• Welcome to the all-new HUG. All content has been converted from the old system, and over the next days we will re-style HUG in a more familiar way.

Getting to grips with "clipping" - an ongoing research project

mhennessy

Member
THIS THREAD RESTATES THE FINAL FIVE POSTS IN THE SOURCE THREAD AND CONTINUES HERE >>>>

[HR][/HR]
Hi,

Sorry it's taken a while. Extracting the gear from the workshop and setting it up does take a bit of time, and taking photographs does take some trial and error...

Yes, to set the scene, I'd really like a picture of your digital scope with a memory setting of peak output on sine wave as an overlay, and a live screen shot of the music below and at clipping (not too much zoomed in - perhaps 2 seconds of x axis. If you can make a PostIt pointer showing the max possible output (or point with a pen etc.) then we can make sense of the display. You should now be able to attach it to this tread, 1280 x is ideal.
OK, if I've understood correctly, this picture should be what you're after:

You must be registered for see images attach



You can see that there is a small amount of "dynamic headroom", as the average level of programme material is less than a sine wave, so the power supply rails will be slightly higher. With this amp, there isn't much in it - less than a decibel.

I'm impressed at how closely the numbers agree closely with my visual estimates from the analogue 'scope the other day. Of course, this 'scope is measuring peak-to-peak, and I recorded peak, so my earlier results would need to be doubled. I said 25 volts for "loud serious", and 30 volts for "clipping", and we see 51.6 and 66.1 respectively. Not bad...

One more thing for clarification: how different would the overall situation be if the speakers were somewhat more sensitive? The ones you used had 83dB sensitivity like LS3/5a or similar quality mini monitor. If the speakers were a more typical 85/86dB efficiency, how would that change the numbers, and the number of clicks on the amp before clipping do you think?
Now, as I'm sure you know, this is not an easy question!

If somehow we were able to find two loudspeakers that were completely identical apart from sensitivity, then we could say that we'd need less power from the amplifier to achieve the same sound levels. And this is not to be sniffed at, as 2 or 3 dB more sensitivity equates to a power requirement of 31 or 25 watts instead of 50. A reasonable saving in terms of power supply, heat sinking, etc...

But of course it's not that simple. For a start, even just measuring sensitivity is a complex business, so how sure are we that the supplied figures are accurate? And our perception of loudness is very frequency-dependant, so if the less efficient loudspeaker had a more "forward" midrange, say, compared to the more sensitive unit, they might sound very similar to us. Of course, many other effects contribute...

So with all that in mind, I decided to repeat the listening with a loudspeaker with a sensitivity of 85dB/watt. But the subjective results were inconclusive. Perhaps the more efficient model was slightly louder for a given volume setting. But perhaps the slightly smoother midrange of the more efficient model convinced me to listen louder? But trying some different tracks confuses the picture entirely - something bass-heavy like Massive Attack sounded better on the less efficient model, whereas something "dry" like Donald Fagen worked slightly better on the more efficient units. I really wouldn't want to call it. And this is no surprise. Too many variables...

P.S. Regarding (3) above: I assume that to determine the maximum power output of the amp (actually, the maximum voltage output of the amp) you play a signal into the amp, connect the load or the speakers and turn up the volume, click by click, observing the increasing vertical magnitude of the display on the 'scope connected across the output. The vertical divisions on the scope are known (so many volts per cm), and you keep turning up the volume until there is no further increase in the height of the waveform traced on the 'scope. When you reach the volume setting where no more output can be extracted from the amp, you read off the height of the displayed waveform in cms, and multiply that by the number or volts per cm displayed, and write down your stated 'absolute maximum output voltage at point of clipping' i.e. the amp is working flat-out. Correct?
That's one way to measure it, and it's what I did to get the numbers on the above image.

One of the most widely-used methods is to increase the level of a 1kHz sine wave until the measured THD+N reaches 1%, then measure the voltage (and calculate the power from that). When clipping occurs, we suddenly see harmonics of the original 1kHz tone, at 3kHz, 5kHz, 7kHz, etc., and additionally, we'll usually see some 100Hz from the power supply (most amplifiers use unregulated power supplies - and there are compelling reasons to do so - so as the output transistors saturate during clipping, they pass the unfiltered 100Hz straight out to the load). The THD+N test will catch harmonics and/or mains ripple. It works well, and resolving 1% doesn't require state-of-the-art equipment.

You must be registered for see images attach


(Note that the 'scope was in analogue mode for this. When I captured the sine wave in digital mode for the overlay in the first image, for clarity I used "averaging" to remove the "fuzziness" at the peaks).

As I had the dummy loads and test set in the room, I measured my amplifier using this approach, and measured 21.45 volts RMS (using a True-RMS DVM - easier and more accurate than a 'scope). This equates to 57.5 watts. And that was with two channels working flat-out. With a single channel driven, the voltage was 22.4V RMS, which is 62.7 watts. Slightly higher, as one would expect, but not much as it has a decently "stiff" power supply...

The "estimate from the 'scope screen' approach gave 56 watts, so it's nice and close - within experimental error.

