I think Goddard's point is that it is impossible for higher sampling rates to have a negative impact on the audible range.
 
In post #253 I show 2 examples of how higher sampling rates (and hence the presence of more ultrasonic energy) can impact the audible range. Since he knows everything, I'd like him to tell me how those two examples are flawed. His main point and those two points can't be correct at the same time.

I can say this thread has turned into an awesome example of logical fallacies in action. I have seen "appeal to authority" , "ad hominem", "strawman" and "red herring" galore!

John T
 
Let's suppose there is some interaction in sound waves, as described. So we play a note on a trumpet, and there's some stuff going on at, say, 40khz that's completely inaudible, but interacts with the other movements in the air and has some influence on what's going on at, say 5khz, which is audible.
 
As an aside, I can't imagine that this isn't the case. You're pushing air around, and it's a complex system.
 
However, if we're capturing all the audible range, then we are by definition capturing every effect that occurs in the audible range, even if we aren't capturing the cause.

John, what you are referring to is intermodulation distortion. However for intermodulation distortion to occur, you need a non-linear transfer medium, whether air, a microphone, a signal path, a plugin or whatever. At the air pressure levels that audio waveforms occur, it is essentially linear, so there will be virtually no intermodulation distortion.
 
The problems start when you record the ultrasonic frequencies and then let them loose in your DAW, where we love to add non-linear transfer functions to the signal. I posted two such examples: a compressor and a tube emulator. These perform their "magic" by behaving differently depending on the level of the signal. If you have ultrasonic energy present, these signal paths will respond to it, which will then also alter the audible frequencies (unless they themselves first filter out the HF in their algorithms).
 
So unless anyone can come up with actual proof for why having ultrasonic energy present is beneficial, it is best to avoid it altogether.

Sure, I understand. All I'm saying is that if we've captured the audible part, we've captured the way it sounds. If there's something inaudible affecting that, we've captured the effect it has. So there's simply no need to capture inaudible stuff, even before you get into whether it does harm. You can't, by definition, lose anything about the way the audible range sounds, as long as you capture the entire audible range. Though it's fairly clear to me you get this, and I think we're more or less in agreement.
 
I don't object to long threads or anything, but it's a shame there's so much verbiage in the thread given how essentially simple the issue is. Once you've got a sampling frequency above double the audible range, with an effective filter, everything is hunky dory. There are a number of ways of approaching the filter problem, one of which is indeed working at higher sample rates. That's what the higher rate is for, in an oversampling convertor; giving room to make the filter. It's not at all for capturing ultrasonics. The converter is getting rid of ultrasonics. That's part of its job.
 
 

There can be situations where there may be energy present that does not seem to be very obvious hearing it but if you filter or don't capture that energy there can be audible differences. I feel this is a very audible situation. I am from that school too. You can have a track that sounds Ok but insert a LPF filter on it and roll off everything (slowly) above say 15KHz and that track suddenly will sound better and clearer. It does to me and this happens more often than not. Same applies to the low end and interesting we are not talking about low end much here. An instrument may never go below (with fundamental that is) 80 Hz. Yet put a HPF filter on that and set it so nothing at 80Hz is touched but below that the signal is removed and that too can sound better. That is really audible. Maybe these concepts support the theory that we simply don't need to capture stuff we cannot hear so much at all.
 
Audible situations are much more interesting to talk about. Things that are very minimal are a waste of time to a certain extent. The reality is if you listen to a 24 Bit 382Khz mix and a 16 Bit 44.1 K mix, if the music is great, the engineering great then the mix will survive and still smell as sweet as a rose right at the end. Emotional impact equal for both mediums. That is what is important. My point about very high quality sources being bottle necked down to 16 bit 44.1K and no one hearing the difference really highlights the uselessness of this stuff. All fun but not that important at the end of the day.
 
Learning how to EQ, compress and limit a mix perhaps is way more important and in that situation a mix can be destroyed. Things that do effect 100% of the sound to me are much more interesting than stuff that might only have an effect of 0.1% or less.
 
I was thinking about analog as well. What is different about analog is the way the frequency response dies off at the high end. It is slow and gradual up there. My Revox B77 reel to reel machine specs at 25KHz or so as the top limit of its freq response (-3dB) but it still produces output at 35KHz! The graph is slow and easy from 25KHz up to 35KHz. It is a beautiful thing. The old Ortofon moving coil pickup cartridges are flat to 50KHz from memory. Now that is top end!
 
Final mixes I think need all the frequency range they can get. Individual tracks not so much so perhaps. I believe there is difference between tracks, buses and final mixes frequency response demand wise that is. It is the complex interaction of tracks that may only be limited out to 15 KHz or far less that might result too in final harmonics that extend up higher. High frequencies interact. If we want that nice analog top end than all we need to do is to go to 24 Bit 96KHz and that is easy to do now. We don't really need anything above that. Some experts say we only need to go to 60KHz digitally obviously to get the response out to 30KHz perhaps. Our nearest rate to that is 88.2Khz which does at least give us a nice 40Khz response.
 
