You seem to have the latest version of the sacd component and I think the 64fp corresponds with the Double-Precision of older versions (fp=floating point?), not shure though.
SACD ISO to DVDA?
- Thread starter s6270
- Start date
Help Support QuadraphonicQuad:
Yes it is the latest as I did not have the SACD plug-in until this weekend when I first gave it a shot
So should I select the 64fp over the 32fp?
So should I select the 64fp over the 32fp?
My guess is yes, I use the same settings as HomerJAU and that takes away the noise I had using other settings. Like I said I never used the latest version so I'm not shure if it does the same.
I use Version 0.7.3 if you want to try that.
I just tried the latest version with 64p applied and heard no difference compared to a rip with my older plug-in so this setting seems ok.
I believe technically 64p is better but like many things in audio whether you will/can actually hear a difference is debatable (and often this is more psychological/placebo than real IMO)
Same with the PCM Sample Rate: I go with 88.2kHz because I cant hear any difference with anything higher, the files are smaller and its still seen as 'high resolution'. But your choice entirely, although if you want to eventually move it to DVD-A then 88.2 is about it for mch. (I'd leave the DVDA at 88.2 as re-sampling to 96kHz adds no additional data)
Yes, but I don't do any 'serious' listening on my PC, just while I'm mucking around in my home office doing conversions etc. I convert to FLAC using those settings and then listen to those files played back in my main room using a media player.
I tried to find some online guidance on this issue a while back. I found a few discussions offering an explanation that DSD somehow has its roots based on the redbook sample rate of 44.1 kHz, and the only real benefit in going with higher sample rates is when it is a multiple of 44.1. I have no idea if that info is correct or internet BS. I do wish someone with good knowledge of the subject could give a definitive explanation.
That the sample rate(s) of DSD are multiples of redbook (44.1) is true. That's because the original intent of DSD was as an archiving format, intended not to be released itself, but to be downsampled (decimated) transparently to 44.1 for consumer CD release.
That multiples of 44.1 are the only beneficial higher rates is false (obviously, the very popular 96kHz rate isn't a multiple of redbook -- it's a multiple of the standard video audio rate, 48kHz).
That higher rates than 44 or 48 are audibly beneficial for the consumer end format, under normal listening conditions, is itself very debatable
That multiples of 44.1 are the only beneficial higher rates is false (obviously, the very popular 96kHz rate isn't a multiple of redbook -- it's a multiple of the standard video audio rate, 48kHz).
That higher rates than 44 or 48 are audibly beneficial for the consumer end format, under normal listening conditions, is itself very debatable
I researched this on the SACD component development site because I was interested in the different filtering options provided. The author of the software says use "Multistage 32fp" if you don't want to filter DSD ultrasonics, use "Direct 32fp" if you do want to filter DSD ultrasonics at 30kHz and use "Installable FIR 32fp" if you want to filter at 40/50/60 kHz. The custom frequency filters are included with the package. He said some PCM DACs amplify the DSD ultrasonics if they aren't filtered out.
The author also mentioned he didn't see much point in the double precision (64fp) versions, though he included them for those that think they might make a difference.
So are you saying that once the SACD is converted this way you have to run it through a filter? I am not sure how that would be done, do you do that in foobar or in an audio editor?
So bottom line question, what do we get by doing it this ripping way that we don't get if we just record the SACD into the PC using an SACD player and capturing the playback at 24/96 or 24/88 in real time? (Obviously the PC recording setting could be wrong so the result files could be messed up, but other than that)
himey
2K Club - QQ Super Nova
- Joined
- Nov 11, 2013
- Messages
- 3,482
Once your SACD files are converted, your fine, no additional filter is needed.
As far as advantages using this method vs the line out, my guess is the additional dac conversion from digital to analog on top of the recording process will add more abnormalities than the simple conversion to pcm within the computer. Also, we are talking multichannel here. The process is going to be cumbersome recording three separate stereo channels and syncing them up in an audio editing program. Less fidelity and more work.
Ideally you want to convert to individual dsd files and play back without converting to pcm using a reciever or dac that does dsd without conversion, but not all of them accomplish this so you do the next best thing which is this conversion to pcm, and lossless compression to flac, or authoring of a DVDA disc.
What I meant was, when converting the DSD to PCM that there is no advantage in going to 96 kHz instead of an 88.2 kHz sample rate. But there would be an advantage in going to the next multiple of 44.1, such as 132.3, 176.4, etc. Having clairified that, I still am not sure if its true. And like you said, beyond that, is it audible?
Last edited:
I think himey answered the question, but I wanted to confirm that the filtering is built in to the conversion if you choose the direct or installable methods. I'd say the advantage of going this route vs recording over analog depends on how good the conversion is. If the analog DACs do a better job than than this tool, then you are probably better off recording from analog. I use Weiss Saracon for conversion, but it looks like this foobar method is getting better. The early versions had a bad rep.
The way to test it is take a PCM wave file, convert to DSD and then convert back to PCM and see how close it is to the original.
This is not an issue if you have a 6 channel recorder such as MOTU 828.
Digital capture is not bound by analog constraints and you get some of your life back. I'd rather do my rip in 15 minutes than 'real time' 40 minutes. With digital rip you are also capturing the album metadata (Artist, Album, Track titles) and both stereo and MCH tracks in one go!
Foobar does the filtering during conversion. I cannot hear any difference between a SACD played directly (Oppo 95) and via a SACD ISO rip converted to FLAC at 24/88.2kHz, but I've got old ears...
The thing is, going analog and the next thing your spending $1000's on esoteric cabling and other gear that's just not needed, but you may think will improve you experience. Don't go there!. Go digital!
When you recode from DSD to,say, 88.2 PCM, you are already omitting everything above 44.1khz.
Any ultrasonic 'DSD noise' between 22 and 44 kHz remains. You'll tend to see it as a haze up near 40 kHz if you look at a spectral view of the PCM file. I have never heard of this being audible. I suppose theoretically if you played such tracks at ridiculous volume, it could damage a limited tweeter.
Adding a 30 kHz low pass filter (this is what two of the options do) simply filters out everything between 30 and 44 khz. The file would still be 88.2 though (has a 44kHz bandwidth), it's just 'dark' above ~30khz in spectral view. Totally tweeter-safe.
FWIW, hardware SACD players, in accordance with Scarlet Book spec, typically had a 50 kHz low pass filter at the end of the output stage. To prevent the sh*t-ton of ultrasonic DSD hash that occurs above that from freaking out consumer systems.
Currently, I just use the default converter for foo_input_sacd (Multistage 32fp [ floating point ]) at 88.2 kHz SR. I fiddle with the PCM Volume level (+0 to +6) until there are no 'overloads'.
Can you please explain what some of those settings in Foobar mean. Specifically, the DSD2PCM mode section, where you chose Multistage 32fp. What do all those settings mean, and why do you choose the one you choose. Again, I'm a noob trying to understand....
Damn....that's an awesome link.
Similar threads
