Home |
Search |
Today's Posts |
#41
Posted to rec.audio.pro
|
|||
|
|||
0dBFS+ Levels in Digital Mastering
geoff wrote: "Called 'Sound Check' . Just a simple normalisation thing, I think. "
Ahh yes - Now you're talking! Although I seriously doubt it peak normalizes. I could expreimient with it by applying it in iTunes to a couple of test tracks: one very dynamic(high peak to avg ratio), the other, highly compressed(low peak to avg). If they both sound as loud as each other, then it is not peak-normalizing. The average is closer to what we hear, and if you align the tracks by peak, the average of the less dynamic track should sound louder. |
#42
Posted to rec.audio.pro
|
|||
|
|||
0dBFS+ Levels in Digital Mastering
On 13/10/2015 5:08 AM, Scott Dorsey wrote:
In article , Trevor wrote: Rubbish, peak limiting that causes flat tops IS clipping. YOU simply don't understand what they have done, or the difference between compression and limiting it seems. How much of a flat top is needed for it to be clipping? One sample? Many samples? At least 3 or more. (genuine square waves excluded) And as I have said, some CD's have HUNDREDS of consecutive samples at maximum level, and often many similar groups in one song. IF there are only 3 in fact I'm not worried in the slightest, when there are hundreds I am. That is far more common with pop CD's these days than many people seem to think, simply because they never look. At what point does limiting end and clipping begin? As soon as the knee ends and maximum level is reached for at least 3 or more samples. I'm surprised you don't understand this already It's not rocket science! On the other hand, let's say I have a drum hit that is only three samples long and it's the only loud thing on the track. I run it through a limiter... maybe it's just a pair of back to back diodes, and I chop it down 20 dB. It sounds fine. Did I clip it, or did I limit it? IF the drum hit is now flat topped for those 3 samples you did BOTH, regardless of what level you "chop it down to". (of course there is NO such thing as a drum hit for 3/44,000's of a second!) I'm totally amazed that someone with your experience still doesn't have a vague notion of what clipping actually is. It can occur anywhere in the chain, and is independent of final level. Just like you can clip a Mackie mixers mix bus, even if the output fader is well below maximum. In fact you can clip a single channel on ANY mixer, without clipping the mixer output, or amplifier input. THAT channel is STILL clipped!!! Trevor. |
#43
Posted to rec.audio.pro
|
|||
|
|||
0dBFS+ Levels in Digital Mastering
On 13/10/2015 7:40 AM, geoff wrote:
On 13/10/2015 2:25 a.m., Scott Dorsey wrote: geoff wrote: No CD that I've purchased in the last decade exhibit clipping that I've noticed, though a few are over-compressed to hell (not many fortunately). And some I've been prompted to actually check ! Not much 'current' pop though I concede. The problem is that in the digital world, clipping is whatever you define it as. I tend to set metering so three consecutive FS samples light the over light, and so that is clipping. I draw the line at 2, though 3 seems to be the commonly accepted criteria. Yep, and yet MANY Cd's have HUNDREDS of consecutive samples at maximum level :-( No-one can argue THAT is *NOT* clipping simply because the CD is normalised to -0.3dBFS, CAN THEY? Or are some people who claim to be "pro's" really that stupid? Trevor. |
#44
Posted to rec.audio.pro
|
|||
|
|||
0dBFS+ Levels in Digital Mastering
Trevor, Scott:
And whether its clipping as in overloading a bus or as in clipping aesthetically(a waveform), both can sound like ****e! |
#45
Posted to rec.audio.pro
|
|||
|
|||
0dBFS+ Levels in Digital Mastering
On 13/10/2015 7:50 AM, geoff wrote:
On 13/10/2015 3:23 a.m., Trevor wrote: On 13/10/2015 12:25 AM, Scott Dorsey wrote: geoff wrote: No CD that I've purchased in the last decade exhibit clipping that I've noticed, though a few are over-compressed to hell (not many fortunately). And some I've been prompted to actually check ! Not much 'current' pop though I concede. The problem is that in the digital world, clipping is whatever you define it as. I tend to set metering so three consecutive FS samples light the over light, and so that is clipping. No, you first have to normalise the gain back to 0dBFS for that to work since many CD's are first *severely clipped* then normalised to about -0.3dBFS, so your clip lights will never come on. BUT the flat tops remain regardless! Aggressive limiting that