Clipping on overheads

[CalariasDead]

i recorded drums with my digi 001 and everything turned out ok, except for the right overhead clipped on every snare hit. i must have not noticed while setting levels. is there any way i can fix this problem with out re recording the drums. i have pro tools 6.4 and the waves dimaond bundle to help

 

[Farview]

The closest thing would be to put a high pass filter on it at 60hz or above.

 

[AardvarkPSW]

i recorded drums with my digi 001 and everything turned out ok, except for the right overhead clipped on every snare hit. i must have not noticed while setting levels. is there any way i can fix this problem with out re recording the drums. i have pro tools 6.4 and the waves dimaond bundle to help

Two options come to mind.

One.

Zoom your waveform in and nuke each of the short, ugly waveforms you see...one for each snare hit! Make certain there is a zero crossing on either side of each edit. Doing this will teach you to make sure to never make that mistake again! (You can try re-drawing...this is a bit of an art to itself)

Two.

Copy your snare track and move it forward a smidge. Do not send it to the mix buss. Put a gate on the offending track and key the gate from the snare track copy...which you could also gate so as to better define the keying element. This allows you to "Look Ahead" in tuning the gate on your overhead track. (This is also a useful way of manipulating analog gear with slower detection circuits.)

There are other ways of mending the track using PT automation features and eqs/de-essers but both of the mentioned techniques are useful to learn for other handy mix and fix situations.


Cheers,
Aardvark

 

[SouthSIDE Glen]

The Sonic Foundry (Sony) Noise Reduction bundle does include a "Clipping Restoration" plug. If you have a DirectX -> RTAS wrapper (I'm not even sure there is such a thing?) you could use that for maybe passable results (it depends on how critical your ears are.)

Better off re-tracking then repairing, though.

G.

 

[Cloneboy Studio]

My opinion--nothing can be done.

Best solution: make the overheads mono by using only the left side.

Next time record at -12 dbfs RMS and -6 dbfs peak and you won't get those types of problems.

 

[AardvarkPSW]

My opinion--nothing can be done.

Best solution: make the overheads mono by using only the left side.

Next time record at -12 dbfs RMS and -6 dbfs peak and you won't get those types of problems.

Bollocks.

Dig in and fix the problem like a man and don't heed the nonsense of those who dare not face a trying task of their own design. This is easy to fix and you won't mess up your soundstage by using the left only overhead.

And there is NOTHING wrong with hyper fast overages in digital land when it comes to transient peaks. If you don't hear a problem THERE IS NOT A PROBLEM!!!

Many high end mastering guys use overage pushing as a "limiting" affect to increase volume....get with the programme!!

It is bad advice like this that really annoys the hell out of me. Forget this nancy shit about -6 or -12 peaks....bullshit...also if you do the resolution math it gets ugly fast. "Take it to the limit...one more time!"

Recording it again is a nice thought but if you are going to learn to mix...learn to mix, not avoid the work involved.



Cheers,
Aardvark



P.S. The clipping may not be digital clipping but the pre-amp or the mic cacking out!!! Learn to identify the problem at source.

 

[SouthSIDE Glen]

Many high end mastering guys use overage pushing as a "limiting" affect to increase volume....get with the programme!!

And another one drinks the koolaid.

It's almost impossible to give worse advice than this one.

Using awful mastering...excuse me, I meant marketing...technique as an excuse to let accidental digital overs slip by is the biggest load of shinola to cross this board since sosob's preaching of the idea of using a reverb as an EQ.

G.

 

[AardvarkPSW]

And another one drinks the koolaid.

It's almost impossible to give worse advice than this one.

Using awful mastering...excuse me, I meant marketing...technique as an excuse to let accidental digital overs slip by is the biggest load of shinola to cross this board since sosob's preaching of the idea of using a reverb as an EQ.

G.

I suggest you reconsider this comment.

I am merely explaining a fact about audio practices as they relate to the real world and the working ear. You might also want to remember that many "over" indicators come on well before clipping. Duh.

A very fast transient overage is rarely, if ever, audible. Listen and learn my truculent friend. If you have the monitoring enviroment of course!!

Cheers,
Aardvark

 

[giraffe]

Copy your snare track and move it forward a smidge. Do not send it to the mix buss. Put a gate on the offending track and key the gate from the snare track copy...which you could also gate so as to better define the keying element. This allows you to "Look Ahead" in tuning the gate on your overhead track. (This is also a useful way of manipulating analog gear with slower detection circuits.)

