Mark, you are correct that this hi-rez thing cannot work well as a "castle in the air", i.e. without a solid base of outstanding records.
The sad reality is that the recording industry is very slow at producing them. It wasn't any better with CDs - in 3 decades they produced what, 100 of them? I remember visiting "hi end" audio shows, just to hear the same tracks all over again, year after year. A very unnerving sentiment of deja vu and stagnation.
The public at large is not dumb at all - like many are trying to blame them for. They noticed the subpar quality and decided it's not worth their financial effort. So they looked at getting their tunes at wholesale prices - and they found that with streaming. That's where most of the action is this days. After some other idiots killed FM radio with incessant advertising, now the public is on the streaming platforms.
With the recording studios concentrating on "cost control", it will take us a few more decades to get 100 audiophile reference records in hi-res.
But the good thing is that those hi-rez players are very good at playing CD material. Unlike the vinyl to CD transition, this one is a smooth one. No rug pulled from under our feet.
The sad reality is that the recording industry is very slow at producing them. It wasn't any better with CDs - in 3 decades they produced what, 100 of them? I remember visiting "hi end" audio shows, just to hear the same tracks all over again, year after year. A very unnerving sentiment of deja vu and stagnation.
The public at large is not dumb at all - like many are trying to blame them for. They noticed the subpar quality and decided it's not worth their financial effort. So they looked at getting their tunes at wholesale prices - and they found that with streaming. That's where most of the action is this days. After some other idiots killed FM radio with incessant advertising, now the public is on the streaming platforms.
With the recording studios concentrating on "cost control", it will take us a few more decades to get 100 audiophile reference records in hi-res.
But the good thing is that those hi-rez players are very good at playing CD material. Unlike the vinyl to CD transition, this one is a smooth one. No rug pulled from under our feet.
Last edited:
How about "audiophile environments"? I don't think this thing hi-rez was targeted at "engineering environments".
With regard to certain of the recent comments, I would like to respond: What make some threads exciting and very active is when sides are taken and tribalism takes over. Each side talks past the other using debating techniques, reasoning, argumentation, name calling, as much and as far as the moderators will allow. It gets people worked up and coming back for more, but it also bears some similarity in human behavior to the two main banned topics. But, for DIYAudio folks, it can suck one in as others are sucked into reality TV or professional wrestling. Is that type of thread good for the DIYAudio website? I guess it brings people to look at ads, and maybe shop at the store, etc. When it gets down to reducing the arguing and working together to do some work and learn something, then it's not fun anymore, apparently so anyway. I guess then the forum may become less well thriving?
When I first came here, I thought, finally a group of smart people who have invested time and effort to understand engineering. A very select group that might turn out to have a lot more on the ball than other forums. A interesting group of people to get to know and to converse with. If this thread, now having become maybe a little less argumentative and polarized, has at the same time become uninteresting to the participants because of those changes, and now fizzles out because it was principally the fighting and controversy that drew interest and participation, then I think I may be out of here too, but not to Blowtorch. Just out. Why waste time to end in fizzles? Not much use to it.
When I first came here, I thought, finally a group of smart people who have invested time and effort to understand engineering. A very select group that might turn out to have a lot more on the ball than other forums. A interesting group of people to get to know and to converse with. If this thread, now having become maybe a little less argumentative and polarized, has at the same time become uninteresting to the participants because of those changes, and now fizzles out because it was principally the fighting and controversy that drew interest and participation, then I think I may be out of here too, but not to Blowtorch. Just out. Why waste time to end in fizzles? Not much use to it.
Last edited:
Mark, ur incorrect.
There are many active and useful threads that pertain to specific designs and circuits.
Look around.
The problem comes (if it does, as in this thread) where there is something that is more general, less specific and conceptual, like this thread.
One of my aims was to reduce the discussion to a well thought out and specific test based on a specific circuit/design... and move in an ordered and reproducible manner from that point. In order to focus and limit the "professional wrestling".
