Audioquest v. Monprice Comparison

[Irvrobinson]

There is a bit of a incongruity between how digital theory is taught and how analog theory is taught. In many (most, probably) cases, the student of digital processing is never taught any analog theory ... they are taught that a one is a one and a zero is a zero, and let's move on. The better way to do it would be to teach some analog theory to go with it, but I've spoken to many a digital processing professional who could not cite Ohm's Law. That is a serious error.

I've never heard of a BSEE university program that doesn't include analog electrical engineering concepts. I don't know what you're talking about, and I'm convinced you don't know what you're talking about.

 

[Irvrobinson]

** For example, a Redbook compliant CD has the music data written to multiple areas of the disk multiple times. When an error is discovered (can't read a scratch in the disk, for example) the CD player can read the data somewhere else and get the correct data out to the rest of the system. This takes time, but there is enough time to do it if the other area isn't damaged as well.

That's not correct. My guess, Dr. Wikipedia, is that you read the following about the Redbook CD error correction strategy and misinterpreted it:

This stream of audio frames, as a whole, is then subjected to CIRC encoding, which segments and rearranges the data and expands it with parity bits in a way that allows occasional read errors to be detected and corrected. CIRC encoding also interleaves the audio frames throughout the disc over several consecutive frames so that the information will be more resistant to burst errors. Therefore, a physical frame on the disc will actually contain information from multiple logical audio frames. This process adds 64 bits of error correction data to each frame. After this, 8 bits of subcode or subchannel data are added to each of these encoded frames, which is used for control and addressing when playing the CD.

 

[highfigh]

Not really the case with HDMI. It's a one-way protocol (there is some two-way communication with the right gear at both ends, but it's not possible to correct the AV transmission that way; it just sets things like resolution). So it really is a matter of either the cable does the job or the picture or audio starts to show signs of stress, or fails completely.

Longer cables have a harder time keeping the error rate within the requirements. There are workarounds to that but they mean added costs and perhaps data limits (so, maybe OK with 1080p but not 4K, for example). Those are mostly used by installers who plan the entire system / room.

1's and 0's are sent via analog discreet voltage levels. It's really important to remember that in all cases these are analog devices working with voltages that represent digital data. The CPU in the computer you are using to read this is an all-analog device. There are no real ones and zeros anywhere in a digital system; they are merely represented by analog signals.* And so on.

There is a bit of a incogruity between how digital theory is taught and how analog theory is taught. In many (most, probably) cases, the student of digital processing is never taught any analog theory ... they are taught that a one is a one and a zero is a zero, and let's move on. The better way to do it would be to teach some analog theory to go with it, but I've spoken to many a digital processing professional who could not cite Ohm's Law. That is a serious error.

So (and this is just illustrative; the actual voltages differ depending on the digital protocol) it's not a matter of 0V = a digital one and 5V = a digital zero, it's more like 2V is the one and 4V is the zero, with a "zero crossing point" of 3V.

If there is sufficient interference or some other issue, it's usually about that zero crossing point. At which voltage above 3V does the interface decide it's dealing with a digital zero? What if the voltage is slow or weak, and and that one never changes to a zero because it's not clearly well above the zero crossing point? As the resolution of your system goes up, the only way to force more digital ones and zeros through the cable is by increasing the frequency of the changes. That requires ever more capable high frequency analog devices and cable transmission capabilities.

That kind of thing is where the problems live.

I've tried to make it easy to understand, you don't need to be an engineer (electrical or digital) to grasp the concepts. The actual workings of all this is quite complex but the principles are simple.

It might also be helpful to note that these protocols are always making mistakes. Most digital systems have robust error correction routines. These generally take care of the error rate and do an excellent job of it.** So the place where errors can be generated is in that analog cable interface, because they cannot always be corrected ... they are assumed to be correct, rightly or wrongly.

* For the sake of correctness, digital ones and zeros do exist, sort of, but in storage media, like the pits and lands of an optical disk. You can use almost any media for storage of digital files ... Vinyl LP records can store digital data, for example, and were once used as such (briefly). Some people might be more familiar with analog tape (cassettes) used in early home computers. Same thing. Note that they are all analog as well, but the data can be very discretely represented, which is a kind of one and zero world.

Once that storage media needs to be used to actually get that data off the disk and to a system, it's all analog electrical signals (versus analog storage in a non-electronic form) from that point.

