Upsampling to anything other than your DAC's internal conversion rate

[Norton]

What I have heard so far is that there are some DACs that do seem to prefer DSD and others that seem to prefer PCM. I think Miska has made a very coherent case for why doing the filtering at much higher sampling rates is capable of reducing noise artifacts in the audible range (see post #19 above). So, based on the responses submitted so far let me take a cut at a DSD pro and PCM pro list (and please feel free to correct my mistakes and I'll republish the chart:

 

Best with DSD input: iFi, Oppo, T+A, Teac, exaSound, Fostex, Sony and most other ESS Sabre-based DACS

 

Best with PCM input: Chord DACS, NAD, Schitt (Yggy), Prism Callila

 

After this list gets a bit further along, I'll try to take a shot at also including some specific input rates for each DAC (knowing that risks much greater disagreement).

 

From my listening there's a category missing - best with bit perfect input, DSD or PCM . I'd definitely put both the Hugo and my current Esoteric in that category -noting that the latter is a very close relation of the TEAC dacs you mention above.

[Melvin]

My ears/brain prefer no software up-sampling with both my Schiit Bifrost Multibit and Chord Qute EX DACs. No big surprise considering both manufacturers are already listed. That being said, both DACs, when used with my Wyred 4 Sound re-clocker, which re-samples everything to 24/96, sound fantastic.

[sdolezalek]

My ears/brain prefer no software up-sampling with both my Schiit Bifrost Multibit and Chord Qute EX DACs. No big surprise considering both manufacturers are already listed. That being said, both DACs, when used with my Wyred 4 Sound re-clocker, which re-samples everything to 24/96, sound fantastic.

 

Melvin: Without trying to single you out specifically, if I can use what you said above to make a broader point:

 

1. Your DACs are both doing internal upsampling and applying the filters the manufacturer chose to do so.

2. Your Wyred 4 Sound reclocker, to the extent it resamples everything to 24/96 is also applying its own and different set of filters

3. HQPlayer or Audirvana/iZotope would each also be applying their upsampling and filter methods

 

If we assume (which may not be a fair assumption) that all of the upsampling is bit perfect in that it applies no DSP or other changes to the signal except duplicating data points exactly, then the differences you hear will all be due to the choices of filters applied (chosen for you in the case of Schitt/Chord/Wyred and chosen by you in the case of HQPlayer/Audirvana).

 

Given the range of opinions expressed about filters/upsampling, I think it is fair to assume that the filters DO have an impact on the resulting sound. Now, if you chose the DAC(s) you have by comparatively listening to them against other DACs, part of what you were doing was showing a preference for that manufacturer's choice of filters. Thus, if you "love" that particular sound, it would not be surprising that you would prefer it over a set of filters that produce a different sound. For most folks who use their DACs with HQPlayer or Audirvana but chose no upsampling, I think that all they are really saying is "I like the filtering choices made by my DAC manufacturer."

 

The next level up though would be to ask, which of the above is closest to the original recording? I think Miska has clearly shown why upsampling data to a higher rate allows a set of filters to be applied (whether by him or by your DAC manufacturer) that produces fewer artifacts in the audible range (so has nothing to do with higher frequency hearing/response beyond 20kHz/steeper transients, etc. that often get referenced to say upsampling can't make a difference). What is obviously much, much harder is to show before and after plots for your specific DAC using its internal upsampling/filters versus your DAC using upsampled input with HQPlayer filters applied. The Meridian folks backing MQA are suggesting they can do something along those lines (ignoring anything else they are doing re compression or copy protection) that ultimately leads to your DAC producing an output more faithful to the original recording. Miska is doing the same in continuously testing/improving his filters.

 

That is the main reason I focus on these software players -- because as time progresses, the software filters can keep getting better (for those of you who are photographers, it is like running a picture you processed in the 1995 version of Photoshop and today rerunning the RAW version of that file through the latest version of Photoshop; it is amazing how much more can be done to the RAW original with the capabilities of the new software). In theory, the same will continue to be true of the music software as we keep learning beter and better filtering methods.

