[bigshot]

In the 70's HK provided to its dealers an excellent LP of the impact of fr on reproduction. Heck, it was evident on the Bose 901s, and even more so one some other more critical loudspeakers.

In the 70s, my brother had a fantastically expensive McIntosh system with custom speakers and a nice Thorens turntable. He had a test record with separate bands of tones covering the full spectrum of human hearing. After a dozen playings the super high frequencies started turning into a mush of distortion. The delicacy of the modulations in the grooves required to reproduce those super high frequencies didn't hold up to the mechanical means of playing the record. Most LPs had rolloffs at the top to prevent premature record wear. Wanna bet that test record you remember was dealing with frequencies well below 14kHz?

 

[Albedo]

Concerning SACD are we referring to the Meyer & Moran Paper? -> http://www.whatsbestforum.com/showthread.php?2812-Two-unresolved-issues&p=39527&viewfull=1#post39527
 
About audible jitter -> http://www.whatsbestforum.com/showthread.php?1151-Audible-Jitter-amirm-vs-Ethan-Winer&p=12249&viewfull=1#post12249
 
Further down the thread, post #142..
 
Quote: http://www.whatsbestforum.com/showthread.php?1151-Audible-Jitter-amirm-vs-Ethan-Winer&p=46982&viewfull=1#post46982

We take digital data that is nicely marked with its timing and data in a computer file and turn it into a real-time stream across a cable. The source is the “master” telling the destination when to play each and every sample. On paper, the timing is perfect vertical pulses that instantly go from zero to one, and with zero variation. Sampling theory says if we did that, indeed we can have perfect reproduction. That is, if we reproduce the samples at the same time they were captured (digitized), we can reconstruct the signal perfectly.

Alas, real world doesn’t work that way. The above definition of our timing signal is that of a square wave. The wiki on square wave says it nicely: “An ideal square wave requires that the signal changes from the high to the low state cleanly and instantaneously. This is impossible to achieve in real-world systems, as it would require infinite bandwidth.” Let me repeat: you need infinite bandwidth. No cable or interface has infinite bandwidth. So we know what gets to the other side has less than perfect edges. As soon as we modify those waveforms, we also start to mess with the timing that can be extracted at the receiver. Yup, horrors of horrors. Your digital cables can have a sound!

We get confused looking at the low-speed audio signals thinking not much accuracy is needed to represent their timing. Arny made that mistake of thinking 1 microsecond of timing should still be great. That is one millionth of a second. By a person not schooled in the science of digital audio and signal processing, that does look like the right metric relative to CD’s 44,100 samples per second with each sample taking 22 microseconds.

 

 

[bigshot]

We were talking about Macintosh computers having 1.5 ns of jitter... Inaudible.

I did my own SACD comparison test. I determined that all the differences I could find were differences in mastering and mixing.

 

[Albedo]

We.. as in you and me? No -> http://www.head-fi.org/forum/thread/560773/does-it-really-sound-the-same/195#post_7608893
 
The first Macintosh was introduced on January 24, 1984, you were in this thread talking about the seventies and McIntosh.

 

[bigshot]

McIntosh is a high end audio brand. The computer is Macintosh. The subject of jitter was brought up because someone claimed Mac computers have "horrible jitter" which was measured by someone (not necessarily reliable) at 1.5 ns.

 

[Prog Rock Man]

I have to copy and paste to get this quote
 
"We take digital data that is nicely marked with its timing and data in a computer file and turn it into a real-time stream across a cable. The source is the “master” telling the destination when to play each and every sample. On paper, the timing is perfect vertical pulses that instantly go from zero to one, and with zero variation. Sampling theory says if we did that, indeed we can have perfect reproduction. That is, if we reproduce the samples at the same time they were captured (digitized), we can reconstruct the signal perfectly.