flat-tops the signal isn't necessarily clipping, it's just aggressive limiting. Rubbish, peak limiting that causes flat tops IS clipping. YOU simply don't understand what they have done, or the difference between compression and limiting it seems. I think this is the only thing where we really disagree. The rest of your debate seems hair-splitting about definitions of terms. Rubbish, When someone claims clipping is not clipping simply because it's level has been post reduced, or that clipping is not clipping simply because it was done by a "limiter" that is FARRRRR from hair splitting about definitions, that is about that person not understanding the basic fundamentals of recording!!!!! I say that 'clipping' is only the result of a digital or analogue *overload*, where the actual mathematical or electrical constraints of the process are exceeded and nothing can exist above. Right, as occurs when a limiter exceeds it's absolute peak level, a mixer or amplifier exceeds it's maximum rail voltage, or a digital device exceeds it's digital full scale level, whether OR NOT, that level is subsequently reduced AFTER clipping!! 'Limiting' is the *controlled* result of a digital or analogue process. And yes, that can be extreme to the point of resembling clipping, but is not the same thing. Bull****!! A limiter by it's very nature will produce CLIPPING when it exceeds it's knee and reaches it's maximum level. That's the whole point of the device after all. One may choose to use it so it never exceeds the knee (and hard limiting may not even have a knee!) but IF it does exceed absolute maximum, then CLIPPING will occur! That you choose to believe it is somehow not clipping just because YOU prefer to call it limiting only shows you ignorance!!! Achieving, or adding to, limiting by clipping (sadly) is done, presumably by those who are incompetent, lack understanding of the implications, or have cynical intent. Yep, but limiting is NOT necessarily the same as compression. You and many others still seem to be totally confused about that. And of course one can drive a compressor into clipping as well! Clipping is clipping no matter how it is achieved. Arguing that it is not is simply TOTAL BS! Trevor. |
#46
Posted to rec.audio.pro
|
|||
|
|||
0dBFS+ Levels in Digital Mastering
On 10/12/2015 6:24 PM, geoff wrote:
My car iPod Touch has a automatic replay level function based on (I think) a peak-level parameter encoded into the file. When it works on a particularly 'different level' track there seems to be a second-long level adjustment at the commencement of a song I would like to have something like that in my car. It might not work as well for me as it does for you, however, because for me, a "song" (one file) is a two hour radio program of music that often has song-to-song level variations. I've been trying to score a cheap couple of generations old iPod Touch so I could use it for some audio apps that aren't available for my Androids, but so far no luck. They seem to hold their value nearly as well as microphones. -- For a good time, visit http://mikeriversaudio.wordpress.com |
#47
Posted to rec.audio.pro
|
|||
|
|||
0dBFS+ Levels in Digital Mastering
On 10/12/2015 10:32 PM, geoff wrote:
Called 'Sound Check' . Just a simple normalisation thing, I think. My TASCAM DR-40 has something like that. I leave it turned off. In the DR-44, it's on all the time. TASCAM seems to be sensitized to rules about hearing protection and this is one of the things that they do (in addition to popping a reminder up on the screen when you turn the playback volume up) to comply with the CE mark testing. -- For a good time, visit http://mikeriversaudio.wordpress.com |
#49
Posted to rec.audio.pro
|
|||
|
|||
0dBFS+ Levels in Digital Mastering
On Tuesday, October 13, 2015 at 9:14:46 AM UTC-4, Mike Rivers wrote:
On 10/13/2015 2:54 AM, wrote: geoff wrote: "Called 'Sound Check' . Just a simple normalisation thing, I think. " Ahh yes - Now you're talking! Although I seriously doubt it peak normalizes. I could expreimient with it by applying it in iTunes to a couple of test tracks: one very dynamic(high peak to avg ratio), the other, highly compressed(low peak to avg). I think there are several setting in iTUNEs (or other programs) that you can use to change the way the normalization is performed. The key point however, is that none of the setting change the dynamics within a song. The ENTIRE song is adjusted by the same constant value. It is no different from setting the volumne control during playback. This does not really change the shape of the envelope or change the dynamics. Mark |