** For example, a Redbook compliant CD has the music data written to multiple areas of the disk multiple times. When an error is discovered (can't read a scratch in the disk, for example) the CD player can read the data somewhere else and get the correct data out to the rest of the system. This takes time, but there is enough time to do it if the other area isn't damaged as well.

Now, that's at 44.1KHz. What if your data is now 88.2KHz (a Hi-Rez audio file). Now the time to make that correction is halved. That is an example where higher resolutions (like 4K) can introduce problems, or to put another way, where a cable that was fine with 1080p can fail with 4K.

What the what?

If HDMI is a one way protocol, explain EDID, CEC, ARC, etc?

You seem to be saying that a CD has data tossed around on the surface as if it were on a magnetic disc, which uses a pickup moving all over to retrieve the bits, wherever they may be. That's not even close- the disc rotates, the laser pickup moves from the center to the outside as the track progresses and the parity bits used for error correction are either in the same area as the data used for the audio (or video, as the case may be) or stored in the error correction circuit but the error correction uses random bits because a few odd ones don't really matter as long as the gap isn't sufficient to cause muting or some other disturbance in the sound or picture. When Sony introduced the CDP-101, we used a wedge of electrical tape on CDs to show that missing some of the data would result in uninterrupted music- this wedge was cut to a point at the inside and was at least 1/2 wide at the outside of the disc. It played fine and if the pickup had to dash wildly around the disc, it wouldn't be on a rail system to keep it aligned.

Analog computers and data on CDs, eh? Bit=binary digit. If the lands and grooves were analog, there would be no reason to have a Digital to Analog Converter, right?

HDMI uses a very small voltage "window" to make the handshake- the upper is about 5V and the lower is about 4.7V- THAT's the reason it often fails- the voltage drop in the cable or connection makes one of them drop the connection.

 

[Johnny2Bad]

With all due respect, read about the Redbook Format specification.

A simplified summary can be found here:
http://www.mediatechnics.com/redbook.htm

I'd link to the actual Redbook spec, except it costs a bunch of money and isn't available online. It isn't that easy to digest anyway, summaries are better for most people.

But suffice to say that the engineers at Philips/SONY were quite clever in creating the error correction functions. Each frame contains some of the data found in other frames, so that without resorting to random access a lost frame can be recovered given enough time to read multiple frames, which for most errors it has. Similar processes are used in other optical disc formats.

CDs are stamped into aluminum foil by a press in much the same way vinyl records are, then coated with polycarbonate. How is that not an analog storage media?

You need an DAC because the data stored in the analog media ... the physical CD ... is in a digital format. In order to use that data as sound waves, it needs to be (re-)translated to an analog format, because sound is in the analog domain. Thus a Digital to Analog Converter (DAC).

EDID, etc cannot correct HDMI data, they perform other functions, such as communicating resolution and manufacturer identification of the connected devices. We refer to these as two-way, but maybe it would be easier to think of them as two separate data streams about two different things. The reality is that EDID stream from display to source doesn't have the bandwidth to perform error correction.

 

[Irvrobinson]

With all due respect, read the Redbook Format documents.

A simplified summary can be found here:
http://www.mediatechnics.com/redbook.htm

CDs are stamped into aluminum foil by a press in much the same way vinyl records are, then coated with polycarbonate. How is that not an analog storage media?

You need an ADC because the data stored in the analog media is in a digital format.

Who are you responding to, and what is your point?

 

[RichB]

So long as we are going off topic, I have always wondered why DB PowerAmp ripper does not always detect the same number for CRC matches against the online database.

I always rip immediately and they are not dirty.
It does seem like the data integrity is not prefect.

I don't think this results in audibility issues, but it is not data grade, by that I mean file or Ethernet transmission.

- Rich

 

[mtrycrafts]

... but I've spoken to many a digital processing professional who could not cite Ohm's Law. That is a serious error.
...

You mean E=IxR? Seriously?

 

[Beave]

Who/what is a "digital processing professional?" I've never heard of that as a job title or even as a description of a job title.

 

[Beave]

You need an DAC because the data stored in the analog media ... the physical CD ... is in a digital format. In order to use that data as sound waves, it needs to be (re-)translated to an analog format, because sound is in the analog domain. Thus a Digital to Analog Converter (DAC).

??

 

[Johnny2Bad]

??

The CD is an analog storage media. Analog simply means "in the world we live in" ... a physical thing you can hold in your hand.*

It's the same with analog electronics ... they do their work in the analog domain (again, the world we live in) by exciting electrons. Signals pass from one point to another in an analog domain, regardless of whether they represent digital or analog information.

