It absolutely is valid for me to mention a belief set which tends to accompany a related technical argument. I have little doubt that a search would locate examples of you having done that. However, I neither assumed nor declared anything about what you, or anyone else in particular, felt about "perfect sound forever". If you feel a member of the group which fall in to that belief catagory, well, then that's self selection on your part. It has nothing to do with my debating technique, as you call it. So, mistake number 1. above belongs to you.
As far as supposed mistake number 2. above. You tacitly ackowedge the requirement for continuous periodicity in the signal with that rather odd example of a forever repeating CD track. However, and what's odder still, is that the periodicity requirement of the sampling theorem applies to signal capture, not playback reconstruction. So, exactly how does the example of a continuously repeating CD track demonstrate that a signal which was not periodic at the time it was sampled can still be treated as if were actually periodic? It that were the case, the sampling theorem requirement of periodicity in the sampled signal would have no implication. You could simply declare any signal to have been periodic after the sampled fact.
Not once have you addressed my observation that that many DSP applications contain time-domain sensitive signals which preclude the use of brickwall filtering. Digital oscilloscopes, EKG, radar, sonar, etc. Better to ignore those, it would seem. Respond if you like, but it's becoming apparent that I should expect little beyond insults and condescension in return.
Last edited:
You appear to believe that recording a piece of music once is somehow fundamentally different from recording it when repeated; so different that mathematical theory which definitely applies to the latter has nothing useful to say about the former. This is a common mistake. The differences turn out to be negligibly small.
As I said, and I think you agree, it is the necessary filtering which can create difficulties. A brickwall filter has phase implications at frequencies below the stopband and it is these which can distort waveforms. The extent to which these phase changes are audible is a separate issue. As our ears are generally much more sensitive to the presence and amplitude of frequency components rather than their phase then the correct option for audio is a brickwall filter. For a digital scope the requirement is quite different, as our eyes are quite sensitive to the shape of a waveform so a much gentler filter can be used (combined usually with faster sampling) - even a little aliasing might be tolerated. Horses for courses.
As I said, and I think you agree, it is the necessary filtering which can create difficulties. A brickwall filter has phase implications at frequencies below the stopband and it is these which can distort waveforms. The extent to which these phase changes are audible is a separate issue. As our ears are generally much more sensitive to the presence and amplitude of frequency components rather than their phase then the correct option for audio is a brickwall filter. For a digital scope the requirement is quite different, as our eyes are quite sensitive to the shape of a waveform so a much gentler filter can be used (combined usually with faster sampling) - even a little aliasing might be tolerated. Horses for courses.
The subjective qualms I have with SACD tie in closely with the objective qualms I have. This could easily be nocebo effect, or could be down to implementation but I note a lack of dynamics when compared to NOS multibit which I lay at the door of noise modulation. Its a similar loss of dynamics that I hear with S-D in general. So the price of fixing up the phase issues is a too high one to pay for me personally.
Just to confirm, is your definition of "dynamic" per the Holt "standard": subjectively having a large volume ratio between loud and soft sounds, punchy, apparent high contrast?
Frank
Yes, I've noted in much CD sound a dullness, or grayness, which for me says that the treble is not working correctly. That part of the spectrum in CD sound is extremely vulnerable to interference, I don't see it as a manifestation of noise modulation or something intrinsic to the DAC's way of functioning. And that's because I can alter that character of the sound very easily by varying the level of effectively injected noise. Simple example: CD is working nicely, bring a switched off mobile phone over next to the player, switch the phone on: instantly the treble quality will fall off the cliff, go all flat and scratchy ...
Frank
Frank
A few minutes spent doing dynamic range analysis (Keith Howard's little software utility is ideal) may convince you that your perception may have more to do with the choices made in production and mastering at the source.
I would believe that, except that I can "recover" dynamics from a seemingly poor CD by appropriate tweaking of the system overall. That's not talking about the deliberate compression of dynamic range to OTT levels in modern pop, that's an entirely different thing! Having said that, even the worst of the worst current excesses can be made listenable to, by pushing the level of refinement a further notch again; I've surprised myself several times recently, thinking that a relatively new release was impossible to rescue, but then finding yet another little bit of leverage to lift the level of SQ, and lo and behold, I could enjoy the recording. Perhaps not as an artistic production, 
, but soundwise it came together ...
Frank
, but soundwise it came together ...Frank
That deliberate compression is not confined to pop, sadly. Once lost, dynamics cannot be recovered, first principles. This is one of the biggest impediments to decent sound in the home.
That's an interesting one ... about 9 months ago I made a reasonably serious attempt to investigate and understand what the current mashing of sound was doing, within the bowels of the beast, so to speak, the raw waveform. And it's a bit of a curate's egg: if the type of compression is relatively straightforward, that done at the last stage of mastering, after the tracks have been assembled then a bit of judicious fiddling with the right tools can quite adroitly reverse the squeezing. Essentially you work out the compression curve that's been applied, where the knee is and suchlike, and the end result can be a remarkably improved version.
However, and a big however, if the track is a complex mix of tracks, all of which have had their own distinct compression curves applied prior to mix down, then this is the nightmare scenario. First you have to separate the instrument and musical element strands, this is DSP at the highest level: one or two companies are attempting such, and presumably being paid much moolah for their efforts ... not something for the DIYer to tackle first up!
Frank
However, and a big however, if the track is a complex mix of tracks, all of which have had their own distinct compression curves applied prior to mix down, then this is the nightmare scenario. First you have to separate the instrument and musical element strands, this is DSP at the highest level: one or two companies are attempting such, and presumably being paid much moolah for their efforts ... not something for the DIYer to tackle first up!
Frank
Let me see if I can follow your thinking here. You're saying the DAC can't be responsible for restricting the perceived dynamics because you can further restrict the perceived dynamics by moving your (alive and kicking) phone adjacent to the CD player?
Yes, it's a bit like saying expensive turntables are not much good for playing records, because the cartridges tend to mistrack if you bang your fist repetitively on the table where the TT is; otherwise ...
Or to be more reasonable about it: I don't have a problem with dynamics on my setups: multibit, or S-D. In fact, if anything the current S-D beast has more dynamics than the Yamaha multi-bit, but that's really a factor of the Yamaha being intrinsically quite "smooth", and that I've learnt more about overall system tweaking along the way. Where I have trouble is in maintaining the "sweetness" or clarity of the treble, which is just another way of saying that distortion can easily intrude. The mobile, like a myriad of other factors, can easily throw the DAC off its food
, in my system because of the low cost nature of the circuitry this manifests as a dirtiness or somewhat glaring quality, pretty unpleasant; on more "refined" systems this may sound like dull, boring sound, or the seeming disappearance of all sound elements below a certain volume level, I find this behaviour quite bizarre!Frank
Last edited:
Not news, just something that's not worried about enough by the designers. For the last 30 odd years audiophiles have been wringing their hands in anguish about digital not being good enough, not coming up to the mark. Well, in my experience interference, especially that related to RF, is a key factor that drags down the subjective quality of CD sound, is what makes people complain that it doesn't have an "analogue quality"; not the number of bits or sampling rate or levels of jitter ...
Frank
Frank
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.