#50
Posted to rec.audio.pro
|
|||
|
|||
0dBFS+ Levels in Digital Mastering
Mike Rivers wrote: "On 10/13/2015 2:54 AM, wrote:
geoff wrote: "Called 'Sound Check' . Just a simple normalisation thing, I think. " Why don't you test it by generating some sine wave test signals, both continuous and with burst level changes, and learn what it's really doing? That way you'd know (and can explain to us) why something sounds louder or not. - show quoted text -" I already KNOW why there's a loudness difference between a more dynamic song and a less dynamic song when both are peak- normalized, I.E. to 0dBfs, given the same playback volume setting. But terminology gets in the way of our understanding each other, Mike. I understand CONCEPTS Mike - but fail like our previous president at verbalizing them. And I seriously DOUBT iTunes' Sound Check is peak- based, or else it wouldn't serve any purpose beyond what has already been done on thousands of CD titles. I know MP3Gain is not peak-based, because I have NEVER had to adjust playback volume ONCE during playing hundreds of mp3 versions of my collection with mp3Gain applied - except when the phone rings, or the wife comes home(!) In fact, one of the biggest complaints I've read about Apple Sound Check is about how much stuff it turns DOWN. Users thus must turn their volumes up higher when listening to a playlist that might span several(pre- and loudness war era) decades. MP3Gain does too, depending upon where one sets its threshold. It uses a scale: 87-105dB, with a recommended setting of 89dB. I normalize at 91, which triggers the clipping warning on some songs, mainly stuff before 1980, but is relatively inaudible to the ear. This level helps the small amps in my mobile devices where the MP3s are played. Of course, for nearly EVERYTHING released since 2000, it applies at least 6dB of negative gain. If I rip two mp3s of the same song off a hot modern compressed CD, mp3 gain ONE of them, and place them both in my DAW, it does show the original kissing full scale, and the re-gained as pathetically small, with between 6-10dB of headroom between the flat-tops and full scale. So I DO understand what's going on Mike, Scott, geoff, and that ASSHOLE thread-crasher who shall remain nameless, it's putting things into words where I stumble. |
#51
Posted to rec.audio.pro
|
|||
|
|||
0dBFS+ Levels in Digital Mastering
wrote: "I think there are several setting in iTUNEs (or other programs) that you can use to change
the way the normalization is performed. The key point however, is that none of the setting change the dynamics within a song. The ENTIRE song is adjusted by the same constant value. It is no different from setting the volumne control during playback. This does not really change the shape of the envelope or change the dynamics. -Mark" ^^ CORRECT! ^^ |
#52
Posted to rec.audio.pro
|
|||
|
|||
0dBFS+ Levels in Digital Mastering
gray_wolf wrote:
0dBFS+ Levels in Digital Mastering http://service.tcgroup.tc/media/Leve..._AES109(1).pdf Interesting. I have noticed that in my better recorded stuff that the maximum levels tend to be at least -3dB and some as low as -6dB. This paper is fifteen years old, and back then this was a new problem that people hadn't really seen before. This is a problem with converter design, and it's most often a problem with trying to run a ladder dac with an I/V stage off the same rails as the ladder. In the modern sigma-delta world the problem is different, but it still is something converter designers need to be aware of. This, though, is not a problem with the sound processing, it's a problem with the converter design. Benchmark has a good white paper on how these problems were solved, I believe. --scott -- "C'est un Nagra. C'est suisse, et tres, tres precis." |
#53
Posted to rec.audio.pro
|
|||
|
|||
0dBFS+ Levels in Digital Mastering
geoff wrote:
On 13/10/2015 2:05 a.m., JackA wrote: http://www.keithhirsch.com/target-cds Jack A truly bizarre website. People collect strange things. I suppose this is no worse than the people who go ga-ga over Columbia six-eyes LP pressings. --scott -- "C'est un Nagra. C'est suisse, et tres, tres precis." |
#54
Posted to rec.audio.pro
|
|||
|
|||
0dBFS+ Levels in Digital Mastering
wrote:
a good way to differentiate clipping from limiting from compression is by considering the attack and decay times relative to the features of the waveform That's half of it... and the other half is the shape of the knee. But how fast is fast? When does soft clipping become limiting and vice-versa? --scott -- "C'est un Nagra. C'est suisse, et tres, tres precis." |