[Melvin]

Melvin: Without trying to single you out specifically, if I can use what you said above to make a broader point:

 

1. Your DACs are both doing internal upsampling and applying the filters the manufacturer chose to do so.

2. Your Wyred 4 Sound reclocker, to the extent it resamples everything to 24/96 is also applying its own and different set of filters

3. HQPlayer or Audirvana/iZotope would each also be applying their upsampling and filter methods

 

If we assume (which may not be a fair assumption) that all of the upsampling is bit perfect in that it applies no DSP or other changes to the signal except duplicating data points exactly, then the differences you hear will all be due to the choices of filters applied (chosen for you in the case of Schitt/Chord/Wyred and chosen by you in the case of HQPlayer/Audirvana).

 

Given the range of opinions expressed about filters/upsampling, I think it is fair to assume that the filters DO have an impact on the resulting sound. Now, if you chose the DAC(s) you have by comparatively listening to them against other DACs, part of what you were doing was showing a preference for that manufacturer's choice of filters. Thus, if you "love" that particular sound, it would not be surprising that you would prefer it over a set of filters that produce a different sound. For most folks who use their DACs with HQPlayer or Audirvana but chose no upsampling, I think that all they are really saying is "I like the filtering choices made by my DAC manufacturer."

 

The next level up though would be to ask, which of the above is closest to the original recording? I think Miska has clearly shown why upsampling data to a higher rate allows a set of filters to be applied (whether by him or by your DAC manufacturer) that produces fewer artifacts in the audible range (so has nothing to do with higher frequency hearing/response beyond 20kHz/steeper transients, etc. that often get referenced to say upsampling can't make a difference). What is obviously much, much harder is to show before and after plots for your specific DAC using its internal upsampling/filters versus your DAC using upsampled input with HQPlayer filters applied. The Meridian folks backing MQA are suggesting they can do something along those lines (ignoring anything else they are doing re compression or copy protection) that ultimately leads to your DAC producing an output more faithful to the original recording. Miska is doing the same in continuously testing/improving his filters.

 

That is the main reason I focus on these software players -- because as time progresses, the software filters can keep getting better (for those of you who are photographers, it is like running a picture you processed in the 1995 version of Photoshop and today rerunning the RAW version of that file through the latest version of Photoshop; it is amazing how much more can be done to the RAW original with the capabilities of the new software). In theory, the same will continue to be true of the music software as we keep learning beter and better filtering methods.

 

I was saying exactly what you so succinctly stated .. yes, I prefer what the DAC designers chose for me. This is also not a slight toward Miska or Damien (amongst others) whose work I admire and have paid for. It's simply a preference. We are fortunate to have so many talented folks offering so much choice.

 

My inclusion and comment on the re-clocker was to simply illustrate even with my "clear cut" preference, I love the SQ when the DACs are fed a cleaned up and ASRC'd signal. So much for absolute preference!

[ericuco]

I was saying exactly what you so succinctly stated .. yes, I prefer what the DAC designers chose for me. This is also not a slight toward Miska or Damien (amongst others) whose work I admire and have paid for. It's simply a preference. We are fortunate to have so many talented folks offering so much choice.

 

It is really a pleasure to have a great discussion with people agreeing to disagree without getting nasty about it which is often the case on CA (and other forums).

 

Also, it strikes me as funny that this whole conversation would have been considered heresy about 5 or so years ago when I first started following CA. At that time, you had to make sure the software was NOT doing anything to the file being sent to the DAC ... must be bit perfect, no funny business. But that was before we (well some of us) learned what was going on inside the DAC's with all the upsampling, etc. Now it seems reasonable if the DAC's are monkeying with the file, why not use a more powerful computer and software to do it prior to heading into the DAC.

 

As the recent Noble prize winner penned, The Times They Are a-Changin' ....

[austinpop]

Great discussion.

 

But I do want to bring this back to the central point about DAC knowledge. How do we know what the magic sample rate is for each DAC that minimizes the DAC's internal proceeding?

 

I doubt DAC vendors are willing to disclose this, are they? For example, my Ayre Codex DAC supports up to DSD128 for DSD, but 32/384 for PCM.

