Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[malice]

What kind of treatment does it to the signal. I understand it is not a simple 6dB boost ... Does it involve some more or less complex tape compression modelisation or is it more simple than that.

I only used it once so far on a rough mix, it sounded rather nice and usable, not as drastic as a cranesong would do, but a bit fuller (in a good way)...


Please, can you tell us a bit more without unveiling a secret receipe

malice

[danlavry]

“What kind of treatment does it to the signal. I understand it is not a simple 6dB boost ... Does it involve some more or less complex tape compression modelisation or is it more simple than that.
I only used it once so far on a rough mix, it sounded rather nice and usable, not as drastic as a cranesong would do, but a bit fuller (in a good way)...
Please, can you tell us a bit more without unveiling a secret receipe  

malice”

Let’s begin with WHY I did that digital saturation:
So many audio people are in a “mad race” to make things louder. I am not talking about turning the volume knob to louder. The race is about being louder relative to another song, or commercial… The louder one gets more noticed, so they say.

Well, so many mastering guys, including some of the very well known, keep pushing the level up, and of course at some point the signal clips. I wish people did not clip but many do. Some music tolerates it better, some worse.

So here is what the digital soft saturation does:
It boosts the signal by 6dB, thus it is a lot louder. Of course if that is all it did, then anything over -6dBFS (the top 6dB) would be clipped.
So in order to avoid clipping we modify the process as follows:
All signals up to -12dBFS are boosted by 6dB. The range between -12dBFS and 0dBFS (full scale) gets some “special treatment”. It gets “squeezed into what is left, which is the top 6dB.

The secret is how to do it (the squeezing) well. Clearly it is not a straight line approach. I looked at tapes and the higher the signal gets, the more magnetized the tape gets so the impact of the peaks is less as the signal grows. That is the story in a nutshell.

BR
Dan Lavry
 

[ammitsboel]

Hi Lavry.

It's nice to finally have a talk with you

When the "big guys" clip their converters it seems like it's an artistic choice that makes the sound more edgy and creating the illusion that the musicians are playing more edgy "with a bad attitude"  than they really are.
My gues would be that the sound they are getting needs the AD chip clipping to get like that and a limiter(or other circuitry) probably can't do it.

but what are your thoughts about this Lavry, does your new circuit make the same edgy sound without digital overs?

What I'm wondering about is why some albums in this "converter clipping" catagory sounds "closed in" when compared to oders in the same catagory and done my the same engineer...? must be a less good mix together with a nice saturated converter mastering.

Best Regards

[jazzius]

ammitsboel wrote on Wed, 20 October 2004 01:52

Hi Lavry. When the "big guys" clip their converters it seems like it's an artistic choice that makes the sound more edgy and creating the illusion that the musicians are playing more edgy "with a bad attitude"  than they really are. My gues would be that the sound they are getting needs the AD chip clipping to get like that and a limiter(or other circuitry) probably can't do it.

This is a great example of the insane mental gymnastics that some people go thru to convince themselves that something that sounds like sh*t actually sounds good.

[bblackwood]

jazzius wrote on Wed, 20 October 2004 01:34

ammitsboel wrote on Wed, 20 October 2004 01:52 Hi Lavry. When the "big guys" clip their converters it seems like it's an artistic choice that makes the sound more edgy and creating the illusion that the musicians are playing more edgy "with a bad attitude"  than they really are. My gues would be that the sound they are getting needs the AD chip clipping to get like that and a limiter(or other circuitry) probably can't do it. This is a great example of the insane mental gynastics that some people go thru to convince themselves that something that sounds like sh*t actually sounds good.

He didn't say it sounded good...

Are you suggesting that there's another reason some mastering engineer's would rather clip the signal than limit it? Too lazy to reach over and dial in the limiter?

[danlavry]

ammitsboel wrote on Wed, 20 October 2004 01:52

Hi Lavry. When the "big guys" clip their converters it seems like it's an artistic choice that makes the sound more edgy and creating the illusion that the musicians are playing more edgy "with a bad attitude"  than they really are. My gues would be that the sound they are getting needs the AD chip clipping to get like that and a limiter(or other circuitry) probably can't do it.

As stated earlier, I did that tape like saturation because I wanted to offer a "compromise" for those that just wanted to push thing to the clip point and beyond. As far as I know, clipping does a lot more than make things edgy. Some people say it ruins the image, but that is a sonic comment, and I rather stay on technical side of the issue.

