Digital signals

[Vaughan Odendaa]

Hello there,

I need some help in understanding digital. A friend of mine told me yesterday that if he copies a dvd it won't be the same quality as the original.

I know this to be wrong but I don't understand fully as to why he is wrong. This is something that I need to know.

Also, my manager told me that digital is inferior to analog and that it cannot create a perfect duplicate of the original recording. He is one of the analog supporters. I need some help in understanding these two statements.

In particular, if one can explain why a digital signal doesn't degrade after copying, like a cd, or a dvd on to another dvd. That would be helpful.

I hope someone can help explain these things to me.

Thank you.

--Sincerely,

 

[TheLaw]

Digital and analog are quite different and there are a lot of factors to consider when converting analog to digital. With sound, the human ear is analog so even digital music gets converted back into an analog signal before getting to your speakers.

The sampling rate is the single most important factor when converting analog data to digital data. Sampling is the measuring of a wave's amplitude at regular intervals to accurately reproduce the signal. For example... an mp3 file playing at 192kb/s will sound better than an mp3 file playing at 128kb/s. This is because the file playing at 192kb/s has been reproduced at a higher sampling rate (a more accurate reproduction of original analog wave) than the file playing at 128kb/s. For digital audio 192kb/s is considered CD quality sound.

So to talk about degrading... Copying digital to digital, there should be no degradation, and the same with analog to analog. Where you typically see degradation in signal is copying from analog to digital and vice versa. One of the problems with copying DVD's is the copyright protection on them. So you download special programs to get rid of that and then compress the file to fit on one disc, and this all takes a toll on the quality of the copied disc.

Hope this helps you a little.

 

[BMXTRIX]

Getting rid of the copyright protection on a disc does not affect the quality at all. When backing up your discs from one DVD to another you typically will back up onto 4.7GB single layer DVDs. Commercial DVDs are almost always dual layer 9GB discs. So, to get 9GB down to 4.7GB, you have to throw some stuff away.

The best way to do this is to get rid of the special features, menus, foreign languages, etc. Just copy the movie, and ONLY the movie. This is what I do when I make my backups. Unfortunately, even after that, you may find that the movie itself is greater than 4.7GB, and this means that you must still get rid of data.

How?

Compression. The digital video is re-encoded in MPEG2 format with greater compression so that it takes up less space. This has the effect of brining down the quality, SLIGHTLY, on the video. Typically the quality drop is not huge unless the movie was huge and requires a lot of compression. So, a Superbit DVD (full 9GB), backed up onto a 4.7GB DVD, will suffer the most. While, most movies, especially animated films at 90 minutes or so, typically have a file size in the 3-4GB range and require zero additional compression so end up pixel-for-pixel identical to the DVD original.

DIGITAL MUSIC VS. ANALOG: Read this, it explains it (with graphics) much better than about anything else. The key concept being that analog CAN be better, but there is often a good bit of noise on the medium from analog which digital typically doesn't have. A great record player and a excellent record, may surpass CD. But, that same CD in a $25.00 walkman that you can run around with will still sound really good and is kind of tough to do with a record.

http://electronics.howstuffworks.com/question487.htm

 

[MDS]

Also, my manager told me that digital is inferior to analog and that it cannot create a perfect duplicate of the original recording. He is one of the analog supporters. I need some help in understanding these two statements.

It is loosely true that digital sampling cannot create a perfect duplicate of a recording because to do so would require an infinite number of samples; remember an analog waveform is continuous and thus has an infinite number of values between any two points. Digital sampling creates 'snapshots' of the values at many points in time.

Think of it this way: how many points are there between the values 1 and 2? There are an infinite number of points. If we take 5 samples, we would have values of 1, 1.2, 1.4, 1.6, 1.8, and 2. If we take 10 samples we would have 1, 1.1, 1.2, 1.3, ....1.9, 2. More samples captures more of the values between two points.

Now with digital sampling of analog waveforms, as long as the number of samples is sufficiently high, you get a very close representation of the original. On playback, a filter reconstructs the analog waveform - essentially by 'connecting the dots' between the samples.

Once you have a digital representation of a waveform, copying it does not change the data. If you rip the tracks from a CD, you are simply copying the data and it will not change [although there can be some complications in accurately reading the data from a CD, for the most part modern drives are accurate and you will get exactly what is on the disc]. Same for DVD, unless you re-encode the data (the case BMXTRIX is describing).

Copying an analog waveform will NOT be identical. Think about recording to tape a person's voice using a microphone. No mic is perfect. You are recording what the mic 'hears'. What your ears hear vs what the mic hears are not necessarily identical. Add to that the complication that the tape itself has a limited capacity to capture the sound - if the level is too high, the tape will saturate and the recording will distort.

A stupid analogy would be this: If I give you a piece of paper with a phone number on it, that is digital data. If you copy the number to another piece of paper and give it to a friend, no data is lost and the number is fully preserved. If I speak the number to you, that is analog. If I speak the number to you, *you* have to make sure you heard it correctly when you write it down. If you mess it up when you write it down before giving to a friend, the data could be changed.

