Is the dynamic range of CD sufficient?
- Thread starter Al M.
- Start date
Dynamic range of Redbook is not only adequate,it is rarely properly exploited. I am referring DDD
Most recordings are compressed to avoid the need to constantly adjust volume.
Edit To me it would appear to be axiomaticic that detail is essential to the realism of any reproduction. Masking blemishes in the reproduction
process notwithstanding.
Most recordings are compressed to avoid the need to constantly adjust volume.
Edit To me it would appear to be axiomaticic that detail is essential to the realism of any reproduction. Masking blemishes in the reproduction
process notwithstanding.
Last edited:
With all due respect, I have not seen a single technical reason why the Trinity DAC sounds so good in the thread you refer. Lots of information, enthusiast subjective reports, but nothing that could tell me why it sounds as described - in other words, no correlation at all between the technical and the subjective. I concluded that the only possible proof is listening - but I appreciate the threads for enjoyment and keeping me informed of what is going around.
BTW, TAS has no measurements ... And most of TAS or Stereophile reviews have much more information than the referred thread. Usually the more interesting debates on WBF are anchored in formal reviews carried by the professional magazines.
By analogy I had the pleasure of seeing Victoria Silversyetd unPlayable.They had made a dress for her out Polaroid pictures of herself. As tantalizing as the photos were,they could hardly rival a centerfold printed from a twelve color negative.
Certainly inanimate cannot convey human emotions. They can however pass through the external stimuli.Organ,drums,bass guitars,the twang of Phoebe Snows voice.Bob Dylan's scratchy voice all provoke emotion. Certain reproduction equipment is just incapable of recreating that detail.
Certainly inanimate cannot convey human emotions. They can however pass through the external stimuli.Organ,drums,bass guitars,the twang of Phoebe Snows voice.Bob Dylan's scratchy voice all provoke emotion. Certain reproduction equipment is just incapable of recreating that detail.
What you say may alas hold for most pop/rock recordings, but not for classical. I listen to a lot of classical and classical avantgarde, and the engineers appear to revel in letting loose the full dynamics of the musical performance. Sure, there may be exceptions; rbbert for example was complaining about compressed dynamics in a recent release of Brahms symphonies under Chailly (compression audible both on CD and SACD). But I think that is a rare occurrence and it will stay that way.
For pop/rock, compression has been the rule of the day since ages, even before the exaggerated loudness wars of today. I found it quite strange to compare in that respect two versions of "Behind Blue Eyes" on a 1995 CD remaster of the 1971 classic "Who's Next". The alternative version, also interesting musically, was much less dynamically compressed than the widely-known version originally released.
I suppose the remaster simply reproduced both versions as is, and it shows that only for actual release the engineer(s) felt the need for much more compression. The alternative version sounds dynamically better precisely because originally it was never commercially released!
Sure, the precise correlation between the technical and subjective is hard to make, and professional reviewers have a hard time with that as well. But nonetheless it is interesting to read both about specific technical implementations of a format and about the subjective results.
Yes, I agree with that.
They usually have much more and better described subjective information of how it sounds. But if you put all the technical info together from all the pages of the Trinity DAC thread, including info that comes from the designer himself, it is quite impressive.
No, but then I don't think there are many performances with a dynamic range beyond 96 dB, see my post #24 on page 3.
And even if there were some compression of the recording of such a performance, you probably would hardly notice it as Esldude in post #23 suggested.
I agree with both points.
I guess I misunderstood you - I thought you were saying that classical recordings might have a greater dynamic range than accommodated by CD/redbook.
I'm sorry but this article from Stuart is very old, and digital audio evolved a lot during the last years, from software to ADC & DAC.
Please read Bob Stuart on: http://www.stereophile.com/interviews/906bob
"Steve Harris: So 16-bit is enough, then?
Stuart: No. It's not enough. On some recordings you can hear the limitations of that. More the limitations of the 44.1kHz than of the 16 bits. Sixteen bits give you a 90dB dynamic range. If the recording is properly made and the A/D converter's good, and at the mixing or mastering stage they didn't forget to dither it properly, within the window between the peak level of the music and the noise floor of the recording venue, then the hiss of correctly done 16-bit audio is around the room-noise level—sometimes it's above it, and that's a problem. But 44.1kHz, there's no doubt that that was a bit mean at the top end, and on certain instruments, violins and so on, you can hear that—just a little edge sometimes, which comes from that sample rate."
To me is very obvious by listening real high resolution (not a fake up sampled).
Roch
Al. I.
Impressive by the quantity, but unfortunately probably because of the need to protect the design, it omitted most of the crucial and innovative characteristics of the design. Most of the critical questions raised by some curious participants were left unanswered. One does not know exactly why it should sound different from any other DAC - but this is true for most existing high-end electronics. Wizardry is still a keystone in high-end, don't you agree?
Al. I.
Impressive by the quantity, but unfortunately probably because of the need to protect the design, it omitted most of the crucial and innovative characteristics of the design. Most of the critical questions raised by some curious participants were left unanswered. One does not know exactly why it should sound different from any other DAC - but this is true for most existing high-end electronics. Wizardry is still a keystone in high-end, don't you agree?
Yes, I agree, but the collective engineering experience in the field from all this "wizardry" (or rather, experimental trial and error) has given us the high-quality CD playback that people could not have dreamed of in the 1980s.
