one begin?
 
How straightforward is the process?
I usually convert DSF to FLAC, but I believe its done the same way. Its pretty easy.

Open Foobar, select add file, choose the ISO. The individual tracks will pop up in the window. (probably the stereo and the MC tracks?? .. again I do the DSF files extracted from the ISO beforehand).

Select the tracks you want to convert and right click. Choose convert, quick convert, FLAC.

Choose your options. Most choose 24 bit depth. There won't be an option for sample rate (see below). click the convert button. Specify where the output is to go.

The sample rate is set in the settings for the decoder. I know its somewhere in the preferences/installed plugins section. You will probably want to specify 88.2kHz. The default will be 44.1kHz

No guarantees here. Like I said, I rip the DSF files off the disk first and then work with those.

Check the end of the tracks for loud pops. I hear lots of people get them, especially on the last track. For whatever reason, I don't.

EDIT: One more thing, if at any point you see something is greyed out or you can't select it, it likely means you don't have the proper plug in installed. You will need the SACD decoder and a FLAC encoder I believe.
 
Last edited:
Fwiw, I need to convert SACD ISO to FLAC because my OPPO 103D supports gapless playback of stereo and multichannel FLAC via USB

I don't have any reason to want to talk you out of FLAC, but if uninterrupted playback is your only real concern, you may be interested to know that you can get it with DSD by using whole album files with cue sheets.

If you already have ISO files, it should be pretty easy to do
 
I need to get my SACD ISO files converted to FLAC and I've been putting it off for ages and ages because it seems more critical - and trickier - to get right for issues outlined in this thread, as well as outlined in lots of threads all over the web. Fwiw, I need to convert SACD ISO to FLAC because my OPPO 103D supports gapless playback of stereo and multichannel FLAC via USB; I also use my OPPO for ripping SACDs along with the Sonore ISO2DSD GUI, but it's the next step I need help with.

I rip and manage all my media in Windows 7 so that's the platform I'll be using. I use a Macbook Air for day to day browsing and stuff and so although I have the option of using mac OS, I'd prefer to stick with Windows if possible because all my music is on my Windows computer.

Ideally, it would be great to have a one-programme solution that just needs a few clicks and it's good to go. Things like normalising and replaygain are well outside my comfort zone. I need a simple solution that pretty much does everything for me. Oh, and as well as having the ISO2DSD GUI on my computer, I've also got DVD Audio Extractor, dbpoweramp and JRiver MC19.

So where and how does one begin?
There’s some good info on this thread Derek-

https://www.quadraphonicquad.com/fo...c-from-multi-channel-discs.29362/#post-526826
 
I don't have any reason to want to talk you out of FLAC, but if uninterrupted playback is your only real concern, you may be interested to know that you can get it with DSD by using whole album files with cue sheets.

If you already have ISO files, it should be pretty easy to do
The big reason to convert to FLAC is if you want to use any DSP or room correction. But if the OP is decoding with his Oppo, he's probably not doing any of that anyway.
 
Back
Top