[bigshot]

That's because we are making recordings designed to be heard by human ears!

 

[TheSonicTruth]

That's because we are making recordings designed to be heard by human ears!

I want to hear what *is*, not something all slick and over-produced. And meters in a DAW or on a board help keep you from clipping or distorting.

 

[pinnahertz]

Mastering that reduces the dynamic range of a symphony composition to that of a pop/rock piece is not what I'd call good mastering.

Do you have an example of that?

Are you drawing that conclusion based on the misuse of the DR meter function, or some other way?

 

[pinnahertz]

I want to hear what *is*, not something all slick and over-produced. And meters in a DAW or on a board help keep you from clipping or distorting.

If you didn't have those meters your chances of recording what *is* would be minimal at best.

 

[pinnahertz]

That would explain your "use your ears" statement. They all say that - mixing engs, mastering engs, producers: "Ignore the numbers".

As an engineer I take strong exception to your generalization.

As an engineer who has worked extensively in classical music, I'm afraid I have to take strong exception with your assumption that it's all compressed as well. Our audience doesn't tolerate compression well, and since all product is market driven, I haven't seen much use of compression on classical recordings since the days of vinyl when you had to do something about HF limiting and keeping the PPP stuff above the surface noise.

 

[bigshot]

I want to hear what *is*, not something all slick and over-produced. And meters in a DAW or on a board help keep you from clipping or distorting.

I don't have to ask if you've ever produced a record! Clipping or distorting are the least of the problems to solve.

 

[TheSonicTruth]

I don't have to ask if you've ever produced a record! Clipping or distorting are the least of the problems to solve.

By that are you suggesting those two are ok/justifiable?

 

[pinnahertz]

By that are you suggesting those two are ok/justifiable?

Come on. Do you really think any producer or engineer would think that related to classical music?

Both are easily avoided, and that's like recording 101...first thing you learn. Non-issue.

So, just arguing are we?

 

[TheSonicTruth]

If you didn't have those meters your chances of recording what *is* would be minimal at best.

And where in post #32 did I negate the need for meters? In post 32, by suggesting that metering keeps you 'out of the red', I was implying that engineers, on forums like Gearslutz and newsgroups like rec.audio.pro actually told me "ignore the meters - USE YOUR EARS". Idiotic advice if you ask me. I say, use your ears, and check the meters to confirm.

 

[TheSonicTruth]

Do you have an example of that?

I stated such example earlier in this thread. I had mentioned that both the first movement of Beethoven's Symphony #5 on CD, and a typical '80s pop tune on CD, registered a DR=12 in Foobar. If you need visible proof of that I can add screen shots. Both CDs were original '80s releases by the way - not reissues or "remasters". Listening to both, that is, "using my ears", led me to assume that the Beethoven selection would naturally - logically - return a significantly higher DR value in the Foobar software than the pop-rock song.

Are you drawing that conclusion based on the misuse of the DR meter function, or some other way?

How does one "misuse" Foobar2000 Dynamic range meter? You drag the songs you want to analyze into the Foobar interface, right-click on the songs you want to check, and select 'Dynamic Range Meter'. A snapshot is then produced and a text document of the DR results is saved in the specific folder the song resides in. Simple! No settings to fudge with, no code.

 

[TheSonicTruth]

Come on. Do you really think any producer or engineer would think that related to classical music?

Both are easily avoided, and that's like recording 101...first thing you learn. Non-issue.

So, just arguing are we?

Not at all. I just wasn't sure what bigshot meant, in post #36, about "Clipping or distorting" being "the least of the problems to solve." Did he mean 'least important'? Or something else?