The information on a CD is digital data ... the musical performance stored there can only be recreated via a digital to analog conversion, a form of translation, because it has been previously translated into that format via an analog to digital conversion, and must be recovered for it to be useful to us re-playing music.

You might think of digital data as having been translated to a language we can't speak ... someone would have to translate it back to one we do speak for us to make sense of it.

You have to be able to conceptualize the distinction between the media itself and the information stored on the media. They are not the same thing.

Going back somewhat on topic, you have to be able to conceptualize the difference between the digital data transmitted by a cable and the vibrating electrons doing the transmission.

Who/what is a "digital processing professional?" I've never heard of that as a job title or even as a description of a job title.

Graduates of Comp Science who have to interface digital and analog devices. As contrasted to graduates of Electrical Engineering who have to interface digital and analog devices. Currently, the EE gets formal training in Comp Science, but not the other way around.

* If you were small enough, and the Laws of Physics allowed you to exist at that size, you could hold one of the electrons vibrating in a cable in your hand. That electron could be vibrating to convey digital information or analog information, but the electron itself exists in an analog domain. Think 'Alice In Wonderland'.

** Which would be " ... word(s) that is used before someone's name, stating their social rank, qualifications, position in an organization, sex, etc. ..."

 

[Beave]

 

[highfigh]

With all due respect, read about the Redbook Format specification.

A simplified summary can be found here:
http://www.mediatechnics.com/redbook.htm

I'd link to the actual Redbook spec, except it costs a bunch of money and isn't available online. It isn't that easy to digest anyway, summaries are better for most people.

But suffice to say that the engineers at Philips/SONY were quite clever in creating the error correction functions. Each frame contains some of the data found in other frames, so that without resorting to random access a lost frame can be recovered given enough time to read multiple frames, which for most errors it has. Similar processes are used in other optical disc formats.

CDs are stamped into aluminum foil by a press in much the same way vinyl records are, then coated with polycarbonate. How is that not an analog storage media?

You need an DAC because the data stored in the analog media ... the physical CD ... is in a digital format. In order to use that data as sound waves, it needs to be (re-)translated to an analog format, because sound is in the analog domain. Thus a Digital to Analog Converter (DAC).

EDID, etc cannot correct HDMI data, they perform other functions, such as communicating resolution and manufacturer identification of the connected devices. We refer to these as two-way, but maybe it would be easier to think of them as two separate data streams about two different things. The reality is that EDID stream from display to source doesn't have the bandwidth to perform error correction.

I never wrote that EDID is for error correction- it's part of the 'handshake'- if it doesn't meet the requirements, we're not going to see video or hear the audio, or both. It's used to identify the device characteristics, so one piece will know how to send the signal (resolution, aspect ratio, etc).

Analog vs digital has NOTHING to do with the disc being stamped into aluminum! NOTHING! You're trying to make one definition of 'analog' apply to the physical disc when everyone else is using 'analog' or 'digital' when referring to the signal. You might want to use 'tangible' when referring to the disc.

An analog signal is continuously variable, with no breaks in, for lack of a better term, the waveform. Can it have sections where no variation occurs? Of course- that's part of being variable. Maybe the stream of information is better. Digital SIGNAL is discrete pieces of information that are assembled in groups of bits that have some meaning (again, binary digits, which are either ones or zeros, on or off), called 'words', in a way that they can be read by the converter and re-made into an analog signal, as it impacted the mics or entered the mixer from the electronic instruments. Morse code is groups of light flashes, dashes and dots, short and long 'pings' of sound, that have meaning. A continuous string of these dots and dashes, etc can be assembled but without pauses between the words, it takes longer to decipher the message. Jumble them together and it's jibberish unless some kind of pre-determined key is used in the event that the order of the words is changed, possibly to prevent others using the information. CDs use 16 bit encoding of signal that was sampled 44.1KHz (actually, 44,056) times per second. DATA from each frame is distributed throughout the disc, but the disc in Orange Book is limited to 99 separate tracks and that precludes randomly placed data.

PCM recording on tape uses/used 14 OR 16 bit- if 16 bit was used, S/N was better but there was no data for error correction and if 14 bit was used, two bits were available, called 'parity bits'. If you look at the video from a PCM recorder, you would see columns of lines that vary with the signal content, separated by a thin gap of noise and at 14 bit- two columns are seen at the right side, neither with the same density of activity, mostly dark. These are used for error correction and it's not a fixed frame- it, too, changes with the signal.

