Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[teodorom]

Hi,
I confess that, since I convert from PCM to DSD, I'm using MC 27.0.85 only as a front-end to HQPlayer.
The quality of the conversion done by MC is quite good, but not as good as the one provided by HQPlayer (ASDM7).
Please make this little effort!
Thanks
Teodoro Marinucci

[tij]

Also curious ... why convert PCM to DSD? ... if DSD format can encode some sonic things that PCM cannot, those things wont be present in PCM anyway ... am i missing something?

I can understand DSD to PCM as not many things support DSD.

[teodorom]

am i missing something?

Yes, you are missing something.
However I'm unable to explain to you the reason why.
Some people in Audio Review (Italian Hi-Fi magazine) discovered that, after the conversion from PCM to DSD, music flowed more "naturally".
I didn't believe that, but I had to try and then I was easily convinced. Even if I agree with you that information cannot be created. It works, simply.
It's a feature that is already present in MC, so if you think that you are missing something you should ask to those having developed the "modulator".
That, I repeat, is quite good but the work of Jussa Laako (a.k.a. Miska) is slightly better (even if much more expensive, and more complicated to handle, too).

[Soundwave]

Yes, you are missing something.
However I'm unable to explain to you the reason why.

See, this is a problem you're going to need to rectify. Because unless you can give a detailed enough explanation of what it is you want made better, how would the developer know what to do if he asked you the same question that the other guy just asked you about what it is you actually need fixing?

Some people in Audio Review (Italian Hi-Fi magazine) discovered that, after the conversion from PCM to DSD, music flowed more "naturally".
I didn't believe that, but I had to try and then I was easily convinced. Even if I agree with you that information cannot be created. It works, simply.

Aside from the potential violation of laws of logic occurring with this statement. Your best bet is for then to provide you with the details of what needs to be done technically. Unless of course you expect the developers to go hound "some people" and ask them..

It's a feature that is already present in MC, so if you think that you are missing something you should ask to those having developed the "modulator".
That, I repeat, is quite good but the work of Jussa Laako (a.k.a. Miska) is slightly better (even if much more expensive, and more complicated to handle, too).

That's not how things work in the realms of requests and claims. If I say I saw someone fly. And you ask me where did I see such person do that. If I reply "if you think you're missing something in the understand of how people fly, go ask the guy I saw". You would laugh your away out of my presence.

Beyond the obvious questionable claim you make about "better conversion" (which you openly state you can't even define yourself beyond vacuous descriptors like "natural"), no one is going to go hounding developers of other programs and ask them "yo dude, can we have the method which your DSD decoder, encoder, converter, and "modulator" actually work?

This goes for everyone, and not just you. If I told someone "yo can you improve the FLAC decoder because when I use X-Player when encoding my files to AAC, I think the FLAC decoder X-company has produces a more natural baseline for the lossy AAC encoding I'm going to need to do" - they would rightfully roll their eyes. The idea of using the word "natural" to describe an entirely technical process is completely uninformative.

This is again, granting the idea there is an audible difference as you claim in the first place - which is simply being entertained for argument's sake, but in reality would need to be blind tested against to see if your experience was psychologically influenced, or you actually are hearing a relevant difference.

[Stellabagpuss]

Hi,
I confess that, since I convert from PCM to DSD, I'm using MC 27.0.85 only as a front-end to HQPlayer.
The quality of the conversion done by MC is quite good, but not as good as the one provided by HQPlayer (ASDM7).
Please make this little effort!
Thanks
Teodoro Marinucci

I totally undertand what you are trying do, HQ player has a lot of fans doing PCM to DSD conversion, with exactly the general response of sounding better, I am no expert, but I am guessing it must be in the software of HQ Player, or DSD has a slight signature. I must point out, I haven't tested this out myself.

However, many years ago, I read the same thing with FLAC vs WAV, and although I found there was nothing partically bad about FLAC, I found WAV to flow better, and I still use WAV to this day, or I convert FLAC back to WAV, perhaps all formats have a slight signature, which sounds crazy being digital, perhaps someone can give a reason behind this?

[tij]

Regarding WAV vs FLAC ... i am at dead end ... both contain exact data ... with difference being FLAC is compressing it losslessly and during playback it gets decompressed to exact "WAV"

With DSD vs PCM ... it is possible that DSD DAC is of better quality than PCM DAC ... maybe made intentionally by manufacturer so that ppl buy DSD DAC

Like Soundwave said ... unless you can quantify the problem ... there is no way developers can do your request ...

