Is it O.K. to Clip instead of Limit?

[LogicDeLuxe]

Thats interesting, I read somewhere people in the 80's wouldn't let the peaks touch the 0.0 celing as some CD players back then would read it as an error, is this still necessary now?

I've never seen such a CD player. And I have several CD's reaching 0.0.

I have in the past loaded bought mp3's into logic and they all seemed to hit 0.0.

Most of those MP3s probably would even go beyond 0.0. It's in the nature of data reduction, ie. peaks can go higher as they were in the source. That's why MP3gain sometimes even fixes problems.

I bought a few albums in MP3 format. Before I burn those to CD audio, I normalize them to 0 dB, which can be a reduction of several dB in case of a hot master.

 

[Massive Master]

I've heard plenty of converters that sound terrible at -0.0 (and lower for that matter) dBFS. A lousy third of a dB of headroom in the grand scheme isn't much to ask.

 

[NYMorningstar]

The problem with the volume war that gets me isn't just all about the crushing but with the loss of dynamics in the music itself. It's like bands are writing rests right out of music. So all I want to say about the volume war is GIVE IT A REST!

 

[masteringhouse]

Thats interesting, I read somewhere people in the 80's wouldn't let the peaks touch the 0.0 celing as some CD players back then would read it as an error, is this still necessary now? I have in the past loaded bought mp3's into logic and they all seemed to hit 0.0.

Consectutive samples at 0 are often a way to measure overs, I'm not clear if this is what you meant by error. This measurement is still the same.

 

[SouthSIDE Glen]

It's like bands are writing rests right out of music.

That's a great point. On a similar vein, one "trick" I have noticed quite a bit is the inclusion of tracks of fuzz/noise/distortion that don't add anything musically to the arrangement or the mix.

Sometimes they double an existing track (I have heard fuzz tone stuck underneath acoustic guitar), but sometimes that make little to no musical sense at all, they are just a track or two of noise layered under (sometimes over, if it's really overdone) the whole mix. It's only purpose seems to be to increase the density of the mix, fill in the silence, and bump up the RMS.

Sometimes it's a cool effect, like that fuzz under the acoustic (if you want to check it out, check out "World Of a King" by David Meade). But often times it's just annoying at worst, or unhelpful to the sound at best.

G.

 

[benage]

I suppose when releasing an album you've got three choices:

A: Try and get the loudest track you can, making it as least as loud as everybody else or possible louder.

B: Try to presere a little dynamic range making your tracks slightly quieter and hope no one really notices.

C: Realease an album with great dynamic range which is way quieter but make a point of why on the C.D. with maybe a sticker on the front like the 'turn it up' campain and try and start a new trend (presuming your a big artist).


I'm toying between B and C.

 

[Massive Master]

B: Try to presere a little dynamic range making your tracks slightly quieter and hope no one really notices.

No one (who's not in a participant *in* the loudness war) ever really does notice. They just turn the volume up a bit and then complain about how crappy (and loud) the next track is.

 

[SouthSIDE Glen]

I'd like to offer option D, if I may, plus a small modification to option C. First, option D:

Push your quietest song as loud as you can before it starts falling apart. Use that as the benchmark level for your album. Then release the album at that volume, knowing you'll sleep well at night because the music is good enough for people to want to listen to and the mastering is done well enough to keep them listening to it for more than a couple of weeks without getting tired of it. Let everyone else worry about their own albums being good enough to compete against yours.

As far as the label business goes, I suggest a "Free Volume Initiative" label that can inform the user that the volume has been optimized for the artistic content, and doesn't conform to anybody else's arbitrary standards for volume, either by being maximized for maximum loudness or by passing the bar of an arbitrary formula decided upon by a couple of engineers that cannot possibly be equally valid for all recordings.

G.

 

[LogicDeLuxe]

I suppose when releasing an album you've got three choices

Or, if you can master with HDCD peak extension, you can have both B and C at the same time.

The downside is, you obviously have to play it back on a HDCD capable DAC in order to enjoy the full dynamic range. Not a thing you usually have in a boom box, car stereo, Walkman, boom box and similar devices.

The advantage is, people who really want the best are willing to pay for a good HDCD capable DAC. It's not even that expensive like all those voodoo stuff. Or, if you can live with Windows, even WMP will do on a 24 bit soundcard.
And at least, the CD won't sound worse on a regular DAC than any other with the same overall volume.

It's not exactly the point HDCD is made for, but since the original dither algorithm HDCD hardware offers is rather dated and doesn't even have noise shaping, it isn't really an improvement over recent dither algorithms, though it offers about 6 dB more headroom. It actually is somewhat over hyped, imho.

Also, I noticed that B doesn't seem to be a problem anymore for at least one big label: Universal. Once among the worse mastered CDs, the average volume of the last few years became about 3 dB quiter than those of SonyBMG.


