Low- and mid-level productions with crazy levels

[UnderTow]

Originally Posted by tonmeister

interesting, funny the wee details they neglect to mention... will have to test this.
One of the original selling points of HD is that it is virtually impossible to max out the bus, not so it seems.stike

What is meant by that is that you can sum many many tracks without maxing out the bus as long as you adjust the output of the bus accordingly.

Alistair

[Aisle 6]

Originally Posted by UnderTow

I didn't mean just any plugins. As I wrote, I specifically meant plugins that clip at 0 dB FS.

Alistair

There are plenty around.
I use a PT HD system and love it or hate it depending on the day and whether I am tracking or mixing. Whatever you think of DAW's in general, you must admit that there is some good dialogue going on in this thread.

I would like to hear from others who have had a similar experience with mixes done on a PT LE system by the home user and brought in to PT HD systems. Generally I have found the mixes to be loud and have plenty of overs. Obviously no attention to gain structure, but the clients bounces did not suffer.

Anyone else???

[Machina]

Hi guys,

good to see a discussion going on here! So you can say that PT LE or other native sytems like Logic or Cubase (that most "small" recording and mix guys use) allow you more clipping without less audible distortion? On single tracks and on the master buss? So maybe a not very "careful" usage of the DAW would lead to louder mixes & masters? Kinda ironic...

Cheers,

Chris

[Aisle 6]

Originally Posted by Machina

Hi guys,

good to see a discussion going on here! So you can say that PT LE or other native sytems like Logic or Cubase (that most "small" recording and mix guys use) allow you more clipping without less audible distortion? On single tracks and on the master buss? So maybe a not very "careful" usage of the DAW would lead to louder mixes & masters? Kinda ironic...

Cheers,

Chris

Considering the premium that PT HD users pay over a PT LE user.....very ironic.

[UnderTow]

Originally Posted by Machina

Hi guys,

good to see a discussion going on here! So you can say that PT LE or other native sytems like Logic or Cubase (that most "small" recording and mix guys use) allow you more clipping without less audible distortion?

But they are not clipping! That is why there is less audible distortion. Or rather, no distortion at all. The signal goes over 0 db FS but it isn't clipping as long as it is pulled back down below 0 dB FS before it gets converted back to a fixed point format. (Rerecording, bounce-to-disk, going to the converters or digital outputs).

Aisle 6, the vast majority of native plugins do not clip at 0 dB FS. Notable exceptions are Waves LinEQ and I think Waves LinMB. Those are the only two that I know of. (That doesn't mean there are not others).

Of course some floating point plugins might behave unpredictably above 0 dB FS simply because the developers didn't spend too much time on that side of things (why should they?).

On single tracks and on the master buss? So maybe a not very "careful" usage of the DAW would lead to louder mixes & masters? Kinda ironic...

Only on the master. The channels can go pretty much as loud as you want. Even something like +500 dB FS isn't a clip in a floating point DAW. The thing the counts is that the signal comes back down to 0 dB FS before the output.

I really don't believe that floating point DAWS have anything to do with louder mixes.

That said, I stick to traditional gain staging even in a floating point DAW. It is just good practise and doesn't take any more time or energy. Just start with lower channel levels.

Alistair

[Machina]

Originally Posted by UnderTow

But they are not clipping! That is why there is less audible distortion. Or rather, no distortion at all. The signal goes over 0 db FS but it isn't clipping as long as it is pulled back down below 0 dB FS before it gets converted back to a fixed point format. (Rerecording, bounce-to-disk, going to the converters or digital outputs).

I really don't believe that floating point DAWS have anything to do with louder mixes.

Oh okay, I guess I understand. So in the end (which is the output) it's all pretty much the same cause you have to pull back down below 0 dB FS anyways which wouldn't result in any more loudness. So my question is: Is there any "advantage" (not speaking of sonic quality of course!) in those systems if you use clippping on purpose (as in mastering)?