One way is to expose parameters used to convert PCM to DSD for users to fiddle with ... but that can be dangerous? ... possible equipment damage?

[dtc]

HQPlayer takes a very different approach to digital music than MC. MC's basic premise is bit perfect audio, an outgrowth of the days when everything went through the unknowns of the Windows Mixer. HQPlayer is not concerned with just bit perfect audio. It uses a Sigma- Delta Modulator (SDM) similar to what is used in many DACs.   The OP's original reference is to ASDM7, which is simply their name for the 7th order asymmetric SDM used in HQPlayer. I believe HQPlayer has both 5th order and 7th order SDMs. For a tutorial on SDM see here.

http://www2.ing.unipi.it/~a008309/mat_stud/MIXED/archive/A2011/Approfondimenti/Understanding_sigma_delta_CUT.pdf

HQPlayer also implements a large number of filters, some shared between PCM and DSD and some specific to the given format.  For a discussion of the sound of some of these filters, see here.

https://audiobacon.net/2021/03/17/hqplayer-better-than-a-5000-upscaler/4/

That article is also a good general purpose view of some of the issues that people who use HQPlayer are concerned about.

For a more technical discussion about  HQPlayer filters when used in PCM to DSD conversion and how the results compare to MC, see

http://archimago.blogspot.com/2019/02/measurements-look-at-hqplayer-325.html

What the OP original ask about is pretty well understood in the HQPlayer community and should be something that JRiver is aware of, even though they are unlikely to implement the type of architecture that would allow the conversion and filtering capabilities that provide the improved sound that the OP referenced.

[dtc]

Jim - Why move this? Is was a direct feature request for JRiver. Yes, it involved a third party as a comparison, but the request was for you to implement a similar feature. 

[JimH]

You're in the realm of audio embellishment and that's not what JRiver does.  Tweak all you want.  It's just not core to our mission.

[dtc]

The OP put in a feature request. I simply tried to explain what the issue was to some who did not seem to understand.  A simple response to the OP from you in the original thread would have put an end to the discussion. I already said you were probably not going to do it. I did not advocate for you to do it. I just explained the background.

[teodorom]

Some comments are really insulting.
Anyway: I don't want to discuss the benefits (if any) of the conversion from PCM to DSD, I simply state that:

• The conversion from PCM to DSD is already there in MC: if you really think it's a useless feature you can drop it (and I shall stop to buy MC)
• A lot of people are spending much more money with Roon and HQPlayer: let me guess that they are all stupid or deaf

As far as I see in version 28 you are more focused on video than on audio. Me (only me?), on the contrary I'm more focused on audio and I'm not interested at all in video, so, I'll try version 28, but "sic stantibus rebus", I shall skip it.

[JimH]

Everyone has a point of view or opinion.  It's not personal.

To your comment about JRiver being "more focused on video than on audio", we get that from all sides.  "You're ignoring [insert subject here]"   It's not so.  We're doing our best to provide a broad and deep product that has a familiar interface for organizing and playing all kinds of digital media.  That you only care about audio doesn't mean you won't find photos or video interesting in the future.

I think if you look at the details of what we do, you'll find loads of work on audio.  We also do a number of things that are in the "infrastructure" category.  Things that make everything work better.  Servers, for example.  Remotes.   

And "audiophile" means person who loves audio.  It has come to sometimes be used as person who loves to tweak and twiddle their audio.  There's room for both, but JRiver has limits.  If you need to go somewhere else to find something you need, well fine.  Just don't slam the door on JRiver.  Don't resent us when we don't do your favorite feature.  You may find something else you haven't discovered or something that's coming in the future.

[teodorom]

Sorry about that, but I've got that opinion ("more focused on video than on audio") by looking at version 28 features.
Perhaps I've not well understood (Netflix Navigation, Rip with MakeMKV, PTV) or the list was not complete, but, please, give me your insight on my two points.
I'm not happy when I read concerns (or misunderstandings) about PCM to DSD conversion.
If you consider this point a minor (or useless) feature, drop it!
If you think that, improving it, you can get the attention of some more "audiophiles", do it!

[JimH]

Those changes are only what we've done so far. 

Take a look here for changes in previous versions:
https://wiki.jriver.com/index.php/Release_Notes

Work on things like Expressions or MCWS can benefit both audio and video users.  Other developers can use them to do new things, so we pay a little more attention to them.

