Does analog move more air. . . ?

[mshilarious]

Those pictures are working OK. This board is shot though, we need to find somewhere new to argue!


So, anyway, yes that is exactly what you'd expect from a 44.1 or 48kHz digital system. At double speed you'll get an extra harmonic, see my Excel sheet for appearance. I call it a "molar" wave.

It's kind of funny that you are now saying that digital turns everything to sine waves when at the beginning people were saying digital turns everything to squares. It can't be both! The reality is digital applies a hard bandwidth limit to a signal; together with a tiny wee bit of aliasing and possibly some measureable jitter that's about all it does.

The problem with the wave view on that scope is you can't see distortion until it exceeds 1% or so. Now, obviously at 10kHz if there was distortion it would have to be 20kHz, which is asymmetrical and thus less likely than 3rd, which of course the system shouldn't pass. Also it's really hard to see aliasing or jitter because those tend to be <0.01%, usually much less. In an old DAT machine though I dunno, those might not test as well.

This is where an FFT is quite useful.

 

[Victory Pete]

Those pictures are working OK. This board is shot though, we need to find somewhere new to argue!


So, anyway, yes that is exactly what you'd expect from a 44.1 or 48kHz digital system. At double speed you'll get an extra harmonic, see my Excel sheet for appearance. I call it a "molar" wave.

It's kind of funny that you are now saying that digital turns everything to sine waves when at the beginning people were saying digital turns everything to squares. It can't be both! The reality is digital applies a hard bandwidth limit to a signal; together with a tiny wee bit of aliasing and possibly some measureable jitter that's about all it does.

The problem with the wave view on that scope is you can't see distortion until it exceeds 1% or so. Now, obviously at 10kHz if there was distortion it would have to be 20kHz, which is asymmetrical and thus less likely than 3rd, which of course the system shouldn't pass. Also it's really hard to see aliasing or jitter because those tend to be <0.01%, usually much less. In an old DAT machine though I dunno, those might not test as well.

This is where an FFT is quite useful.

Well, it sure is funny that a 1K square wave in produced a 1K square wave out, but a 10K square wave in produced a perfect 10K sine wave out. The analog experiment looked quite different.

VP

 

[Victory Pete]

Well there is still something wrong with the pictures, they are not going in as I wanted them too. Here are the pictures of the analog test. Square wave in does not come out square, just as I expected.

VP

 

[Victory Pete]

Here are some links to "Digital". After reading these I still have to think the upper frequencies are "Fabricated" in some manner. I cant believe a whole group of people are imagining the differences between Analog and Digital.

Digital Recording

Planet Of Tunes - How do audio analogue to digital converters work?

VP

PS Here is a quote from the 2nd link

"Sample rate is constant

Once set, sample rate does not vary during a recording, although different audio files recorded at different sample rates may be used together in a multitrack system if the software permits it. Usually, as in the case of a DAW, audio files of differing sample rates will need to conform (be converted too) a single sample rate, typically 44.1KHz, 48KHz, 96KHz or 192KHz. This sample rate is usually set in the application preferences for the recording session.
■Higher sample rates produce better quality recordings but also bigger file sizes which demand greater space on storage devices (such as hard drives), and faster processors (CPUs) to manipulate.
■Lower sample rates produce poorer quality but also smaller file sizes which demand less of storage systems, CPUs and will transfer over networks (internet) faster."

 

[Victory Pete]

I am going to try to load a picture of the square wave Digital output at 1K. It remained a square except for the superimposed sample ripple.

VP

 

[mixsit]

Well there is still something wrong with the pictures, they are not going in as I wanted them too. Here are the pictures of the analog test. Square wave in does not come out square, just as I expected.

VP

(from post 403..
That's 10k? It's pretty good isn't it? What kind of bandwidth then does it take to do that? Quite high presumably?

Odd that some of the pics fly and some don't.
If they do, could you name them too? Amd thank you guys, this turning into a pretty good thread.

 

[grimtraveller]

Since these quote were kind of ... taken out of context, allow me to 're-contextualize' them:

I didn't take any of the comments out of context. It was a long, long thread and within any thread of that length, multiple contexts will often abound.

