Time for an update regarding my trials with HQPlayer, this is promted in part by @LBameule's recent post on another thread, but I was planning to write something anyway. I find that it is useful write up your observations periodically as a kind of archive record of your thoughts, I quite often refer back to old posts I have made, it is surprisingly useful sometimes.
Anyway, if you take the long road from being HQPlayer 'Newbie' to HQPlayer 'expert', I would say that I am still firmly in the first half of that journey, but learning all the time. At first I was doing many comparisons between upsampling to DSD vs PCM. To be clear, this was using the microRendu and Mutec MC3+USB, so the Mutec ends up doing a final DSD to PCM conversion, before reclocking the PCM and sending to the Devialet. All source material was 16/44.1. There was a lot I liked when trying DSD, maybe sounding more natural, in some instances slightly magical, improvements to sound staging, lots of good stuff. However, I was fairly consistently finding DSD upsampling to be a little less dynamic and also I was finding a clearly discernible loss of bass definition and accuracy. After much experimentation I made the decision that DSD upsampling was not for me, and that I should spend my time concentrating on finding the best PCM filters. Basically, I could not live with the loss of bass definition, that said, a different system, different tastes in music, there is some real magic in DSD upsampling, so it may well work for others.
So I then started running through all the PCM filter options. Eventually I found that I likes the various 'poly-sinc' filters, poly-sinc-mp, poly-sinc-shrt, and poly sinc-shrt-mp. These ended up on a short list with a couple of other options. When running through my short list I did have some 'wow' moments, those kind of moments when you really start hearing those details, and a new insight into familiar music emerges, these came with poly-sinc-ext (which is a linear phase filter). After some careful listening, I think the 'audiophile magic' that came from poly-sinc-ext is basically a manifestation of slightly better upper mid to treble definition. This is about as far as I had got by last weekend, I had not yet has time to experiment properly with the various 'dither' options. It is interesting to note that after I had been running through the various filters and posting a few comments and questions on 'Computer Audiophile', Dr Tone (Late of this parish) posts something along the lines of 'I found poly-sinc-ext NS9 worked best with the Devialet'. (Dr Tone has now moved to the pleasant land of McIntosh and blue VU meters) I have a lot of respect for Dr Tone, but is fair to say that there have been a few things we have disagreed upon over the years, however, I do find it interesting that we both pretty much indipendantly found poly-sinc-ext to work well with a Devialet. For the Record, he has moved to different settings with his new McIntosh rig. Anyway, I knew that last weekend there would be much time to listen to music but no opportunity to play with HQPLayer filters, so HQPlayer was set to poly-sinc-ext NS9 and much casual listening to whatever anyone wanted to listen to ensued. All I can say is that with all kinds of music everything sounded great. The 1000 Pro has sounded excellent from day 1, but right on the margins there have been thoughts that something is not quite right, that tiny bit of upper mid digital fuzz that gets you thinking that the Pro is so revealing that it needs a better source, maybe I do need an Aurender N10 or dCS upsampler or something. The effect of all these HQPlayer filters are on the margins, small tweaks, often just discernible or maybe just discernible, but last weekend I found myself simply enjoying the music, and indeed enjoying the music beautifully reproduced without worrying about the system. Lets put it this way, in absolute terms, I do not think HQplayer has done that much to change the sound quality, it has made a change, but it is small. However, that change has eliminated some of the tiny little annoying niggles that get to you sometimes. The niggles too are small, but they can have a big effect on whether or not you sit down and get into the music or if you sit down and get worry about if the system is quite right. The effect of these tiny little changes, if you get it right, can actually be huge.
So that is where I am now, I still have not had any time to properly experiment with the 'Dither' settings, and perhaps revisit the 'minimum phase' poly-sinc filters versus the 'linear phase' options. Although I can take my time with this now as I am pretty happy with poly-sinc-ext NS9. Also, I see DSD as unfinished business. Aural memory is notoriously hopless, but notwithstanding this, I cannot help but think back to the demo I had of the dCS Upsampler and clock. During one demo of this with a D800, the dCS rep was on hand. He said it would sound best feeding the Devialet DSD64, I argued that the Devialet ADH is PCM, so this is converting to DSD simply for the Devialet to convert back to PCM, pointless and inefficient. So, we ran though a playlist upsmpled to DSD 64, then upsampled to PMC. From what I recall of the demo, Upsampled to DSD sounded best.
So, either my memory of the dCS demo is poor, or the dSC DSD upsampling does not suffer the loss of dynamics and bass definition that I find with HQPLayer. I would need another demo of the dSC to sort this one out for sure. The only problem here is that if I do request another demo of the thing, I will start to become obliged to buy it! Oh well.... One thing I may try when I get some time and sufficient motivation is to try HQPlayer DSD upsampling via AIR, so using the Devialet DSD MAT Engine to do the DSD to PCM conversion, rather than the Mutec, which is more equivalent to the dCS demo. That can wait though, I am more than happy with the more purist PCM only approach, poly-sinc-ext for now, and the next trials will be to establish what these 'Dither' settings might do.
