GroupDIY

General Discussions => Studio A => Topic started by: pucho812 on August 13, 2018, 08:19:00 PM

Title: Clipping converters
Post by: pucho812 on August 13, 2018, 08:19:00 PM
I read a lot of so called mastering guys.  They all talk of clipping converters.  I can only assume they are mis-using the terminology, or something.  Am I missing something here?
Title: Re: Clipping converters
Post by: JohnRoberts on August 13, 2018, 08:26:01 PM
Early A/D convertors behaved badly then saturated (they rolled over to 0 again, FS rail to rail loud clicky stuff). Modern convertors are more well behaved so saturation (clipping) is relatively clean with quick recovery and no unexpected artifacts.

Not good practice (IMO) but what would I know about music these days?

  JR
Title: Re: Clipping converters
Post by: Gold on August 13, 2018, 09:07:22 PM
Nope, you heard right. Lots of MEs do it. There are long threads on GS about which converter clips best.  I don’t do it on a regular basis but if extreme level is needed sometimes there is no choice. Sometimes it actually helps. The distortion can give a snap to the leading edge which helps on some material.
Title: Re: Clipping converters
Post by: pucho812 on August 13, 2018, 09:10:08 PM
 Doesn't that give you digital distortion ? or is it just a matter of over driving the Analog stages
Title: Re: Clipping converters
Post by: EmRR on August 13, 2018, 10:04:09 PM
I don't think there's been any hip-hop or rap done without converter clipping in mastering in 2 decades, it's expected as part of the process. 
Title: Re: Clipping converters
Post by: Gold on August 13, 2018, 10:36:22 PM
I don't think there's been any hip-hop or rap done without converter clipping in mastering in 2 decades, it's expected as part of the process.


Yeah, it’s part of the sound.  Clipping a Prism AD2 is a sound that is part of the lexicon so to speak.

You are clipping both the analog and digital stages.
Title: Re: Clipping converters
Post by: pucho812 on August 14, 2018, 12:49:37 AM
Sounds awful to me.  lol
Title: Re: Clipping converters
Post by: Gold on August 14, 2018, 02:25:00 AM
Sounds awful to me.  lol

Not if you do it right. 
Title: Re: Clipping converters
Post by: Rochey on August 14, 2018, 08:57:22 AM

There are still plenty of products on the market that use old converters that have horrible saturation behaviour. (Rollover, where it goes from max positive to max negative, is one of the most horrid).
The last converter I worked on (PCM1865) was defined very clearly to "clip like an opamp", so  you'd clip at 0x0FFF (24 bit) until the signal comes back within range.

However, there are a few arguments to avoid doing this

A) Go much above the input signal range and you'll start shorting the signal through the device ESD diodes to the supply rail.  Not good for your source
B) THD near clipping increases exponentially. Given the extended dynamic range of mastering ADC's, you have more dynamic range than you have THD. Better to sacrifice 2 or 3 dB of headroom and get both good SNR and THD.
C) Why commit distortion to a recording? Wouldn't you be better recording clean and adding distortion as a post process?

Title: Re: Clipping converters
Post by: JohnRoberts on August 14, 2018, 09:51:10 AM
There are still plenty of products on the market that use old converters that have horrible saturation behaviour. (Rollover, where it goes from max positive to max negative, is one of the most horrid).
The last converter I worked on (PCM1865) was defined very clearly to "clip like an opamp", so  you'd clip at 0x0FFF (24 bit) until the signal comes back within range.
Not surprising, the customer expects it to clip cleanly or doesn't even think about it.
Quote
However, there are a few arguments to avoid doing this

A) Go much above the input signal range and you'll start shorting the signal through the device ESD diodes to the supply rail.  Not good for your source
prudent design could limit over drive capability with design of analog side input stage.
Quote
B) THD near clipping increases exponentially.
not sure what you are describing. Some analog paths saturate fairly suddenly, some over a small range. I would expect saturation in a A/D convertor to be more sudden, than gradual. Note: there is one digital console with a soft limiter built into the mic preamp that some customers apparently like.
Quote
Given the extended dynamic range of mastering ADC's, you have more dynamic range than you have THD.
not sure what this means?
Quote
Better to sacrifice 2 or 3 dB of headroom and get both good SNR and THD.
that is the classic trade off between noise floor and saturation. For years operators with a little(?) digital knowledge hit the convertors hard to avoid quantization noise. This is not an issue with modern convertors that have an analog sounding noise floor.
Quote
C) Why commit distortion to a recording? Wouldn't you be better recording clean and adding distortion as a post process?
Mastering is the last step of the process where we stick a fork in it...  In my personal opinion clipping is undesirable, but they aren't mixing for me as their paying customers.

