We don't need no stinking hi-rez

[Sal1950]

Have YOU made a report to the site Admin over these issues ?

LOL, No but ya'll started this. But no problem I'll have it done in just a minute.

 

DONE

[kumakuma]

Was "anciet" a typo too ?

 

Was "anciet" a typo too ?

 

Yup, the system is creating bit identical posts.

[Paul R]

Was "anciet" a typo too ?

Probably- I am using voice transcription as my eyes are acting up a bit. You never know, between the voice transcription and spelling checker, some really odd things slip in. I do not misdoubt it is me and not the system.

[The Computer Audiophile]

Have YOU made a report to the site Admin over these issues ?

Sandyk- What browser are you using?

[esldude]

Jud,

 

Here is something of interest that a friend sent to me, that seems on topic. I hope he doesn't mind my sharing it here.

---------

Why we need PCM 352 Khz and DSD 256.

 

It has been fairly well established that the human hearing can detect timing differences greater than 4 micro seconds, 4 x 10^-6

What timing resolution is achievable in the various PCM format sample rates ?

 

44 KHz 22.7 micro seconds

96 KHz 10.4 micro seconds

192 KHz 5.2 micro seconds

352 KHz 2.8 micro seconds

 

For DSD, assuming we want 20 bits to achieve the peak level, then the achievable timing resolution is

 

DSD 64 8.8 microseconds

DSD 128 4.4 micro seconds

DSD 256 2.2 micro seconds

 

Hopefully this stirs the pot!

 

Larry

 

Larry,

 

Please don't post this load of garbage information. It is all incorrect. The timing between samples is not the timing resolution limits of any of those mediums.

 

Not trying to be a jerk Larry. But this is an old myth that gets repeated over and over. Misinformation isn't helping anything.

 

Human timing discernment between ears is on the order of 10 microseconds maybe even a bit lower. Redbook CD has timing resolution with a good enough clock of about 55 picoseconds. Even lower with good dither. Yes 24 bit and higher sample rates lowers this number even further. Even redbook however far exceeds the microsecond range.

 

1/sample rate/number of levels(65,536 for 16 bit)/2pi will give you the timing resolution.

[sandyk]

Sandyk- What browser are you using?

 

I.E.11 , but I am not the only one reporting problems. I think that even Superdad may have experienced a duplicate post a few hours ago. The forum can even be very slow at times in the last day or 2 just responding to page changes.

I also experienced the same thing as sal1950 when the post appeared after I answered yes to leaving the page. See also reply 776 by esldude

[esldude]

Sandyk- What browser are you using?

 

I did not make a report myself yet.

 

I am using Firefox on Windows 8.1 the last couple weeks.

 

The site works normally and fine 90 % of the time. Then every so often the last few days, when you submit a post there is a long wait on the CA site to respond. 30 seconds or a minute later it does. Even then everything works okay. But a couple or three times I saw a double post show up.

 

Also had one act this way posting via tapatalk which double posted.

 

I also have twice tried quick reply and the page kept asking if I wanted to leave or stay on the page. Had to close the page and navigate back to it.

[esldude]

I have difficulty understanding exactly what you were hearing. With this be similar to the result of a comb filter?

 

Well, I haven't heard for sure what Jud did and vice versa. But I believe so. He said one half of the head wasn't synched with the other half. I would have described it as the aural version of a double stamped image. So it sounds similar apparently.

[sandyk]

A reply that I just made in another thread went straight in without any delay. This is the first time this has happened in the last day or so.

 

This one did too !!!

[esldude]

Yep, I agree about simply trying stuff out.

 

About the 16/44.1 vs. 24/96 - There's certainly the thought that if one will hear a difference anywhere, it will be at the extremes, as you are describing with the looping experiment.

 

But I distrust extremes. Archimago offered a linear phase vs. minimum phase test at his site, where each example was taken to extremes of ringing, the thought being to maximize any difference. I had thought to take part (even though I mentioned on the site some reasons why I didn't think the test would necessarily be valid), but never got around to it. I doubt I would have been able to pass the test.