Some people seem to have come into this thread/forum simply to 'ruffle our [regular readers/posters] feathers', so to speak. I found it to be rude. There is generally a certain level of respect for one another, as well as a certain level of respect for tape recording, that seems to have been largely absent from this thread.

I didn't catch that. I didn't feel anyone was being rude or provocative. When you move around some of the forums here at HR, you see rude, condescending and provocative. Even those that didn't come down on the side of analog, I thought, were pretty civil about it. Granted, there were moments of friction between Lt Bob and Victory Pete but Pete was a willing contributor to that. I don't think there was deliberate feather ruffling going on, but it is sometimes a feature of some people in the analog forum to have their feathers ruffled if any 'non analog praise' viewpoint comes up.
Compare this thread with the one about "Will analog multitracks be made again" in which there are deliberate swipes at tape users and deliberate feather ruffling. Two completely different vibes.

Aside from that, some of the information being presented as fact just seems outright silly.

On both sides of the tracks. In every camp, there are those that want to 'win the argument'.

Analog/digital debates are plentiful elsewhere; I still stand by the opinion that they don't serve much of a purpose in this sub-forum.

I disagree. As I pointed out earlier, there is still quite a bit of crossover between the two mediums for a number of users. I'm one of those myself. So when digital recording constantly receives little sideswipes like the ones that were in the opening post and have continued throughout, the door is opened to refutation. Preference is one thing. Trying to validate one's preference by presenting it as irrefutible, die hard fact, is another thing altogether.
For the record, I defend analog recording when, in other forums, people take a smarmy view of it.

it's just watering down the relevancy of the forum. I'm not sure anything useful from this thread has been added to the database of knowledge here.

It isn't the debate or comparisons between the two recording mediums that that have watered down the relevancy of the forum. The forum's own members have done that.

How Many Regular Members Are No Longer Here?

This forum is relatively dead compared to years past

As to the second point about what useful stuff may or may not have been added to the database of knowledge, that really depends on whether or not one wants to remain fixed in an entrenched position come what may or whether or not one can allow the other side to have a legitimate stance while keeping one's own preference.
Although there was lots of stuff I didn't understand/give a damn about/couldn't verify, equally, there was useful stuff there. I generally work on the basis that someone, somewhere will one day benefit from the debates.

You should be a journalist, my friend!

I seriously thought about it in my late teens. Then I discovered what an editor did and that turned me right off !

 

[mshilarious]

Well, it sure is funny that a 1K square wave in produced a 1K square wave out, but a 10K square wave in produced a perfect 10K sine wave out. The analog experiment looked quite different.

VP

It's not funny, it's the way it should look. Review my Excel chart again--with 10kHz, you have no harmonics at 44.1kHz so you get a 10kHz sine wave. At 1kHz you have 3, 5, 7, 9, 11, 13, 15, 17, and 19kHz, which will look like a square with overshoot and ringing.

 

[mshilarious]

(from post 403..
That's 10k? It's pretty good isn't it? What kind of bandwidth then does it take to do that? Quite high presumably?

I think that's 1kHz, I count 12 divisions for a half-wave at 0.2ms/div = 1/.0024 * 2 = 833Hz. But then I am bad at reading scopes.

HR is clearly not helping us; I've had to resort to hosting everything on my site

 

[Victory Pete]

(from post 403..
That's 10k? It's pretty good isn't it? What kind of bandwidth then does it take to do that? Quite high presumably?

Odd that some of the pics fly and some don't.
If they do, could you name them too? Amd thank you guys, this turning into a pretty good thread.

Sure it is a nice sine wave but what went in is a square wave.

VP

 

[Victory Pete]

It's not funny, it's the way it should look. Review my Excel chart again--with 10kHz, you have no harmonics at 44.1kHz so you get a 10kHz sine wave. At 1kHz you have 3, 5, 7, 9, 11, 13, 15, 17, and 19kHz, which will look like a square with overshoot and ringing.

What is funny are the 2 different outcomes for Digital and Analog.