 

Even if I accepted your point that something like HQPlayer is going to sound better, what would I set the upsampling to?

 

 

Sent from my iPhone using Computer Audiophile

[YashN]

But I do want to bring this back to the central point about DAC knowledge. How do we know what the magic sample rate is for each DAC that minimizes the DAC's internal proceeding?

 

You'll have to have a versatile DAC, something like HQ Player and then listen carefully.

 

On the theoretical side, you could look for a DAC in which a particular mode defeats a filter internally or something like that. Or if you can have a look at the DAC Chip datasheet and the DAC schematics, that can provide more clues.

 

On the other hand, there's always also a subjective element involved as well.

 

So, you have to try out the combinations, and to reduce that number, perhaps ask other members what they prefer and start from there.

 

Finding your system's sweet-spot is inherently a pursuit of an audiophile IMO.

 

That includes the discussion above, but can also extend to other components. For instance, getting my SET Tube amp built was quite a revelation (8W Tubes sounding better than 85W SS??!!), but I still have to find the proper sweet spot, as currently I am using the SET with a set of speakers which are not highly efficient and that's supposed to be where the magic is.

[徐中銳]

In short, some of you are misleading with your approach and conclusions to the topic.

 

DAC, digital-to-analogue-converter. Analogue is the goal ? Amplification of the-analogue-signal out ?

 

Like I've previously quoted of Low :

Recalling Low's comment :

The digital HiRes initiative is further confusing this issue. While high-resolution files can offer significant audio advantages' date=' the most important thing they offer is that when properly processed, there are less audio band artifacts, less misinformation to corrupt our music. Unfortunately, higher speed processors are noisier, so a 24/192 file sometimes sounds better when processed at 24/96.[/font']

 

Had reason to recall what WL said at

From a 2 hours interview at recent High End Munich 2016' date=' edited down to 53 minutes :

[/font']

 

It's best to start by contacting your manufacturer of choice and ask how they're implementing their chosen DAC chip. The sweet-spot of upconverting may well be less than its maximum processing availability—due to sensitivity of preserving analogue signal's integrity from noisey environmental interferences.

 

I need to start my new New Zealand day, will write more when I next can.

 

Quickly, of Ayre, Charles Hansen's view, remember The Ear - DSD under Fire Thread

 

And Low also said :

The difference between good and bad hi-fi is how long it takes before one gets tired, not whether you can count 75 or 76 people in the chorus. Successful emotional stimulation over time is the result of minimal misinformation, not maximum information -- it is the misinformation (distortion, the moving of energy to the wrong frequency and/or time) which interferes with why we listen to music, not a lack of information.

[semente]

Great discussion.

 

But I do want to bring this back to the central point about DAC knowledge. How do we know what the magic sample rate is for each DAC that minimizes the DAC's internal proceeding?

 

I doubt DAC vendors are willing to disclose this, are they? For example, my Ayre Codex DAC supports up to DSD128 for DSD, but 32/384 for PCM.

 

Even if I accepted your point that something like HQPlayer is going to sound better, what would I set the upsampling to?

 

 

Sent from my iPhone using Computer Audiophile

 

The Teac UD-H01 that I used to own upsampled everything to 192k with its ASRC, it says so in the website literature.

But when I moved from BitPerfect to HQPlayer and fed the DAC with upsampled redbook the improvement in sound quality was quite noticeable, which means that the ASRC was rubbish.

 

I have recently ordered a custom designed and built NOS filterless DAC which can be fed PCM at 384k, making the most of HQPlayer's PCM processing capabilities.

The alternative would have been a DSD-able NOS DAC.

 

Which commercial DACs besides the Teac 501 can have filtering and possibly the ASRC turned off?

 

R

[Jud]

Yeah as far as I know . I could be wrong but I think the Yggy upsamples all files to 24/192 .

 

I believe Yggy upsamples to 358/384KHz, but the max it will accept is 176.4/192KHz, so it does do at least one round of upsampling on any signal fed to it. Miska's software offers two "closed form" filters, which are the same general type as those used in Schiit's multibit DACs, so if you have one of these Schiit DACs you might wish to experiment with them.

