Measurement of total system behavior (response, distortion, phase etc) to the reproduced sound at listening position.
Well, I agree with that, and can be done on an individual basis, but surely that's an example how measurements can help, not hinder enlightenment?
The point is that "advertising is image" and irrelevant measurements are widely used to create a certain "image".
I don't understand why you don't answer my simple question about your assertion that measurements can lead us astray, do you not consider it a sensible question?
I think what he was getting at was fairly obvious.......if you try to apply existing measurement to a ‘perception’ you might not be measuring what actually matters. This is more than likely due to the fact alot of folks think there’s no further frontier in audio and it’s all ‘good enough’ and sounds the same......anything that sounds different from ‘good enough’ is broken.
So basically the measurements that infer ‘good enough’ might not be good enough. 😛
That's probably what he ment, so the same old might not, uncertainly, doubt thing expressed in a novel way?
That's probably what he ment, so the same old might not, uncertainly, doubt thing expressed in a novel way?
There are many things I was Taught as gospel that I’ve either come to know as misunderstood or just plain wrong to start with.......to say ‘this is it’ pack it up ‘nothing left to see here’ is just plain wrong.
I know that's why I'm not interested in wasting time with someone who can't or won't learn - I've been down that rabbit hole with him before & what's posted 2 pages ago (or even a post or two before) is completely ignored or has flown completely over his head. When people post about something I know little about, I either ask sensible questions in an attempt to learn or stay quiet & listen & learnHe calls FUD quite often, doesn't know enought to ask sensible and leading questions.
More like buffoon than clown who jumps in whenever he see an opening for some attempt at a put-down. He doesn't know enough himself so rides on other posters tails in his put-downsSJ you might think you are the 'class clown' interjecting constantly, that your inputs are in someway humorous.
AgreedPlease see it from others' POV that this behaviour of yours of constant interruption and clogging of the forum encourages the like minded to do the same and amounts to an incredibly selfish waste of other members time.
Also please consider that while you keep interrupting, sensible and educational posts get lost and good opportunities for advancement are squandered......your selfish attention seeking behaviour would not be tolerated in the household or the workplace, please stop bringing it here.
If only 😀Yes he repeatedly is not interested in answers, this thread is his primary form of 'entertainment' it seems.
SJ enough is enough, most of us have had a gutfull of your nonsense, please tone it down or take a holiday from this forum for for a while, the forum and you will both be better for you doing so.
I think what he was getting at was fairly obvious.......if you try to apply existing measurement to a ‘perception’ you might not be measuring what actually matters. This is more than likely due to the fact alot of folks think there’s no further frontier in audio and it’s all ‘good enough’ and sounds the same......anything that sounds different from ‘good enough’ is broken.
So basically the measurements that infer ‘good enough’ might not be good enough. 😛
It has been said numerous times already but some people (not you, MM) don't absorb information
I'll boil it down to essentials - we listen to a dynamically changing soundwave (i.e music) & use our auditory system to evaluate it according to the full spectrum of techniques used internally & yet to be discovered. Perceived accuracy is a perception of the soundstage being very well defined & steady, with layering & each voice/instrument being in the correct stable, unwavering place with all the timbre of the instruments & voices being realistic. We know what this accuracy is because we experience it in the real world & the soundscape we are exposed to every day at every moment or when we go to a concert. So we are judging our replay system based on this.
Now someone comes along & does the usual measurements using a sine wave signal to map THD, phase, frequency, amplitude, etc. And using FFT to extrapolate this signal into a repeating signal in order that it can be analyzed to levels that impress the pants off us. But step back & ask what could possibly be wrong with this? Is there something missing in this scenario?
How can these possibly tell us much about the behaviour of the system when it is handling music-like signals - about the stability of the soundstage, the layering, the timbre, etc.
There is just such a big gap between the perceptual side & the measurements side that asking people to accept measurements as defining perceptual accuracy is laughable & those that call this FUD or scoff at the idea are similarly buffoons who think they know it all.
You know this is the great scam perpetrated by those-that-know-it-all - they expect us to swallow this scam on two fronts - they use simplistic measurements & say that it fully represents perception - they use flawed blind testing & say this is what we are able to hear normally
The sensible ones who have some experience of how really good audio can sound (as JC, RM, MM & others have said) based on their own experience know this is a scam but a lot of people who have never experienced really good audio fall for the scam. Falling for the scam unfortunately prevents these people from ever being curious enough to investigate really good audio - that is the great shame in all of this
So Ethan Winer is a buffoon?
