Metering in digital domain

But if you don't like the taste or texture of your pudding you have to know something about cooking to decide what you need to change in the recipe!

(Oh, and there's an old British tradition of putting sixpences into Christmas pudding for people to find when eating it!)
 
regardless of the files that you linked, nobody that has commented in this thread was there when you recorded it or how or what you did specifically. It is a pointless post with to many variables to be confirmed. That is not saying you could be lying but it does leave some room for skepticism,
:laughings:Actually, that's exactly what you were saying without actually saying it. That's what a sceptic does, they habitually doubt. If you doubt what someone has said, you are saying you do not believe them. If you do not believe they are being truthful.......

OffCenter, you originally asked a question completely based around science, not art, and after receiving scientific answers, you conclude with it being a matter of art. If anything, I would argue it's the other way around. Making MUSIC is art, not science. However, recording that music is science with art thrown in. Even the most "artistic" of engineers still follow principle rules governed by the science of acoustics and electronics.
^^^^^^^^^^This, in spades.
Personally, I've found this to be a fantastic thread. Sure, sometimes, the science has gone over my head, but it's all been helpful. There's a big difference in saying "I track at low levels" and "Tracking at high levels will guarantee your sound will be lousy". One does not inevitably follow the other. As with many things, I find that there are paradoxes afoot in making and recording music.
On my DAW, I've noticed time and again that if one records the sound continuously into the clip zone, it sounds utterly awful. It's worse than nails down a blackboard. But if on the odd occasion, the sound peaks for a split second into the clip zone, you can't hear any of that digital clipping sound. In fact, you can't tell anything untoward has happened.
@Offcentre;
You've asked a couple of times why, if you record at high or low levels and find acceptable results either way, keep hammering home the point that you don't have to record at low levels because they are not intrinsically better. I thought the answer to that would be obvious. If the accepted wisdom is that it does make a difference to 'go low' because higher levels are worse, then if you know that's not true, you perpetuate something knowing it's not true. Some people can live with that. Some can't. When a child cuts or grazes themselves, they automatically believe a plaster makes it better. It doesn't. I've always made the point of telling them that. Go wash it with water and let it get air so a scab can form. It'll do more for you than any plaster unless you're involved in some activity that risks pouring dirt into the cut/graze. But for decades the stock response to a cut or graze is to put a plaster on it. No one tells the child that it actually keeps the cut moist and hinders drying and scabbing. Usually, the child stops squawking. And the falsehood is perpetuated.
As for threads going off in unforseen directions, you, I and no one else has any control over that - and that's a good thing. I've learned more from this intense debate than if someone had directly 'answered' your original question by pointing you to a painless link. Besides which, although you asked for no arguing {which no one can guarantee}, you asked for advice because you felt your gain staging through plug ins was wrong. That alone was opening the lid on a Pandora's box, which I would argue was a pretty healthy one.
 
I didn't find it controversial--but then you can count me amongst the numbers who enjoyed this thread and the points raised.

(But then I've seen things from Ethan Winer--for example his AES presentation--for several years now and enjoy his insight into things.)
 
Yeah, bit what you said could lead to someone asking you "why are you attacking me?". It's been known to happen. :)
Good point.
I wasn't attacking though. I did find the first bit I quoted genuinely funny. I still do. I have a bizarre sense of humour at the worst of times.
 
:laughings:Actually, that's exactly what you were saying without actually saying it. That's what a sceptic does, they habitually doubt. If you doubt what someone has said, you are saying you do not believe them. If you do not believe they are being truthful.......


^^^^^^^^^^This, in spades.
Personally, I've found this to be a fantastic thread. Sure, sometimes, the science has gone over my head, but it's all been helpful. There's a big difference in saying "I track at low levels" and "Tracking at high levels will guarantee your sound will be lousy". One does not inevitably follow the other. As with many things, I find that there are paradoxes afoot in making and recording music.
On my DAW, I've noticed time and again that if one records the sound continuously into the clip zone, it sounds utterly awful. It's worse than nails down a blackboard. But if on the odd occasion, the sound peaks for a split second into the clip zone, you can't hear any of that digital clipping sound. In fact, you can't tell anything untoward has happened.
@Offcentre;
You've asked a couple of times why, if you record at high or low levels and find acceptable results either way, keep hammering home the point that you don't have to record at low levels because they are not intrinsically better. I thought the answer to that would be obvious. If the accepted wisdom is that it does make a difference to 'go low' because higher levels are worse, then if you know that's not true, you perpetuate something knowing it's not true. Some people can live with that. Some can't. When a child cuts or grazes themselves, they automatically believe a plaster makes it better. It doesn't. I've always made the point of telling them that. Go wash it with water and let it get air so a scab can form. It'll do more for you than any plaster unless you're involved in some activity that risks pouring dirt into the cut/graze. But for decades the stock response to a cut or graze is to put a plaster on it. No one tells the child that it actually keeps the cut moist and hinders drying and scabbing. Usually, the child stops squawking. And the falsehood is perpetuated.
As for threads going off in unforeseen directions, you, I and no one else has any control over that - and that's a good thing. I've learned more from this intense debate than if someone had directly 'answered' your original question by pointing you to a painless link. Besides which, although you asked for no arguing {which no one can guarantee}, you asked for advice because you felt your gain staging through plug ins was wrong. That alone was opening the lid on a Pandora's box, which I would argue was a pretty healthy one.