Cheers,
Aardvark

learned something neat, yea

 

[giraffe]

A very fast transient overage is rarely, if ever, audible. Listen and learn my truculent friend. If you have the monitoring enviroment of course!!

Cheers,
Aardvark

2 sample transients are the best argument for tape that i can think of.

 

[AardvarkPSW]

2 sample transients are the best argument for tape that i can think of.

Yes...tape has a nice way of dealing with this issue. It becomes something musical under these circumstances and less of a math concern. I still like to print to both the daw and the 1/2 inch for this reason among others. I think many home recordists would do well to get a 1/4 inch machine and try mixing to it for shits and giggles...much to be learned for sure!



Cheers,
Aardvark

 

[Cloneboy Studio]

It is bad advice like this that really annoys the hell out of me. Forget this nancy shit about -6 or -12 peaks....bullshit...also if you do the resolution math it gets ugly fast. "Take it to the limit...one more time!"

Recording it again is a nice thought but if you are going to learn to mix...learn to mix, not avoid the work involved.

Wow, worst recording advice ever.

 

[shackrock]

ha. any EE will tell you that clipping is clipping, it's an overload... maybe we can hear it and maybe we can't, but staying away from it prevents hearing it...it's kinda like abstanince...for clipping. HA!

anyway, my overheads used to do the same thing, I had to get a couple of pads to clip on the end of the mic before the cable... -10 and -20 db pads, I tihnk the 10's worked fine though...usually had to put the -20 on the kick, fixed things right up.

 

[AardvarkPSW]

Wow, worst recording advice ever.

Ask yourself what the resolution of your signal is if you peak at -12.

Before you open up your mouth and make another a stupid remark consider the math and consider my previuos post. It also helps if you recognise the difference between a mic overload, a pre-amp overload and an overload at the converter stage.

You can record your drums with some transient peaks as overs at the converter stage and not go to recording hell. Yes, you basement recordists should avoid overages in general but that does not mean you cannot use a good understanding of the physics involved to improve your recordings.

So before you go all silly sit down and calculate how much resolution is to be had on a drum track that peaks at -12.

It is very helpful in digital recording to know these things...I suggest you start learning and stop typing.


Cheers,
Aardvark

 

[mshilarious]

Ask yourself what the resolution of your signal is if you peak at -12.

Before you open up your mouth and make another a stupid remark consider the math and consider my previuos post. It also helps if you recognise the difference between a mic overload, a pre-amp overload and an overload at the converter stage.

You can record your drums with some transient peaks as overs at the converter stage and not go to recording hell. Yes, you basement recordists should avoid overages in general but that does not mean you cannot use a good understanding of the physics involved to improve your recordings.

So before you go all silly sit down and calculate how much resolution is to be had on a drum track that peaks at -12.

It is very helpful in digital recording to know these things...I suggest you start learning and stop typing.


Cheers,
Aardvark

Oh good, now we have pedantic pros

Let's see, we'll be generous and assume the average homereccer is using a 24 bit converter with 110dB dynamic range. If we give away 12dB, then we have 98dB left. Cutting it pretty close to the final 16 bit mix, but then how many of us are really using 96dB of dynamic range anyway Especially after we send it to a "pro" mastering house that crushes the life out of it Ethan Winer argue that the acoustic noise floor in our "basement" rooms is way high as it is, but he is nice about it.

Note: this post does not refer to the mastering guys who hang out here, because they are cool

 

[Cloneboy Studio]

Ask yourself what the resolution of your signal is if you peak at -12.

I said RMS -12 dbfs, peaks no greater than -6 dbfs.

Before you open up your mouth and make another a stupid remark consider the math and consider my previuos post.

Heh... oh teach me master.

It also helps if you recognise the difference between a mic overload, a pre-amp overload and an overload at the converter stage.

Wow, spoken like someone who just graduated out of a 2 year recording school! You speak of the obvious as if it were some great revelation. Let me give you a clue--someone that knows what they are doing doesn't overload anything unless it is on purpose for a special effect (and they never overload their AD's).

You can record your drums with some transient peaks as overs at the converter stage and not go to recording hell. Yes, you basement recordists should avoid overages in general but that does not mean you cannot use a good understanding of the physics involved to improve your recordings.

Um... AUDIO EXAMPLES of this "technique" please.

Let's see... what you are saying is this:

1.) Because resolution data is so important we have to record really hot, even to the point of clipping on transients.

2.) However, clipping equals--NO MORE DATA, you've peaked... everything over that is lost forever. It's 100% data loss over 0db.