#55
Posted to rec.audio.pro
|
|||
|
|||
0dBFS+ Levels in Digital Mastering
In article , Trevor wrote:
Many samples? At least 3 or more. (genuine square waves excluded) And as I have said, some CD's have HUNDREDS of consecutive samples at maximum level, and often many similar groups in one song. IF there are only 3 in fact I'm not worried in the slightest, when there are hundreds I am. That is far more common with pop CD's these days than many people seem to think, simply because they never look. Well, that's the degenerate case. Those CDs are clearly clipped. But I'm talking about the borderline cases, because that's where it gets interesting. At what point does limiting end and clipping begin? As soon as the knee ends and maximum level is reached for at least 3 or more samples. I'm surprised you don't understand this already It's not rocket science! That's a good definition. Although... I might decide it is 2 samples or 8 samples and be able to make a good argument for those too. I'm totally amazed that someone with your experience still doesn't have a vague notion of what clipping actually is. It can occur anywhere in the chain, and is independent of final level. Just like you can clip a Mackie mixers mix bus, even if the output fader is well below maximum. In fact you can clip a single channel on ANY mixer, without clipping the mixer output, or amplifier input. THAT channel is STILL clipped!!! Oh, I have a vague notion of what clipping actually is.... and a vague notion of what limiting actually is.... but I can think of a LOT of examples that are sitting directly on the border. I don't want a vague notion, I want a precise mathematical description. My line might be "if you can hear it, it's clipping, if you can't, it's limiting." That's no less vague, though. --scott -- "C'est un Nagra. C'est suisse, et tres, tres precis." |
#56
Posted to rec.audio.pro
|
|||
|
|||
0dBFS+ Levels in Digital Mastering
(Scott Dorsey) writes:
gray_wolf wrote: 0dBFS+ Levels in Digital Mastering http://service.tcgroup.tc/media/Leve..._AES109(1).pdf Interesting. I have noticed that in my better recorded stuff that the maximum levels tend to be at least -3dB and some as low as -6dB. This paper is fifteen years old, and back then this was a new problem that people hadn't really seen before. This is a problem with converter design, and it's most often a problem with trying to run a ladder dac with an I/V stage off the same rails as the ladder. In the modern sigma-delta world the problem is different, but it still is something converter designers need to be aware of. This, though, is not a problem with the sound processing, it's a problem with the converter design. You keep saying this. Exactly what is the problem? Are you referring to the 5 items on p.4 of that reference? I have a couple of remarks regarding those items (if anyone wants to hear them...): 1. If the digital filter is distorting/clipping, it's implemented wrong. Any diligent DSP designer would look out for that sort of thing. 2. I think most of the rest of these problems arise from the basic sampling theory explanation given on p.2. Several years ago I proved that, with the theoretical brick-wall lowpass interpolation filter, it is possible that a limited-range digital signal can give rise to an infinite analog output. This was an analytical result, but it can have practical implications, namely, that, given there are no infinite-length filters in the real-world, the DAC designer should be able to specify the absolute maximum analog output level the DAC will produce, -- Randy Yates Digital Signal Labs http://www.digitalsignallabs.com |
#57
Posted to rec.audio.pro
|
|||
|
|||
0dBFS+ Levels in Digital Mastering
Scott Dorsey wrote: "My line might be "if you can hear it, it's clipping, if you can't, it's
limiting." That's no less vague, though. " And that brings up a valid point: Clipping, as I've understood it for the last 20 years, is when the electronics of a device become momentarily overloaded. And how audible it sounds is dictated largely by duration. Limiting is more an aesthetic property, but again, its duration determines its audibility. A SIDE NOTE: Via google groups, this thread, for about 5 minutes, around 11AM Eastern time, "disappeared" from the R.A.P. header list. It is now back, as I am replying to it. |
#58
Posted to rec.audio.pro
|
|||
|
|||
0dBFS+ Levels in Digital Mastering