I would also anticipate that when the music/test signal takes the amp into clipping, the displayed wave momentarily becomes a brighter, whiter, spot on the screen, making clipping point visually unmistakeable. True? If so, a photo of that would explain the issue.
Yes, very much so - on an analogue 'scope. Be careful with digital 'scopes because many of them display the image quite differently, and every "pixel" is shown at the same brightness. Some, like Tek's "Digital Phosphor" series attempt to simulate the effect of analogue CRT displays...

But having said that, it's really, really hard to photograph!

You must be registered for see images attach


I hope this helps. I'll leave the gear in the music room for a while longer, so if you need anything more, please shout :)

All the best,

Mark
 
Attachments only viewable to members

A.S.

Administrator
Staff member
Analysis of test CD (Diana Krall)

Analysis of test CD (Diana Krall)

Mark,

I have received in the post this morning a copy of Diana Krall's from this moment on purchased on line (HMV retail store didn't have it). Catalogue B0007323-02. I have no idea if this disc is genuine or not - it looks perfectly normal. I purchased it to repeat your measurements.

Unfortunately, regarding track 5 (the title track), a quick visual examination of the waveform on this CD shows that from the first bar, the audio is markedly level limited and asymmetrically so, not that asymmetry is unusual with musical instruments or voice. I attach a few screen shots with Audition set-up as I normally use it, with 0dB, peak level, full modulation occupying the entire vertical scale. Left channel on top of the right channel. I have used two different ripping tools with identical results.

I suppose that it could be argued that the audio is not actually clipped (although it looks like it is) but severely limited. It has clearly been adjusted during production/mastering with intent to be banging hard against the end stops right throughout this track. We can use software tools to reduce the loudness and let the software predict (by maths) how much the clipping/limiting has removed, and an attempt can be made to restore what has gone.

Can you compare this with your disc please?

The essence of the work you have undertaken in this thread is to illustrate how the amplifier is readily driven into clipping by real-world music. That surely mandates that whatever source audio is used to illustrate that point (especially to a non technical audience) must be beyond technical reproach. If the source audio is itself level maximised (i.e. clipped or seemingly clipped) then that has the potential for great confusion. It also makes it more difficult (or impossible) to demonstrate in isolation the amplifier's clipping action alone.

>
 
Attachments only viewable to members

mhennessy

Member
Limited yes, but of value

Limited yes, but of value

Hi Alan,

Can't immediately find a catalogue number, but the number on the bar-code is 602517050426.

No matter; from a close examination of the opening bars, it looks identical to mine.

Yes, it's not clipped on the CD - else Audacity would show red lines over the waveform - but there's no doubt that some limiting was expertly applied during mixing and/or mastering, and that occasionally looks like clipping in the editor (although on an analogue 'scope, it does look slightly different to amplifier clipping). It's a busy track, especially compared with the rest of the album, and it certainly makes an impact!

And I certainly agree that we should be using material that should be more "blameless" in this regard. This particular track only entered the discussion because it's the track that first convincingly opened my ears to this particular issue.

It's a shame that we only have access to the published CD. But, as I said earlier, comparing the released version to a version that has been clipped by an amplifier does yield audible differences, so it's a valid - albeit non-perfect - demonstration. A good starting point.

For me, there are two distinct threads to this investigation:

1. Clipping happens more often that we might expect - especially with well-recorded material.
2. When it occurs, it isn't always noticeable as obvious distortion, but it can change the character of certain instruments.

Earlier I mentioned "Money For Nothing" - the original 1980s release has an astonishing Peak to Mean Ratio for a contemporary release, and I'm pretty sure that there is no clipping inherent in the recording. Trying to play that at reasonable volume via my 50 watt setup definitely resulted in very visible clipping - and it was audible too. But I'm not convinced that a snare drum - no matter how well recorded - will help to demonstrate point 2 here :)

I think your knowledge of classic music is wider than mine - my collection is roughly 5:1 in favour of non-classical works - so I'm thinking that having identified brass as a potentially good way to demonstrate this effect, there must be some examples of brass-rich music from the classical repertoire that would have been recorded in a more sympathetic manner, and these might be a more appropriate choice for further investigations. I'll have a think.

Any suggestions? I welcome the chance to broaden my CD collection!

All the best,

Mark
 

A.S.

Administrator
Staff member
Hot plus more hot ....

Hot plus more hot ....

... And I certainly agree that we should be using material that should be more "blameless" in this regard. This particular track only entered the discussion because it's the track that first convincingly opened my ears to this particular issue.

It's a shame that we only have access to the published CD. But, as I said earlier, comparing the released version to a version that has been clipped by an amplifier does yield audible differences, so it's a valid - albeit non-perfect - demonstration. A good starting point.

For me, there are two distinct threads to this investigation:

1. Clipping happens more often that we might expect - especially with well-recorded material.
2. When it occurs, it isn't always noticeable as obvious distortion, but it can change the character of certain instruments.

Earlier I mentioned "Money For Nothing" - the original 1980s release has an astonishing Peak to Mean Ratio for a contemporary release, and I'm pretty sure that there is no clipping inherent in the recording. Trying to play that at reasonable volume via my 50 watt setup definitely resulted in very visible clipping - and it was audible too. But I'm not convinced that a snare drum - no matter how well recorded - will help to demonstrate point 2 here :)

I think your knowledge of classic music is wider than mine - my collection is roughly 5:1 in favour of non-classical works - so I'm thinking that having identified brass as a potentially good way to demonstrate this effect, there must be some examples of brass-rich music from the classical repertoire that would have been recorded in a more sympathetic manner, and these might be a more appropriate choice for further investigations. I'll have a think.