And funny, C, the way you describe it, reminds me a bit of Superbit DVD. Actually nothing new at all, but just use an existing standard in a better way. In case you don't know: On the none Superbit titles, there are many on which Sony filtered out details on purpose, so the movie can be better compressed to fit on a DVD-5 rather than the usual DVD-9, just to save a few cents.
If the CD requires a special sticker to inform the user that it actually uses the entire dynamic the mix has to offer, eventhough the CD is capable of doing so with ease, it's very similar, isn't it?

Push your quietest song as loud as you can before it starts falling apart.

Don't you mean the loudest song? If you push the quitest song to the limit, the others would be in real trouble if the relation in volume is to be preserved.

 

[SouthSIDE Glen]

Don't you mean the loudest song? If you push the quitest song to the limit, the others would be in real trouble if the relation in volume is to be preserved.

I don't quote follow your reasoning. Note that I said use the quietest song as the benchmark - or maybe better word might be reference - for the album. Part of the job of album master is usually to NOT preserve the relative volumes, but rather to even them out, at least to some degree.

The louder songs would not be in trouble because they would not have to be pushed as hard to meet that level, and if you wanted to make them louder, you still have "room" to do so. [WARNING: Another Analogy Ahead] It's like a fleet of boats traveling as fast as the maximum speed of the slowest boat. [/ AAA Warning]

Note also that we should be talking about perceived level, not necessarily actual RMS numbers.

G.

 

[masteringhouse]

Just curious, for bands that you really like, do you stop listening to them because of any sort of audio production issues? If so, do you not listen to them on mp3 and earbuds?

I would like to propose an option E. Create music so compelling that all of the above is secondary. While I'm not a fan of crushing music to death, I tend to listen even less to music that sucks no matter how well produced.

 

[SouthSIDE Glen]

Just curious, for bands that you really like, do you stop listening to them because of any sort of audio production issues?

If that were the case, Robert Johnson would be in real trouble

G.

 

[masteringhouse]

If that were the case, Robert Johnson would be in real trouble

G.

A beautiful example!

Rather than a "stop the loudness" campaign I would like to start a "stop the shitty music" campaign.

Anyone with me? I can see the logo now ...

 

[benage]

Option D was a better way of what I was trying to say with B in terms of make it quite loud but not so loud the music suffers, And if the music's not suffering noticably Is there any real point to option C?

 

[SouthSIDE Glen]

Rather than a "stop the loudness" campaign I would like to start a "stop the shitty music" campaign.

Anyone with me? I can see the logo now ...

LOL, I'm with ya, Tom. Only problem with the label logo I'm envisioning is that the label logo might itself incur an adult content warning label

And if the music's not suffering noticably Is there any real point to option C?

Ahhh, now you're starting to come around young Grasshopper* .

That's all many of us have been saying all along; if the customer feels the music itself is worth listening to, and the quality of it's presentation doesn't take away from that, that's all anybody needs.

Excess volume is for those who don't believe in their music.

G.

*Only joking. No condescension or patronizing intended.**

**The fact that I have to stick that stupid politically correct disclaimer in there makes me almost as sick as having to even talk about the volume wars does.

 

[masteringhouse]

Option D was a better way of what I was trying to say with B in terms of make it quite loud but not so loud the music suffers, And if the music's not suffering noticably Is there any real point to option C?

Classical music where dynamics is a much larger concern than in Pop music and where pretentious labels regarding sound quality may actually have an effect on some of the people purchasing it.

 

[LogicDeLuxe]

I don't quote follow your reasoning. Note that I said use the quietest song as the benchmark - or maybe better word might be reference - for the album. Part of the job of album master is usually to NOT preserve the relative volumes, but rather to even them out, at least to some degree.

Well, it seems a misunderstanding then. I usually adjust the volumes to each other before I even think of the final level. After that, I have a relation which makes sense musically, and of course, which I want to preserve during further editing. Thus pushing the hottest part of the entire album to a reasonable maximum (not excessively over the top, of course), and then adjust the volume of the other tracks by the same amount is the usual way I do it.

Just curious, for bands that you really like, do you stop listening to them because of any sort of audio production issues? If so, do you not listen to them on mp3 and earbuds?

Yes, there are. Some albums I really like musically, but very rarely actually listen to them, because they sound like a turd:

Rosenstolz - "Zucker": Mirror frequencies in the strings, dull sounding drums, clicks everywhere. And the bus apparently is compressed that much that any tremolo (which is a notable trademark of her voice) is turned into obvious modulation artifacts.
I once offered them to remix and remaster that album. They are actually pleased by my offer, but they had a contract and gave me the address of their producer who refused to even answer me. I hate those contracts where the band isn't even allowed to have a word about the way a CD is mastered. And this is the majority, I afraid.