 

[Vaughan Odendaa]

Thank you guys for the explanations. I really appreciate this information. Are there any explanations (theory-wise) that would explain why digital signals do not degrade if copied from a digital source ?

Thanks again.

--Sincerely,

 

[jonnythan]

Because digital data looks like this:

01010100100101111010101010101010101001111010111110010100111001011

The data is a stream of 1's and 0's. That exact stream can be copied exactly an infinite number of times.

 

[avnetguy]

01010100100101111010101010101010101001111010111110010100111001011

Here's an example of a digital copy, the data above is exactly the same as the original post.

Steve

 

[jonnythan]

Because digital data looks like this:

01010100100101111010101010101010101001111010111110010100111001011

The data is a stream of 1's and 0's. That exact stream can be copied exactly an infinite number of times.

And here's another.

This is how digital data is stored and transmitted. 1's and 0's. Copies transcribe these bits perfectly.

It doesn't matter whether the data is the latest Diana Krall CD, your letter to Grandma, a downloaded game, or whatever. It's all the same.

 

[MACCA350]

Originally Posted by jonnythan
01010100100101111010101010101010101001111010111110 010100111001011

And the third generation digital copy is still the same as the original etc. etc. etc

cheers

 

[mtrycrafts]

Also, my manager told me that digital is inferior to analog and that it cannot create a perfect duplicate of the original recording. He is one of the analog supporters. Thank you.

--Sincerely,

Actually a digital recording can duplicate the electric signal that is being converted to a digital signal perfectly. Nyquist theorem.

http://www.google.com/search?sourceid=navclient&ie=UTF-8&rls=GGLG,GGLG:2005-38,GGLG:en&q=nyquist+theorem


Don't forget, the hearing capability is very limited and cannot hear everything that is being recorded. One reason perceptual coding works so well.

Acoustics and digital is not a simple issue to understand so briefly. Perhaps you can do some google search further on this.

http://electronics.howstuffworks.com/framed.htm?parent=cd.htm&url=http://www.ee.washington.edu/conselec/CE/kuhn/cdaudio2/95x7.htm

http://www.deluxemedia.com/images/cdintroduction.pdf

If he is into vinyl:

http://groups.google.com/group/rec.audio.high-end/browse_frm/thread/b5b3d74ebf5195d9/653de88007541118?dq=&hl=en&lr=&ie=UTF-8&prev=/groups?hl=en&group=rec.audio.high-end#653de88007541118

http://groups.google.com/group/rec.audio.high-end/browse_frm/thread/b5b3d74ebf5195d9/653de88007541118?dq=&hl=en&lr=&ie=UTF-8&prev=/groups?hl=en&group=rec.audio.high-end#653de88007541118

 

[mtrycrafts]

The sampling rate is the single most important factor when converting analog data to digital data. Sampling is the measuring of a wave's amplitude at regular intervals to accurately reproduce the signal. For example... an mp3 file playing at 192kb/s will sound better than an mp3 file playing at 128kb/s. This is because the file playing at 192kb/s has been reproduced at a higher sampling rate (a more accurate reproduction of original analog wave) than the file playing at 128kb/s. For digital audio 192kb/s is considered CD quality sound.
.

Are we sampling here or using perceptual coding, throwing info out deemed not necessary?

 

[WmAx]

DIGITAL MUSIC VS. ANALOG: Read this, it explains it (with graphics) much better than about anything else. The key concept being that analog CAN be better, but there is often a good bit of noise on the medium from analog which digital typically doesn't have. A great record player and a excellent record, may surpass CD. But, that same CD in a $25.00 walkman that you can run around with will still sound really good and is kind of tough to do with a record.

http://electronics.howstuffworks.com/question487.htm

The howstuffworks.com link is erroneous and should be removed from the website. That graph is not represenative of the analog signal produced from digital audio. The only time such a signal would exist is if you did not have an anti-alias filter in the ADC stage, which prevents exactly this from occuring in the signal output. The article is full of half-truths and unsupported claims. I expected better from howstuffworks.com.

If you want a factual article, you can refer to this article from Lavry Engineering:

http://www.lavryengineering.com/documents/Sampling_Theory.pdf

-Chris

 

[WmAx]

Also, my manager told me that digital is inferior to analog and that it cannot create a perfect duplicate of the original recording. He is one of the analog supporters. I need some help in understanding these two statements.
,

Ask your manager to post here. I will address his concerns with digital audio. Please refer to the Lavry Engineering article that I posted previous to this reply.

-Chris

 

[WmAx]

It is loose
Think of it this way: how many points are there between the values 1 and 2? There are an infinite number of points. If we take 5 samples, we would have values of 1, 1.2, 1.4, 1.6, 1.8, and 2. If we take 10 samples we would have 1, 1.1, 1.2, 1.3, ....1.9, 2. More samples captures more of the values between two points.

