Hi,
Given the threads title, mentioning CD-Players, it is natural to consider the DAC being the DAC chip in the CDP. In fact, I cannot find any way to reconcile option 2 with the title, nor do see any such digital processing after the CD-Player but before a standalone DAC being discussed. So I guess I'm a little confused as to what we are discussing?
Ciao T
Given the threads title, mentioning CD-Players, it is natural to consider the DAC being the DAC chip in the CDP. In fact, I cannot find any way to reconcile option 2 with the title, nor do see any such digital processing after the CD-Player but before a standalone DAC being discussed. So I guess I'm a little confused as to what we are discussing?
Ciao T
Hi Wombat,
There is something missing in our understanding and modeling of human hearing. Each digital filter has its own sound rendering and the linear phase digital filter is one of the most destructive digital filter used in digital audio.
To understand all impact of digital audio filtering, come to the practice; any incomplete theories are not very useful.
I which to everyone a happy new year 2012
Following the theory everything you said should be true and any digital filter acting outside the audio band might NOT be audible.
There is something missing in our understanding and modeling of human hearing. Each digital filter has its own sound rendering and the linear phase digital filter is one of the most destructive digital filter used in digital audio.
To understand all impact of digital audio filtering, come to the practice; any incomplete theories are not very useful.
I which to everyone a happy new year 2012
As I see, my english is too bad.
Now I ask simple Question:
What are the sonic differences between different RF amp topologies ?
What are the sonic differences between different focus error amp topologies ?
What are the sonic differences between different tracking error amp topologies ?
What are the sonic differences between different EFM demodulator and PLL topologies ?
What are the sonic differences between different master clock frequencies (16.9344 MHz and 11.2696MHZ - 256fs/384fs)?
Are there sonic differences between various software for the tracking error control (firmware for operating MCU - mostly inside by the VFD control IC)?
What about of different ERCO variants?
All these parts are for me in front of the DAC and not in front of the DAC IC.
Thus DF and upsampler ic's should not be part of this discussion
Last edited:
DF and upsampler are not technically part of the Digital-Analog Converter, the precondition the digital signal that is delivered to the DAC in order to make the analog filtering easy to build.
Also, IMO, the EMF demod/PLL chip (and the available RAM buffer) has a very important role in dealing with low-frequency jitter, inherent due to mechanical imperfections in the motors that spin the disc, in focus and traking systems and imperfections of the disc itself. Lots of people say that HDD is solvng the problem - forgetting that 99% of that HDD music comes from an optical drive too with the same problems.
Personally I was floored what sonical difference is between a 16kB (±4 frames of jitter fluctuation) and a 32kB buffer in those chips - extracted signal played via SPDIF to the the same DAC.
Also, IMO, the EMF demod/PLL chip (and the available RAM buffer) has a very important role in dealing with low-frequency jitter, inherent due to mechanical imperfections in the motors that spin the disc, in focus and traking systems and imperfections of the disc itself. Lots of people say that HDD is solvng the problem - forgetting that 99% of that HDD music comes from an optical drive too with the same problems.
Personally I was floored what sonical difference is between a 16kB (±4 frames of jitter fluctuation) and a 32kB buffer in those chips - extracted signal played via SPDIF to the the same DAC.
Last edited:
Hi,
I did spend a fair bit of time on this topic in the last millennium.
If we can make sure a number of strictures (below) are obeyed, the answer is: "I do not find any differences if...
1) All systems equally recover the full original data äfter error correction and do not engage error concealment.
2) All systems re-time the output data with a clock that is not subject to crosstalk through the PSU or other means (bad ground layout, bad layout creating antenna's for noise, acoustic feedback etc). from the servo/digital processing subsystem
3) The subsystems following the re-timing are isolated from any crosstalk from the digital sectionin the various domains (EMI/RFI, ground pollution, other coupling of noise/signals)
Of course, the ability of different servo subsystems to compensate for CD Errors or just plain acoustic feedback does vary, with some being able to handle much larger losses of datatracks without invoking error concealment than others. These differences may be the most critical ones when it comes to audible differences in well designed CD-Transport solutions.
Sadly usually commercial solutions evidence severe problems with powersupply design (usually cost related, such running all subsections of a given transport from one supply), clock design and layout, so that problems are almost assured.
Thankfully CD is essentially history so much of this is not any longer relevant...