JR
Title: Re: Clipping converters
Post by: EmRR on August 14, 2018, 10:03:13 AM
C) Why commit distortion to a recording? Wouldn't you be better recording clean and adding distortion as a post process?

Mastering is the last step of the process where we stick a fork in it...  In my personal opinion clipping is undesirable, but they aren't mixing for me as their paying customers.

While you guys are raising these questions, I'm telling you IT IS AN OPERATIONAL STANDARD within particular genre, and has been for a very long time.  It's all over everything you hear in those genre.


Title: Re: Clipping converters
Post by: JohnRoberts on August 14, 2018, 10:12:12 AM
While you guys are raising these questions, I'm telling you IT IS AN OPERATIONAL STANDARD within particular genre, and has been for a very long time.  It's all over everything you hear in those genre.
I am not raising this as a new question... Long before there were A/D convertors people saturated their analog audio paths, so good design involved fast/clean overload recovery.

The customer is always right, even when.....

While I avoid listening to some genres of music intentionally, from what I hear blaring out open car windows as they drive by, clipping on top of clipping is acceptable to many listeners.

JR
Title: Re: Clipping converters
Post by: Gold on August 14, 2018, 10:23:33 AM
Making a technical argument against clipping is  a situation where things around here get hilarious. Non users pulling numbers out of space to describe sound. Yeah, right.
Title: Re: Clipping converters
Post by: Script on August 14, 2018, 10:48:08 AM
I wake up to the sound of music
AD converter comes to me
Speaking words of wisdom
Let it clip

Let it clip, let it clip, let it clip, yeah, let it clip
There will be distortion
Them dudes don't give a...(bliep)
 8)
Title: Re: Clipping converters
Post by: JohnRoberts on August 14, 2018, 10:55:49 AM
Making a technical argument against clipping is  a situation where things around here get hilarious. Non users pulling numbers out of space to describe sound. Yeah, right.
I find both sides of this debate(?) humorous... that horse left the barn a long time ago.

JR
Title: Re: Clipping converters
Post by: Gold on August 14, 2018, 11:07:01 AM
It like a luthier saying someone isn’t playing their  guitar right .  Who cares?
Title: Re: Clipping converters
Post by: Rochey on August 14, 2018, 11:40:00 AM
John,

this is what i'm talking about: (https://i.imgur.com/80cTeII.png)
At high amplitudes, the THD swamps the N in the THD+N :)

My recommendation with customers designing audio systems is to only allow brief transients (where clarity etc isn't really a requirements) to enter that space.

Finally - I still don't understand why you would distort using the converters instead of in a post processing algorithm?
Title: Re: Clipping converters
Post by: Gold on August 14, 2018, 12:09:09 PM
Finally - I still don't understand why you would distort using the converters instead of in a post processing algorithm?

Because professionals are able to make decisions and commit to them. Maybe it sounds better? Maybe the engineer wants to hear the finished sound before committing to the rest of the processing?

That graph illustrates perfectly why you would do it. The sharp rise in distortion happens right below 0dBfs. Material with strong transients like hip hop for instance will stay clean except for the heavy transients like bass drum or snare drum. The added distortion fools the ear into thinking the 6db you just lopped off is still there.
Title: Re: Clipping converters
Post by: Pip on August 14, 2018, 12:20:26 PM
Mastering engineers I work with will, depending on the request of myself or the artist and it is a technique some use at their own discretion, hit certain converters hard, clipping them, usually rhythmically on the downbeat so it is musical. It is not a constant or random over.
Title: Re: Clipping converters
Post by: Gold on August 14, 2018, 12:29:45 PM
I've never had a client request that I do it or don't do it. They will say they want it louder. How that happens is my job.  Anticipating the level that is expected and delivering it is also my job.
Title: Re: Clipping converters
Post by: Script on August 14, 2018, 11:24:37 PM
Which audio interfaces (equipped with which chips) are the ones that people think clip 'gracefully'? Just being curious.