I can't see how converting from PCM to DSD would improve the sound.  And I don't see how a different method would work.  Turning up the volume is a simple way to improve the sound, especially if you're in your 50's or older.

[dtc]

I can't see how converting from PCM to DSD would improve the sound.  And I don't see how a different method would work. 

Take a look at the Archimago article I linked to. He clearly shows that using different modulators and different filters produces different results from HQPlayer.

And look at the Audiobacon article I linked to. He reports on the sound, to him, of different combinations of modulators and filters in HQPlayer.

Shannon-Nyquist is a theoretical equation. But, in the real world the conversion from Digital to Analog is not a closed form equation. Different techniques produce different results. That is, at least in part, why different DACs sound different and why some people prefer one DAC over another. HQPLayer is basically just using software to do things similar to what  DAC designers use in their designs.

The differences are subtle and not everyone can hear the differences. But some people can. That is why there are different DACs and different options in players like HQPlayer.

You may not want to get into that type of development. But it would be nice if you at least acknowledged that it exists and that some people can hear the differences.

[teodorom]

Take a look at the Archimago article I linked to. He clearly shows that using different modulators and different filters produces different results from HQPlayer.

Thanks, this is precisely what I wanted to say.
But now I must realize that the conversion from PCM to DSD is something that people in MC added as as feature without believing in it, without understanding it.
Why you added it?
You can simply eliminate it, so that I shall not bother you anymore.

[dtc]

Thanks, this is precisely what I wanted to say.
But now I must realize that the conversion from PCM to DSD is something that people in MC added as as feature without believing in it, without understanding it.
Why you added it?
You can simply eliminate it, so that I shall not bother you anymore.

The routine that MC uses is pretty standard. The method that HQPlayer uses is a much more complicated one, based on their SDM architecture. For most of the users here, the difference is probably not noticeable nor important. Many people here are happy with the PCM to DSD conversion.   People who value the small differences in sound migrate to HPlayer. The ones who value the user experience and the integration of images and video stay here.  That is why Roon decided to provide HQPlayer as an output. They wanted to provide their user environment to people who valued HQPlayer.  JRiver has decided that that integration is not something they want to do and that is there choice.   The author of the Audiobacon article concluded that his DACs were a better option than the HQplayer options. Different products exist for difference audiences. No one product can satisfy everyone.

My point is that it is good to understand the various different options and make your choice based on what is most important to you.  And don't belittle the products who make different design choices.  Miska's advantage is not in front ends and multimedia. JRiver advantage is not in SDM and filter design.  Two excellent products that serve separate audiences.

[Soundwave]

Take a look at the Archimago article I linked to. He clearly shows that using different modulators and different filters produces different results from HQPlayer.

And look at the Audiobacon article I linked to. He reports on the sound, to him, of different combinations of modulators and filters in HQPlayer.

Shannon-Nyquist is a theoretical equation. But, in the real world the conversion from Digital to Analog is not a closed form equation. Different techniques produce different results. That is, at least in part, why different DACs sound different and why some people prefer one DAC over another. HQPLayer is basically just using software to do things similar to what  DAC designers use in their designs.

The differences are subtle and not everyone can hear the differences. But some people can. That is why there are different DACs and different options in players like HQPlayer.

You may not want to get into that type of development. But it would be nice if you at least acknowledged that it exists and that some people can hear the differences.

It's still not apparent how this improves audible sound quality. I took a look, and saw no audible band difference of note.

Maybe I'm tired and missed it. But where do you see an audible improvement? For FR I see nothing, and for Noise I see differences below human threshold of hearing.. Lastly, is there a blind test between HQPlayer's DSD from PCM conversion being tested against any other PCM to DSD coverters?

If neither of these are provided. You now have the answer to why he doesn't see how the sound could be improved.. Nor would he have reason to acknowledge the last claim you make, the one being: "People can hear the differences". I'm just not seeing where this is actually proven anywhere. Also, do take a look at what DSD is in general a bit, https://troll-audio.com/articles/pcm-and-dsd/ converting to it is practically pointless. I understand if a studio wishes to record with it, and simply having DSD playback for that purpose without having to incur a lossy hit by way of conversion. But to take PCM to DSD, as I said, practically pointless. Especially with the injection of so much noise into the ultrasonic band.