Any suggestions? I welcome the chance to broaden my CD collection!

All the best,

Mark
All noted. To be fair though, brass that is recorded/reproduced that 'hot' (i.e. with absolutely no dynamic headroom at all) is likely to take on a sonic characteristic unlike a real brass instrument played in a nice acoustic, with the microphone a distance away, and captured at a loudness which leaves enough dynamic headroom in reserve regardless of the capabilities of the amplifier. That a small amplifier driven hard by such a 'hot' recording is going to add another layer of mangling to the sound isn't to be doubted as 'clipping' by definition, will take the audio experience away from what one would hear live and into an unpredictable sonic region.

The problem as I see it is that listening to this track on your small amp (driven hard) gave you a subjective experience that you did not like. But it would be a very bold man that could say with certainty, that what he didn't like was just the sound of the amp in clipping, as opposed a combination of very hot recording pushing the sound to be as loud, and sellable, as possible plus the limitations of the amp. The amp and the recording together would, I suggest, be sonically intertwined and inseparable from each other. Isn't that the case?

I believe that a simple visual peek at the waveform in Audition or similar waveform editor tells you at a glance whether a recording is likely to be in the high fidelity class or not. If the levels are rammed hard up against the wall - the modern way - then one has to ask if that recording is worthy of replay on even the finest high fidelity system, which is capable of so much more dynamics potential. A great high fidelity recording should only rarely use the full dynamic range i.e. touch the 0dB limit lines just a few times during a performance.

I am often told how the HUG is read as a factual, objective, impartial, non-marketing insight to the truths of hearing and audio. If that is the case, we've achieved our objective of creating content of lasting value to the ordinary, non-technical reader, a task which, I truth, is surely the remit of the audio media, not us. However, we are here now and we have an opportunity to maximise. I think what we should do is fork this thread, back up a bit, and make a fresh start having accounted for confounding variables that have unwittingly crept into this otherwise thought-provoking analysis. If we can set this analysis on solid foundations, other researchers can replicate the experiments, and we put ourselves beyond petty, ill-informed criticism.

I am still of the opinion that we have within our grasp an insight to the matter of amplifier sonics of real value.

How does that seem to you?
 

mhennessy

Member
The recording, or the amp - or both?

The recording, or the amp - or both?

The problem as I see it is that listening to this track on your small amp (driven hard) gave you a subjective experience that you did not like. But it would be a very bold man that could say with certainty, that what he didn't like was just the sound of the amp in clipping, as opposed a combination of very hot recording pushing the sound to be as loud, and sellable, as possible plus the limitations of the amp. The amp and the recording together would, I suggest, be sonically intertwined and inseparable from each other. Isn't that the case?
Just to clarify, while I noted a difference between clipped and unclipped, I didn't have a subjective preference for either, and couldn't tell by ear which was which. I needed a 'scope to reveal that the difference between the two versions was clipping.

I tend to feel that a clipped recording might sound different if clipped some more; you only have to listen to a transistor radio with a 1 watt output stage to see how this hypothesis could hold water. As a counter-argument though, you could suggest that the bulk of the damage has been done by the initial clipping, and perhaps a bit more clipping might not alter things all that much? Intuitively, I can't help feeling that this might be the case for a solo instrument, but that things are much more complex when you have a complete mix as your source material.

Trying to pin this down with hard science and numbers would be challenging to say the least. That's not to say that we shouldn't try :)

I believe that a simple visual peek at the waveform in Audition or similar waveform editor tells you at a glance whether a recording is likely to be in the high fidelity class or not. If the levels are rammed hard up against the wall - the modern way - then one has to ask if that recording is worthy of replay on even the finest high fidelity system, which is capable of so much more dynamics potential. A great high fidelity recording should only rarely use the full dynamic range i.e. touch the 0dB limit lines just a few times during a performance.
Yes, but we are getting into the murky areas of genre with this. For many "rock and pop" recordings, the engineers will want the loudness to be essentially the same throughout - perhaps with variations during choruses or the break. Others will have a wider range for artistic reasons. Even a good classical recording will be "compressed" (usually manually by a musically trained engineer who is intimately familiar with the score) as few would welcome the entire dynamic range of an orchestra at home in their listening rooms.

I suppose it's a question of accepting that the released CD is what it is. The finished product might not meet the standards we hope for, but it might still have other merits. Ultimately, some non-ideal CDs could be suitable for subjective testing, but I certainly agree that it would be nice to separate out the recording from the amplifier if that were possible.

I am often told how the HUG is read as a factual, objective, impartial, non-marketing insight to the truths of hearing and audio. If that is the case, we've achieved our objective of creating content of lasting value to the ordinary, non-technical reader, a task which, I truth, is surely the remit of the audio media, not us. However, we are here now and we have an opportunity to maximise. I think what we should do is fork this thread, back up a bit, and make a fresh start having accounted for confounding variables that have unwittingly crept into this otherwise thought-provoking analysis. If we can set this analysis on solid foundations, other researchers can replicate the experiments, and we put ourselves beyond petty, ill-informed criticism.