I mean, what do they have/do in technical electronics terms, that other chips don't do?
Title: Re: Clipping converters
Post by: Zander on August 15, 2018, 05:01:35 AM
I didn’t read the whole thread. But I think what a lot of ME’s do is use the “overkill” protection/limiters in the convertors. At least I see the led indicators blinking when I do this in my Prism. There may be overshoot beyond that.
Title: Re: Clipping converters
Post by: JohnRoberts on August 15, 2018, 10:12:10 AM
John,

this is what i'm talking about: (https://i.imgur.com/80cTeII.png)
pretty familiar looking chart... in ideal world that onset of distortion caused by clipping would be a straight vertical line. Any slope there is perhaps an artifact of measurement. For example, tiny variation in test signal amplitude can make measureable difference in distortion at the margin, perhaps even a steep slope like shown.
Quote
At high amplitudes, the THD swamps the N in the THD+N :)
yes... and with noise still almost 80dB below -30dBFS signal is unlikely to be a problem.
Quote
My recommendation with customers designing audio systems is to only allow brief transients (where clarity etc isn't really a requirements) to enter that space.

Finally - I still don't understand why you would distort using the converters instead of in a post processing algorithm?
As has already been shared, people saturate final mixes to raise the average loudness.  As long as the clipping recovers cleanly (no burps or farts), and is limited to narrow transient events, the distortion added is HF and may even complement the transient impact. Where we get perceived audible distortion from clipping is when LF content clips, creating distortion products that are audible as distinct signals.

Objectively I hate clipping, subjectively sometimes it works, and apparently is well tolerated in some musical genres. As a convertor designer, you need to do what the customer expects (clip clean and recover fast, with no additional audible artifacts). YMMV

JR
Title: Re: Clipping converters
Post by: JohnRoberts on August 15, 2018, 10:27:10 AM
Which audio interfaces (equipped with which chips) are the ones that people think clip 'gracefully'? Just being curious.

I mean, what do they have/do in technical electronics terms, that other chips don't do?
I recall an old attempt by dbx (IIRC) in a line of GEQ where the treated the MSB with some non-linear law... This was early days for digital, and mooted by modern convertors we have now.
===
The only current example i heard about is Midas M32 but I have never seen a schematic so do not know any details. I ASSume they are using one of the sundry digitally gain controlled mic preamps so may incorporate that into some dynamic overload soft limiting.

FWIW I have been out of these design trenches for a couple decades so may not be completely up on all available technology.

JR
Title: Re: Clipping converters
Post by: EmRR on August 15, 2018, 11:12:25 AM
I didn’t read the whole thread. But I think what a lot of ME’s do is use the “overkill” protection/limiters in the convertors. At least I see the led indicators blinking when I do this in my Prism. There may be overshoot beyond that.

Prism is a diode clipper isn't it?  Very different sound from converter clipping. 
Title: Re: Clipping converters
Post by: Script on August 15, 2018, 11:52:52 PM
Not an expert in converter topology, but don't converters too feature an amp input stage with on-board (or rather in-chip) diodes or transistors? So how is that different? Better specs than thru-hole diodes/transistors? Asking cos I'm really dumb on the subject.
Title: Re: Clipping converters
Post by: Zander on August 16, 2018, 03:51:29 AM
Prism is a diode clipper isn't it?  Very different sound from converter clipping.
Yes, it is. I see now. I need to investigate the difference  with and without the overkill in circuit. I will report back. Thanks
Title: Re: Clipping converters
Post by: Whoops on August 22, 2018, 04:38:15 AM
Doesn't that give you digital distortion ? or is it just a matter of over driving the Analog stages

Depends on the converter, most high end converters will need quite a lot of input level after the first clipping Led appears, for you to listen Digital distortion.

Think of it as a Free Hard Limiter with very fast attack and very fast release.

Paul Gold is right and gave some good insight in this thread
Title: Re: Clipping converters
Post by: Whoops on October 25, 2018, 11:18:13 PM
Which audio interfaces (equipped with which chips) are the ones that people think clip 'gracefully'? Just being curious.

Any Old standard Apogee for example

AD8000
AD16
Rosetta
Title: Re: Clipping converters
Post by: hitchhiker on October 26, 2018, 01:29:03 PM
I've read that the converter I own, the UA 2192, is good for this. I've not tried it though.
Title: Re: Clipping converters
Post by: weiss on October 26, 2018, 01:44:03 PM
interesting topic!

i thought burl audio and the dave hill / crane song gear were famous for their colour and clipping abilities..
Title: Re: Clipping converters
Post by: EmRR on December 26, 2018, 10:50:01 AM
I just read a thread elsewhere about using converter clipping rather than limiting to achieve RMS levels in the -4 to -6 dBFS range, in which it was suggested that many/most Nashville mixers are doing this and printing that hot, and that’s whats sent on to labels and mastering. 