The comment about the AD doing the clipping: Remember that a clipped signal contains huge bandwidth and clipping in the wrong place can cause a lot of aliasing. Processing the signal either before or after the AD can be done with much less aliasing.

Also, with sigma delta AD's you really do not want to get to a limiting point. The user does not know it, but the front end of the AD (the modulator) is still in its active region even if the digital output is already clipping. That extra margin is there to try an avoid the sigma delta modulator from losing it's tracking. If you push too much signal and run over that margin, you have to re establish proper operation. It is similar to having to reset your gear. In fact some chips do reset the modulator and it takes time, thus pretty noticeable...

BR
Dan Lavry

[ammitsboel]

danlavry wrote on Thu, 21 October 2004 01:10

ammitsboel wrote on Wed, 20 October 2004 01:52 Hi Lavry. When the "big guys" clip their converters it seems like it's an artistic choice that makes the sound more edgy and creating the illusion that the musicians are playing more edgy "with a bad attitude"  than they really are. My gues would be that the sound they are getting needs the AD chip clipping to get like that and a limiter(or other circuitry) probably can't do it. As stated earlier, I did that tape like saturation because I wanted to offer a "compromise" for those that just wanted to push thing to the clip point and beyond. As far as I know, clipping does a lot more than make things edgy. Some people say it ruins the image, but that is a sonic comment, and I rather stay on technical side of the issue. The comment about the AD doing the clipping: Remember that a clipped signal contains huge bandwidth and clipping in the wrong place can cause a lot of aliasing. Processing the signal either before or after the AD can be done with much less aliasing. Also, with sigma delta AD's you really do not want to get to a limiting point. The user does not know it, but the front end of the AD (the modulator) is still in its active region even if the digital output is already clipping. That extra margin is there to try an avoid the sigma delta modulator from losing it's tracking. If you push too much signal and run over that margin, you have to re establish proper operation. It is similar to having to reset your gear. In fact some chips do reset the modulator and it takes time, thus pretty noticeable... BR Dan Lavry

Thanks Dan,

There is no doubt about that converter clipping destroys the audio image bit by bit...

But what i was saying was that it was my impression that the suff that comes together with converter clipping was to some extension a desired effect.
How else can you make a such "flying" high end as some of these converter clipped albums have.
I'm not saying that it sounds great, it actually sounds diffuse and unatural, but also funny and interesting to some extent.

I think that these guys with 20+ or more in the buisness would have found another way to limit than this if they wanted to.
Or do you think they are just making the same mistake again and again?

How does the "re establish of proper operation" sound? is there a break up?

Best Regards,

[jazzius]

The only thing you could say in favour of clipping is that it'll have more punch then something else that has been limited to achieve the same RMS value......for the simple reason that although the digital bits stop at 0dBFS, the speaker cones don't....they keep on goin'. A limter will slow the transients as they approach 0 in order to prevent flat-topping.

But of course the clipped stuff will sound less punchy then something that sounds good....and we'll tire of it very  quickly because of the musically unrelated distortion.

[bobkatz]

danlavry wrote on Wed, 20 October 2004 20:10

Processing the signal either before or after the AD can be done with much less aliasing.

As in "clipping". A big "left handed compliment"!

BK

[danlavry]

bobkatz wrote on Thu, 21 October 2004 16:34

danlavry wrote on Wed, 20 October 2004 20:10 Processing the signal either before or after the AD can be done with much less aliasing. As in "clipping". A big "left handed compliment"! BK

Well, I should have stated it differently. If you run into the supply rail before the AD, of course it is bad news. The analog signal will have some serious distortions. Some of the distortion energy is within the hearing range, and some is above it. If you listen to that sound, it will be distorted, but the distortion is "limited" in some way. A steady state tone, for example, will still have an integral relationship between the fundamental and the harmonics.

Take that analog signal and run it into an AD. It first gets to pass through and anti aliasing filter, thus removing that high frequency energy. Doing that, the part of the signal you hear is converted correctly (in theory) and the high frequency was removed thus caused no harm.

On the other hand, if the clipping occurred AFTER the analog filter, the high frequency energy is NOT removed and will alias into the audio range. That kind of a distortion is so much worse, with tones at "non harmonic locations"...

So I should have said: Clipping is a bad problem. Clipping inside the AD is much worse.

BR
Dan Lavry