Absolutely!! No better example could you have chosen. A guy who sells room treatments & guess what all competent audio electronics sound the same - patently obvious marketing spiel? He uses exactly what I am talking about - simplistic thinking & simplistic testing to fool the people who swallow the scam that I am talking about - a big supporter of blind ABX testing. Quick question did he ever do a blind test of his room treatments??
BTW, he's the one I was referring when I said, ironically, all audio electronics have been sorted since 1950 - something he stated before.
You may remember you queried this post of mine before - well it comes from him - a buffoon, par excellence
Last edited:
Finding a tiny difference between two things which ought to be the same will always be difficult - unless the test is sighted in which case it could sometimes be fairly easy. In order to show that ABX has a significant bias towards a null result you would need to find a difference which is blindingly obvious yet does not show up in ABX. Something measurable would be good, such as different frequency response.mmerrill99 said:Yes, it is likely a difference in the DAC processing between 16/44 & 24/192 but so what? The point is how difficult it is to get a positive result in a Foobar ABX test & thereby 'prove' that there is an actual audible difference. It is an example of a real world Foobar ABX positive result & what is needed to achieve this. Remember that he had already established his preference for 24/192 over long term sighted listening without difficulty but still found that he needed to find a specific aspect of the sound that would allow him to differentiate between 16/44 & 24/192 & this wasn't an easy thing to find - then he needed to do a substantial amount of training/warming up before he could reliably identify differences & finally he found that retaining focus during the testing was arduous.
Your apology is accepted. I realised that we think in quite different ways because I found that almost anything I said seemed to annoy you, so I tried to avoid commenting on your posts.tapestryofsound said:DF96,
Sorry I kicked off with you awhile back, and said a few things that maybe I should not have. I have since come to realise our intellects are so diametrically opposite that in many ways we are equals, and as such, I now consider you as part of my people.
I thought so at first, but then I realised that a really expert pedant could construe the apparent denial as meaning 'it is untrue that I was with them in the past but I make no comment about whether I am with them now'. His later denial was much clearer. For what it is worth, I had decided a few months ago that he probably was not with MA but the lack of a clear denial still left open the possibility - which others here were still trying to pursue.Joe Rasmussen said:I never saw this ambiguity when he answered Nope!
No. By definition the original data was bandlimited (to half the original sampling frequency) so can fully be described by any new faster sampling frequency. Going from 16 to 24 bits provides plenty of room for rounding errors in the interpolation, so for all practical purposes the upsampled stream contains exactly the same information as the original stream. Any audible differences must therefore come from the subsequent DAC process and any filtering there.indra1 said:Upsampled stream should contain interpolated data and some smoothing, also different cutoff freq in output filter would add to the difference.
I read some of the website RF page & he seems to contend that RF signals being handled by ICs cause a drain on the PS which reduces the power available for the actual signal processing & reduced dynamics results? When a device oscillates due to RF handling then this is very much obvious as overheating of IC & eventual failure of active component so what about a level of RF that doesn't cause this oscillation - is it just benign? He says not, what are others view on this?
Everything here is simply nonsense and made up. He does not even have a basic grasp of the concepts. For instance he says "20,000 Volts of RF" when he does not even understand the concept of volts per meter as the measure of field strength. Will some one here who is a ham or RF designer please inject some sanity?
I thought so at first, but then I realised that a really expert pedant could construe the apparent denial as meaning 'it is untrue that I was with them in the past but I make no comment about whether I am with them now'. His later denial was much clearer. For what it is worth, I had decided a few months ago that he probably was not with MA but the lack of a clear denial still left open the possibility - which others here were still trying to pursue.
So "nope" does not mean no? One of us is confused, pretty sure it is not me.
I have a question for you: Are you one of those who believe there is a 'holy war' against blind testing?
