Has anyone played with the USB DAC DSD settings and heard changes in sound for non-DSD tracks? Just using Qobuz, I would think this would make no difference since everything is FLAC, but I started adjusting those yesterday and each sounds different. This is going to a Classe Delta Pre. Of the three, I have a strong preference for DSD over PCM. ‘Native’ loses a lot of dynamics, especially in the bass and mid-bass. There is a lot of bass, but very little attack and slam, almost like it is crossed over too low. DSD to PCM sounds thin in those areas. Per definitition of what these settings do, I would expect no difference, so curious if anyone has noticed the same or can explain why. I am using an RS-130. Thanks.
i try dsd over pcm, the sound just unaccepted if compare to dsd native. 
the other thing that affected ur sound is are u using i2s with hdmi to connect ?
From my understanding, my DAC only does DSD Native if a specific driver is used on a Windows PC. it definitely still plays the track in the Native setting, but I’m not playing DSD tracks either. I’m not aware if this driver exists in the RS130 USB implementation. The DAC is USB, AES or Optical only, unfortunately.
DSD over PCM is exactly the same thing as Native DSD, just packaged differently, so whatever difference one hears is purely psychological.
2 Likes
Fair enough, but I am not playing any DSD tracks at all. Would you expect DSD TO (not over) PCM to sound different if I were playing DSD files (which I am not)?
Depends, since PCM and DSD are entirely different ways of encoding sound, and there is no One True Way of converting one to the other. Depending on how that conversion is done (whether by DAC’s internals, the streamer, or some external software) it could sound different, although in practice as long as one isn’t applying too much extra processing during conversion, the difference should not be in any way audible. For a specific DAC and your own preferences… you could try it. It’s still more likely to be more psychological perception and lessany real difference though.
I find it strange that the settings would have any impact on a FLAC track at all. The only reason I messed with the default, which is Native, is because I put a much cheaper preamplifier in for the sake of experimenting and immediately felt it sounded better. I went back and forth a few times and finally in disbelief started trying to find if I was missing something and noticed the HiFi Rose was set to DSD over PCM versus the Native on my Classe, and that the Classe owners manual says for native the streamer/PC has to have a specific driver. After changing it for the Classe my feelings toward the cheaper preamp changed and I’m in love with the Classe again. Maybe it’s all in my head, but I guess that’s all the matters since it’s just me sitting there listening.
Classe could be doing something… strange… depending on how this is configured, or maybe Rose would be doing something it should not depending on the setting (a bug in Rose software… would not be unheard of), but if everything were working right it should not have made any difference. On the other hand, if you like the sound, that’s what’s important.
Problem with Native DSD isn’t so much the driver per se, but the sending device knowing what the receiver can handle. USB Audio only defines standards for PCM, so no matter what you plug into it, as long as it supports UAC2.0, devices can establish what is the maximum PCM rate that can be sent to the DAC. For DSD, there is no standard, so at best the sender can figure out what is the highest PCM rate that can be sent, and try sending DSD over PCM at that rate and hope that the DAC can deal with it. If the source either “knows” (i.e. the software has been built with capabilities of a certain DAC coded into it)OR if it has a custom driver then it can also figure out what the DSD capabilities are and send it appropriately. For obvious marketplace reasons, manufacturers generally only make drivers for Windows. People maintaining the audio side of Linux had made configurations for many DACs available, and you can compile the Linux kernel with proper DSD support for many devices. But chances of a random DAC working 100% with DSD from a source running a very customized Android are pretty low…
All that said, if you are making these changes in the RS130, and it affects how PCM tracks sound, this is most likely some edge case bug that Rose would need to look into. Since there does not appear to be any setting in the RS130 to convert PCM to DSD, PCM from Qobuz should be sent exactly the same regardless of how these are set…
1 Like
Is your comment based on fact. Is your comment based on audio file playback or on music streamed audio?
If it’s based on streamed audio or NAS audio file, did you try it with SFP Fiber Optic Ethernet? Did you try it with a good (possibly custom pin configuration) i2s HDMI cable? Was your cable DVI to HDMI or was it Mode A or B i2s HDMI 2.1 to custom pinned HDMI cable?