I am still of the opinion that we have within our grasp an insight to the matter of amplifier sonics of real value.

How does that seem to you?
OK - lead the way!
 

A.S.

Administrator
Staff member
Opening gambit

Opening gambit

OK. Like you, I have very limited time to devote here, so we should aim for short, sharp, contributions that 100% of our largely non-technical audience can absorb.

To set the scene:

(1) I have repeatedly stated that under controlled conditions, the huge subjective differences I read are claimed to about this or that amplifiers seem to diminish to almost nothing, or indeed be nothing. The essence of this belief is that controlled conditions place all contenders on exactly the same playing field for loudness etc.. By implication then, if these sonic differences are real, rather than being imaginary and part of the audio folklore, it must be that only when the amplifiers are operated outside of controlled conditions that these differences can even conceptually emerge. The determination of whether or not an amplifier is being fairly compared against other amps mandates the use of simple test equipment, otherwise it cannot be said that controlled conditions have prevailed for the comparisons.

(2) Your working hypothesis is that the loudness and dynamics of recorded music are such that (small) amplifiers are routinely stretched to their limits when driving a loudspeaker loads, generating an in-room loudness that their users find just about satisfying. Operating at their limits, when they just do not have any more power to deliver to the speakers, is a function of the limitation (and especially cost and complexity) of amplifier circuit design, especially that of the power supply and the output stages. You also suggest that the ordinary listener is not sensitive to the sonic effects of an amp (or even a recording) in clipping, and that, to paraphrase you, most listeners (with small amps) are very often but unwittingly listening to their amps when they are, demonstrably with test equipment, being driven at the very edge of their capabilities, and beyond into clipping.

My observation (1) and yours (2) are not mutually independent; they can coexist perfectly happily if, for example, one condition (C) exists in reality, namely:

(C) If operating an audio amplifier at the very limits of its power capability drives the amplifier into what is called 'clipping', we can say, and measure, that the amplifier is no longer operating under my controlled conditions (1) but in a lawless, unpredictable, operational region where the normal internal circuit linearisation processes have, during the period of clipping, broken down and no longer operate. Pushed to the very limits of its capability in the absence of internal feedback, the amplifier is momentarily unstable.

Can we agree this starting point without splitting significant hairs?!

Assuming we are in agreement, I suggest the first real issue to consider is what sort of audio should be use to make these clipping-audibility tests. Why not use test tones? When an amp is driven into clipping, a sine wave would generate lots of harmonics (easy to demonstrate), so why not just use test tones, not music?
 

tedwin

New member
Standardise on Harbeths?

Standardise on Harbeths?

Hi.

Just a quick chip in from an end user.

Regards agreeing on some starting points. Mark's speakers.. Could Harbeth lend him something midrange, SHL5, M30.1, C7? It seems obvious from discussion so far that the speakers being used will have a large effect on the outcome of this exploration, wouldn't it be sensible to use a Harbeth product for a Harbeth forum? Considering most of us probably have some or want some it would at least make results directly relevant and aid in the ongoing queries regards minimum amplifier requirements for Harbeth products.

Fascinating subject btw, I'd love to know if I'm likely to be regularly listening to a clipped signal :)

Ted.
 

mhennessy

Member
Action plan

Action plan

Can we agree this starting point without splitting significant hairs?!
Yes, I think this is good place to begin.

Assuming we are in agreement, I suggest the first real issue to consider is what sort of audio should be use to make these clipping-audibility tests. Why not use test tones? When an amp is driven into clipping, a sine wave would generate lots of harmonics (easy to demonstrate), so why not just use test tones, not music?
Yes, as a way of easing a non-technical audience into the subject, using a pure 1kHz tone is a really good starting point. We routinely do this at work as a demonstration, and find that mild clipping that causes around 1% THD+N is audible to pretty much anyone, irrespective of their background and training. Presumably you'd be able to mock this up in Audition for people to try themselves? I could produce some screen-shots from the oscilloscope using real amplifiers if that helps...

Thinking ahead, should we write a "primer" that introduces the frequency-domain, so that people can understand words like "fundamental" and "harmonics". The only web sources that I know about are probably too technical for our purposes. Perhaps it's already on the HUG somewhere? Making the jump from tone to music will be tricky. Perhaps when we get to that point, we need to start by investigating real instruments in their undistorted form first?

All the best,

Mark
 

A.S.

Administrator
Staff member
First steps - sine wave?

First steps - sine wave?

Yes, I think this is good place to begin.

Yes, as a way of easing a non-technical audience into the subject, using a pure 1kHz tone is a really good starting point. We routinely do this at work as a demonstration, and find that mild clipping that causes around 1% THD+N is audible to pretty much anyone, irrespective of their background and training. Presumably you'd be able to mock this up in Audition for people to try themselves? I could produce some screen-shots from the oscilloscope using real amplifiers if that helps...

Thinking ahead, should we write a "primer" that introduces the frequency-domain, so that people can understand words like "fundamental" and "harmonics". The only web sources that I know about are probably too technical for our purposes. Perhaps it's already on the HUG somewhere? Making the jump from tone to music will be tricky. Perhaps when we get to that point, we need to start by investigating real instruments in their undistorted form first?