 

However, I have been able to pass blind tests at home, where I put Audirvana Plus on shuffle play with files upsampled using Audiophile Inventory's minimum phase and linear phase filters. With these very good quality files, what I was able to identify blind through my speakers, which carefully preserve phase and timing, was a slight degradation of imaging with the minimum phase filtering. I attribute this to minimum phase filters having group delay, which messes up (technical term) very slightly the phase and timing attributes of the speakers. The difference is a delicate thing (though I feel better imaging does help create a sense of realism), and I'm not at all confident ringing filters would preserve it.

 

That's an example of people believing that the difference being sought must lie in one direction (pre-ringing with linear phase filters, no pre-ringing with minimum phase), and thus devising a test designed to maximize such a difference, which would likely destroy the difference I heard in my system (better imaging with a linear phase filter that preserves phase and timing integrity through my speakers). Thus my overall preference would be to (1) listen to the highest quality files possible (2) using a protocol designed to absolutely minimize differences in any other factor besides the one being tested, because any audible differences may lie in a direction other than one I've thought of before listening.

 

The point was not so much to maximize differences. The point is often a particular type of distortion is easier to hear if it has been increased. Then knowing what it sounds like you might more sensitively identify it at smaller levels. If you already hear linear vs minimum phase, then you need merely to make one conversion with that type filtering. See what it sounds like in your mono test.

[Jud]

I have difficulty understanding exactly what you were hearing. With this be similar to the result of a comb filter?

 

No. Forget comb filtering, because everything was mono. Think of playing a mono track, all instruments and vocals right in the center. Now imagine being at a bizarre fantasy concert where the lead vocalist standing center stage is a composite whose right side is a different person with a different appearance than his left side. That's what it sounded like.

[Jud]

Well, I haven't heard for sure what Jud did and vice versa. But I believe so. He said one half of the head wasn't synched with the other half. I would have described it as the aural version of a double stamped image. So it sounds similar apparently.

 

The two sides were perfectly synced time wise, and from a mono source, so no imaging effects. Imagine painting the left side of a mannequin white and the right side black. Its location won't appear to waiver, but the two sides will look quite different.

[Jud]

Sandyk- What browser are you using?

 

I experienced the same thing early last night posting from my iPhone.

[Jud]

Well, I haven't heard for sure what Jud did and vice versa. But I believe so. He said one half of the head wasn't synched with the other half. I would have described it as the aural version of a double stamped image. So it sounds similar apparently.

 

I obviously haven't been describing this very well.

 

Here's an example of a double stamped image:

 

 

That's not the visual equivalent of what I heard. This is:

 

 

You see that the left and right sides of the face don't cause any interference effects with each other, they just look different. The dividing line is obvious, so the image of the face is not a seamless whole, which is what I meant when I said the center image I heard was not a well-integrated, coherent one. Rather, it was like the TV image above, centered but obviously divided into two halves.

 

Edit: By the way, what I just described was my listening experience with two different brands of capacitors on the two output channels of my DAC, not any experience with 16/44.1 in one ear and 24/96 in the other (which I haven't been able to listen to for the reasons that Audacity won't output such a file, and even if it could I'm pretty certain my DAC and player software wouldn't play it).

[PopPop]

A reply that I just made in another thread went straight in without any delay. This is the first time this has happened in the last day or so.

 

This one did too !!!

 

There may have been a problem with the web server last night. A couple times I posted things I got a screen that there was a problem and that my post had failed. Then, after refreshing the screen, there it was. Not to worry about the duplicate posts mate, I have to read a lot of these things twice any way. [emoji5]️

[Sal1950]

I did not make a report myself yet.

 

I am using Firefox on Windows 8.1 the last couple weeks.

 

The site works normally and fine 90 % of the time. Then every so often the last few days, when you submit a post there is a long wait on the CA site to respond. 30 seconds or a minute later it does. Even then everything works okay. But a couple or three times I saw a double post show up.

 

Also had one act this way posting via tapatalk which double posted.

 

I also have twice tried quick reply and the page kept asking if I wanted to leave or stay on the page. Had to close the page and navigate back to it.

 

If you get that question just answer to leave the page and the posting will continue normally, At least that has been my experience when I got that response page.