I have just mastered a very high quality Jazz CD. The mixes came to me at 24 Bit 88.2 Khz and yes they sound great. But even after mastering and dithering down to 16 Bit 44.1Khz the difference between the two sounding formats is so miniscule. It is really inaudible in fact. I did a little test and just changed down the original mixes from their original format to 16 Bit 44.1KHz and did no mastering, just compared the two and believe me I would challenge anyone to hear anything between them! The issue is the music is just so good and the moment I hear either of those formats I am off listening to the playing so once that happens I think there is no hope of hearing anything much else!

Jeff, I agree with everything you said, although I would point out that for old analog equipment, if you want to be flat up to 20Khz and have minimal phase shifts that high, they would require a frequency response that goes much higher than 20Khz because none of them had any sort of filter with a sharp enough cutoff (whether intentionally added to the signal path or just a result of the analog circuit itself).
 
For analog tape, you really don't want to be able to record too high frequencies because the next thing you'll find the record bias signal in your audio path, which itself is at a pretty high level.

John T
Everyone's entirely aware that this stuff has been being discussed for years.

 
And yet rather than reading up on past discussions (some of which I even provided links to) and getting up to speed on their own, people prefer to remain oblivious to all that's been raised and debated at length already elsewhere, and then, having too little knowledge of the topic to understand what I've posted or its relevance, feel entitled and justified in complaining about me being obscure or circular or elliptical or not rigorous enough and accuse me of posting misinformation or of being flat out wrong or of being a douche or of being confused, when they still can't even understand how 32-bit floating-point data represents 24-bit integer PCM data after I've not only explained it but also pointed them to a color-coded diagram showing exactly which bits do what, and then still had to provide further even more basic explanation, and even another poster had to post a color-coded diagram?
 
I'm sorry, but if you can't even make an effort to educate yourself to get up to speed, don't act like babies expecting to be spoon-fed info and complain that it's obscure or irrelevant or whatever if it goes over your heads. Do some homework on your own, consult more reliable and knowledgeable sources of info than forum posts or blogs of dubious expertise or validity and maybe you might just acquire some quality knowledge on your own which you can share, which you may then find a far greater reward than any "platinum" post count and which can serve you and your production prowess well into the future.
 
Here's a simple way anyone might check their level of knowledge, a 1998 article by Martin Walker of SOS on the options available for "Optimising PC Digital Audio Quality In Software". If you don't already know and understand the stuff Martin talks about in that article, which remains in large part just as valid and vital today, then you definitely need to do your homework, because this thread has been cluttered with far too many very basic questions about stuff which anyone who's been using DAWs and on this forum for years really should already know, especially if they are working professionally in this field.
 
I can see now that it was a mistake to even join this thread, and that I may have erred by overestimating the general comprehension level of many of the participants, but I was growing concerned due to that Noel's original post referred to that blog as a "great article" and as "well worth the read" and a number of people including Garrigus had then concurred that it was a good article/info, when it was quite obvious to me that it was written by a less-than-knowledgeable poser and riddled with misinfo and scaremongering and that there was a serious risk of folks less knowledgeable being taken in by it.
 
But now, having seen once again the snideness and drive-by insults and bandwagon jumping and outright schoolyard bullying attempts that have become far too common on the CW forums, let the gullible and uninformed find their own way along the path to knowledge without me from here on. I've already given enough precious time to this thread, and although I do appreciate that some folks have expressed their thanks along the way, I only participate in the forum nowadays as my way to give back for what others were kind enough to share with me in the past, and I sense that my karma cup has been filled as concerns this thread.
 

Cry me a river.  i asked you a simple question and you go playing victim.  I can't say I'm particularly close to anyone on here enough to team up with them.  At most, maybe there are two in this thread clearly on opposite ends of the argument than you.  Some of us have independently been turned off by your posts without necessarily being on one side or the other but you are too hellbent on being "right" to pay attention to that.  Every time you are asked a question you respond with a different variation of "i'm right, you are wrong, here, read this article" .  I say good riddance.  There are plenty here willing to spew out the exact same info without being so combative about it. The fact that you were not nearly as belligerent to Noel, even when you disagreed shows that you are at least somewhat  aware of how combative you are being.  if you said the same things to Noel that you said to John and John, you would be making an involuntary exit.

Who farted?

John T
Sure, I understand. All I'm saying is that if we've captured the audible part, we've captured the way it sounds. If there's something inaudible affecting that, we've captured the effect it has. So there's simply no need to capture inaudible stuff, even before you get into whether it does harm. You can't, by definition, lose anything about the way the audible range sounds, as long as you capture the entire audible range. Though it's fairly clear to me you get this, and I think we're more or less in agreement.
 
I don't object to long threads or anything, but it's a shame there's so much verbiage in the thread given how essentially simple the issue is. Once you've got a sampling frequency above double the audible range, with an effective filter, everything is hunky dory. There are a number of ways of approaching the filter problem, one of which is indeed working at higher sample rates. That's what the higher rate is for, in an oversampling convertor; giving room to make the filter. It's not at all for capturing ultrasonics. The converter is getting rid of ultrasonics. That's part of its job.
 
 

This is exactly right. The only thing I would add is in the subsonic zone. We can't hear it but we can feel it. 
 
Supersonic is not anything we need to worry about. We can neither hear it or feel it.