Yes, this I understand now. I think you should have been a wee bit more explicit about this.
Ciao T
I did spend a fair bit of time on this topic in the last millennium.
If we can make sure a number of strictures (below) are obeyed, the answer is: "I do not find any differences if...
1) All systems equally recover the full original data äfter error correction and do not engage error concealment.
2) All systems re-time the output data with a clock that is not subject to crosstalk through the PSU or other means (bad ground layout, bad layout creating antenna's for noise, acoustic feedback etc). from the servo/digital processing subsystem
3) The subsystems following the re-timing are isolated from any crosstalk from the digital sectionin the various domains (EMI/RFI, ground pollution, other coupling of noise/signals)
Of course, the ability of different servo subsystems to compensate for CD Errors or just plain acoustic feedback does vary, with some being able to handle much larger losses of datatracks without invoking error concealment than others. These differences may be the most critical ones when it comes to audible differences in well designed CD-Transport solutions.
Sadly usually commercial solutions evidence severe problems with powersupply design (usually cost related, such running all subsections of a given transport from one supply), clock design and layout, so that problems are almost assured.
Thankfully CD is essentially history so much of this is not any longer relevant...
Yes, this I understand now. I think you should have been a wee bit more explicit about this.
Ciao T
How is CD history now? Most of the releases are still on CD format of FLAC originating from a CD. Ripping a CD in a computer doesn't get past the transport reading error correction and jitter issues.
Most the stores that sell digital music have original sources CD format. And even the so-called hi-res files it was proved that sometimes are just upsampled from CD.
Personally I buying all the new SACD releases (like the last Pink Floyd) and avoiding the bluray because of crappy HDMI connection.
Last edited:
Hi,
As physical media, sure.
Alas, ripping correctly in a PC does get past these issues.
In most cases (there are exceptions of very bad condition CD's) the audio data is recovered completely and without error.
Jitter is a different story, there are solutions as well.
I generally avoid DS DAC's so I also avoid DS Music Formats. But if you like it, who am I to tell you otherwise...
Ciao T
As physical media, sure.
Alas, ripping correctly in a PC does get past these issues.
In most cases (there are exceptions of very bad condition CD's) the audio data is recovered completely and without error.
Jitter is a different story, there are solutions as well.
I generally avoid DS DAC's so I also avoid DS Music Formats. But if you like it, who am I to tell you otherwise...
Ciao T
I am sorry for you, since all the modern ADC (in recording studios) are delta-sigma based, you must miss a lot of music.
Hi,
I am pragmatic, I listen to the music I get. Still I find that well made older recordings that avoided DS ADC's are preferable, just as I find regardless of ADC Type I prefer DAC's that are classic PCM not DS.
That is a preference of mine and disagrees with your preference, I guess we are both entitled to ours.
Ciao T
I am pragmatic, I listen to the music I get. Still I find that well made older recordings that avoided DS ADC's are preferable, just as I find regardless of ADC Type I prefer DAC's that are classic PCM not DS.
That is a preference of mine and disagrees with your preference, I guess we are both entitled to ours.
Ciao T
really all without exception ?
What about the tracks from follow URL:
https://www.hdtracks.com/index.php?file=staticpage&pagename=aboutus
Do you know other online portals, from the residual 1% of that HDD music comes not from an optical drive ?
Last edited:
Of course, we can express our prefferences, it's a free froum 
I think that the older recordings are preferable too, but just because of the loudness war. Usually, it's almost imposible to know hat ADC they used during the digital transfer.
Exemple: I liked better the "Avalon" SACD (2ch) remaster than the older japanese HDCD (decent sounding too) or the "Best Of" SACD (that is bad, might have used the PCM from previous CD release). How much is that the mixing and how much the ADC? Hard to say.
I think that the older recordings are preferable too, but just because of the loudness war. Usually, it's almost imposible to know hat ADC they used during the digital transfer.
Exemple: I liked better the "Avalon" SACD (2ch) remaster than the older japanese HDCD (decent sounding too) or the "Best Of" SACD (that is bad, might have used the PCM from previous CD release). How much is that the mixing and how much the ADC? Hard to say.
Interesting discussion, but, as we want to squeeze out the best possible sound quality out of CD (or any digital source for that matter), shouldn't we look around and check what's there right now (this thread started over two years ago)?