Randy Yates writes:
(Scott Dorsey) writes: gray_wolf wrote: 0dBFS+ Levels in Digital Mastering http://service.tcgroup.tc/media/Leve..._AES109(1).pdf Interesting. I have noticed that in my better recorded stuff that the maximum levels tend to be at least -3dB and some as low as -6dB. This paper is fifteen years old, and back then this was a new problem that people hadn't really seen before. This is a problem with converter design, and it's most often a problem with trying to run a ladder dac with an I/V stage off the same rails as the ladder. In the modern sigma-delta world the problem is different, but it still is something converter designers need to be aware of. This, though, is not a problem with the sound processing, it's a problem with the converter design. You keep saying this. Exactly what is the problem? Are you referring to the 5 items on p.4 of that reference? I have a couple of remarks regarding those items (if anyone wants to hear them...): 1. If the digital filter is distorting/clipping, it's implemented wrong. Any diligent DSP designer would look out for that sort of thing. 2. I think most of the rest of these problems arise from the basic sampling theory explanation given on p.2. Several years ago I proved that, with the theoretical brick-wall lowpass interpolation filter, it is possible that a limited-range digital signal can give rise to an infinite analog output. This was an analytical result, but it can have practical implications, namely, that, given there are no infinite-length filters in the real-world, the DAC designer should be able to specify the absolute maximum analog output level the DAC will produce, OK, a little crow I'll eat... With a delta-sigma DAC, there is non-linearity (requantization) involved. This can make it impossible or much more difficult to analyze the maximum output level. -- Randy Yates Digital Signal Labs http://www.digitalsignallabs.com |
#59
Posted to rec.audio.pro
|
|||
|
|||
0dBFS+ Levels in Digital Mastering
On 10/13/2015 10:10 AM, wrote:
But terminology gets in the way of our understanding each other, Mike. I understand CONCEPTS Mike - but fail like our previous president at verbalizing them. This is why I suggested testing something that you could quantify rather than depending on what you think you know and hear. It's much easier to see what a level control is doing when looking at a steady or periodic signal than when looking at a music waveform. -- For a good time, visit http://mikeriversaudio.wordpress.com |
#60
Posted to rec.audio.pro
|
|||
|
|||
0dBFS+ Levels in Digital Mastering
Mike Rivers wrote: "On 10/13/2015 10:10 AM, wrote:
But terminology gets in the way of our understanding each other, Mike. I understand CONCEPTS Mike - but fail like our previous president at verbalizing them. This is why I suggested testing something that you could quantify rather than depending on what you think you know and hear. It's much easier to see what a level control is doing when looking at a steady or periodic signal than when looking at a music waveform. - show quoted text -" Or for a more real-life test: LISTEN to a song without level control and then with it it, to HEAR what is going on. As others have said in this group, loudness is personal and hard to quantify, I.E. to express in mathematical or tech terms, such as "avg" "RMS", and "peak". |
#61
Posted to rec.audio.pro
|
|||
|
|||
0dBFS+ Levels in Digital Mastering
Randy Yates wrote:
You keep saying this. Exactly what is the problem? Are you referring to the 5 items on p.4 of that reference? The whole point of the problem described in that paper is that you can have a waveform whose peak is greater than 0dBFS even though the highest sample in the waveform is only at 0dBFS... there can be an intersample peak which is higher than any of the samples. And... some converters were not designed with that in mind, and they clip the peak. I have a couple of remarks regarding those items (if anyone wants to hear them...): 1. If the digital filter is distorting/clipping, it's implemented wrong. Any diligent DSP designer would look out for that sort of thing. Yes. 2. I think most of the rest of these problems arise from the basic sampling theory explanation given on p.2. Several years ago I proved that, with the theoretical brick-wall lowpass interpolation filter, it is possible that a limited-range digital signal can give rise to an infinite analog output. This was an analytical result, but it can have practical implications, namely, that, given there are no infinite-length filters in the real-world, the DAC designer should be able to specify the absolute maximum analog output level the DAC will produce, Yes. Fifteen years ago, though, people weren't thinking so much about this. --scott -- "C'est un Nagra. C'est suisse, et tres, tres precis." |