Alas, real world doesn’t work that way. The above definition of our timing signal is that of a square wave. The wiki on square wave says it nicely: “An ideal square wave requires that the signal changes from the high to the low state cleanly and instantaneously. This is impossible to achieve in real-world systems, as it would require infinite bandwidth.” Let me repeat: you need infinite bandwidth. No cable or interface has infinite bandwidth. So we know what gets to the other side has less than perfect edges. As soon as we modify those waveforms, we also start to mess with the timing that can be extracted at the receiver. Yup, horrors of horrors. Your digital cables can have a sound!

We get confused looking at the low-speed audio signals thinking not much accuracy is needed to represent their timing. Arny made that mistake of thinking 1 microsecond of timing should still be great. That is one millionth of a second. By a person not schooled in the science of digital audio and signal processing, that does look like the right metric relative to CD’s 44,100 samples per second with each sample taking 22 microseconds."
 
What sound would that be? Decresed sound stage? Better treble? Or just not working and break ups and crackles?

 

[nick_charles]

 
Quote:

Well that settle's it!

Try this simple experiment. Obtain a high resolution recording of a simple, solo violin. Use a device that allows you to cut off (not roll off) fr about 18K, 16K and so on. Let us know what happens.

In the 70's HK provided to its dealers an excellent LP of the impact of fr on reproduction. Heck, it was evident on the Bose 901s, and even more so one some other more critical loudspeakers.
 

 
Actually back in the late 70s before the CD standard was finalised a group of researchers from JVC did tests on the effect of hard cut-off filters at 14, 16, 18 and 20K using trained  audio engineers
 
Quote:

We asked about 30 audio engineers to be judges. They have a wide experience in listening to sounds and are
sensitive to musical material. For example, some of them are loudspeaker designers, amplifier designers, mastering
operators, tape-recorder designers, some of them play instruments, etc.

 
- they had speakers capable of FR in excess of 35K and chose music with much information above 20K, some of which they described as synthesizer noodling.  the paper was called Sampling-Frequency Considerations in Digital Audio by TERUO MURAOKA, YOSHlHlKO YAMADA, AND MASAMI YAMAZAKI , J . Audio Eng. Soc.,  vol. 26, p. 54 (Jan/Feb. 1978).
 
They found that none of their listeners managed a 95% reliability at detecting the 20K low pass, or the 18K low pass or the 16K low pass - it was only at 14K that it was clearly audible. They concluded that leaving it at 20K was okay.

 
[size=xx-small][size=xx-small]　[/size][/size]
[size=xx-small][size=xx-small]o·L- -------illL------- ------ 2 7----[/size][/size]
[size=xx-small][size=xx-small]14KHz 16KHz [/size][/size][size=xx-small][size=xx-small]IBKHz 20KHz[/size][/size]
[size=xx-small][size=xx-small] [/size][/size]
 

 

 

 

[Chromako]

Quote:

Alas, real world doesn’t work that way. The above definition of our timing signal is that of a square wave. The wiki on square wave says it nicely: “An ideal square wave requires that the signal changes from the high to the low state cleanly and instantaneously. This is impossible to achieve in real-world systems, as it would require infinite bandwidth.” Let me repeat: you need infinite bandwidth. No cable or interface has infinite bandwidth. So we know what gets to the other side has less than perfect edges. As soon as we modify those waveforms, we also start to mess with the timing that can be extracted at the receiver. Yup, horrors of horrors. Your digital cables can have a sound!

Sure. The square wave won't be perfect. However, digital receiving ICs are designed to correct for this. 
 
See the image: It's of a digital signal, and it's not even close to an exact square wave. But the receiving chip only cares about that hexagon to tell a one or zero. Excursion into the hexagon, it's a zero. Signal goes around the hexagon, it's a one. The microprocessors aren't even designed for a perfect square wave.
 

 
 
 
 

 

[bigshot]

14 kHz is the line I found in my own informal testing too. Using an equalizer will teach you a lot about sound.

Square waves are about as far from "natural sound" as you can get. The real world has no place for square waves.