VP

 

[Tim Gillett]

What is funny are the 2 different outcomes for Digital and Analog.

VP

Neither the analog recorder nor the digital recorder has a ghost of a chance of faithfully reproducing a 10khz square wave because such a wave contains frequencies well below and well above their design passbands.

Why not inject the frequencies both recorders were designed to handle?

Tim

 

[Victory Pete]

Neither the analog recorder nor the digital recorder has a ghost of a chance of faithfully reproducing a 10khz square wave because such a wave contains frequencies well below and well above their design passbands.

Why not inject the frequencies both recorders were designed to handle?

Tim

I only see one "frequency" on my scope, the 10KHz square wave. I am going to record a Cymbal soon. My experiment points out how the Digital and Analog have very different outcomes. I am interested in what happens to the "fabricated" high frequency wave forms produced by A/D & D/A converters.

VP

PS An over driven guitar signal comes close to being a square wave, I can use that as a test signal also.

 

[mshilarious]

PS An over driven guitar signal comes close to being a square wave, I can use that as a test signal also.

Not that close unless maybe you take a Metal Zone distortion direct in, and besides the highest fundamental tone on a guitar is about 1.3kHz so even it you got a square off of that you've already done that test.

Most guitarists don't like much above 6kHz in their signals which is why electric guitar amps don't have tweeters. You can't make a high frequency square wave with no content above 8kHz. Your 24th fret E string will only have 1.3kHz, 3.9kHz, some 6.5kHz, and maybe very little 9.1kHz. Smooth-topped triangle wave at best.

What is funny are the 2 different outcomes for Digital and Analog.

Rather I would say what is sad is your poor understanding of signal and filter theory.

 

[Tim Gillett]

I only see one "frequency" on my scope, the 10KHz square wave. VP

Yes it has a 10khz "frequency" because that is the number of complete cycles per second as they cross the zero point. But look at the various parts within that square wave. Remember, the horizontal axis represents time and the vertical, amplitude.

1. The horizontal part of the wave is essentially DC and so well outside what the recorders are designed to reproduce.

2. The vertical part of the wave is (theoretically) infinite in frequency and so it too is way outside of the abilities of the recorders to reproduce.

3. The "corners" of the wave are where the frequency radically changes from DC to "infinite" and then from "infinite" back to DC. This is the part of the wave where every frequency in between DC and infinite occurs, even if only for a tiny instant.

A true square wave is therefore an impossibility. It's just an idea.

An overdriven guitar signal comes no closer to being a true square wave than anything else.

When we talk about audio we normally mean nothing below 20hz and nothing above 20khz, only what's in between. That replicates roughly the bandwidth of healthy human hearing. It was a commonly accepted rule of thumb for designers long before digital recorders came along.

I've been on this forum on and off for some years. Every now and then someone asserts that analog magnetic tape recording methods have better bandwidth capabilities than digital methods. It's the exact opposite. Digital creams analog magnetic tape methods for bandwidth. You just have to use a suitable sample rate.

That is to say, even though no recorder could ever properly reproduce a true square wave, and no signal generator could ever inject one in the first place, digital methods would come far far closer to reproducing it that analog magnetic tape methods ever could -if that's what you're chasing. I'm not chasing it. I'm one of those people who's quite happy with normally accepted audio bandwidth, whether from an analog tape recorder or a digital recorder.



Tim

 

[Victory Pete]

Not that close unless maybe you take a Metal Zone distortion direct in, and besides the highest fundamental tone on a guitar is about 1.3kHz so even it you got a square off of that you've already done that test.

Most guitarists don't like much above 6kHz in their signals which is why electric guitar amps don't have tweeters. You can't make a high frequency square wave with no content above 8kHz. Your 24th fret E string will only have 1.3kHz, 3.9kHz, some 6.5kHz, and maybe very little 9.1kHz. Smooth-topped triangle wave at best.



Rather I would say what is sad is your poor understanding of signal and filter theory.

Nothing is sad here, except the way I feel when I listen to Digital.

VP

 

[mshilarious]

OK then, do your guitar test, see if your hypothesis or my hypothesis is correct. If I am right you owe me a beer!