From what I’ve read: Many people have stated i²s is the best Sound Quality – i2s carries the sound, with clock, etc… Other people stated they did i2s/USB tests and preferred the sound quality from the USB SFP for their system.
Do you own a HiFi Rose product? If so, what have you done to improve sound quality? Please no more blah, blah, blah negative comments without supporting facts (like tests you did on your system).
Please add factual value to the posted topic. Other posts have stated what they have done to improve Sound Quality like using the Internal SSD,…
Now, back to topic
I don’t yet own the HiFi Rose RS130 – I’m still doing research. I too am interested in hearing the answer: which has better sound & what are the sound differences between: Native DSD, DSD to PCM (processed by RS130) or, in my current case, FLAC file playback?
1 Like
Thanks for the detailed explanation, it makes sense. It could very well be something with the Rose software as well. I am pretty confident that each of the three settings sound different with FLAC files on the Classe over USB, but I know the mind can do crazy things as well. I guess l’ll just enjoy it and not stop analyzing it!
1 Like
You’re welcome.
Yeah, it’s really strange. Of course if you decide once that there is some difference between these settings, it is very hard to unhear it without running a bunch of blind tests, although it would be quite interesting to find out whether there is something (and Rose has a bug to fix) or it’s just a psychoacoustic illusion. On the other hand, Rose should have equipment to capture digital output and see if it is at all affected by DSD settings…
1 Like
Hi
I always do this for most audio files, regardless of the original format. I am upsampling all at DSD 512. For me, it sounds better to upsample because the internal processes of my DAC for playing DSD are just providing better final audio results compared to FLAC decoding. It is not a massive difference, but a difference is a plus for me.
These are completely independent and unconnected processes, but sure, if you use something like HQPlayer you can tweak conversion to create a different sonic signature. It’s quite debatable which one is more “correct” though.
In my opinion, it is not about what is correct, more or less. It is about what you like, more or less.
Exactly. If you can really hear the difference, and prefer it that way, that’s fine. It’s just not necessarily re3producing what was recorded, which is also a valid preference.
I really struggle with this. I bought what are considered as “revealing” speakers because I do like hearing what was intended, or as close to it as possible. I’m not sure if what I’m hearing (or think I’m hearing) in DSD over PCM is true to the source or not, but I do feel its closer than what I was getting with Native. I guess I need to visit a HiFi store soon and play the same tracks. 
The most critical component of your complex audio system is your room.
According to my personal experience, the room is the most omitted aspect of the audio system despite being by far the most important. If you go to a hi-fi shop, you will find a very different room than yours. This will be for sure. What you will hear there It will not be the same as your place.
This is no 1. In hifi shops, some good rooms are very little sonically prepared. Only very few shops have invested in the correct room.
But in any case, According to my experience till now. Any decent DAC provides better output in DSD decoding mode.
2 Likes
I am not sure if doing your own PCM to DSD conversion is really advantageous to anything but Jussi’s bank account (most contemporary DACs do it internally, in highly optimized hardware), but you can certainly tweak the sound to your liking.
But @bogdan.murarescu is absolutely correct – the room (and speakers, too) are most important and most overlooked components, especially if you want “accurate” rather than “pleasing” sound. This days it is reasonably easy to get quite flat/transparent electronics that will add no or minimal colorations all the way to the amplifier output terminals. No existing speakers are quite flat though, and wwith added room interactions what you actually hear might be quite far from the “accurate” output of the equipment. A loot of it can be fixed by convolution filters, but some things can only be taken care of by careful positioning of speakers, and adding acoustic treatment.
Definitely won’t argue on the importance of room treatment, speaker placement and equipment. I just moved my speakers back .5 inches last night and added a 1 degree tilt. Even small adjustments like this make some difference. I actually think it’s because I’ve put so much time into these areas and listening to the changes that I’m able to notice, or think I notice, changes liked referenced with the DSD settings.