And I am not sure if you can read about the true 'trade secrets' in designs in Stereophile either.
I'm sorry as well, because your response leads me to believe that you may not have actually read the paper. If you had, then you might know that Stuart's research led him to conclude that 20 - 21 bit of resolution was probably necessary to deliver a perceived absolute transparency, that, along with a maximum 60KHz native sample rate. This research is partly based on Stuart's academic foundation in human psycho-acoustics.
Some random tidbits that may or may not be interesting...
16 bits ideally provides 98 dB SNR, and range from max output to quantization noise floor (spurious-free dynamic range, SFDR) of ~144 dB.
24 bits ideally provides ~146 dB SNR and ~216 dB SFDR.
Both SNR and SFDR are independent of sample rate for an ideal converter. Note that circuit noise and clock jitter sets the noise floor of many wideband (MHZ - GHz+) converters. And the amount of jitter that can corrupt a signal depends upon the signal frequency, not the sampling rate. The limitations of real-world converters (ADCs and DACs) are such that very few approach these limits over the audio bandwidth.
Rest in my experience/in my opinion:
Top recording mics have a dynamic range ~140 - 160 dB. Records and tape have a dynamic range around 60 - 70 dB'ish. CD's 80+ dB set by the recording chain more than the medium.
Most audio systems IME, including speakers, electronics, and room, have a dynamic range around maybe 80 dB.
What you need to hear every nuance in a recording in your room, I have no idea.
16 bits ideally provides 98 dB SNR, and range from max output to quantization noise floor (spurious-free dynamic range, SFDR) of ~144 dB.
24 bits ideally provides ~146 dB SNR and ~216 dB SFDR.
Both SNR and SFDR are independent of sample rate for an ideal converter. Note that circuit noise and clock jitter sets the noise floor of many wideband (MHZ - GHz+) converters. And the amount of jitter that can corrupt a signal depends upon the signal frequency, not the sampling rate. The limitations of real-world converters (ADCs and DACs) are such that very few approach these limits over the audio bandwidth.
Rest in my experience/in my opinion:
Top recording mics have a dynamic range ~140 - 160 dB. Records and tape have a dynamic range around 60 - 70 dB'ish. CD's 80+ dB set by the recording chain more than the medium.
Most audio systems IME, including speakers, electronics, and room, have a dynamic range around maybe 80 dB.
What you need to hear every nuance in a recording in your room, I have no idea.
Very interesting that you would mention the room contributing to dynamic range. Perhaps you just meant the noise floor, but one of the many things that I noticed after installing my ASC room treatment was that the perceived dynamic range went up significantly in terms of the explosiveness of dynamic transients at loud levels. I suppose this may have to do with the removal of time-delaying reflections that smear one signal into another (?).
When you say that CD's 80+ dB is set by the recording chain more than the medium, which factors in the recording chain would limit dynamic range?
Don,
Even the entry of Wikipedia on Dynamic range refers that "In 1981, researchers at Ampex determined that a dynamic range of 118 dB on a dithered digital audio stream was necessary for subjective noise-free playback of music in quiet listening environments.[23]" I have not read the paper, and do not know the context of the sentence, but I share your opinion - I also have no idea I can support with good evidence. Besides, IMHO this question can not be solved in a few lines just using a calculator and a sound-meter using the formula of the maximum achievable dynamic range of digital audio. It is much more complicated than people think.
20 bits and 60 kHz is probably adequate for playback, but like so many other things (like the dynamic range of a dithered 16 bit recording) depends on 1) perfect implementation of that resolution and 2) recording at higher resolution to avoid filtering artifacts.
Actually those guidelines were given partly considering those artifacts you speak about. Another expert in psycho-acoustics has almost the exact parameters in mind in the person of JJ Johnston. What gave them those opinions is the fact some very small number of people may hear to near 25 khz. You may get some artifacts from filters, and need a little space to take care of that and be sure of no possible artifacts. One has said 60 khz and another 65 khz for sampling. Both also were premising this on the best hearing humanly known, under absolutely optimum conditions to completely guarantee full transparency to anyone. In the real world in probably 99% and more of all scenarios and probably 99% of all listeners, 48 khz 20 bit would do just as well.
Back on topic, how is a medium's ability to reproduce the dynamic range of recordings "useful within the context of connecting emotionally with the music?" Really?
Tim[/QUOTE]
Can someone please translate.
GG
Tim[/QUOTE]
Can someone please translate.
GG
Can someone please translate.
GG[/QUOTE]
Well, the premise or assumption if you will, is live music allows more emotional connection than anything else. (That premise itself might need examining) So in reproducing it, though it won't equal being there, the closer you get, the closer to the emotional connection of live you can get. Anything that degrades fidelity would possibly reduce that emotional connection. One way to degrade playback is to shortchange dynamic range. It might not be the most important, it is even conceivable it is not important in reproduced music being emotional. But that is the idea.
A basic idea of fidelity to the recording. Full possible dynamic range in playback is higher fidelity than a lesser level of dynamic range.
Similar threads
 | Steve Williams Site Founder | Site Owner | Administrator |  | Ron Resnick Site Owner | Administrator |  | Julian (The Fixer) Website Build | Marketing Managersing |