I feel this may confuse some readers. It should be pointed out that the only relevance the number of samples have considering PCM audio, is the bandwidth. Therefor, 44.1kHz is perfectly sufficient for human perception, according to known credited perceptual research, since this extends the bandwidth to approximately 22kHz.

-Chris

 

[krabapple]

Digital and analog are quite different and there are a lot of factors to consider when converting analog to digital. With sound, the human ear is analog so even digital music gets converted back into an analog signal before getting to your speakers.

The sampling rate is the single most important factor when converting analog data to digital data.

Actually, the bit depth is vitally important too.

Sampling is the measuring of a wave's amplitude at regular intervals to accurately reproduce the signal. For example... an mp3 file playing at 192kb/s will sound better than an mp3 file playing at 128kb/s. This is because the file playing at 192kb/s has been reproduced at a higher sampling rate (a more accurate reproduction of original analog wave) than the file playing at 128kb/s. For digital audio 192kb/s is considered CD quality sound.

No, you are confusing (kilo)bits per second with samples per second. MP3 kbps does not refer to a sampling rate (at least not directly), it is a measure of data flow in a digital system. A sampling rate of 44.1 kHz, (CD standard) on the other hand, simply means that a continuous (e.g. audio) signal was digitally sampled 44,100 times each second. By the Nyquist theorem this is is sufficient to accurately digitally capture frequencies up to ~22 kHz, which is greater than the upper limit of hearing for the vast majority of people.

So to talk about degrading... Copying digital to digital, there should be no degradation, and the same with analog to analog.

Analog to analog will certainly result in degradation, quite likely audible.

Where you typically see degradation in signal is copying from analog to digital and vice versa.

No, you will much more typically see degradation in analog-to-analog, whereas D/A and A/D conversion can both be 'transparent' if properly implemented.

I second the recommendation for the Lavry article. Also highly recommended: Nika Aldrich's book 'Digital Audio Explained'. The howstuffworks article is ill-informed junk.

 

[krabapple]

Think of it this way: how many points are there between the values 1 and 2? There are an infinite number of points. If we take 5 samples, we would have values of 1, 1.2, 1.4, 1.6, 1.8, and 2. If we take 10 samples we would have 1, 1.1, 1.2, 1.3, ....1.9, 2. More samples captures more of the values between two points.

But beyond a certain point, as demonstrated by the work of Shannon and Nyquist, it is unnecessary to capture 'more values' to digitally capture and reconstruct the original bandwidth-limited signal with utmost fidelity. For a bandwidth-limited signal -- a category that includes all audible sound -- it is not true that the more values we capture, the more accurate our digital copy of an analog wave will be. We need only capture at a sampling rate of twice the highest frequency.

 

[avnetguy]

You'll always find some that like the sound of analog over digital, maybe they'll say something is missing or is has a different sound that they don't like but for the vast majority CD quality audio passes with flying colors as far as reproduction goes. It certainly works well enough for my ears.

Now having said that, its time for me to open a can of worms , or at least I still believe it to be (flame suit on). In order to "properly" represent an analog waveform, this is not just the retention of the frequency (determined by the nyquist) but also a realistic representation of the amplitude, one must have a sampling rate 10 times higher than that of the minimum frequency digitized.

For example, would a 14 kHz sine wave digitized at 44.1 kHz be truely represented by only 3 data points?

But getting back to the real world ... would we ever hear the difference due to the limitations of our own ears?

Steve

 

[mfabien]

With all of that in mind, can someone here answer my question which remains unanswered in the "HD DVD Tour" thread:

"Steely Dan "Everything Must Go"

tell me if the High Resolution in 5.1 multi channel rendition beats the DTS surround rendition (using a digital connection) by a long shot that would justify lossless DTS-HD over current DTS?

I play that particular DVD-A in my normal DVD player in DTS surround and love it."


I'm trying to determine if the HD-XA1 is a must over the HD-A1 in selecting a HD DVD player. DD TrueHD and DTS-HD would come over the 6 analog cables.
Would the difference be "Priceless" as the MasterCard ad says?

 

[MACCA350]

Anyone see in the product specs page of both the Toshiba HD-DVD players here

Dolby Digital Plus support for up to 5.1 channels, Dolby True HD support for up to 2 channels, and DTS-HD support for up to 5.1 channels of DTS core only

As others have said, I to will wait for second or even third generation players(maybe I'll just wait for Denon to release a new Universal Player that does everything )

cheers
EDIT: oops, wrong thread

 

[WmAx]

Now having said that, its time for me to open a can of worms , or at least I still believe it to be (flame suit on). In order to "properly" represent an analog waveform, this is not just the retention of the frequency (determined by the nyquist) but also a realistic representation of the amplitude, one must have a sampling rate 10 times higher than that of the minimum frequency digitized.

This is not true(10x sample frequency required). 44.1kHz, 16 bit, is sufficient to perfectly store the data for time, amplitude and frequency, up to approximately 22 kHz. For details, refer to the Lavry Engineering paper that I linked to earlier in this thread.

-Chris