[Jud]

The Teac UD-H01 that I used to own upsampled everything to 192k with its ASRC, it says so in the website literature.

But when I moved from BitPerfect to HQPlayer and fed the DAC with upsampled redbook the improvement in sound quality was quite noticeable, which means that the ASRC was rubbish.

 

I have recently ordered a custom designed and built NOS filterless DAC which can be fed PCM at 384k, making the most of HQPlayer's PCM processing capabilities.

The alternative would have been a DSD-able NOS DAC.

 

Which commercial DACs besides the Teac 501 can have filtering and possibly the ASRC turned off?

 

R

 

The easy alternative is to find a DAC whose highest input rates aren't upsampled internally. In other words, say you have a PCM DAC that would upsample lower rate inputs to 352.8/384KHz, but allows you to feed it 352.8/384KHz and doesn't do any further resampling. Or, more typical these days, you have a DAC that accepts DSD128 and won't do any further resampling to input at that rate. So you just feed these DACs the max rate and it's the same as using an NOS DAC.

[semente]

The easy alternative is to find a DAC whose highest input rates aren't upsampled internally. In other words, say you have a PCM DAC that would upsample lower rate inputs to 352.8/384KHz, but allows you to feed it 352.8/384KHz and doesn't do any further resampling. Or, more typical these days, you have a DAC that accepts DSD128 and won't do any further resampling to input at that rate. So you just feed these DACs the max rate and it's the same as using an NOS DAC.

 

So it's easier to find a DSD-able commercial DAC that doesn't upsample?

 

R

[YashN]

Which commercial DACs besides the Teac 501 can have filtering and possibly the ASRC turned off?

 

I know my iFi iDSD Nano has the digital filter turned off when fed DXD (analogue filter is still there).

 

So, in that particular case, one would compare with A+ or HQ Player whether a Redbook file plays better sent as is, or else up-sampled to a multiple of 44.1KHz, up to DXD and compare these for a potential sweet-spot for this DAC and Redbook with the player.

[Jud]

 

If we assume (which may not be a fair assumption) that all of the upsampling is bit perfect in that it applies no DSP or other changes to the signal except duplicating data points exactly, then the differences you hear will all be due to the choices of filters applied (chosen for you in the case of Schitt/Chord/Wyred and chosen by you in the case of HQPlayer/Audirvana).

 

It's never bit perfect, since part of the idea is to change the sample rate (i.e., add more bits). And this is true of DAC internal sample rate conversion and filtering as well. But now think of what Miska said regarding the fact that the data is not the signal. As you mention below, what is being done is that the (digital) data is being changed in service of getting the (analog) signal closer to the performance, whether that was live or as it came from the studio.

 

Most of these conversions are "lossy" in the mathematical sense that the original data points cannot be recovered by a mathematical operation from the converted data. A few (the Schiit filters and a couple of Miska's) are "closed form," which means the original data points can be recovered. So they are not "lossy" mathematically, but they are still not bit perfect. As to whether such filters sound better, I'll leave this to the listeners and developers.

 

Given the range of opinions expressed about filters/upsampling, I think it is fair to assume that the filters DO have an impact on the resulting sound. Now, if you chose the DAC(s) you have by comparatively listening to them against other DACs, part of what you were doing was showing a preference for that manufacturer's choice of filters. Thus, if you "love" that particular sound, it would not be surprising that you would prefer it over a set of filters that produce a different sound. For most folks who use their DACs with HQPlayer or Audirvana but chose no upsampling, I think that all they are really saying is "I like the filtering choices made by my DAC manufacturer."

 

The next level up though would be to ask, which of the above is closest to the original recording? I think Miska has clearly shown why upsampling data to a higher rate allows a set of filters to be applied (whether by him or by your DAC manufacturer) that produces fewer artifacts in the audible range (so has nothing to do with higher frequency hearing/response beyond 20kHz/steeper transients, etc. that often get referenced to say upsampling can't make a difference).