Anyway, if you take the long road from being HQPlayer 'Newbie' to HQPlayer 'expert', I would say that I am still firmly in the first half of that journey, but learning all the time. At first I was doing many comparisons between upsampling to DSD vs PCM. To be clear, this was using the microRendu and Mutec MC3+USB, so the Mutec ends up doing a final DSD to PCM conversion, before reclocking the PCM and sending to the Devialet. All source material was 16/44.1. There was a lot I liked when trying DSD, maybe sounding more natural, in some instances slightly magical, improvements to sound staging, lots of good stuff. However, I was fairly consistently finding DSD upsampling to be a little less dynamic and also I was finding a clearly discernible loss of bass definition and accuracy. After much experimentation I made the decision that DSD upsampling was not for me, and that I should spend my time concentrating on finding the best PCM filters. Basically, I could not live with the loss of bass definition, that said, a different system, different tastes in music, there is some real magic in DSD upsampling, so it may well work for others.
So I then started running through all the PCM filter options. Eventually I found that I likes the various 'poly-sinc' filters, poly-sinc-mp, poly-sinc-shrt, and poly sinc-shrt-mp. These ended up on a short list with a couple of other options. When running through my short list I did have some 'wow' moments, those kind of moments when you really start hearing those details, and a new insight into familiar music emerges, these came with poly-sinc-ext (which is a linear phase filter). After some careful listening, I think the 'audiophile magic' that came from poly-sinc-ext is basically a manifestation of slightly better upper mid to treble definition. This is about as far as I had got by last weekend, I had not yet has time to experiment properly with the various 'dither' options. It is interesting to note that after I had been running through the various filters and posting a few comments and questions on 'Computer Audiophile', Dr Tone (Late of this parish) posts something along the lines of 'I found poly-sinc-ext NS9 worked best with the Devialet'. (Dr Tone has now moved to the pleasant land of McIntosh and blue VU meters) I have a lot of respect for Dr Tone, but is fair to say that there have been a few things we have disagreed upon over the years, however, I do find it interesting that we both pretty much indipendantly found poly-sinc-ext to work well with a Devialet. For the Record, he has moved to different settings with his new McIntosh rig. Anyway, I knew that last weekend there would be much time to listen to music but no opportunity to play with HQPLayer filters, so HQPlayer was set to poly-sinc-ext NS9 and much casual listening to whatever anyone wanted to listen to ensued. All I can say is that with all kinds of music everything sounded great. The 1000 Pro has sounded excellent from day 1, but right on the margins there have been thoughts that something is not quite right, that tiny bit of upper mid digital fuzz that gets you thinking that the Pro is so revealing that it needs a better source, maybe I do need an Aurender N10 or dCS upsampler or something. The effect of all these HQPlayer filters are on the margins, small tweaks, often just discernible or maybe just discernible, but last weekend I found myself simply enjoying the music, and indeed enjoying the music beautifully reproduced without worrying about the system. Lets put it this way, in absolute terms, I do not think HQplayer has done that much to change the sound quality, it has made a change, but it is small. However, that change has eliminated some of the tiny little annoying niggles that get to you sometimes. The niggles too are small, but they can have a big effect on whether or not you sit down and get into the music or if you sit down and get worry about if the system is quite right. The effect of these tiny little changes, if you get it right, can actually be huge.
So that is where I am now, I still have not had any time to properly experiment with the 'Dither' settings, and perhaps revisit the 'minimum phase' poly-sinc filters versus the 'linear phase' options. Although I can take my time with this now as I am pretty happy with poly-sinc-ext NS9. Also, I see DSD as unfinished business. Aural memory is notoriously hopless, but notwithstanding this, I cannot help but think back to the demo I had of the dCS Upsampler and clock. During one demo of this with a D800, the dCS rep was on hand. He said it would sound best feeding the Devialet DSD64, I argued that the Devialet ADH is PCM, so this is converting to DSD simply for the Devialet to convert back to PCM, pointless and inefficient. So, we ran though a playlist upsmpled to DSD 64, then upsampled to PMC. From what I recall of the demo, Upsampled to DSD sounded best.
So, either my memory of the dCS demo is poor, or the dSC DSD upsampling does not suffer the loss of dynamics and bass definition that I find with HQPLayer. I would need another demo of the dSC to sort this one out for sure. The only problem here is that if I do request another demo of the thing, I will start to become obliged to buy it! Oh well.... One thing I may try when I get some time and sufficient motivation is to try HQPlayer DSD upsampling via AIR, so using the Devialet DSD MAT Engine to do the DSD to PCM conversion, rather than the Mutec, which is more equivalent to the dCS demo. That can wait though, I am more than happy with the more purist PCM only approach, poly-sinc-ext for now, and the next trials will be to establish what these 'Dither' settings might do.
1000 Pro - KEF Blade - iFi Zen Stream - Mutec REF10 - MC3+USB - Pro-Ject Signature 12