All the best,

Mark
I think the secret really is to hit a key concept along the road and then gently work with it. Rather like making a trek across a wilderness without too much planning, and every now and again turning a stone over, crouching down and taking a close look at what's under it. If concepts can be explained with a picture, preferably a moving one, so much the better.

So, why don't you explain in very simple language what a sine wave is, how on earth it relates to music, and I'll make an example or two. Sine waves are at the heart of all audio equipment design, and when we talk of 'frequency response' we're doing nothing other than commenting how good the equipment is at reproducing a range of sine waves. The assumption is, and has been since the dawn of recorded sound, that if a system can faithfully reproduce sine waves, it must be capable of faithfully reproducing music. You can imagine how much more convenient sine waves are for the audio designer than having to have a vast library of recorded music. Keep the explanation very simple please! Your challenge is to do that in one sentence.

BTW: looking for the real sound of brass? Look no further than BBC Prom 2013 #30, Prokofiev's Piano Concerto No.2, clip here. Quite something!

 

mhennessy

Member
Defining sine waves, and their usefulness

Defining sine waves, and their usefulness

Explain a sine wave in one sentence? Impossible - not a chance! I eventually got it down to 3...

Sine waves:

Sine waves are ubiquitous; they are found in the natural world, in engineering, in mathematics, and even in music! If you can imagine a weight on a spring bobbing up and down at a steady pace, then you've got a pretty good idea of what a sine wave looks like.

Sine waves are rather boring to listen to - much like a tuning fork - but when analysed, it transpires that a sine wave at a constant volume contains energy at just one frequency - and that's what makes them so interesting to audio engineers.

It's worth taking a look at the Wikipedia article. Yes, there is a lot of theory there, but ignore that if you wish - I'd recommend you simply look at the animated images, and have a listen to the audio clip provided.


Let's think about music:

One of the key properties of a sine wave is the repetition rate - how many vibrations do we get in a given time frame. For an audible sine wave, this determines the pitch that we perceive. If you press the middle-C key on a piano keyboard, a string inside the piano is set into motion, and it vibrates back and forth roughly 260 times per second. If you press the A key to the right of that, a different string is made to vibrate at 440 times per second.

The unit for pitch - or frequency, as engineers like to call it - is sometimes called "cycles per second", but is more usually called Hertz (abbreviated to Hz). We are pretty sensitive to pitch - some people more than others, of course.

For anyone with a musical background, you might be interested to look at this table of piano key frequencies. As you can see, a wide range of frequencies are covered by this instrument.

Another property is the size of the vibrations - how hard did we hit the keyboard, how much is the string moving. In engineering terms, we call this amplitude. A very important concept that we will no-doubt be building on.

Does all this make sense?
 

A.S.

Administrator
Staff member
Actually using sine waves

Actually using sine waves

Sine waves:

Sine waves are ubiquitous; they are found in the natural world, in engineering, in mathematics, and even in music! If you can imagine a weight on a spring bobbing up and down at a steady pace, then you've got a pretty good idea of what a sine wave looks like.

Sine waves are rather boring to listen to - much like a tuning fork - but when analysed, it transpires that a sine wave at a constant volume contains energy at just one frequency - and that's what makes them so interesting to audio engineers....
So, the audio sine wave is one (spot) frequency picked from a range of frequencies ranging from low to high, that cover the human hearing range. That places it in the range of vibration from about 20 times per second to about 20,000 times per second; the deepest church organ note (which felt rather than heard) right up to the highest tone from a cymbal.

You say that audio engineers find sine waves useful. They can pass a single-pitch sine wave or series of sine waves, or even a non-stop sweep of sine waves with a sequence start and stop frequency, through a piece of audio equipment and it will convey useful information about how that equipment is handling the sine wave. I guess the beauty of the sine wave is that its energy and the observer's attention is entirely focused on the behaviour of the equipment at the frequency or frequencies of interest. In other words, by using and interpreting a sine wave, the audio engineer has eliminated confounding variables that might have led to inappropriate conclusions.

That means sine waves are an excellent investigative tool. If 1000 audio engineers are asked to test a piece of audio equipment with sine waves and simple test equipment, if they are careful and serious, the results will be identical. If 1000 listeners are asked to critique a piece of audio equipment with their choice of music at their choice of loudness, 1000 different subjective opinions could be expected. That's not a lot of help to an audio engineer. Who would he believe? How would he know if he had designed a market-suitable piece of equipment? How would he know the limitations of the equipment so that warranty returns could be minimised? How would he be able to publish any sort of specification that would stand up in court?

So if sine waves are so useful to engineers, why doesn't the home audio enthusiast listen to sine waves, and have simple test equipment to see for himself the behaviour of sine waves through the equipment in his system? Wouldn't it be logical that if any element along the signal chain from the microphone to the loudspeakers significantly altered a perfect sine wave along the route, that that equipment could and should be identified as the 'weak link' and replaced?

What does a sine wave sound like? Here is an example; I've generated successive sine wave tones as 100Hz, 500Hz, 1000Hz (1kHz), 5kHz and 10kHz. Although these tones are technically absolutely pure, they may sound a little 'rough' depending on your replay setup. My plastic PC speakers rattle a bit on the low tones.