I agree with you on just about every level of your post. I dint say he was lying but think about the fact that there are many variables. Its an easy enough test to try myself, and will, but i did at least give my feedback on the links at the risk of being wrong. But, if there weren't skepticism then this useful information would never have been brought to light. I don't mind being the guy that opens the box. At least something is being let out that is helping me regardless. I think that for the most part im pegged as confrontational when in reality im inquisitive. Questioning things and trying different approaches get answers and results.

My point with the levels is, if you are in the digital realm and it doesn't matter if you have a hot signal or not, why not stay safe then? Pushing it isnt resulting in better sound and either is the opposite. But if you stay lower you have less of a chance to clip which is not good. Also it has been pointed out that you may not even notice that you are hearing that little 1db clip ITB until you are back at the analog stage. Plus that little clip will be very noticeable when it comes time to boost the overall track, possibly even effect a gate or compressor in its signal path. Ive seen Dave Pensado manually go into tracks and tame those spikes just for that reason. Maybe im wrong but that is what im here for, set me straight.

And its good to see all my old friends are still on here to keep things interesting.
 
Id also like to add that I also said i respect Ethan and watched his videos. But if someone is passionate enough to go to the trouble he has to debunk audio myths he should not have a problem with a little harmless debate. This is how the world works even if some dont agree with that outlook. Enlighten me please. I want to know what you know so i can better my recordings, that is all i am out for.
 
you may not even notice that you are hearing that little 1db clip ITB until you are back at the analog stage. Plus that little clip will be very noticeable when it comes time to boost the overall track, possibly even effect a gate or compressor in its signal path ... Maybe im wrong but that is what im here for, set me straight.

Most of your post is not wrong, but this one part is. If the input clips briefly while recording, then that simply becomes part of the sound of the recorded Wave file. If you don't hear it, or it's not bad enough to be a real problem, then it won't be a problem later when you go back out through the converter unless the overall output clips a second time. You can lower either the master bus or that one track to avoid clipping the main output. This too is very easy to test!

--Ethan
 
My point with the levels is, if you are in the digital realm and it doesn't matter if you have a hot signal or not, why not stay safe then? Pushing it isnt resulting in better sound and either is the opposite. But if you stay lower you have less of a chance to clip which is not good.
But, if it makes no difference either way, you could advocate staying safe, but it begs the question why, if it makes no difference. So it becomes a circular argument. Whereas if there's no implied "better" or "worse" by taking one side or the other {because it often comes across that one way is better than the other}, then people are free to state why they go one way or the other or that they go one way or the other without the nuclear fallout that so often accompanies these kind of discussions.
 
.... if you are in the digital realm and it doesn't matter if you have a hot signal or not, why not stay safe then?

I don't think that was the crux of the earlier debate.
I believe some were saying that your audio quality changes even with *increased* digital levels (-18 VS -8)...and not just when they are hitting/passing the "0" clip point.
That view comes from the belief that to get from -18 to -8....your analog front end has to work real hard and THAT is where the audio quality is suffering.....NOT in the digital realm....but that may not always be the case. Your analog front end might just fine, and still pushing the digital levels up pretty high.

I think I and other folks covered the whole analog front end perspective, and as I said, I would never set my analog signal level just by going for "safe" digital levels.
I would set them by the front end, analog levels...and the sound I want from that front end.
Actually...since I often track to tape and dump to DAW...I never see my digital levels until I'm making the transfer. I just record away as hot as I like in the analog world. When I make the transfers...some digital channels are on the hot side, some are not....but I've never clipped the digital signal, and once in the DAW, I adjust as needed.

Yes, often if you keep your pres in their "middlin" operating range, chances are good that the digital meters will fall in the -18 to -14 range, and that's fine.
The real point I am making is that you seem to be just focused on "safe digital levels"...which is the analytical view....
...and I'm saying let your preamps fall where they sound right, even if the digital level ends up being on the hot side....
...which is the artistic view, and what you want to go for...right? :)
 
Most of your post is not wrong, but this one part is. If the input clips briefly while recording, then that simply becomes part of the sound of the recorded Wave file. If you don't hear it, or it's not bad enough to be a real problem, then it won't be a problem later when you go back out through the converter unless the overall output clips a second time. You can lower either the master bus or that one track to avoid clipping the main output. This too is very easy to test!

--Ethan

Fair enough. Im just wondering hypothetically that if i would go in and tame just that little spike would that be an efficient way of fixing the problem without lowering that channels place in the mix? Ive done this kind of editing just for sibilance instead of using a DE-Esser and seemed to have a more natural effect to my ears.
 
But, if it makes no difference either way, you could advocate staying safe, but it begs the question why, if it makes no difference. So it becomes a circular argument. Whereas if there's no implied "better" or "worse" by taking one side or the other {because it often comes across that one way is better than the other}, then people are free to state why they go one way or the other or that they go one way or the other without the nuclear fallout that so often accompanies these kind of discussions.