In life we cannot have full information on everything yet we still have to make decisions. If someone talks nonsense about electricity and RF yet claims to be able to deal with RF problems then we assume he is not able to do this. This saves us wasting our time and money. Now, in theory, could he be right in the sense that his mods do something genuinely useful in spite of his confusion? Yes, in theory that is possible. How likely is it? An alternative hypothesis is that his nonsense is just marketing and he doesn't believe it either. Finally there is the hypothesis that he believes the nonsense (as do some of his customers) but his mods do nothing useful (for his customers). I really don't care which of the latter two options are true; in either case I would not encourage people to employ him.mmerrill99 said:Dan, there are a core group on here that always look for a flaw, a reason to reject, a conviction that they are always right , a lack of curiosity of any other possibility.
Nonsense. Excessive RF can cause bias shifts and these might in extremes cause problems but these would be direct problems, not mediated via extra PSU drain. He would be better off keeping quiet; by talking nonsense he alerts people who know about electronics that he doesn't know what he is doing.I read some of the website RF page & he seems to contend that RF signals being handled by ICs cause a drain on the PS which reduces the power available for the actual signal processing & reduced dynamics results? When a device oscillates due to RF handling then this is very much obvious as overheating of IC & eventual failure of active component so what about a level of RF that doesn't cause this oscillation - is it just benign? He says not, what are others view on this?
Have you ever heard of Mr. Fourier or Mr. Laplace? "Music-like signals" come within the scope of signals which Fourier can handle - all that is needed is a finite number of finite discontinuities; music has none, so therefore fits the criterion. To a transistor, music is no more complex than a sine wave. What you seem to want is the electronics equivalent of an alchemist; someone who can a little 'eye of newt' to add the sparkle and musicality to a circuit and components which have mostly been put together using conventional engineering, handling a signal which has been produced using the same engineering.Now someone comes along & does the usual measurements using a sine wave signal to map THD, phase, frequency, amplitude, etc. And using FFT to extrapolate this signal into a repeating signal in order that it can be analyzed to levels that impress the pants off us. But step back & ask what could possibly be wrong with this? Is there something missing in this scenario?
How can these possibly tell us much about the behaviour of the system when it is handling music-like signals - about the stability of the soundstage, the layering, the timbre, etc.
Is there? How big is "big"? Where do you think the measurements came from? Did evil engineers conspire together to invent measurements out of thin air just to annoy the alchemists? Did we wrongly apply to audio some measurements which we found useful elsewhere?There is just such a big gap between the perceptual side & the measurements side
Is ABX testing not meant for small differences or what are you guys promoting it as?Finding a tiny difference between two things which ought to be the same will always be difficult - unless the test is sighted in which case it could sometimes be fairly easy. In order to show that ABX has a significant bias towards a null result you would need to find a difference which is blindingly obvious yet does not show up in ABX. Something measurable would be good, such as different frequency response.
I showed the difficulty involved in Foobar ABX testing with a real-world example of someone who used it & got positive results. It shows what's actually involved, not talking hypothetically or abstractly.
So you agree that using Foobar ABX testing for small differences "will always be difficult"
Blindingly obvious differences don't need ABX testing & I never suggested that blindingly obvious differences will show Foobar ABX will return nulls - I always said ABX testing was only good for gross differences - small differences are where the null bias is evident. You sta ethe same "a tiny difference between two things which ought to be the same will always be difficult"
I seem to remember Jakob & I both proposing that controls should be used in Foobar ABX tests so as to evaluate this bias towards nulls but I wouldn't suggest using gross differences as controls - controls are meant to evaluate the test & tester at a level somewhat equivalent to the DUT.
Are we still talking about this? - FFS, everybody needs to get over it.......
I thought so at first, but then I realised that a really expert pedant could construe the apparent denial as meaning 'it is untrue that I was with them in the past but I make no comment about whether I am with them now'. His later denial was much clearer. For what it is worth, I had decided a few months ago that he probably was not with MA but the lack of a clear denial still left open the possibility - which others here were still trying to pursue.
I believe there are people who wish to denigrate blind testing. Some of them do so because of commercial interests. Some do so because they wish to preserve the 'magic' of audio; they simply prefer alchemy to chemistry and believe that one day lead will turn into gold. Perhaps some of them find chemistry too difficult to learn? Others do so for other reasons, which might be unclear.Joe Rasmussen said:I have a question for you: Are you one of those who believe there is a 'holy war' against blind testing?