 

Exactly. There are still disputes over whether some aspects of filter differences are audible (for example, differences of thousandths of a second in transient response). I don't have the tools or knowledge to carry out measurements, but let me say two things from a subjective point of view:

 

- The first time I used HQPlayer, I was shocked at how different from each other the various filters sounded to me. I was not expecting anywhere near that level of difference.

 

- Something very similar happened when I first tried the "Plus" version of Audirvana. I was surprised at how much difference even small adjustments in some of the parameters could make in the audible results. Now since I didn't know what I was doing (and still don't in any real sense), it's quite likely some of the settings I experimented with were "pathological," i.e., stuff that people who know about filters would never do. So of course some of the changes I made were going to create very audible differences. Still, it was very interesting to me that I felt I could hear changes of a percent or two in some parameters.

[Jud]

So it's easier to find a DSD-able commercial DAC that doesn't upsample?

 

R

 

Yes, just because there are more delta-sigma than PCM DACs these days.

[Jud]

Great discussion.

 

But I do want to bring this back to the central point about DAC knowledge. How do we know what the magic sample rate is for each DAC that minimizes the DAC's internal proceeding?

 

I doubt DAC vendors are willing to disclose this, are they? For example, my Ayre Codex DAC supports up to DSD128 for DSD, but 32/384 for PCM.

 

Even if I accepted your point that something like HQPlayer is going to sound better, what would I set the upsampling to?

 

 

Sent from my iPhone using Computer Audiophile

 

It's not really that complicated. As I said before upthread somewhere, the choices are essentially two:

 

- For most DACs, set the upsampling rate to the highest input rate the DAC will accept. Easy peasy.

 

- For a few DACs that will accept DSD, you may want to send the highest PCM rate.

 

That's pretty much it. The only other decisions you would need to make, and probably the harder ones, are what filter and modulator to choose if using HQPlayer; what filter parameters and modulator to choose if using A+; or with XXHighEnd on Windows (since the filtering is pretty well set and it doesn't do DSD), your choices from among the myriad OS tweaks XXHE allows you to make.

[austinpop]

 

So, you have to try out the combinations, and to reduce that number, perhaps ask other members what they prefer and start from there.

 

 

Wasn't that how this particular discussion started? It would be great to have a thread where people would post their DACs with their findings and observations with upsampling.

 

I would create such a thread if I actually used HQPlayer!

 

For now, I'm just watching and learning. I'm a streamer/UPnP kind of guy, so for me adopting this approach requires a big operational change, to which I'm not really inclined.

[Miska]

Which commercial DACs besides the Teac 501 can have filtering and possibly the ASRC turned off?

 

You mean digital filtering? Analog output has analog filter as it should...

 

iFi, TEAC -503 models, T+A DAC8 DSD, hiFace DAC, etc... Metrum DAC's don't even have any digital filters (PCM NOS DACs).

 

Chord Mojo also when you run it at 16x input rates, I don't count the following linear interpolation (argh) as a filter.

 

Many also have adjustable analog filter.

[Miska]

From my listening there's a category missing - best with bit perfect input, DSD or PCM . I'd definitely put both the Hugo and my current Esoteric in that category -noting that the latter is a very close relation of the TEAC dacs you mention above.

 

Chord also performs better when DSD is converted to PCM before sending there, because they have less than great DSD to PCM converter inside.

 

How do you know how your "bitperfect DSD" you are sending to your DAC came to be? (or PCM for that matter)

 

Chord Mojo definitely goes into category "disable it's volume control and feed it 16x 32-bit PCM".

[sdolezalek]

It's never bit perfect, since part of the idea is to change the sample rate (i.e., add more bits). And this is true of DAC internal sample rate conversion and filtering as well. But now think of what Miska said regarding the fact that the data is not the signal. As you mention below, what is being done is that the (digital) data is being changed in service of getting the (analog) signal closer to the performance, whether that was live or as it came from the studio.

 

Most of these conversions are "lossy" in the mathematical sense that the original data points cannot be recovered by a mathematical operation from the converted data. A few (the Schiit filters and a couple of Miska's) are "closed form," which means the original data points can be recovered. So they are not "lossy" mathematically, but they are still not bit perfect. As to whether such filters sound better, I'll leave this to the listeners and developers.