Looking for a DAC I tried the various popular over/upsampling options as provided by my Harman Kardon CDP (Anagram technique - AD1955 DAC which can be used separately), a good quality BurrBrown based CDP, and a Twisted Pear Audio Sabre DAC.
NOS never did it for me, until I tried this one:
NOS Mini DAC Octave
For the first time I heard a real upgrade to my digital based sources with this NOS thing with it's non-audio based dac chips and "lack" of analog output stage.
All former candidates sounded more or less the same; switching to the Octave just improves attack and definition by a degree leaving the over/upsampling types further behind than I expected.
With all respect, but for me the technical discussion on pre-dac over/upsampling techniques has become pointless. My ears tell me what sounds most natural, and isn't that what we are looking for?
With pre-dac techniques I would not exclude the quality of SP/DIF data transfer; many CD players have a miserable digital output. Removing the totally unnecessary pulse transformer from my CD player transferred the SP/DIF signal to what it should look like.
Just wanted to let you know, to (maybe) put things in the right perspective.
Looking for a DAC I tried the various popular over/upsampling options as provided by my Harman Kardon CDP (Anagram technique - AD1955 DAC which can be used separately), a good quality BurrBrown based CDP, and a Twisted Pear Audio Sabre DAC.
NOS never did it for me, until I tried this one:
NOS Mini DAC Octave
For the first time I heard a real upgrade to my digital based sources with this NOS thing with it's non-audio based dac chips and "lack" of analog output stage.
All former candidates sounded more or less the same; switching to the Octave just improves attack and definition by a degree leaving the over/upsampling types further behind than I expected.
With all respect, but for me the technical discussion on pre-dac over/upsampling techniques has become pointless. My ears tell me what sounds most natural, and isn't that what we are looking for?
With pre-dac techniques I would not exclude the quality of SP/DIF data transfer; many CD players have a miserable digital output. Removing the totally unnecessary pulse transformer from my CD player transferred the SP/DIF signal to what it should look like.
Just wanted to let you know, to (maybe) put things in the right perspective.
Last edited:
I'm curious to ask - do you lay the improvement over previous incarnations of NOS at the feet of these two attributes of the Metrum? If so, any ideas for how they achieve it? I'm totally in agreement with your description of how NOS improves attack and definition, I experience this with my TDA1543 modded designs when compared to AD1955 and ESS9022 (both oversampled/SD)
I'm curious to ask - do you lay the improvement over previous incarnations of NOS at the feet of these two attributes of the Metrum? If so, any ideas for how they achieve it? I'm totally in agreement with your description of how NOS improves attack and definition, I experience this with my TDA1543 modded designs when compared to AD1955 and ESS9022 (both oversampled/SD)
To be honest, I don't know.
I auditioned TDA1541A and 1543 based NOS dacs longer ago, but never found them to sound significantly "better".
Maybe the lack of the output stage of the Octave is beneficial. The dac itself is designed by an engineer with a HF background so I guess he did his PCB design homework well.
I do not notice anything of the inherent -3 dB at 20kHz typical for NOS dacs.
Hi,
Why would you say that -3dB at 20KHz is typical for Non-Os DAC's? I would argue the presence of such forms a design flaw, more so one easily remedied...
Ciao T
Why would you say that -3dB at 20KHz is typical for Non-Os DAC's? I would argue the presence of such forms a design flaw, more so one easily remedied...
Ciao T
-3.2dB at 20kHz is typical for NOS - this is the sinc function's response. It can be corrected in analog or digital form - the original SAA7220 has a smaller correction (for 4X and a bessel post-filter) built in to the digital filter. I agree with you pieter that its barely noticeable subjectively - perhaps my ears are getting too old
Hi,
I am unsure what you define as "correction filter".
I am not sure what you have tested and remember, but as we have pretty good representation in the Benelux Area it should be to go and audition one of the products I designed, which include such a filter (optional). In fact, it may be fun to hear how you rate the Metrum DAC against them, but that is another story...
Ciao T
I am unsure what you define as "correction filter".
I am not sure what you have tested and remember, but as we have pretty good representation in the Benelux Area it should be to go and audition one of the products I designed, which include such a filter (optional). In fact, it may be fun to hear how you rate the Metrum DAC against them, but that is another story...
Ciao T
I mean the network to correct the -3 dB drop at 20 kHz in order to get a flat response.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.