 

[Rdr. Seraphim]

 

 
Actually back in the late 70s before the CD standard was finalised a group of researchers from JVC did tests on the effect of hard cut-off filters at 14, 16, 18 and 20K using trained  audio engineers
 
 
- they had speakers capable of FR in excess of 35K and chose music with much information above 20K, some of which they described as synthesizer noodling.  the paper was called Sampling-Frequency Considerations in Digital Audio by TERUO MURAOKA, YOSHlHlKO YAMADA, AND MASAMI YAMAZAKI , J . Audio Eng. Soc.,  vol. 26, p. 54 (Jan/Feb. 1978).
 
They found that none of their listeners managed a 95% reliability at detecting the 20K low pass, or the 18K low pass or the 16K low pass - it was only at 14K that it was clearly audible. They concluded that leaving it at 20K was okay.

 
[size=xx-small][size=xx-small]　[/size][/size]
[size=xx-small][size=xx-small]o·L- -------illL------- ------ 2 7----[/size][/size]
[size=xx-small][size=xx-small]14KHz 16KHz [/size][/size][size=xx-small][size=xx-small]IBKHz 20KHz[/size][/size]
[size=xx-small][size=xx-small] [/size][/size]
 

 

 

I wonder what the average age of the engineers were. My little experiment was during my High School years, with zero formal training, but using all "state of the art" hi-fi gear (HK Citation, Marantz, Thorens, etc) of the period, with a bunch of my buddies from our Physics class. It was obvious and apparent to all the kids of that ilk. Nothing so fancy as what you're quoting. Can't help but wonder if things null as a result of the test itself. To this day, a buddy and I can still hear 14.5K, but at a much lower level than way back when

 

[bigshot]

It isn't the ability to hear, it's the importance of those frequencies to music. Even if you have absolutely perfect hearing, it doesn't matter because there just isn't all that much going on up there. Look at how your favorite CD displays on a frequency spectrum chart in your sound editing program. It's a great big hump right down the middle of the audible range and almost nothing at the very top.

Worrying about square waves and super high frequencies ignores what we buy stereo equipment to play... MUSIC. This stuff looks great on paper, but in the playback of recorded music it's as necessary as teats on a bull hog.

 

[Rdr. Seraphim]

Tha's where I'll hold to my own experience, and understanding of music. I'd rather retain my fr "up and beyond." Not so worried 'bout square waves except to show that an amp is stable and well designed. Heck, every Marantz, under Saul anyway--including his tube stuff--had very respectable if not state of the art square wave performance.

I'll keep my fr beyond 15K, thank you. You can keep your teats on a bull hog.

 

[BlackbeardBen]

Quote:

Tha's where I'll hold to my own experience, and understanding of music. I'd rather retain my fr "up and beyond." Not so worried 'bout square waves except to show that an amp is stable and well designed. Heck, every Marantz, under Saul anyway--including his tube stuff--had very respectable if not state of the art square wave performance.

I'll keep my fr beyond 15K, thank you. You can keep your teats on a bull hog.

And that's why 44.1 kHz was chosen as the sample rate of CD audio.
 

 

[Prog Rock Man]

Quote:

Sure. The square wave won't be perfect. However, digital receiving ICs are designed to correct for this. 
 
See the image: It's of a digital signal, and it's not even close to an exact square wave. But the receiving chip only cares about that hexagon to tell a one or zero. Excursion into the hexagon, it's a zero. Signal goes around the hexagon, it's a one. The microprocessors aren't even designed for a perfect square wave.
 

 
 
 
 

To clarify, I had to copy and paste a quote made by Albedo, not myself.
 
Anyway here are three tests of HDMI cables that show they either work or don't work. They do not and cannot affect sound or picture quality
 
http://www.eurogamer.net/articles/digitalfoundry-vs-hdmi
 
http://www.audioholics.com/education/cables/long-hdmi-cable-bench-tests/hdmi-cable-testing-results
 
http://news.cnet.com/8301-17938_105-20056502-1/why-all-hdmi-cables-are-the-same/?tag=nl.e702
 
and a blind test as well that had an even split of preferences between cheap and expensive
 
http://www.expertreviews.co.uk/home-entertainment/1282699/hdmi-investigated-are-expensive-cables-a-scam