I’d like to know what % of clipped masters like this  are seen by mastering houses, and what you do with them.   It seems any changes probably require additional limiting or clipping. 
Title: Re: Clipping converters
Post by: weiss on December 26, 2018, 01:06:20 PM
i'm trying to understand this. why would one want to clip or limit before mastering?
Title: Re: Clipping converters
Post by: JohnRoberts on December 26, 2018, 02:12:52 PM
i'm trying to understand this. why would one want to clip or limit before mastering?
There is no good reason... the bad reason is to be louder.

JR
Title: Re: Clipping converters
Post by: Gold on December 28, 2018, 02:52:51 PM
There is no good reason... the bad reason is to be louder.

JR

I get clipped masters on a fairly regular basis. At least it isn’t like it was about ten years ago when in indie rock it was common to absolutely destroy the mix with massive clipping. As much for the sound as for level.

Louder is always better until it isn’t.
Title: Re: Clipping converters
Post by: Dualflip on January 04, 2019, 01:23:07 AM
So many manufacturers are spending money on R&D, buying expensive equipment like an AP with extremely low residual distortion so they can squeeze the lowest distortion possible, they manufacture everything with premium ultra low noise and low distortion parts, all of that to end up with a guy clipping the converters to make things loud.   ::).
Title: Re: Clipping converters
Post by: JohnRoberts on January 04, 2019, 10:38:56 AM
So many manufacturers are spending money on R&D, buying expensive equipment like an AP with extremely low residual distortion so they can squeeze the lowest distortion possible, they manufacture everything with premium ultra low noise and low distortion parts, all of that to end up with a guy clipping the converters to make things loud.   ::).
I've told this story (too) many times, but back when I was designing consoles I was diligent about providing clip detection at multiple points within each console strip. Some competitors were less diligent, spending the money they saved instead on advertising that they had more headroom.  ::)  Since a little clipping is not very noticeable**** that advertising had many consumers drinking their kool-aid and believing the lie. I was in the unpleasant position of appearing to have less headroom because of my honest representation of signal status. :-[ 

Here is a true anecdote told me by a Peavey sales rep... He visited a dealer in his territory who had one of these "concealed" clipping mixers, feeding a Peavey amp and speakers (this was before that mixer company was also selling amps and speakers). The signal sounded horrible (obviously clipped) and the dealer explained that "the mixer had so much headroom it was blowing away the Peavey gear"  :o. Since I had already schooled the Peavey reps about this mixer dirty trick at a seminar in Meridian, the Peavey rep walked over to the mixer and cleaned up the sound quality by correctly trimming the input channel gain structure.

JR

*** I am also aware of competing amplifier manufacturers who slowed down the attack time of their amplifier peak/clip LEDs again because it made their amps appear stronger. They literally played louder (higher average power) while ignoring modest amounts of clipping.
---
I have even performed single blind listening tests between allowing a power amplifier to clip, or enabling active clip limiting. I was very disappointed to hear how many listeners preferred the clipped amp sound.  >:(  That would be why there are switches to defeat limiting on Peavey's premium amps but DDT clip limiting was always hard wired "on" inside cheap top box amp sections, because those customers ride them hard and put them away wet.   
Title: Re: Clipping converters
Post by: abbey road d enfer on January 04, 2019, 12:47:54 PM
i'm trying to understand this. why would one want to clip or limit before mastering?
I would say that it's because clipping (whether it's analog or digital) and saturation plug-ins are the amulets that sound butchers use to make believe they are shamans. Show-off is a very strong motivator in a crowded industry.
Title: Re: Clipping converters
Post by: Whoops on April 04, 2019, 11:58:20 PM
i'm trying to understand this. why would one want to clip or limit before mastering?

That can be part of the sound of the mix.  Bands and Artists may push for it, for some styles of music it may be part of the sound, in the same way as a distorting guitar sound is part of the rock guitar sound.

Is not always to make it louder, people may enjoy what clipping some converters imparts on their mix.

I Mastered a record recently were the mixes were clipped in a Lavry converter.
I asked the band to send me the same mixes but not clipped for the mastering process, so I had both.
In that particular band and music the clipped in the Lavry mixes sounded better for their style of music.
I used the unclipped mixes first and worked from there.

In the end the band asked to re-do the mastering using the Clipped mixes.
With both Masters at the same Loudness level, the band was of the opinion the clipped mixes sounded punchier for their type of sound.

I can understand them, and I'm also of the opinion it worked better, even if it's not something I would normally do.