Ex.2

An audio engineer using sine tones to investigate a piece of audio equipment would write out a little table with the frequencies from top to bottom and a level reading (from his test instrument) along the other axis - see attached. As he's working in audio, by convention, he'd convert voltages to decibels - see attached.

If the investigator is using modern test equipment, he won't need the paper and pencil any more. He wont need to study the reading on a meter, tone by tone. He can use a computer to generate a sine sweep, connect the output of the equipment under test to the soundcard input, and let the computer do the hard work of capturing and displaying the swept tone. Here is the same 100Hz to 10kHz tone range (and all the intermediate frequencies) covered in just three seconds:

Ex.3

And can you believe it ... using a fast modern computer, the sine sweep can be covered even faster! Here is the same 100Hz to 10kHz range (and all the intermediate frequencies) covered in one tenth of a second:

Ex.4

That's not all! We can cover that range in even one tenth of that time. Here is 10Hz to 10kHz (and all intermediate frequencies) covered in one hundredth of a second ....

Ex.5

But wait! We can go faster yet! We can cover our 100Hz to 10kHz (and all intermediate frequencies) in just one thousandth of a second. Now it has lost all sense of pitch or tones: it's become a click to the human ear:

Ex.6

Note: if the human ear had perfect resolution, you should be able to hear and identify 100Hz, 101Hz, 102Hz, 103Hz and all the way up to 9997Hz, 9998Hz, 9999Hz and finally 10,000Hz in examples 3-6. You can't because that would be an unnecessary ability in our evolutionary development. A $10 soundcard would be able to resolve these individual frequencies with ease; it would not sound like a click to the computer 'ear' but an almost infinite series of wholly identifiable tones, passing by very fast.

Regardless of how much time we are prepared to sit studying our test meter and writing down readings or using the computer to whip through the sine test, the measurement results should be identical, providing that the equipment under investigation is not 'corrupting' the sine waves.

Q: If sine waves are so useful for objectively technically evaluating audio equipment, why would anyone ever bother to listen to music and then reach personal subjective opinions about the merits of this or that equipment? Thoughts?

A supplementary question is: what does a sine wave look like when displayed on test equipment? What does it look like when not cleanly passed through a piece of audio equipment? Is that 'corruption' visually distinctive? Does that correlate with a change in its sonic qualities, to the ordinary ear? Once a sine wave has been 'corrupted' can that process be reversed and the perfect, pure sine wave recovered?

>
 
Attachments only viewable to members

mhennessy

Member
Interpreting sine waves by looking at them ...

Interpreting sine waves by looking at them ...

Q: If sine waves are so useful for objectively technically evaluating audio equipment, why would anyone ever bother to listen to music and then reach personal subjective opinions about the merits of this or that equipment? Thoughts?
An excellent question, and lots of possible answers.

A lot of audiophiles distrust measurements, claiming that they don't correlate with what they hear. But I'm not sure if they take the time to compare equipment under the sorts of controlled conditions that Alan advocates (precisely matched levels, complete freedom from clipping, and a controlled A-B test). Many electronic products perform so well that it can be hard to find a measurement that discriminates between them. Again, providing controlled listening testing is performed, we also can't discriminate between them by listening! I could say more, but I feel that this question requires a complex answer, and that we should return to this later - once we have covered more of the basics...

A supplementary question is: what does a sine wave look like when displayed on test equipment? What does it look like when not cleanly passed through a piece of audio equipment? Is that 'corruption' visually distinctive? Does that correlate with a change in its sonic qualities, to the ordinary ear? Once a sine wave has been 'corrupted' can that process be reversed and the perfect, pure sine wave recovered?
I've taken some more snaps.

I'm using an oscilloscope to observe a pure sine wave. An oscilloscope looks like an intimidating instrument at first glance, but they are generally pretty easy to operate with practice. It simply plots a graph of voltage verses time. The vertical axis represents voltage, and the horizontal axis is time. The oscilloscope has various control knobs and switches (not shown in these photos) which allow the user to adjust the way the 'scope displays an incoming voltage waveform.

The screen has (red, printed) markings that represent divisions. Just like graph paper. And because the oscilloscope has calibrated controls, we know how many volts each division is "worth", and we also know how much time it takes to move from left to right. The example I'm using today actually writes this set-up information on screen, so it's easy for us to understand what is happening.

This first image shows a pure sine wave:

You must be registered for see images attach
A

We can see that it is 4 divisions high, from a bottom trough to a top peak (there are several ways to measure AC signals, but for now, we'll "park" that, and just stick to peak to peak). When this picture was taken, the vertical sensitivity was set to 0.5 volts per division, so this is a 2V pk-pk sine wave.

Regarding the pitch of this tone, we need to know how long one complete oscillation takes. On a display like this, it's easiest to look for the points were the waveform crosses the centre line. Start on the far left, just to the left of the "1-" indication. We see the "weight on the spring" bob up, reach a peak, fall down again, continue past the "rest" point, and head towards the bottom trough. Once it's gone as far down as it can, it starts on an upward course, and reaches the rest position one more time. At this point, we can say that it has completed one whole "cycle". This point is reached at the exact centre of the screen.