 

Jud: as you can see my feeble lawyer/VC mind struggles with the precision of these engineering concepts What I was mentally trying to separate were those changes that should be non-destructive but perhaps also non-constructive and those that should clearly have an influence on sound (whether good or bad). So thank you for the clarifications.

 

The part I found interesting in playing around with the A+ parameters in iZotope were the necessary tradeoffs and learning much more about my own personal sensitivities to a) ringing, b) decay, c) transient attack, d) frequency flatness and others. Quite often things that sounded great on day one became tiresome after a few days or weeks. The longer I listened, the more less was more (I think you experienced some of the same in your progressive leaning toward less-invasive flatter filter slopes). It was that learning with iZotope that then allowed me much more quickly to choose which filters I liked in HQPlayer (even if I didn't know what fine-tuning parameter tradeoffs Jussi had made).

 

I also have to assume that in choosing some of these filters we are "making up" for limitations elsewhere in our system -- for example a bookshelf sized set of speakers might seem much larger with a bit more reverb and slower decay. With my giant Magnepans, exactly the opposite was true -- I tended toward leaner, more nuanced presentations.

 

All that being said, I do think there is value in us collectively trying to come up with a list of starting points for different DACs. Put differently, I'd like to make sure I'm going through filter choices that make the greatest possible impact because I have chosen a DSD512 or 24/384 starting point wherever that is possible and thus knowing where to start then pushes the learning over to the other filter choices rather than choice of upsampling.

[One and a half]

The DAC cannot know if for example 192 kHz data is result of upsampling or happens to be hires source originally created at 192 kHz. For the DAC it is just data sampled at 192k.

 

The DAC may not understand where the 192 came from, but 'we' know the file has undergone sample rate conversion. It may be all numbers in the mathematical domain, but we are not multiplying any easy equation, like 2 x 4 =8. The algorithms for SRC are surely as complex for upsampling process as they are for the downsampling process.

 

There maybe more processes out in the wild but I only know of Izotope that's built into Sound Forge and Audirvana +. The Izotope SRC can be change the steepness, max filter length, cutoff scaling, alias suppression and Pre-Ring. I don't have much of a clue how to set them (the point of another thread), let alone make conclusions how they work. On the occasion I do SRC to create a CD from a 96/24 file, there's nothing much lost, so by luck SRC works.

 

By inference, you're telling me that upsampling SRC is transparent? If you apply mathematics, no matter how 64bit precision you make it, there will be a difference. Others have posted that they prefer not to upsample and I would agree after listening to various sample rates for HQPlayer, including DSD256 for 18 months. I don't believe it is problem that the computer is working harder and is noisier as a result, it's the FLAC (uncompressing) vs WAV argument, no. Maybe Izotope is better. Maybe I drag out the TASCAM DA-3000 recorder and make some comparisons. Now, which filter to use?

[semente]

If I understand correctly, upsampling affects the audibility of filtering more than anything.

Current knowledge states that 2fs puts the filtering too close to the top end of the audible band and because filtering requires some slope there will be audible consequences.

 

My question is when we upsample, shouldn't the filtering become inaudible?

I thought that one would be filtering (the upsampled redbook) at 4fs or 8fs, etc.

 

R

[Miska]

If I understand correctly, upsampling affects the audibility of filtering more than anything.

Current knowledge states that 2fs puts the filtering too close to the top end of the audible band and because filtering requires some slope there will be audible consequences.

 

My question is when we upsample, shouldn't the filtering become inaudible?

I thought that one would be filtering (the upsampled redbook) at 4fs or 8fs, etc.

 

There are two things involved, digital filtering and analog filtering. Digital filtering exists to help analog filtering, which is always mandatory, to do it's job better. Reason is that it is feasible to create accurate steep filters in digital domain and avoid extra problems of filters operating in analog domain (noise, distortion, etc). This way the analog filter corner frequency can be moved away from the audio band and the slope (filter order) can be relaxed.