My point is: Those productions I'm speaking of are not very well done IMHO. So I ask myself how this kind of loudness is achieved there? I'm sure it's not about great mastering skills and experience or a L2 cranked on the master buss cause it just sounds different... It sounds more like clipping on different stages (maybe in mastering AND mixing?!) which results in audible distortion (not very nice I have to admit) but not in a great loss of impact (like a vanishing snare or kick a.s.o.) if you e.g. use excessive limiting. Hard to explain, but I hope you guys know what I mean

Thanks and greetings,

Chris

[Aisle 6]

Originally Posted by UnderTow

But they are not clipping! That is why there is less audible distortion. Or rather, no distortion at all. The signal goes over 0 db FS but it isn't clipping as long as it is pulled back down below 0 dB FS before it gets converted back to a fixed point format. (Rerecording, bounce-to-disk, going to the converters or digital outputs).

Aisle 6, the vast majority of native plugins do not clip at 0 dB FS. Notable exceptions are Waves LinEQ and I think Waves LinMB. Those are the only two that I know of. (That doesn't mean there are not others).

Of course some floating point plugins might behave unpredictably above 0 dB FS simply because the developers didn't spend too much time on that side of things (why should they?).



Only on the master. The channels can go pretty much as loud as you want. Even something like +500 dB FS isn't a clip in a floating point DAW. The thing the counts is that the signal comes back down to 0 dB FS before the output.

I really don't believe that floating point DAWS have anything to do with louder mixes.

That said, I stick to traditional gain staging even in a floating point DAW. It is just good practise and doesn't take any more time or energy. Just start with lower channel levels.

Alistair

That is all correct. They are not clipping natively...but that is the problem the way you can push the inputs into plug ins such as tape head etc, get a much cleaner "drive" path through the plug. You then take that same mix into PT HD and it clips the plug. You then alter the gain structure to compensate, but it does not hit the plug the same. Granted this does not effect all plugs, but it does change the way some respond. When you have to cater for the last plug in the chain (in a PT HD system) being 24bit back into the 48 bit mixer, then you have to be careful and you cannot as cleanly get the same result. I realize that the math does not necessarily support this, but do some tests.

In the end I agree with you and it is just good practice to maintain a good gain structure. I do it, but it does not mean that the amateurs do.
Cheers,
Scott.

[Machina]

See post #36...

Ideas? Anybody?

Cheers,

Chris

[Online-Mastering]

To clear something up: Something is mashed up here which doesn't fit together. The internal processing headroom of a DAW and or certain Plug-ins and the output resolution which is created by the DAW when you finally bounce the project to hd. For the loudness / level the ONLY relevant parameter is the OUTPUT / MIX BUSS of your system as THAT indicates if clipping in the audio in going on or not.

The individual channels are processed with higher precision and greater wordlength so the 'internal' processing does not overload even if the sequencer says 'it's clipped'. But that doesn't influence the whole audio project, because: If the processing buffers and prevents single sounds from being clipped, they ARE in fact not clipped. Therefor it only changes the relative volume level differences between all audio files in the project. Many plug-ins for example do 64-Bit internal processing to provide a transparent audio stream, but if a plug-in clips, it clips. The sequencer is the next story, it has lots of processing headroom as well, but if you clip a plug-in on an individual channel, it's mixed in with that clipped sound. This has nothing to do with any DAW output resolution or processing depth. In fact, you just clipped a plug-in. This is what you hear.

Again, it does not saying anything about the mix out. So it not true that you can achieve louder or less loud mixing depending on the arithmetic depth of the DAW.

2) To your question about the low-quality / mid-quality MYSpace productions: In fact many of them are totally squashed. Totally! If you clip something digitally till it distorts and produces lots of audible artifacts, you will surely get it as loud or even louder than the top productions out there. Because the limit is the digital medium. And all efforts to make something louder than loud HAVE to lead in distortion, as limit itself is -technically speaking- a form of distortion.

So, if you like your mix louder than METALLICA you just have to open WAVELAB and add another 6db gain on the mastered mix. Et voila!

Arne

[StereoAtLast]

it sounds like you guys are just pissed because bands don't have to have be on a major label anymore to get poorly mastered, loud ass records

[Machina]

Hi Arne,

so you think it's just about clipping digitally by raising volume in the DAW? Could be an explanation... With or without a limiter in the end which pushes the level down to 0 or -0.2 db FS?