#62
Posted to rec.audio.pro
|
|||
|
|||
0dBFS+ Levels in Digital Mastering
Randy Yates wrote:
OK, a little crow I'll eat... With a delta-sigma DAC, there is non-linearity (requantization) involved. This can make it impossible or much more difficult to analyze the maximum output level. This is why we have breadboards and scopes and computer simulation, so we can actually test things before burning silicon! --scott -- "C'est un Nagra. C'est suisse, et tres, tres precis." |
#63
Posted to rec.audio.pro
|
|||
|
|||
0dBFS+ Levels in Digital Mastering
On Tuesday, October 13, 2015 at 3:24:41 AM UTC-4, Trevor wrote:
On 13/10/2015 7:40 AM, geoff wrote: On 13/10/2015 2:25 a.m., Scott Dorsey wrote: geoff wrote: No CD that I've purchased in the last decade exhibit clipping that I've noticed, though a few are over-compressed to hell (not many fortunately). And some I've been prompted to actually check ! Not much 'current' pop though I concede. The problem is that in the digital world, clipping is whatever you define it as. I tend to set metering so three consecutive FS samples light the over light, and so that is clipping. I draw the line at 2, though 3 seems to be the commonly accepted criteria. Yep, and yet MANY Cd's have HUNDREDS of consecutive samples at maximum level :-( No-one can argue THAT is *NOT* clipping simply because the CD is normalised to -0.3dBFS, CAN THEY? Or are some people who claim to be "pro's" really that stupid? Trevor. My concern is the use of a graphic equalizer by the listener, if you allow no headroom. But, let's all consider... When I mixed that Joan Jett song, one stereo channel has a decent size peak in it that limited the rest of the audio. Was I to leave it be or suppress it. It did nothing for the audio. Jack |
#64
Posted to rec.audio.pro
|
|||
|
|||
0dBFS+ Levels in Digital Mastering
At what point does limiting end and clipping begin? there __is__ a definitive answer to that question with clipping , the gain changes fast enough to follow the cycle by cycle waveform ____and therefore creates harmonic and intermodulation distortion___. If you clip a sine wave, you can hear the harmonics. Mathematically, this is a non-linar process. Anything slower than that, is limiting, compression or AGC. These are mmathematically linear process and do not create harmonics or intermod. If you limit or compress or AGC a sine wave, you will not hear harmonics. (I'm not saying these are good, they can still ruin the aesthtics by squashing the dynamic range, but they do not cause harmonics or intermod. Clipping casues harmonics and intermod. Matematically, clipping alone creates new frequencies, The others do not. Mark |
#65
Posted to rec.audio.pro
|
|||
|
|||
0dBFS+ Levels in Digital Mastering
:
Hmmm.. When I limit(flat top) a pure sine wave, even by just .5dB, it sounds different to me |
#66
Posted to rec.audio.pro
|
|||
|
|||
0dBFS+ Levels in Digital Mastering
On 10/13/2015 2:34 PM, wrote:
Hmmm.. When I limit(flat top) a pure sine wave, even by just .5dB, it sounds different to me That's good, because when you clip off the tops it has more frequencies in it than without. But this isn't always a bad thing. Adding harmonics can make a boring timbre sound more musically interesting. It's done all the time, sometimes even naturally. But whether it's beneficial or harmful, it's still distortion. Try looking at the before and after with a spectrum analyzer and you'll have a better understanding of _why_ they sound different. If you're using a DAW that will host VST plug-ins (most can) SPAN from Voxengo is an excellent and free spectrum analyzer plug-in. -- For a good time, visit http://mikeriversaudio.wordpress.com |
#67
Posted to rec.audio.pro
|
|||
|
|||
0dBFS+ Levels in Digital Mastering
Mike Rivers wrote: "On 10/13/2015 2:34 PM, wrote:
Hmmm.. When I limit(flat top) a pure sine wave, even by just .5dB, it sounds different to me That's good, because when you clip off the tops it has more frequencies in it than without. But this isn't always a bad thing. Adding harmonics can make a boring timbre sound more musically interesting. It's done all the time, sometimes even naturally. But whether it's beneficial or harmful, it's still distortion. Try looking at the before and after with a spectrum analyzer and you'll have a better understanding of _why_ they sound different. If you're using a DAW that will host VST plug-ins (most can) SPAN from Voxengo is an excellent and free spectrum analyzer plug-in. - show quoted text -" One area it's not good is in many reissues labeled as "remastered", where a lot of the highest peaks are hard-limited, on top of the whole being compressed. It definitely sounds different, not just louder. |
