Do ME's purposely clip?

  • Thread starter Thread starter fenix
  • Start date Start date
think of it this way:


any wave, squared and flat or not, is a representation of the sum of a (usually) huge amount of perfect sine waves operating at different frequencies. and when it is played back, it SOUNDS like a huge amount of perfect sine waves operating at different frequencies.

a clipped wave creates an effect similar to a square or sawtooth wave.

these wave types are the sum of many, many different sine waves that are layered so precisely with exactly the right frequencies that the overall wave looks square.

however, when it is played back, all the harmonics of these individual sine waves sound out.
these harmonics are what makes the wave unique from a pure sine wave.
a 60Hz square wave is a 60Hz sine wave + many many other higher-frequency sine waves. if you listen closely, you can pick out each individual frequency being sounded.

Thus, when a wave gets flat-topped for many samples in a row, it is played back like it is the sum of a bunch of different harmonics that weren't actually there in the original.

These harmonics are what we hear as distortion.
 
I guess you're talking about Christopher Marlowe, right
Click to expand...

LOL - yes of course ! Ha Ha or maybe Marlow Thomas - Ha Ha Thank you ! kylen
 
kylen said:


OK fenix - great thread you started here ! I owe you a chili dog next time I'm in Cincy ! Ha Ha - or are you a 4-way guy ? :)
kylen
Click to expand...

We don't call em "chilli dogs"--they're Coneys! :)
 
http://www.nowhereradio.com/artists/album.php?aid=1410&alid=-1

Check out the song "routine". This is an example where I slammed the song with the L2. After examing the file, there were NO flat tops that I could see. I raised the volume of the mastered track by 2db which CLIPPED the wave. I examined the clipped wave at questionable points and there were no flats tops that had more than 5 samples.

There's no audible distortion at these flat topped points either. The song is VERY close if not as louder (maybe louder than some) commercial recordings.


It seems to me this is what is going on in the pro Mastering realm in order to squeeze a couple more decibels from a track. When I look at Audioslave or similar hot cds and see the amount of flat tops, I can't imagine a limiter doing this.


And someone said earlier I think that a cd duplication plant would be pissed if they had a CD with overs. You can't have any information above 0db in the digital world!
 
sjoko2 said:
Do mastering engineers clip things on purpose? Nope, definitely not.
Click to expand...
This is incorrect.

I am dealing with one such thing at the moment - the opening of a track clips out, only when transferred to a CD - it does not happen at its original 24/96 state, it does not happen when converted to 16/44.1, but always on a CD, even with the part eq'd and reduced in volume. The reason? Its not clipping.......it a harmonics issue the CD format cannot deal with.
Click to expand...
Umm, that's impossible. If you were talking about vinyl, you'd be fine, but with dig this will never happen.

Next one. The “flat tops” as seen in the examples. Don’t forget this does NOT have to be clipping, it simply can be a limiter. I’m doing some dance/trance stuff at the moment, and I can guarantee you that the tops are all flat due to the huge amount of processing on the kick mainly, but it does NOT clip out on a disk, its simply the waveform.
Click to expand...
Once again, I think you' re mistaken. Clipping and full scale are not tied together - clipping is when something is 'flat-topped' - this does not happen naturally and does not happen with a properly designed limiter.
 
sjoko2 said:
Without meaning to talk down to you,
Click to expand...
Off to a good start...

If you had that, you would know that its possible to create harmonic that, when transferred to the CD medium, sounds like clipping and popping, as the media cannot cope with the sound.
Click to expand...
Ummm, huh?

An effect you can also recreate by, for instance, recording a particular note on a Grand, and dubbing it with the same note, detuned by an extremely very fine margin.
If you would burn the same files to a different medium, for instance encode it to DVD format, the clipping effect would not occur.
Click to expand...
I would like for you to present an example of this, please. You can upload them to my FTP so that I may witness it for myself.
 
Mr. bblackwwod,
Very cool of you to come over here and straighten us out on the clipping thing ! Earlier I thought that was the last thing an ME would want to do ! Ha Ha - a little more research proves otherwise. :cool:
kylen
 
yeah, i'd like to see that example too.


and also, i'd be interested in seeing this "wave with 90% flat-topped samples that has no distortion"

show me i'm wrong and i'll back off!
 
3 samples or more....

I've read some material from Bob Katz years ago in one of his articles and he stated that 3 samples or more is considered an "over", so technically there are overs in digital audio.

One other thing that I'd like to mention is that those sawed off wave forms that are being shown are done with a limiter and I can prove it, if for some reason someone doesn't think so.