Chris

[Online-Mastering]

Both can be the case: Both clipping and limiting exceeds the level where audio waveforms remain intact, keeping their dynamic structure. Basically, limiting is a hard form of audio compression, but -compared to a standard compressor- with high-ratio.
The "harder" a limiter, the higher its ratio and the faster its attack and release times. For 'brickwall effects' / brickwall limiting the ratio is ∞:1, that results in an audio waveform with flatened tops, where all peaks turn into a square waveform.
The typical compressed audio that you can see when you open up a modern pop song with crazy loudness levels. The resolution of the medium cd for example wil always remain limited to 16Bit. A linear 16bit quantization leaves 65536 different representations of an audio signal level. This IS in fact the physical 'border' of the medium. If you want to bring your level above a certain point, you can just bring in limiting or clipping to raise the signal level even more, but you always introduce artifacts with these processing types as well. The side effects are bit different: While plain clipping goes out of specification and results in that harsh and unpleasant sound we all now, brickwall limiters keep the level at the set threshold (e.g. -0.2 dbfs), but introduce 'pumping' as audible side effects, the more you feed into them.

[Machina]

Yeah, I'm relatively familiar with what you are saying but my point is why it is so easy for pretty unexperienced engineers to reach those loudness levels (as you said louder than loud - I'm NOT saying GOOD). You can't do it just with limiting without losing all your transients so you have to use clipping, too. Of course that's a compromise between loudness and distortion/artifacts. I guess the most important thing is to "hide" those artifacts as good as possible. By the way...many MEs I know or work with in Germany don't use clipping - either because they don't want to (soundquality), don't care or don't know about it. But in result they can't get where e.g. an album like Green Day's American Idiot is (in addition to mixes they get that certainly don't have the "potential" to be as loud as an Andy Wallace or Lord-Alge mix I have to admit ).
I guess the inexperience of those low- and mid-level-guys results in the simple use of "turn up the volume" which eventually is clipping.
What really bothers me about it is: You spend years trying to get the best sounding results you can as a "professional" engineer using great MEs (sound-wise, not loudness-wise) and getting competitive results but if you spend a little time surfing the net and myspace you have all those low- and mid-level-productions that are NOT - by any means - professional but outrageously loud. And many (most) people (musicians, bands, clients, ...) just don't get it if they compare: same loudness = equally good, louder = better. I think that has a lot to do why they are somehow not willing to pay more for professional recording, mixing and mastering. They just don't really HEAR the difference anymore and don't see the benefit in it...

Cheers,

Chris

[Online-Mastering]

Two important points in your reply:

By the way...many MEs I know or work with in Germany don't use clipping - either because they don't want to (soundquality), don't care or don't know about it. But in result they can't get where e.g. an album like Green Day's American Idiot is (in addition to mixes they get that certainly don't have the "potential" to be as loud as an Andy Wallace or Lord-Alge mix I have to admit ).

True. Every mix has an inherent loudness potential which results as part of the arrangement, the used sounds, the use of the stereo field and so on... For example, hip-hop productions which use snappy percussive sounds and one kick which does not play ALL the time (instead from time to time) leaves room for compression / limiting it to louder levels compared to a track where steady bass impulses are going on. Same for the stereo field: You can excessively use the side channel to increase the amount of signal energy that can fit into a production. Try to make a Green Day with guitars coming from the center (mono) only... The more the energy is 'spread' over the stereo the better the possibility to achieve higher loudness levels... Same for transients / percussive elements: There are an important part of the 'loudness feel' a track generates. And don't forget the dynamic range of a production itself which is basically the difference between the loudest and lowest level signal content parts of a musical production.

I think that has a lot to do why they are somehow not willing to pay more for professional recording, mixing and mastering. They just don't really HEAR the difference anymore and don't see the benefit in it...

Yeah, that's right. If a band focusses on MySpace releases and mp3 distribution they often willingly destroy sweetness and consider loudness to be most important. MP3 for example crushes the musical information in a quite excessive way (Listen to the side channel of a song BEFORE and AFTER mp3 decoding (depending on the rate the artefacts vary)). Also 'low signal level' details as well as percussive / transient signal content are reduced as part of the psycho-acoustic model mp3 is based on (simultanious and temporal masking effects). The result is -we all knew- decreased audio quality when using lossy audio formats...

It's all about an artists point of view and it's also a generation thing. If kids blast their ears with mp3 iphone / ipod speakers at high volumes they will loose their ability to judge these audible sonic differences objectively. In fact, if someone grew up with decoded music only, he does not know WHAT he should hear as difference. Not good, but indeed a development...