I am not promoting ABX. I think part of the problem arise from the way that differences which are claimed to be 'huge' when sighted somehow disappear under blind conditions (ABX or something else). These 'huge' differences are sometimes ones which physics says should be negligibly small (e.g. differences between adequate cables) while psychology says they could be 'huge' (e.g. differences between 'high-end' cables and ordinary cables).mmerrill99 said:Is ABX testing not meant for small differences or what are you guys promoting it as?
I believe there are people who wish to denigrate blind testing. Some of them do so because of commercial interests. Some do so because they wish to preserve the 'magic' of audio; they simply prefer alchemy to chemistry and believe that one day lead will turn into gold. Perhaps some of them find chemistry too difficult to learn? Others do so for other reasons, which might be unclear.
Some of whom couldn't care less, haven't you thought of that? To most people it is a non-topic, maybe you can't come to grips with that? There is no need for war, why incite one for a reason that most don't give a hoot about.
But then, I am only giving an opinion, but it seems to clear to me. There is no 'holy war' in this matter.
Sure, that's reasonable enough & others are curious enough to investigate his mods/tweaks - in the case of RM he reports positive audible effects - don't know how many others report the same about his mods but given that he has a 30 day returns policy it would seem reasonable to assume he has satisfied customers?In life we cannot have full information on everything yet we still have to make decisions. If someone talks nonsense about electricity and RF yet claims to be able to deal with RF problems then we assume he is not able to do this. This saves us wasting our time and money. Now, in theory, could he be right in the sense that his mods do something genuinely useful in spite of his confusion? Yes, in theory that is possible. How likely is it? An alternative hypothesis is that his nonsense is just marketing and he doesn't believe it either. Finally there is the hypothesis that he believes the nonsense (as do some of his customers) but his mods do nothing useful (for his customers). I really don't care which of the latter two options are true; in either case I would not encourage people to employ him.
Ok, I'm not an RF expert so you tell me that as a component approaches oscillation from RF signal processing that it is well behaved up to that tipping point at which it oscillates - nothing, nada is happening with the component processing RF signals up to this point?Nonsense. Excessive RF can cause bias shifts and these might in extremes cause problems but these would be direct problems, not mediated via extra PSU drain. He would be better off keeping quiet; by talking nonsense he alerts people who know about electronics that he doesn't know what he is doing.
Or are you saying a bias shifts may happen? What bias shifts are you talking about & can you say more about how this might effect the actual audio signal processing/handling.
You see there it is - the attempt to explain that a music signal is just composed of a finite number of sine waves so let's just measure a sine wave & extrapolate the results to how the SYSTEM will behave with music. You talk about a transistor when what is underlying why complex music-like signal testing is required, is the behavior of the SYSTEMHave you ever heard of Mr. Fourier or Mr. Laplace? "Music-like signals" come within the scope of signals which Fourier can handle - all that is needed is a finite number of finite discontinuities; music has none, so therefore fits the criterion. To a transistor, music is no more complex than a sine wave. What you seem to want is the electronics equivalent of an alchemist; someone who can a little 'eye of newt' to add the sparkle and musicality to a circuit and components which have mostly been put together using conventional engineering, handling a signal which has been produced using the same engineering.
Playing the poor misunderstood engineer as victim will not get you brownie points.Is there? How big is "big"? Where do you think the measurements came from? Did evil engineers conspire together to invent measurements out of thin air just to annoy the alchemists? Did we wrongly apply to audio some measurements which we found useful elsewhere?
It's the streetlight effect in action & always has been - easier to do measurements of DUTs with simple well-defined test signals & try to extrapolate the results to how the full interconnected SYSTEM will behave with music.
Sorry but playing the pity me sympathy card doesn't work
I am not promoting ABX. I think part of the problem arise from the way that differences which are claimed to be 'huge' when sighted somehow disappear under blind conditions (ABX or something else). These 'huge' differences are sometimes ones which physics says should be negligibly small (e.g. differences between adequate cables) while psychology says they could be 'huge' (e.g. differences between 'high-end' cables and ordinary cables).
Again you demonstrate the disconnect between the understanding of auditory perception & how a small change which effects the perception of the whole playback is described as "huge" but when trying to find a specific area in a ABX test which can reliably be discerned under repeated ABX runs needed fro statistical significance
Read up on auditory perception & how dynamic auditory scenes are created internally.
- Status
- Not open for further replies.
- Home
- Member Areas
- The Lounge
- John Curl's Blowtorch preamplifier part III