I've never experienced a double post, just the slowness and mentioned response page.

[esldude]

I obviously haven't been describing this very well.

 

Here's an example of a double stamped image:

 

[ATTACH=CONFIG]22673[/ATTACH]

 

That's not the visual equivalent of what I heard. This is:

 

[ATTACH=CONFIG]22674[/ATTACH]

 

You see that the left and right sides of the face don't cause any interference effects with each other, they just look different. The dividing line is obvious, so the image of the face is not a seamless whole, which is what I meant when I said the center image I heard was not a well-integrated, coherent one. Rather, it was like the TV image above, centered but obviously divided into two halves.

 

Edit: By the way, what I just described was my listening experience with two different brands of capacitors on the two output channels of my DAC, not any experience with 16/44.1 in one ear and 24/96 in the other (which I haven't been able to listen to for the reasons that Audacity won't output such a file, and even if it could I'm pretty certain my DAC and player software wouldn't play it).

 

Okay, I think I have a better idea of what you are describing. I have been altering files and listening to see what happens. Closest to what you describe would be a minor frequency tilt of one channel vs the other. Assuming I have something close to the right idea about what you were hearing after seeing your additional illustrations.

[esldude]

Oh well, only took 3 attempts to get a post in which then double posted. Is CA seeing unusually high loads of traffic?

 

Lets see first got the stay or leave page thing that would not go away. Decided to hit go advanced. That seemed to work okay except it would do nothing. I killed the page after copying my post. Went back where everything looked okay only to get a double post.

[Sal1950]

Oh well, only took 3 attempts to get a post in which then double posted. Is CA seeing unusually high loads of traffic?

 

Lets see first got the stay or leave page thing that would not go away. Decided to hit go advanced. That seemed to work okay except it would do nothing. I killed the page after copying my post. Went back where everything looked okay only to get a double post.

 

Funny I've never had a double post? Running Linux here on the Chrome browser.

[Sal1950]

Funny I've never had a double post? Running Linux here on the Chrome browser.

When posting above response was very slow, then I got the stay or leave query, answered leave and all ok

[JSWolf]

With 16-bit sampling of an analogue signal, the input maximum amplitude range is divided into 65536 (2 to the 16th power) levels, and the value of the input signal at each sample point is approximated by the nearest level. With 24-bit sampling, you're dividing the input range into 16777216 (2 to the 24th power) steps, which clearly gives a much more accurate representation of the sampled signal.

 

This process is known as quantisation, and the differences between the 'real' analogue input values and their quantised approximations ('quantisation error') effectively adds some noise, which is known as 'quantisation noise'. If we have smaller steps, we have, on average, less quantisation error per sample, and hence less quantisation noise.

[mansr]

With 16-bit sampling of an analogue signal, the input maximum amplitude range is divided into 65536 (2 to the 16th power) levels, and the value of the input signal at each sample point is approximated by the nearest level. With 24-bit sampling, you're dividing the input range into 16777216 (2 to the 24th power) steps, which clearly gives a much more accurate representation of the sampled signal.

 

This process is known as quantisation, and the differences between the 'real' analogue input values and their quantised approximations ('quantisation error') effectively adds some noise, which is known as 'quantisation noise'. If we have smaller steps, we have, on average, less quantisation error per sample, and hence less quantisation noise.

 

True, but not the whole truth. With proper dithering, there is no effective improvement in the representation of an input signal with bounded dynamic range beyond a finite number of bits. For most music, 16-bit sampling is more than enough to push the quantisation noise below audible limits.

[Kees de Visser]

more than enough

The word "enough" isn't very popular amongst audiophiles, probably because it implicates a compromise

[mansr]

The word "enough" isn't very popular amongst audiophiles, probably because it implicates a compromise

 

Does it though? Once you have enough money to buy a Ferrari, adding more money won't make it more Ferrari.

[Kees de Visser]

Does it though? Once you have enough money to buy a Ferrari, adding more money won't make it more Ferrari.

But adding more money could make it two Ferraris. Although you can't drive them both at the same time, it might be fun to have them. Rather similar to those extra bits in hi-res audio.