#68
Posted to rec.audio.pro
|
|||
|
|||
0dBFS+ Levels in Digital Mastering
On Tuesday, October 13, 2015 at 2:21:21 PM UTC-4, wrote:
At what point does limiting end and clipping begin? there __is__ a definitive answer to that question with clipping , the gain changes fast enough to follow the cycle by cycle waveform ____and therefore creates harmonic and intermodulation distortion___. If you clip a sine wave, you can hear the harmonics. Mathematically, this is a non-linar process. Anything slower than that, is limiting, compression or AGC. These are mmathematically linear process and do not create harmonics or intermod. If you limit or compress or AGC a sine wave, you will not hear harmonics. (I'm not saying these are good, they can still ruin the aesthtics by squashing the dynamic range, but they do not cause harmonics or intermod. Clipping casues harmonics and intermod. Matematically, clipping alone creates new frequencies, The others do not. Mark I tend to think, if I could actually clip a sine-wave, I would not hear anything for a brief moment. You couldn't clip with magnetic tape, since it just keeps saturating more. However, you may be able to do it with digital, but maybe not, where it ends up as DC, then I'd think you'd hear harmonics. Great subject. Jack |
#69
Posted to rec.audio.pro
|
|||
|
|||
0dBFS+ Levels in Digital Mastering
On Tuesday, October 13, 2015 at 2:34:38 PM UTC-4, wrote:
: Hmmm.. When I limit(flat top) a pure sine wave, even by just .5dB, it sounds different to me if you flat top it, then it is not limiting, it is clipping that is the essesnce of the difference Mark |
#70
Posted to rec.audio.pro
|
|||
|
|||
0dBFS+ Levels in Digital Mastering
On Tuesday, October 13, 2015 at 2:55:31 PM UTC-4, wrote:
Mike Rivers wrote: "On 10/13/2015 2:34 PM, wrote: Hmmm.. When I limit(flat top) a pure sine wave, even by just .5dB, it sounds different to me That's good, because when you clip off the tops it has more frequencies in it than without. But this isn't always a bad thing. Adding harmonics can make a boring timbre sound more musically interesting. It's done all the time, sometimes even naturally. But whether it's beneficial or harmful, it's still distortion. Try looking at the before and after with a spectrum analyzer and you'll have a better understanding of _why_ they sound different. If you're using a DAW that will host VST plug-ins (most can) SPAN from Voxengo is an excellent and free spectrum analyzer plug-in. - show quoted text -" One area it's not good is in many reissues labeled as "remastered", where a lot of the highest peaks are hard-limited, on top of the whole being compressed. It definitely sounds different, not just louder. of course music, when limited sounds different because the dynamic range is changed... not becasue there are harmincs added.... I was talking about a lone sine wave If a lone sine wave sounds (and looks) different, then it is clipping. Mark |
#71
Posted to rec.audio.pro
|
|||
|
|||
0dBFS+ Levels in Digital Mastering
|
#72
Posted to rec.audio.pro
|
|||
|
|||
0dBFS+ Levels in Digital Mastering
|
#73
Posted to rec.audio.pro
|
|||
|
|||
0dBFS+ Levels in Digital Mastering
|
#74
Posted to rec.audio.pro
|
|||
|
|||
0dBFS+ Levels in Digital Mastering
On 14/10/2015 8:19 a.m., JackA wrote:
O I tend to think, if I could actually clip a sine-wave, I would not hear anything for a brief moment. You couldn't clip with magnetic tape, since it just keeps saturating more. However, you may be able to do it with digital, but maybe not, where it ends up as DC, then I'd think you'd hear harmonics. Great subject. Jack " I tend to think " - yeah right .... Easy to clip program with the electronics prior to the actual tape. geoff |
#75
Posted to rec.audio.pro
|
|||
|
|||
0dBFS+ Levels in Digital Mastering
geoff:
STOP telling me what I don't understand! You don't know me at all. That said, here is a good diagram illustrating clipping: http://www.gmarts.org/pix/fx/fx_clip1.jpg When I utilize "hard-limiter" in a DAW, the result zoomed in looks exactly like the red lines in that diagramSo does anyone know what the F- limiting looks like in a DAW?! |