Counting from left to right, you should see that it took 5 divisions to trace one complete cycle. Indeed, there are two complete cycles across the width of the screen.

Reading off the bottom of the screen, we can see that the main timebase is set to 200 microseconds per division. And 5 lots of 200 microseconds adds up to 1 millisecond. So, in one whole second, we can fit in 1000 of these complete cycles. To put that another way, there are 1000 cycles per second. It's a 1kHz tone!

Here is another image showing the same waveform, but this time I've used the inbuilt cursors to show these features:

You must be registered for see images attach
B

Now that we're getting good at reading oscilloscopes, here's an interesting image. It uses the 'scopes inbuilt digital memory to store and recall several different waveforms, and here I've shown them all at the same time:

You must be registered for see images attach
C

I've simply shown the same 1kHz sine wave at differing amplitudes, representing differing loudness levels...

OK, so what happens when a sine wave is subject to clipping? I've connected the output of my pre-amplifier to the input of the 'scope now so I can turn up the level. First, a pure sine wave, at just below the preamp's clipping point:

You must be registered for see images attach
D

This is 26 volts pk-pk output from the preamp - a large signal. And look at how low the distortion is - lower than I can actually measure with the simple kit I was using today.

Next, I nudge up the preamp volume control by just 1 decibel. This commands an increase in output signal voltage of around 10%, which the amplifier can't do (because its internal power supply is fixed). This image shows the characteristic "flat-top" clipping. Note the cursors, which were unchanged from the previous image. I've also overlaid the non-clipped, pure sine wave onto the right hand display with the clipped version for reference.

You must be registered for see images attach
E

I hope all this makes sense, and helps to illustrate what sine waves look like in both their pure and clipped form.

All the best,

Mark
 
Attachments only viewable to members

tedwin

New member
Display interpretation

Display interpretation

Hi.

Just a quick query, sorry if answer obvious I may well have just missed something.

In the image of overlaid sine waves of the same frequency but different amplitudes, why is the intersection of the various waves above and below the centre line, not on it?

Ted.
 

mhennessy

Member
Time base

Time base

Hi.

Just a quick query, sorry if answer obvious I may well have just missed something.

In the image of overlaid sine waves of the same frequency but different amplitudes, why is the intersection of the various waves above and below the centre line, not on it?

Ted.
Hi Ted,

Well-spotted! I didn't notice until I'd annotated the picture, but didn't have the time earlier to go back and correct it.

It's nothing untoward though. It's simply related to the triggering of the oscilloscope. The triggering circuitry decides when to start moving the spot from left to right, and to do that, it waits for the input waveform to reach a certain voltage. If you look at the extreme left, where the trace begins, you'll see that none of them start on the centre line. When using the "digital" mode of the 'scope, it seems to use a "smart" trigger mode that auto-adjusts according to the signal received. The pictures taken in analogue mode were all OK.

Triggering is a complex subject, and when teaching people to use 'scopes, it's normally triggering that people struggle with the most. So, please don't worry if that brief explanation doesn't make sense! If you remember fiddling with "vertical hold" controls on the backs of really old TV sets, it's a remarkably similar mechanism at play...

Anyway, I'm very happy to repeat and re-shoot tomorrow if it helps.

Cheers,

Mark
 

mhennessy

Member
More on clipping, and photography

More on clipping, and photography

A supplementary question is: what does a sine wave look like when displayed on test equipment? What does it look like when not cleanly passed through a piece of audio equipment? Is that 'corruption' visually distinctive? Does that correlate with a change in its sonic qualities, to the ordinary ear? Once a sine wave has been 'corrupted' can that process be reversed and the perfect, pure sine wave recovered?
So, we've seen examples of pure sine waves, and sine waves that have been clipped in my preamp. Is clipping visually distinctive? I think so...

It's really easy to hear clipping when you use a pure tone as the source. But, things get a lot more complex when we discus "real" audio, and I'm sure we'll be saying more about this soon. Can it be "repaired"? Can you repair an over-exposed photograph? Once the information has been removed in the camera, then it's gone.

Clipping an image for artistic purposes is sometimes desirable, and sometimes the same is deliberately done to an audio signal in the studio - again, for artistic reasons. But, at the risk of preaching to the converted, the job of a hi-fi is to reproduce the recording as it exists, and not to add a "sonic signature" of its own :)

Mark
 

A.S.

Administrator
Staff member
How does clipping sound?

How does clipping sound?

I suppose then the next thing is to demonstrate what a clipped sine wave actually sounds like. I generated 3 seconds of moderately loud 1kHz sine wave, then greatly increased the level until the sine wave has its tops and bottoms chopped off so that they are no longer nicely shaped. Now, to save your ears from having to listen to a clipped sine wave at maximum level, which is very nasty, after clippin, I then reduced the loudness (of the damaged, clipped wave) to approximately match that of the pure wave and joined them together.

The screensnap shows the combined waveform, 6 seconds long. To the eye, it doesn't look much different either side of the join. How about the loudness? That's about the same, pure or clipped. But how about the sound quality? Is that the seemingly innocuous, barely visible difference in the shape of just those very peaks and troughs of the sine wave going to make any audible difference to the sonic character of the 1kHz tone? You decide....