 

If you start with RedBook, there's no way you can avoid having the filtering (or lack of) implications right at the edge of the audio band. You just need to pick something that is as optimal as it can be. I've spent enormous amount of time to come up with filter design methods that are as optimal as possible in both time and frequency domains simultaneously. That is tough challenge to push closer to the limits, because ultimately the two are mathematically bound by the 1/x relationship and the Nyquist fs/2 limit puts hard boundary for the bandwidth. I take objective-subjective approach to my work, so I want things to measure well and once they measure well they also need to sound good. Otherwise I'm not happy and wouldn't have peace of mind. But ultimately you need to make your own choice based on your particular hearing sensitivities and the material you listen. That's why there are all those options with linear- and minimum-phase responses and such. Different types of content emphasize different kind of filter properties, so it is good idea to select a filter that fits both your hearing (different people are particularly sensitive to different sonic properties) and your material.

 

The (anti-alias) filter properties become embedded to the source material when the sampling rate is reduced to 44.1k. With apodizing digital filters it is possible to replace/modify these filter properties that have been embedded to the source material. With suitable apodizing filter you can get "de-blur" effect.

 

For example rock recorded in studio puts demands on the transient response, so a minimum phase filter that doesn't have pre-ringing is usually good fit, since there's usually not much real acoustics involved anyway. While classical music recorded in real acoustics puts demands to the sound field/space, so a linear phase filter is usually good fit, since there are no strong transients or at least very few.

 

As the source sampling rate goes up to higher resolutions, the effect of filters also gradually diminish. So for hires content you have less to think about in that respect. It is much easier to do good D/A conversion for hires than it is for RedBook! For delta-sigma DACs the effect of modulator however is persistent at the same level regardless of the source sampling rate. And that is the another 50% of the DSP-side performance involved.

[semente]

Thanks, that makes it clearer.

 

R

[semente]

There are two things involved, digital filtering and analog filtering. Digital filtering exists to help analog filtering, which is always mandatory, to do it's job better. Reason is that it is feasible to create accurate steep filters in digital domain and avoid extra problems of filters operating in analog domain (noise, distortion, etc). This way the analog filter corner frequency can be moved away from the audio band and the slope (filter order) can be relaxed.

 

If you start with RedBook, there's no way you can avoid having the filtering (or lack of) implications right at the edge of the audio band. You just need to pick something that is as optimal as it can be. I've spent enormous amount of time to come up with filter design methods that are as optimal as possible in both time and frequency domains simultaneously. That is tough challenge to push closer to the limits, because ultimately the two are mathematically bound by the 1/x relationship and the Nyquist fs/2 limit puts hard boundary for the bandwidth. I take objective-subjective approach to my work, so I want things to measure well and once they measure well they also need to sound good. Otherwise I'm not happy and wouldn't have peace of mind. But ultimately you need to make your own choice based on your particular hearing sensitivities and the material you listen. That's why there are all those options with linear- and minimum-phase responses and such. Different types of content emphasize different kind of filter properties, so it is good idea to select a filter that fits both your hearing (different people are particularly sensitive to different sonic properties) and your material.

 

The (anti-alias) filter properties become embedded to the source material when the sampling rate is reduced to 44.1k. With apodizing digital filters it is possible to replace/modify these filter properties that have been embedded to the source material. With suitable apodizing filter you can get "de-blur" effect.

 

For example rock recorded in studio puts demands on the transient response, so a minimum phase filter that doesn't have pre-ringing is usually good fit, since there's usually not much real acoustics involved anyway. While classical music recorded in real acoustics puts demands to the sound field/space, so a linear phase filter is usually good fit, since there are no strong transients or at least very few.

 

As the source sampling rate goes up to higher resolutions, the effect of filters also gradually diminish. So for hires content you have less to think about in that respect. It is much easier to do good D/A conversion for hires than it is for RedBook! For delta-sigma DACs the effect of modulator however is persistent at the same level regardless of the source sampling rate. And that is the another 50% of the DSP-side performance involved.

When you have time, could you elaborate a bit more on the effects of the modulator?

 

Cheers,

Ricardo