I think there has been some fair-ish of criticism about the decision to make the audio CD sample rate 44.1kHz. Doing that does demand more complex filtering, and it may not be sufficiently effective for very sensitive listeners. For the general population though, it is probably good enough. For sensitive listeners, things get much better at 48kHz or higher, as filter performance requirements are eased. Have you tried different antialiasing filters for higher sample rates?
Complex for the ADC or complex for the DAC? If the former you have an argument, if the latter then oversampling blows the argument out the water.
Note there are many good reasons to record at 24/96, very few to playback above 16/44.1 🙂
Note there are many good reasons to record at 24/96, very few to playback above 16/44.1 🙂
Depends upon the Early Voltage of Motorola 2N4401s and 2N4403s manufactured in 1971-3, which I do not know. This determines the JC-2 Phono Amp 2nd stage gain. (First stage gain is 4X, output stage gain is 0.98X).
That might be so, but according to the guys over at Benchmark there is still a problem with playback. Digital oversampling and filtering to provide most of the processing for playback is subject to errors and distortion from intersample overs. Given the way music is often mastered today to achieve high perceived average loudness, intersample overs occur fairly frequently. Benchmark did not address that problem in the DAC-1, but they did in the DAC-2. In reduces the digital signal by 3.5 db before the D/A so that intersample overs effectively don't occur. They say, according to ABX listening tests, that the design change is audible.
Yeah, so put in 6dB of digital attenuation before DAC and stop listening to MP3s and jobs a goodun? Doesn't change the fact that 16/44.1 is good enough for domestic replay.
Most digital filters in DACs still allow for some aliasing. They had to have 20kHz in the passband for marketing and they don't enter the stopband by Fs/2. Not saying it's a big deal, but it's not ideal.
The really sad thing is that while some people are doing that, other equally earnest people are making up circuits as they go along which may have 10% distortion at peaks and frequency response which depend on volume control position and gain which depends on how loud the music was two seconds ago. Both groups sometimes claim to be doing hi-fi; both are kidding themselves and trying to kid others.billshurv said:
That is a different question. Why do we often hear things which are not there?bear said:
Whoever said that intelligence and wisdom were the same thing? They are not, although I suspect they are not wholly uncorrelated as both involve thinking. Life success can be uncorrelated with either! Memorisation of facts may be a small precursor to intelligence (so you have something to think with) but it certainly is not the whole of education. Wisdom, of course, also has some connection with moral judgement - which could quickly take us to places off-topic for this forum.Markw4 said:
Without good listening tests we don't know which equipment shortcomings matter and which do not; we don't know which measurements we should have taken. We know we can measure things we can't hear; there is less evidence that we can hear things we can't measure. As I keep saying, it was listening tests which established that 20-20kHz was the frequency range needed to reproduce music in a way which almost everyone would find realistic. A priori we could not know this, because we know that some musical instruments produce sound outside this range. Only listening tests tell us what range we need. Note that many people don't need this range, maybe a few need a wider range.
I don't think anyone has suggested that he is irrational. He may be mistaken, but that is different.
That may be true. It may not. We don't yet know.
I disagree. If someone managed to find the original research I suspect he would find that there was some attempt to investigate the tails. The 20-20kHz bandwidth was not what the average person needs; it exceeds what the average person needs but meets what almost everyone needs i.e. it already includes the less extreme outliers. Exactly how far down the tail it goes I don't know. I am not aware of any research which has changed the picture on frequency response; occasional claims of more HF needed usually turn out to be based on flawed experiments. I am not saying that someone might not one day do a good experiment and prove that 44.1kHz sampling is not good enough; I am just saying that this has not happened yet.
Distortion would be similar, but there we have the complicating factor of distortion order and the difference (in perception, but not mechanism) between IM and harmonics.
But could this not be a good reason for a few extra bits depth? Use a few dB digital atten before the DAC to avoid overs, and still have good S/N because you have bits to spare.
Jan
Sampling beyond 44.1 kHz is very stupid. Waste of bandwidth and resources and electrical power. Most 30+ agers don't hear 14 kHz
Last edited:
I disagree completely. The issue has to do with easing the requirements on the reconstruction / antialiasing filters. It's not about hearing ultrasonics. We don't need 192khz but 88.2 or 96 make life easier than 44.1 in terms of a quality filter implementation.
Is 44.1 audibly transparent? Probably, but we can afford to sample faster these days.
Last edited:
Not if you are doing a lot of processing. It can help then. Even when ripping vinyl. Nothing to do with your hearing on that side, that is playback!
I think you'll find a large percentage of the -160dB brigade are doing it for the challenge because they can. 🙂
I understand years ago the oscillator thing became an obsession for a while. I'm waiting for them to give up on tweeking the gain control/leveling and start with huge air-coils and capacitors for post filtering. 😉
Anyone who doesn't think that sampling above 44.1 is a good thing, either has not yet heard it, or can not hear the difference, OR is listening on a system with insufficient ability.
----------------
I may have reasonably high "listening accuity" but frankly it's rather a lot less than it was 25 years back. And the things you'd find me hearing here, are likely to be audible to a majority of random listeners here too. OTOH there are people who are unable to hear well enough. So, to them, as I said, there's no point in debating and discussing, espousing your beliefs, since for you what you think/hear is 100% correct.