Ed has dipped his toe in a few times now, not to much result.
The other thing is that it is not necessary to post a response to EVERYTHING (Mr. Kirchoff? and some others...), and keep up a high rate of posts for the thread to progress and have benefit and value. The two things are unrelated, except in a negative correlation.
There are many active and useful threads that pertain to specific designs and circuits.
Look around.
The problem comes (if it does, as in this thread) where there is something that is more general, less specific and conceptual, like this thread.
One of my aims was to reduce the discussion to a well thought out and specific test based on a specific circuit/design... and move in an ordered and reproducible manner from that point. In order to focus and limit the "professional wrestling".
Ed has dipped his toe in a few times now, not to much result.
The other thing is that it is not necessary to post a response to EVERYTHING (Mr. Kirchoff? and some others...), and keep up a high rate of posts for the thread to progress and have benefit and value. The two things are unrelated, except in a negative correlation.
You've got it backwards.
One works from the general TO the specific.
Had anyone who is interested in being "specific" - and we have at least a dozen or more participants who are "qualified" as either EE and/or known circuit designers of proven ability - THEN had they expressed such an interest, I'd have (or one of them would have) opened and linked a thread in Solid State pertaining to the design and implementation - then brought the results back to a thread such as this one.
The "Everything Else" category does not mean "nothing at all" or "spout off endlessly" (afaik, that's closer to the "Lounge") this category is for things that are worthy and do not fall well into one of the other more narrow categories.
But, you knew that?
_-_-
Really, I have to stop responding to what is essentially foolishness...
The additional problem is lines get drawn across topics that are no longer easy to suss out. It's not a novel front end or error-correcting OPS, it's a discussion of imprecise listening preferences. To make progress, especially in terms of subtle listener preferences, requires a LOT of legwork and expense (time and $$). Far beyond what I'd expect a non-academic/non-industry (large company and/or consortium) group to accomplish. Small effect (subtle differences) + human beings = very big N with lots of controls to get much reliable.
And then there's the fact that people's (mine most definitely included) preconceived notions will be challenged.
So it falls back to uncontrolled listening impressions, with their enormity of caveats applied.
I have been thinking about a simple experiment to run. We would need to be able to produce hi-rez wave files with known and verified levels of distortion. I have talked about using recorded cymbals as test sounds before, so I will not repeat everything again at the moment except to say I would probably use single cymbal hits as test samples. Because memory of very subtle sound differences can be very brief, I was thinking of trying creating wave files consisting of a reference cymbal hit, followed by a sequence of hits each at some specific distortion level. Having the reference hit just before the test hits may allow for adequate memory retention of the reference hit sound subtle details. I would like to try some people who we think have pretty good sensitivity to distortion and try some variations of cymbal hit distortion level sequences and see how little distortion they can detect with at least some reasonable statistical significance. This would still be in the design, development, and testing phase of the experimental design. When we find what seems to work best for detectability of low, but known distortion levels, we could roll out the test to a new set of human subjects to see if they can correctly identify patterns of cymbal distortion in some set of wav files. In other words, I would like to explore a little bit what might be a good way to measure any ability at all to detect low levels of distortion. I don't think comparing long sections of a song against each other in ABX style, with a long time delay between cueing up songs would likely be the most sensitive test. I'm not sure what would be the most sensitive, so I don't know what else to do but try some experimenting on that.
What I have described are very preliminary, off-the-top-of-my-head type thoughts. Constructive criticism and ideas for improvements are of course welcome.
What I have described are very preliminary, off-the-top-of-my-head type thoughts. Constructive criticism and ideas for improvements are of course welcome.
Last edited:
Mark--might this get you closer already?
https://www.quantasylum.com/content/Home/tabid/40/Post/342/ABX-Testing-and-Distortion
https://www.quantasylum.com/content/Home/tabid/40/Post/342/ABX-Testing-and-Distortion
I said audio forums, any other suggestions for "audiophile environments"? What does not perceiving sufficient benefit have to do with being stupid or not?