If you break up the data on a CD in the middle of each word and jumble it all together, will it come out as music? No, it won't. Why? Because it wasn't intended to be placed on the disc, randomly. That's how data is stored on a hard drive, but that was too expensive to use at when CDs were developed and scanning from the center to the outside of a disc (linear scanning) was an easier proposition. However, that method doesn't make it 'analog'. The first Sony CD player retailed for $900- a hard drive at the time was far more more expensive.

You, yourself, wrote that the data is stored in digital form and must be converted to analog- again, it's not the medium that makes a CD digital or an LP analog- being read with light makes digital storage possible but a stylus in a groove can't track high speed data as anything close to square waves, on/off pulses. At a slow speed, with some way to prevent the stylus being damaged, it's definitely possible to store a digital signal on a disc that needs physical contact to be read at a slow speed.

https://www.philips.com/a-w/research/technologies/cd/technology.html

Here's a link to info about the Reed-Solomon Cross-interleaved Coding-

https://www.usna.edu/Users/math/wdj/_files/documents/reed-sol.htm

 

[highfigh]

You mean E=IxR? Seriously?

I like P=IE. Just sayin'.

 

[highfigh]

Graduates of Comp Science who have to interface digital and analog devices. As contrasted to graduates of Electrical Engineering who have to interface digital and analog devices. Currently, the EE gets formal training in Comp Science, but not the other way around.

Every Physics class I took covered Electricity and in that section, we covered Ohm's Law. What engineering school DOESN'T require previous or concurrent Physics study in a curriculum?

 

[Irvrobinson]

Graduates of Comp Science who have to interface digital and analog devices. As contrasted to graduates of Electrical Engineering who have to interface digital and analog devices. Currently, the EE gets formal training in Comp Science, but not the other way around.

What compels you to make assertions about topics you know nothing about? No wonder you use an obscure user name. It would be too embarrassing for you to use your real name.

* If you were small enough, and the Laws of Physics allowed you to exist at that size, you could hold one of the electrons vibrating in a cable in your hand. That electron could be vibrating to convey digital information or analog information, but the electron itself exists in an analog domain. Think 'Alice In Wonderland'.

I think I get it now. You're stoned when you're posting this nonsense.

If your hand was small enough to hold an individual electron, what would your hand be made of? And then there's the relatively minor problem by comparison of the Heisenberg Uncertainty Principle, which says the exact position and momentum of an electron can't be simultaneous determined.

 

[Irvrobinson]

CDs use 16 bit encoding of signal that was sampled 44.1KHz (actually, 44,056) times per second.

Is that correct? Everything I've ever read says the CD sampling rate is 44,100 samples per second. Do you have a reference for this?

 

[highfigh]

Is that correct? Everything I've ever read says the CD sampling rate is 44,100 samples per second. Do you have a reference for this?

44.1K is used, but since some people need to go their own route, 44.056 was used for a while as 'the number'- one uses the actual number, one rounds up.

According to this link, it (44.056KHz) comes from the use of videotape for PCM recording- 60f/sec, 245 lines, 3 samples/line (60x245x3=44,100). The difference comes from the fact that the NTSC rate is actually 29.97 frames/second.

https://cardinalpeak.com/blog/why-do-cds-use-a-sampling-rate-of-44-1-khz/

 

[Irvrobinson]

44.1K is used, but since some people need to go their own route, 44.056 was used for a while as 'the number'- one uses the actual number, one rounds up.

According to this link, it (44.056KHz) comes from the use of videotape for PCM recording- 60f/sec, 245 lines, 3 samples/line (60x245x3=44,100). The difference comes from the fact that the NTSC rate is actually 29.97 frames/second.

https://cardinalpeak.com/blog/why-do-cds-use-a-sampling-rate-of-44-1-khz/

I double-checked just to be sure, and CDs actually use 44,100 samples per second, so your statement above about it being actually 44,056 is incorrect. I have no doubt there were other sampling rates commonly used in early digital recording, especially prior to the Redbook being published. For example, early Telarc digital recordings used the Soundstream digital recorder which used a 50KHz sampling rate. Given Johnny's habit of being entirely inaccurate, I think it's important the rest of us are accurate.

 

[mtrycrafts]

I like P=IE. Just sayin'.

I like that as well for power calc.

To not know such basics is unforgivable.

 

[Mikado463]

I like P=IE. Just sayin'.

Apple PIE is my favorite !