The distortion factor when doing level correction is present when you start limiting below the 6db average to peak ratio. Though this is the case great results can be achieved even at this level.

sonicpaint
:D
 
Re: 3 samples or more....

sonicpaint said:
I've read some material from Bob Katz years ago in one of his articles and he stated that 3 samples or more is considered an "over", so technically there are overs in digital audio.
Click to expand...

I understand what you're saying, but that's oversimplifying. 3 samples at 0db is generally considered to be a clip (though it might not be a clip every time.) And clips are caused by audio that goes "over" 0db.
But in the digital realm, once an ADC or program sees material over 0db, it turns it into a clip, and it is no longer "over."
So you could say the terminology applies insofar that the material of those 3 samples was once "over." but it is not, in itself, over.

One other thing that I'd like to mention is that those sawed off wave forms that are being shown are done with a limiter and I can prove it, if for some reason someone doesn't think so.
Click to expand...

Which sawed off wave forms are you talking about?
No one's said that you CAN'T distort or make flat-tops with a limiter, it's just that if you have a good look-ahead limiter and you use it properly (this doesn't mean "weakly"), it will never create flat-tops.

The distortion factor when doing level correction is present when you start limiting below the 6db average to peak ratio. Though this is the case great results can be achieved even at this level.

sonicpaint
:D [/B]
Click to expand...

the distortion level has absolutely nothing to do with the average to peak level. there is no magic number there. if the music was, to begin with, less than 6db RMS-to-peak, how could you call it distorted even after limiting?
distortion is caused by some manipulation of the shape of a waveform, generally when the waveform becomes more square-edged. it has nothing to do with the volume or dynamic range. you could have a recording with 2db dynamic range that isn't distorted.
 
nessbass said:
@regebro:

Yes, that scale is in samples. And it's, as I said, supposed to be a kick drum hit with bass and guitars going on at the same time.
BTW, that 'sawy' waveform starts right where you see it and also ends within that picture shown, so it's really just

up-down-up-down,

that's it.
Click to expand...

Could the kick drum be sampled?
 
regebro said:
Could the kick drum be sampled?
Click to expand...
I don't think so, The kicks all look different and they all have different amounts of sawyness. That's also why I thought it had to be the processing, cause it seems to be volume dependent.
 
it could be wacky compression with like 2 ms attack and 0 release, or perhaps a transient processor.
 
I understand what you're saying, but that's oversimplifying. 3 samples at 0db is generally considered to be a clip (though it might not be a clip every time.) And clips are caused by audio that goes "over" 0db.
Click to expand...

I was simply sharing what I read in an article that Bob Katz wrote some time ago. It is quite simplified but it make sense to me. If he says 3 samples are considered an over, then I would assume that he means every time there is 3 consecutive samples that "go over" then it's considered as an "over". He also went on to say (if I remember correctly) that the only way to truly measure digital overs is with over counters.

Which sawed off wave forms are you talking about?No one's said that you CAN'T distort or make flat-tops with a limiter, it's just that if you have a good look-ahead limiter and you use it properly (this doesn't mean "weakly"), it will never create flat-tops.
Click to expand...

Someone mentioned that they ran a file through a Waves L2 compressor and tried to get that flat tops with out success and went on to ask how it was done? So I mentioned that is was done with a limiter. Your meaning of "proper" use of a limiter is humorous at best for the simple reasons that your thoughts of how to use a limiter, is your own personal thoughts and feelings and have now bearing on the thoughts and opinions of others who use this tool. I think you know and have heard that there is no rules to using compression even at high ratios of 10:1 or greater. I like to ask you a question though. What type of setting are you using when applying a limiter to any given signal? Just curious by the way.

the distortion level has absolutely nothing to do with the average to peak level. there is no magic number there. if the music was, to begin with, less than 6db RMS-to-peak, how could you call it distorted even after limiting?
Click to expand...

When I made that comment I was referring to a whole song not individual instruments, though if you did mean a whole song and the average to peak ratio is already below 6db then the dynamics are shot already. Limiting the file anymore will just finish killing what little dynamics are left. Hmm..lets see if I can explain myself a little better.

Let's say you have a mixdown stereo file and the average to peak ratio is greater than 6db with the largest transients peaking at 0db. In this case I suggested that limiting 6db at this point would start causing noticeable artifacts, though still being able to achieve a reasonable sounding file. I never said that I had a "magic number" those are your words. What I do know is that many songs today have sawed off transients leaving the average to peak ratio much less than 6db (in some cases as low as 3db)and thus degrading the file, the more the "loud" factor is pushed.

The 6db average to peak ratio is what I try to consider my cap in regards to the "loudness wars". The reason for this is because I can still achieve a hot master without giving up too much quality of the original file.

I don't know if you misunderstood some of my points or I wasn't specific enough but I hope I cleared things up.

later,
sonicpaint
:D
 
You must log in or register to reply here.
Back
Top