It takes analog in and outputs analog. I would have to play back a wav file of a cymbal hit though my DAC into that thing, then record it's output into my A/D to make the wave file I want. By that time there is more distortion than what the analyzer says, since my A/D distortion is added. Also, my nice cymbal sample I started with has gone through multiple unnecessary data conversions. That is, at least as far as I can tell from what I see on the website and the specifications. Also, not sure exactly how it produces distortion at this point, since they say they are looking at adding new ways in the future. Not convinced that spending $429 on this would be the best investment for what I would like to do.
I think what I would like to see would be better implemented all in software. Start with a digital wav file, end with some distorted digital wav files. Assemble them into sequences for attempts at distortion pattern identification. Something like that. Again, I don't know what would provide best testing sensitivity, so I would like some capability to experiment on that. I would also like to experiment with higher order harmonic distortion, the more objectionable type, and with some IMD. Perhaps some hysteresis distortion to emulate one property of ferrites. I would probably settle for higher order harmonic distortion though, at least to begin with.
Also, if someone had a convenient way for them to do it analog, digitize it, and measure the result in the wav file, that would be worth a try. Don't know without trying it. Again, not willing to spend $429 to try whatever that thing does. Maybe if someone else already owns one, they could try to produce some cymbal hits for evaluation.
Last edited:
What many here define as 'distortion' may not be all there is to it. I have been making designs for decades that pretty much have low enough distortion, (usually almost un-measurable at typical operating levels), and yet I have found distinct differences in every one of my designs, both amp and preamp. There appears to more to it. IC op amps are by their very nature, imperfect. One of the LAST factors may well be the output stage quiescent current setting, that is way too conservative for Class A operation. Even the AD797 has a Class B output stage, and most other op amps are usually even worse.
Those of you who do not live with these subtle differences on a regular basis, might be immune to hearing differences between CD and extended recording techniques. So be it, but please don't tell us what is audible or not.
Those of you who do not live with these subtle differences on a regular basis, might be immune to hearing differences between CD and extended recording techniques. So be it, but please don't tell us what is audible or not.
It seems to me that if they sound different, then they are reproducing the input signal differently, and that then is by definition a type of distortion. It may be a type not well measured by your distortion testing equipment though. I know some people don't think distortion can exist that their equipment isn't good at measuring, but I don't know. Physicists used to think something could fall into a black hole and general relativity would apply. Now maybe not so sure. Maybe the thing falling gets stuck at the event horizon or the firewall. But, man they sure used to be confident about what they thought before!
Last edited:
Mark, not sure where that link sent you to, but the blog post I meant it to go to is an application that takes a wav file and injects distortion into that wav per your configuration. Not the hardware itself. If you have to navigate to it from the home page, then go to the "September 3rd, 2016" blog post.
Last edited:
Okay, tried it. Funky app. Plays wave file once through then quits and you have to start over. Outputs only to default Windows wav device. Fortunately the Lynx-2 mixer I have on the desktop system allows me to reroute that digitally. The smallest crossover distortion they offer is .99, which they refer to as "slight." Don't know what it is expressed as a percentage (maybe .01?), but it is audible although fairly slight by casual listening standards. Mastering engineers would have no trouble with identifying it, I don't think. We would need something more subtle than that to do a distortion sensitivity threshold search experiment, IMHO. Another possible issue is that it doesn't tell you if A or B is the distorted version, not sure if its always the same or if it randomizes. For assembling a test wav file with distortion less than I can hear, but the others might be able to hear, I would need to know for sure which file is which.
Last edited:
This mode of thought has been dragged trough the mud over and over. Look where this thread is now, truly forlorn.
Only JC gets it.
"Insanity: doing the same thing over and over again and expecting different results." - Albert Einstein
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.