Mike Oldfield - "Tubular Bells III": Actually the album is very listenable. Quite hot, but not over the top. Until the final part where the tubular bells come in. It's close to white noise and very distorted. There are worlds between this and the earlier "Tubular Bells" which I listen to much more frequently.

Santana - "Supernatural" (the studio album): It's that distorted, I never even played it to the end. One of the worst examples ever.

Red Hot Chili Peppers - "Californication": I used to avoid this "excessively clipping everywhere and midrange pushed far up"-album, until I got the bootleg, which I did remaster (excerpts on my homepage), and fans really wanted to buy a copy, which I had to refuse for legal reasons, unfortunately.

Eminem - "The slim shady lp": Distortion everywhere. The wave looks like this happened during tracking.

Any Abba remastered albums: That band has pretty dynamic mixes, and most of them are excellent examples of good mixing as well, so you really don't want to go higher than about -14 dBFS rms sine during mastering, or you will notice dynamic loss. All remasters are far over the top leaving a steady white noise sound. I wouldn't buy those for the world. (There's an example on my homepage how it can sound remastered properly)

I also won't buy any later Bob Dylan albums, and he very sure knows why. It's that same stupid contract thing.

Gov't Mule - "Mighty High": It's a reggae album, and as you probably know, reggae lives from bass intense rhythms. This CD has no such thing whatsoever, which isn't a surprise at a -6 (minus six) dBFS rms sine overall level.

It doesn't matter if I listen to those in a descent home stereo, a PA or on my MP3 Walkman with headphones. (I don't use those earplugs which usually come with the device)

I would like to propose an option E. Create music so compelling that all of the above is secondary. While I'm not a fan of crushing music to death, I tend to listen even less to music that sucks no matter how well produced.

Don't ruin good music, just because you're thinking that it's good enough so that doesn't matter! It does!

Sure, no one wants to listen to very well mastered CD's if the music is not worth it to begin with. However, given the music is worth it, a decent mastering job is important to really make you able to enjoy this.

 

[SouthSIDE Glen]

Well, it seems a misunderstanding then. I usually adjust the volumes to each other before I even think of the final level

Even if you do adjust relative levels before deciding final gain, you still have to use *something* as a reference to adjust those levels to.

I guess the question I have there (I'm not saying it's wrong, I'm just trying to understand a different technique) is how you decide to adjust the volumes to each other? Somewhere along the line you are setting a calibration level to compare them against; i.e. you gotta start somewhere, right?

For example, if you have two songs, a loud one and a soft one (by sonic content), do you adjust the loud one in reference to the soft one or the soft one in reference to the loud one? Or do you pick some arbitrary point in the middle?

Or am I missing something (wouldn't be the first time )?

G.

 

[masteringhouse]

Don't ruin good music, just because you're thinking that it's good enough so that doesn't matter! It does!

Sure, no one wants to listen to very well mastered CD's if the music is not worth it to begin with. However, given the music is worth it, a decent mastering job is important to really make you able to enjoy this.

I don't disagree with your statement, it's more a matter of priorities, opinion, marketing, image, etc. Who is to say that a master is too loud any more than a piece of music is shit? A regulatory committee? A group of MEs?

I recently picked up a best of Black Sabbath album because it had a track that I haven't heard in a long time and it was in the cheap bin. When I played this in my car I was amazed at how bad the overall EQ was, total mid, and very inconsistent across the album. I almost wanted to remaster the thing myself initially. Ya know what though, I been playing this CD more (at least this week) than anything else laying around in my CD case currently. I have to ask why? Great music is great music no matter what the media and audio production quality of it is. We can bitch about the loudness wars for the next 10 years (and probably will) as well as other issues in audio production that are being sacrificed, it still doesn't make the music itself any better.

There has been a lot of emphasis on loudness wars in forums and elsewhere ad nauseum. It just seems to me that there are issues which are just as or more important that are being severely ignored.

 

[SouthSIDE Glen]

Mya favorite way to put it into perspective is that the music and the musicians are always going to be far more important than the engineers and the engineering, and that there is one simple way to prove it.

Take one away and ask if the other can survive.

Musicians thrived long before Edison and Bell came along, and will continue to thrive - perhaps even more so - long after the Age of Technology has expired into the past.

Engineers, OTOH, don't last long without someone and something to engineer.

However, neither one will last long unless people actually want to listen to the music. Even back in Medevil times the good music survived and the bad stuff didn't. There's a reason "Greensleeves" has made it this far but a hundred other Medevil and baroque ballads by the Medevil versons of Flash In The Pan died out long ago.

Yet who remembers who actually originally performed Greensleeves? Some guy named "Traditional" .

G.