Ex.7

You must be registered for see images attach


If the perfect shape of the sine wave is, for whatever reason, disturbed, intentionally or not, and if we can see that corruption on the oscilloscope or the digital audio editor, we are probably going to hear it. There is a correlation between what we can actually see with the unaided eye, graphed for us on our test equipment, and what we can hear. As a generalised working rule, if the sine waves shape, especially at the peaks and troughs of the wave, looks about normal, we are not in the clipping zone and the system 'corruption' (another word for that is distortion) is probably low. A sine wave is a very simple and useful analytical tool that we can have confidence in.

So, this is leading to a key point: if our objective is to faithfully reproduce the multitudinous sine wave tones that are generated by musical instruments (and human voice), the primary requirement of the reproduction chain from the microphone to the speakers must be of avoiding clipping at any point along that chain. No matter how exotic, expensive and technically perfect subsequent equipment after the clipping point in the chain, once clipped, the sound will take on a different character and that character will be passed along the chain to the listener.

If the CD recording is clipped intentionally, artistically, or accidentally, no matter how expensive or well reputed the CD player, amplifier and speakers are, the sonic fingerprint of a clipped sound cannot be reversed or cancelled out. The clipping will have permanently and irreversibly changed the tonality of the music you will hear over the speakers. Clipping is an absolute no-no at any point along a high fidelity audio chain. To let the signal or equipment accidentally clip, is beginners mistake #1. Music that has been clipped at any point along the audio chain is not true to what you would hear live, and a reproduction chain demonstrating clipping distortion cannot be considered 'high fidelity'.

Clipping could be mechanical, electrical or from software or a combination of all. Mechanical clipping could be caused by a faulty microphone, a microphone too close to the performer and driven beyond its maximum loudness specification (for example, using a normal microphone within inches of a bass drum), by a performer who sings or plays unexpectedly loud, or electrical clipping such as faults or incorrect settings in the recording mixer, or software clipping during the production of the CD either intentionally, for artistic effect, or accidentally. So clipping of the sound recording may have taken place before we even start to play it through our audio system. There is absolutely nothing we can do to reverse or 'unclip' the recording we are provided with. Taking my sound example (7) above, we cannot make the second half of the example ever sound like the first half again. The evidence of cliping is proof that the sound wave does not have adequate head room somewhere in the recording or replay chain, and the signal loudness is 'banging' on the end stops.

Q: So where does clipping creep into the home audio chain?

How can clipping creep into our own replay systems, assuming that we at home are the guilty party, not the recording company?




(Note to self: Audition 1.0 & 1.5 in study do not generate audibly clean sine waves; use 2.0
)
 
Attachments only viewable to members

tedwin

New member
Explanation ok

Explanation ok

Anyway, I'm very happy to repeat and re-shoot tomorrow if it helps.

Cheers,

Mark
Thanks for the explanation :)

No need to re-shoot on my behalf, I'm amazed (and grateful) that you've put in this much effort already.
 

tedwin

New member
How permanent is a clipped audio?

How permanent is a clipped audio?

Clipping an image for artistic purposes is sometimes desirable, and sometimes the same is deliberately done to an audio signal in the studio - again, for artistic reasons. But, at the risk of preaching to the converted, the job of a hi-fi is to reproduce the recording as it exists, and not to add a "sonic signature" of its own :)

Mark
..Music that has been clipped at any point along the audio chain is not true to what you would hear live, and a reproduction chain demonstrating clipping distortion cannot be considered 'high fidelity'.

Clipping could be mechanical, electrical or from software or a combination of all. Mechanical clipping could be caused by a faulty microphone, a microphone too close to the performer and driven beyond its maximum loudness specification (for example, using a normal microphone within inches of a bass drum), by a performer who sings or plays unexpectedly loud, or electrical clipping such as faults or incorrect settings in the recording mixer, or software clipping during the production of the CD either intentionally, for artistic effect, or accidentally. So clipping of the sound recording may have taken place before we even start to play it through our audio system. There is absolutely nothing we can do to reverse or 'unclip' the recording we are provided with.
Sorry, me again! Can we confirm that if a clipped signal is used intentionally by the artist, but in the final mix it is used at a lower level we can still expect to reproduce it faithfully, despite its 'squared off' waveform.

Ted.
 

A.S.

Administrator
Staff member
Sorry, me again! Can we confirm that if a clipped signal is used intentionally by the artist, but in the final mix it is used at a lower level we can still expect to reproduce it faithfully, despite its 'squared off' waveform.

Ted.
That is what I said. Once audio is clipped is sonic character is permanently changed. Like a demo?
 

tedwin

New member
The studio or at home

The studio or at home

....Like a demo?
All good thank you. I get the clipping I think. I've listened to the test tone clipped and unclipped. Obvious difference there. I was just trying to confirm that we're talking reproduction at home regardless of what happens in the recording studio. So a clipped waveform used in a mix by a recording artist or engineer will (hopefully) sound the same on my stereo as it does on their studio monitors. I got confused when you started talking about microphones and drums, I'm guessing any clipped signals (if they are happy with the sound) will get through to the final mix but be within the dynamic range of the cd we buy. Hence we only need concern ourselves with faithful reproduction at home irrespective of how the artist achieved the sound they where after...(?)
 
Top