3.) So in the interest of preserving data you should lose data? Doesn't make much sense buddy boy because you are PERMANENTLY losing information on transient peaks just so you can better hear stuff like the microphone bleed, line noise, etc...???

So who needs to STFU and learn something now?

So before you go all silly sit down and calculate how much resolution is to be had on a drum track that peaks at -12.

Seeing as there are 6 db's of range per bit this is hardly the "calculation" you make it out to be. It's a "mere" 22 bits.

It is very helpful in digital recording to know these things...I suggest you start learning and stop typing.

Oh brother!

Spoken with the self-righteousness of the insecure.

I'd love to read your exciting theories on how you deal with all these HOT AND CLIPPING channels of audio on the mixdown buss, what your "ideal" mixdown is like for mastering, and how often you find yourself being either in the red and LOSING data or yanking your master faders down--once again losing the oh-so-precious resolution you clipped your signals (losing information) for.

The world would be SUCH a better place if it weren't for half-informed, half-a-clue types like you going around spreading their misinformation, disinformation and outright idiocy. But hey, most people with some common sense can tell who's a moron and who isn't regardless of whatever newspeak attempt at technical jargon they manage.

2.5K'er

 

[Farview]

Technically, if you peak hits -12dbfs, it is 23 bits. If you hit -6dbfs, you have used all 24 bits. Go back and think some more AardvarkPSW, you seem to have missed the point.

 

[reshp1]

Copy your snare track and move it forward a smidge. Do not send it to the mix buss. Put a gate on the offending track and key the gate from the snare track copy...which you could also gate so as to better define the keying element. This allows you to "Look Ahead" in tuning the gate on your overhead track. (This is also a useful way of manipulating analog gear with slower detection circuits.)

Not only are you completely wrong about everything else, I'd just like to point out that the above would not work since the key input of the gate will OPEN the gate on every snare hit, not close it.

EDIT, and even if it did work the way you described, having an overhead track that cuts out everytime your snare hits is going to do wonders for your stereo image.

 

[SouthSIDE Glen]

I suggest you reconsider this comment.

I - and everybody else who has been on this board for more than a few days - have been considering, discussing and arguing this should-be-self-evident point ad nauseum. I personally have been considering it since I moved from the analog to the digital domain almost 7 years ago after about 20 years in the analog domain. It gets tiring beating a dead horse every time a new fader jockey comes on and digs it up again.

I am merely explaining a fact about audio practices as they relate to the real world and the working ear. You might also want to remember that many "over" indicators come on well before clipping. Duh.

OK, let's talk about the real world. The real world is there is not an audio engineer - tracking, mixing or mastering - whose opinion is worth more than the keyboard on which it is typed who likes or agrees with the recent "RMS wars" fad. Every word I have ever read or heard spoken by the top-shelf engineers in our industry all say the same thing; they all truely disdain the practice of turning perfectly good songs into square waves. The only reason they do it is because the subset of producers who are knuckle-scrapers that sign their checks demand it. And the only reason those monobrows demand it is because they are of the long-ago disproven belief that volume sells. They could give a rat's about the sound quality as long as as many people as possible can hear it (they've ovbiously never heard of volume controls.)

And, BTW, in the digital realm, any editor worth owning does not ignite the "Clip" or "Over" indicator until it hits a voltage over that which can be translated to a digital word of solid "1s". There is no gray area or built-in buffer in real digital clip indicators.

A very fast transient overage is rarely, if ever, audible.

That is true, I grant you that much; especially when you get into the sub-millisecond range.

But I ask you where in this thread's creator's original post that he mentioned anything about that fast of a transient? He said that his snare hits were clipping; no further information was provided as to how much of the decay may have also been clipping.

More to the point, you somehow made a jump from the "transient audibility" argument to a justification for clipping because of the "RMS volume wars." discussed above. Not only are those almost completly unrelated situations with entirely different sonic properties, but it's also making the perjorative implication that The Big Boys are increaasing volume by driving their signals into clipping. They are not. They are compressing the dynamic life out of their waveforms and hard limiting at or near zero dBFS. But they are not cutting the tops off of them by saturating the digital data into clipping.

G.

 

[mixsit]

... Only to add that if one is still having a rough time putting the myth of 'record hot' lest one 'loose resolution' to bed, (let alone the advantages of moderate digi levels), there are some very good, and well detailed threads over on ProsoudWeb that would be worth the time.
FYIW..
Wayne