[dtc]

It's still not apparent how this improves audible sound quality. I took a look, and saw no audible band difference of note.

Maybe I'm tired and missed it. But where do you see an audible improvement? For FR I see nothing, and for Noise I see differences below human threshold of hearing.. Lastly, is there a blind test between HQPlayer's DSD from PCM conversion being tested against any other PCM to DSD coverters?

If neither of these are provided. You now have the answer to why he doesn't see how the sound could be improved.. Nor would he have reason to acknowledge the last claim you make, the one being: "People can hear the differences". I'm just not seeing where this is actually proven anywhere. Also, do take a look at what DSD is in general a bit, https://troll-audio.com/articles/pcm-and-dsd/ converting to it is practically pointless. I understand if a studio wishes to record with it, and simply having DSD playback for that purpose without having to incur a lossy hit by way of conversion. But to take PCM to DSD, as I said, practically pointless. Especially with the injection of so much noise into the ultrasonic band.

First, there is more to this than just conversion from PCM to DSD. There are also the processes after than that actually play the DSD.  It is similar to playing just PCM files. The whole process matters.

Miska's graph show differences, which you may or may not think cause audio differences.  But there definitely are differences. He admits he hears only very minor differences. 

Now, couple that with the Audiobacon review where a trained listener reports that he hears differences with HQPlayer and reports on them, filter by filter.

And, if you go to the HQPLayer forums you will find many who claim they hear a difference.

Sorry, but many people report they hear differences with these various processes, even if you do not believe it.

[JimH]

Nobody doubts that people are sure they hear differences.  But logic is logic and bits are bits.  We're sticking with the science of the subject.  Please find another forum to debate audio mythology.

[Soundwave]

First, there is more to this than just conversion from PCM to DSD. There are also the processes after than that actually play the DSD.  It is similar to playing just PCM files. The whole process matters.

Miska's graph show differences, which you may or may not think cause audio differences.  But there definitely are differences. He admits he hears only very minor differences. 

Now, couple that with the Audiobacon review where a trained listener reports that he hears differences with HQPlayer and reports on them, filter by filter.

And, if you go to the HQPLayer forums you will find many who claim they hear a difference.

Sorry, but many people report they hear differences with these various processes, even if you do not believe it.

This discussion, will, of course, go nowhere and Jim will lock it. I just wish people would have some respect for listeners who say they hear differences.

I'll repeat, I'm looking for audible differences between PCM to DSD converter comparisons. The guy on Audiobacon isn't doing a blind test, nor are his descriptors indicative of any technical merrit, nor is it comparing converters, he's just comparing filters.

Also, I'm not going to go on forums hunting for something you claim exists, for a basic question requesting the comparison between converters. Ideally I'd hope you can provide that, since you're a believer of the claims.

You also just strawmanned me when you said: "Sorry, but many people report they hear differences with these various processes, even if you do not believe it." I do believe they are making a claim of hearing differences. Whether they're born out of genuine auditory ability, or sighted bias is an entirely different story, and the main topic of contention.

Also, what do you mean "have some respect for people who say they hear differences". Who's disrespecting anyone here? OP asked for an update, a few people wanted details on what he was talking about, he wasn't able to provide any detail beyond vacuous descriptors of the sound being "more natural" when listening to HQ DSD vs JRiver DSD. He gets irked somewhat. And here we are now, having you tell us that we're not respecting someone who's asking for something he can't offer details on what exactly he wants, or how JRiver could even get there.

This is all before even granting any of the staggeringly dubious claims about the sensibility of taking native PCM and converting it into DSD for whatever reason (well we know the stated reason why, but no proof of such is provided in terms of audibility by any party, not just OP).

[dtc]

Nobody doubts that people are sure they hear differences.  But logic is logic and bits are bits.  We're sticking with the science of the subject.  Please find another forum to debate audio mythology.

Same old discussion. It does get tedious. You said you did not understand the differences I was pointing out. I just pointed you to the articles that I had already posted.

Yes, bit are bits. And bits put through different algorithms can produce different results.  The output of your PCM to DSD conversion is different than what comes out of the HQPlayer process. So, it is quite possible they sound different. Can you really deny that? And DACs can sound different because they use different algorithms.

I'll debate science any time, any where. I have a Ph. D in chemical physics, did post doctoral research in physics and am published in many physics journals. Learned Shannon-Nyquist about 50 years ago, long before it was a popular topic in audio. Also taught computer science at the college level.