I am surely not trying to take a side on what is better but what outcome each process effects the overall good of the track.
 
I don't think that was the crux of the earlier debate.
I believe some were saying that your audio quality changes even with *increased* digital levels (-18 VS -8)...and not just when they are hitting/passing the "0" clip point.
That view comes from the belief that to get from -18 to -8....your analog front end has to work real hard and THAT is where the audio quality is suffering.....NOT in the digital realm....but that may not always be the case. Your analog front end might just fine, and still pushing the digital levels up pretty high.

I think I and other folks covered the whole analog front end perspective, and as I said, I would never set my analog signal level just by going for "safe" digital levels.
I would set them by the front end, analog levels...and the sound I want from that front end.

Yes, often if you keep your pres in their "middlin" operating range, chances are good that the digital meters will fall in the -18 to -14 range, and that's fine.
The real point I am making is that you seem to be just focused on "safe digital levels"...which is the analytical view....
...and I'm saying let your preamps fall where they sound right, even if the digital level ends up being on the hot side....
...which is the artistic view, and what you want to go for...right? :)

When I asked originally I was just wondering if the sound would suffer with a lower level after the signal was in the box. Like i said i was making the mistake of trying to get as close to odbfs and to me it didnt seem to sound good. I started a new track and used the -18 to -8 and it sounds better to me. Im not worrying about the master clipping that way when the tracks get summed to the master. I am truly not trying to act like im right because i know that im not as experienced as many are but with the help here i could get there. Rami, Greg L and I have a bit of bad blood from other threads that got out of hand but im over that. I want to learn more. This is where im most at home, behind the desk with all my audio toys.
 
Rami, Greg L and I have a bit of bad blood from other threads that got out of hand
Dude, I hate to break this to you, but I have no idea who you are (before this thread) and what you're talking about. You obviously hold grudges a lot more than most people because you were a total stranger to me. That does explain why you went ballistic when all I did was tell you to calm down on Ethan. You came into any response to a post of mine with a chip on your shoulder that I wasn't aware of.

The last 6 pages has been arguing just for the sake of arguing.

Miroslov summed it up best. It's your analog chain that should dictate the level you're recording at.
 
I think most agree that trying to always track up next to 0dBFS is not the best SOP.

My only concern is that if you just always shoot for low, safe levels...sure, things will be fine in the digital world, and often will be fine in the analog front end...
...but you may end up always tracking just for the low, safe levels, and using your eyes to set the front end levels....know what I mean? :)

There are many times when you might want to really hit the analog front end hard, which will kick out a hot level to the digital side.

I still think much of this gets confused/compounded by all these pre/DI/converter combination boxes, where the only thing one has to go by is the digital meter...not to mention, some actually allow you to raise the gain of the *digital input*...which is certainly an easy way to end up near 0dBFS.
My converters don't have the option to change the signal gain...only to chose from -10 or +4...so level adjustment is always happening at the analog front end.
 
Dude, I hate to break this to you, but I have no idea who you are (before this thread) and what you're talking about. You obviously hold grudges a lot more than most people because you were a total stranger to me. That does explain why you went ballistic when all I did was tell you to calm down on Ethan. You came into any response to a post of mine with a chip on your shoulder that I wasn't aware of.

The last 6 pages has been arguing just for the sake of arguing.

I think there was some snipping earlier...from a few sides.
I mean, I saw all the posts that gecko zzed had to delete, yours, his, etc...but why don't we all just move on from that.
Let's not re-stir a new debate about who was arguing with who. :)

This is still a good topic, and some solid on-topic info here.
There were/are some strong views...people get passionate and caught up in the moment.
I think everyone did it to some extent.
 
I think there was some snipping earlier...from a few sides.
I mean, I saw all the posts that gecko zzed had to delete, yours, his, etc...but why don't we all just move on from that.
Let's not re-stir a new debate about who was arguing with who. :)
I agree. I'm over it because I was never really under it to begin with. :D But when I see my name mentioned in a post, I simply respond to that. No big woop. :)
 
I think most agree that trying to always track up next to 0dBFS is not the best SOP.

My only concern is that if you just always shoot for low, safe levels...sure, things will be fine in the digital world, and often will be fine in the analog front end...
...but you may end up always tracking just for the low, safe levels, and using your eyes to set the front end levels....know what I mean? :)

There are many times when you might want to really hit the analog front end hard, which will kick out a hot level to the digital side.

I still think much of this gets confused/compounded by all these pre/DI/converter combination boxes, where the only thing one has to go by is the digital meter...not to mention, some actually allow you to raise the gain of the *digital input*...which is certainly an easy way to end up near 0dBFS.
My converters don't have the option to change the signal gain...only to chose from -10 or +4...so level adjustment is always happening at the analog front end.

Maybe im not doing it right on the front end but ill use the PFL on the channel and set the gain to peak between 0 and +3, then will adjust the pre on my converter to peak around -12 to -6 in the DAW. I dont like the pres on my converters so they are generally set very low. I get a very nice clean signal and go with it. Does this sound like a good start?
 
Back
Top