Another way to think about this: if a Bosendorfer, a Bechstein, and a Steinway are all "pianos" to you, then we've got no way to discuss, nor can I convince you that there are real differences in their sound AND the way they play - OR that all models and years of each are not the same!!
In almost all instances that I have worked with opamps, they do NOT sound the same. I'd be more than thrilled to find/get/use two different opamps that DO sound the same in a given use/circuit/application.
So, who among ALL of the readers and participants wants to tell me that their system has nil siblence of the "objectionable" nature (per my earlier post's description)??
_-_-
PS. in my recent DAC box output opamp swaps the LM4562 and the AD797 sounded rather close and similar to each other. I'd be hard pressed to
tell them apart, unless I was able to identify something peculiar and specific in their "sound" that was a "tell". But I did try another one, that I deemed to be
mo' bettah to pipe the sound through... I was primarily listening for that nasty objectionable sibilence at the time so I ignored soundfield and things like mid bass and bass... fyi, and all that...
(anyone who wants to send me over a pair of opamp singles for a private report on how they compare, PM me! 😀 )
----------------
I may have reasonably high "listening accuity" but frankly it's rather a lot less than it was 25 years back. And the things you'd find me hearing here, are likely to be audible to a majority of random listeners here too. OTOH there are people who are unable to hear well enough. So, to them, as I said, there's no point in debating and discussing, espousing your beliefs, since for you what you think/hear is 100% correct.
Another way to think about this: if a Bosendorfer, a Bechstein, and a Steinway are all "pianos" to you, then we've got no way to discuss, nor can I convince you that there are real differences in their sound AND the way they play - OR that all models and years of each are not the same!!
In almost all instances that I have worked with opamps, they do NOT sound the same. I'd be more than thrilled to find/get/use two different opamps that DO sound the same in a given use/circuit/application.
So, who among ALL of the readers and participants wants to tell me that their system has nil siblence of the "objectionable" nature (per my earlier post's description)??
_-_-
PS. in my recent DAC box output opamp swaps the LM4562 and the AD797 sounded rather close and similar to each other. I'd be hard pressed to
tell them apart, unless I was able to identify something peculiar and specific in their "sound" that was a "tell". But I did try another one, that I deemed to be
mo' bettah to pipe the sound through... I was primarily listening for that nasty objectionable sibilence at the time so I ignored soundfield and things like mid bass and bass... fyi, and all that...
(anyone who wants to send me over a pair of opamp singles for a private report on how they compare, PM me! 😀 )
Last edited:
Yeah, I'd say my system has essentially nil 'false' sibilance.
Female vocals and cymbals come through with really nice liveness and clarity and zero ear bleeding/tinnitus/shut down.
There are tricks on how I got to this.
Dan.
Bear, the question of sibilance is interesting and I think any true audio system will display it from time to time since it can also be in the source material.
Actually, I cannot remember amplification having been the source of it. As far as my observations go, it is in the transducers (mike, speaker, cartridge) where it is generated. I would be very interested if anyone has experienced an amplification stage that introduced sibilance on its own device.
Actually, I cannot remember amplification having been the source of it. As far as my observations go, it is in the transducers (mike, speaker, cartridge) where it is generated. I would be very interested if anyone has experienced an amplification stage that introduced sibilance on its own device.
I agree 100% I once was testing a ribbon tweeter with a chirp (Dick Heyser style) at a certain level it just stuck at one frequency point in some just wrong sound.
Yes, it may be in the source material.
But what would you say IF some source material that you are/were certain was simply "no good" in that it had too much unnatural sibilence and was problematic in that regard on every system you had ever tried it on. And then, one day, it got played back on a system where somehow those issues were SUBSTANTIALLY RESOLVED, to the point where it was actually "good" - AND the rest of the things that usually sounded good or excellent, were still excellent or good without any change whatsoever to the amount of HF energy present in the recordings, just that the "misplaced" sibilence no longer seemed misplaced or standing out, it merely "FIT" naturally?
What would you say then?
But what would you say IF some source material that you are/were certain was simply "no good" in that it had too much unnatural sibilence and was problematic in that regard on every system you had ever tried it on. And then, one day, it got played back on a system where somehow those issues were SUBSTANTIALLY RESOLVED, to the point where it was actually "good" - AND the rest of the things that usually sounded good or excellent, were still excellent or good without any change whatsoever to the amount of HF energy present in the recordings, just that the "misplaced" sibilence no longer seemed misplaced or standing out, it merely "FIT" naturally?
What would you say then?
Goes both ways no difference at all between saying no two op-amps ever sound the same or that silver wire has direction.
Probably no system sounds the same with 20 people in the same room rather than 1 so it might depend on if you want friends or not.
Last edited:
Sorry, disagree.
Nobody has said a word about the red herring you bring up.
It is entirely irrelevant, to try to distract by talking about the number of people in the room or if you want or have friends?
You know it and I know it and everyone reading ought to know it too.
As far as I am concerned you can sit by yourself or with a friend or several and the thing is that the sound isn't going to change or be what I want it to be if it is not if it isn't what it is. I'm not an "emperor has no clothes" type.
Note: no one has commented so far on the issue of testing these super low distortion amps on a "real load" so far??
Of course this has little to do with opamps, other than that power amps are basically big opamps... would everyone's head explode if it was found that the silly no global feedback tube amp produced less icky distortion artifacts than the high feedback/super low distortion amp into a real load?? Is this even possible?? How about if looked at as a percentage OF distortion?
- Status
- Not open for further replies.