 

[Victory Pete]

Yes it has a 10khz "frequency" because that is the number of complete cycles per second as they cross the zero point. But look at the various parts within that square wave. Remember, the horizontal axis represents time and the vertical, amplitude.

1. The horizontal part of the wave is essentially DC and so well outside what the recorders are designed to reproduce.

2. The vertical part of the wave is (theoretically) infinite in frequency and so it too is way outside of the abilities of the recorders to reproduce.

3. The "corners" of the wave are where the frequency radically changes from DC to "infinite" and then from "infinite" back to DC. This is the part of the wave where every frequency in between DC and infinite occurs, even if only for a tiny instant.

A true square wave is therefore an impossibility. It's just an idea.

An overdriven guitar signal comes no closer to being a true square wave than anything else.

When we talk about audio we normally mean nothing below 20hz and nothing above 20khz, only what's in between. That replicates roughly the bandwidth of healthy human hearing. It was a commonly accepted rule of thumb for designers long before digital recorders came along.

I've been on this forum on and off for some years. Every now and then someone asserts that analog magnetic tape recording methods have better bandwidth capabilities than digital methods. It's the exact opposite. Digital creams analog magnetic tape methods for bandwidth. You just have to use a suitable sample rate.

That is to say, even though no recorder could ever properly reproduce a true square wave, and no signal generator could ever inject one in the first place, digital methods would come far far closer to reproducing it that analog magnetic tape methods ever could.

Tim

Okay, but what about the fact that A/D & D/A converters sample the Analog waveform at finite intervals, and at high frequencies there is a compromise?

What about this quote:

"Once set, sample rate does not vary during a recording, although different audio files recorded at different sample rates may be used together in a multitrack system if the software permits it. Usually, as in the case of a DAW, audio files of differing sample rates will need to conform (be converted too) a single sample rate, typically 44.1KHz, 48KHz, 96KHz or 192KHz. This sample rate is usually set in the application preferences for the recording session.
■Higher sample rates produce better quality recordings but also bigger file sizes which demand greater space on storage devices (such as hard drives), and faster processors (CPUs) to manipulate.
■Lower sample rates produce poorer quality but also smaller file sizes which demand less of storage systems, CPUs and will transfer over networks (internet) faster."

VP

 

[mshilarious]

Higher sample rates produce more bandwidth, not necessarily better quality recordings. The only issue is the required bandwidth. If you need more, you use a higher sample rate.

The rest of that quote is dribble about mixing different sample rates within one project, which some software allows and other software does not. Generally it's not a great idea as multiple SRC would have to be done on the fly and good SRC is a DSP-expensive routine. So you'd want to do the SRC as an offline process and then proceed with your project at a single rate.

It would be somewhat akin to syncing a bunch of tape machines all running at different speeds. Could you do that? I suppose. Would you really want to if you had the choice to run them all at the same speed instead? Probably not.

 

[Victory Pete]

Higher sample rates produce more bandwidth, not necessarily better quality recordings. The only issue is the required bandwidth. If you need more, you use a higher sample rate.

The rest of that quote is dribble about mixing different sample rates within one project, which some software allows and other software does not. Generally it's not a great idea as multiple SRC would have to be done on the fly and good SRC is a DSP-expensive routine. So you'd want to do the SRC as an offline process and then proceed with your project at a single rate.

It would be somewhat akin to syncing a bunch of tape machines all running at different speeds. Could you do that? I suppose. Would you really want to if you had the choice to run them all at the same speed instead? Probably not.

Higher sample rates will sample the waveform more times per cycle, resulting in a more accurate "rendition" of the recorded signal. Of course that would make a more accurate output signal, that is a fact. That quote was from a credible website and has a lot of merit despite your "Naysaying"

Once again:

"■Higher sample rates produce better quality recordings but also bigger file sizes which demand greater space on storage devices (such as hard drives), and faster processors (CPUs) to manipulate.
■Lower sample rates produce poorer quality but also smaller file sizes which demand less of storage systems, CPUs and will transfer over networks (internet) faster."


VP