#76
Posted to rec.audio.pro
|
|||
|
|||
0dBFS+ Levels in Digital Mastering
On 13/10/2015 8:42 p.m., Trevor wrote:
Bull****!! A limiter by it's very nature will produce CLIPPING when it exceeds it's knee and reaches it's maximum level. That's the whole point of the device after all. One may choose to use it so it never exceeds the knee (and hard limiting may not even have a knee!) but IF it does exceed absolute maximum, then CLIPPING will occur! That you choose to believe it is somehow not clipping just because YOU prefer to call it limiting only shows you ignorance!!! Bull**** you ! A crude limiter might - a sophisticated limiter will produce a 'knee' region that should not resemble clipping if you look closer. Glad my ignorance keeps such good company. Yep, but limiting is NOT necessarily the same as compression. You and many others still seem to be totally confused about that. And of course one can drive a compressor into clipping as well! Clipping is clipping no matter how it is achieved. Arguing that it is not is simply TOTAL BS! Limiting should be a process closer to extreme compression that clipping. geoff |
#77
Posted to rec.audio.pro
|
|||
|
|||
0dBFS+ Levels in Digital Mastering
On 13/10/2015 8:24 p.m., Trevor wrote:
Yep, and yet MANY Cd's have HUNDREDS of consecutive samples at maximum level :-( No-one can argue THAT is *NOT* clipping simply because the CD is normalised to -0.3dBFS, CAN THEY? Or are some people who claim to be "pro's" really that stupid? Trevor. Was I ever suggesting that some (even many) CDs or other digital media does not exhibit clipping ? Certainly not many of the sort that I purchase. geoff |
#78
Posted to rec.audio.pro
|
|||
|
|||
0dBFS+ Levels in Digital Mastering
On Tuesday, October 13, 2015 at 4:20:19 PM UTC-4, wrote:
On Tuesday, October 13, 2015 at 2:34:38 PM UTC-4, wrote: : Hmmm.. When I limit(flat top) a pure sine wave, even by just .5dB, it sounds different to me if you flat top it, then it is not limiting, it is clipping that is the essesnce of the difference Mark i shoudld clarfiy this... if you flat-top the WAVEFORM, it is clipping.. if you see a sine wave waveform, and it is flat topped, that is clipping... as discussed in another thread, the WAVEFORM is not the same as the ENVELOPE. Limiting can and will flat top an envelope but not the waveform. The difference between clipping and limiting, is closely related to the difference beween a WAVEFORM and an ENVELOPE. Mark |
#79
Posted to rec.audio.pro
|
|||
|
|||
0dBFS+ Levels in Digital Mastering
On 14/10/2015 2:07 a.m., Mike Rivers wrote:
On 10/12/2015 6:24 PM, geoff wrote: My car iPod Touch has a automatic replay level function based on (I think) a peak-level parameter encoded into the file. When it works on a particularly 'different level' track there seems to be a second-long level adjustment at the commencement of a song I would like to have something like that in my car. It might not work as well for me as it does for you, however, because for me, a "song" (one file) is a two hour radio program of music that often has song-to-song level variations. I've been trying to score a cheap couple of generations old iPod Touch so I could use it for some audio apps that aren't available for my Androids, but so far no luck. They seem to hold their value nearly as well as microphones. I bought an iPod Touch (4G) 4+ years ago primarily for Signal Suite , Signal Scope Pro, and Guitar Toolkit. Pretty sure there will be Android equivalents by now. ..... with the bonus of direct USB connection/control to my car stereo. All music exclusively ripped from CD to ALAC. geoff |
#80
Posted to rec.audio.pro
|
|||
|
|||
0dBFS+ Levels in Digital Mastering
..
That said, here is a good diagram illustrating clipping: http://www.gmarts.org/pix/fx/fx_clip1.jpg When I utilize "hard-limiter" in a DAW, the result zoomed in looks exactly like the red lines in that diagramSo does anyone know what the F- limiting looks like in a DAW?! Yes, you must Zoom in to see the WAVEFORM rather than the ENVELOPE. And if the WAVEFORM appears as in the .jpg, then i would call that clipping (as the link does) even if the box that did it is called a limiter. Mark |
Reply |
Thread Tools | |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Forum | |||
Reference Levels for Editing, Broadcasting and Mastering | Pro Audio | |||
Digital Levels on CD's | Pro Audio | |||
Mastering output levels. | Pro Audio | |||
Mixdown Levels--Mastering? | Pro Audio | |||
"0dBFS+ Level in Audio Production." | Pro Audio |