 

[bigshot]

I meant that avoiding clipping is drop dead easy. Engineering music is a heck of a lot more than watching the meters for peaks. You could train a chimp to do that. It's also a heck of a lot more than just trying to reproduce the sound the way it sounds in the booth with the musician. A good producer and engineer work together to create the sound. They don't just hold a mike up to it and lay it down bald like that. The hard part of mixing is carving holes in the sound with EQ and subtle level adjustments to allow everything to be heard at one time clearly. If you just stack everything up, it'll sound like mush even if you aren't clipping and your bit rate is through the moon. I think a lot of audiophiles listen to equipment with meters and specs instead of listening to music with their ears. If they put as much effort into learning how to listen to music with their ears as they do trying to train them to hear inaudible frequencies or numbers on a page, they might actually figure out what makes a good mix good and what makes a bad mix bad.

Less dynamic range might mean over compressed, but more dynamic range doesn't necessarily mean a better mix. It's like Goldilocks and the Three Bears.... This mix is too compressed, this mix is too dynamic... this mix is just right.

 

[pinnahertz]

I stated such example earlier in this thread. I had mentioned that both the first movement of Beethoven's Symphony #5 on CD, and a typical '80s pop tune on CD, registered a DR=12 in Foobar. If you need visible proof of that I can add screen shots. Both CDs were original '80s releases by the way - not reissues or "remasters". Listening to both, that is, "using my ears", led me to assume that the Beethoven selection would naturally - logically - return a significantly higher DR value in the Foobar software than the pop-rock song.

You're using the DR meter for something very different than it's intended purpose, so your conclusions are wrong. And this is not the first time that fact has been posted.

How does one "misuse" Foobar2000 Dynamic range meter? You drag the songs you want to analyze into the Foobar interface, right-click on the songs you want to check, and select 'Dynamic Range Meter'. A snapshot is then produced and a text document of the DR results is saved in the specific folder the song resides in. Simple! No settings to fudge with, no code.

You misuse it by trying to use it to determine total dynamic range of an entire symphony when it's purpose is to determine relatively short-term dynamic range as it relates to aggressive "loudness war" processing that affects crest factor. Its integration time of the DR meter is much too short to include an entire Beethoven symphony. Way too short. There are other tools, for example the "Amplitude Statistics" tool found in Audition, which would give you results you're looking for, so long as you understand which parameters to be concerned with.

For example, I just scanned the entire 4th movement of the Beethoven Symphony #5 (Cleveland/Von Dohnányi) in Audition, I got (the first figure is left, second is right, all figures are re: 0dBFS):

Peak Amplitude: -0.24 dB -0.21 dB
Total RMS Amplitude: -16.90 dB -16.56 dB
Maximum RMS Amplitude: -8.54 dB -9.28 dB
Minimum RMS Amplitude: -58.37 dB -57.91 dB
Average RMS Amplitude: -20.72 dB -20.21 dB
Dynamic Range: 49.83 dB 48.64 dB
ITU-R BS.1770-2 Loudness: -16.03 LUFS
(some data edited out for simplicity)

Is that more like what you are looking for? See how different that is from the DR meter?

What you've done is use the wrong tool to make errant conclusions, then used those conclusions to attack the professional practices of those in the industry. That's really unfair.

Perhaps you might consider taking a breath and reading up on a few basic principles. As an example, within the audio industry there are completely appropriate uses for compression, limiting, peak limiting and clipping. And yes, they're all different. Even within compression and limiting there are thousands, hundreds of thousands of variables that make these processes effective without severe side effect, or aggressive with radical side effect, and anywhere inbetween. And yes, you can clip sometimes, and it can result in a better outcome without any audible distortion, but there's a lot of learning and intelligent decision to be done to make that happen. And yes, you can even compress, often quite a bit, without a listener ever suspecting anything. The secrets lie in complex attack and release algorithms, release gating, variable slopes, and much, much more. You're deep into an area where people have spent lifetimes of research.

 

[pinnahertz]

As a sanity check, here's a 1980s tune (More Than This - Brian Ferry) amplitude stats, taken from the first note to just before the board fade at the end.
Note, this file was an .mp3, so there are peaks above 0dBFs. Sorry, I don't have a lot of 80s pop around, I was working in classical then.

Again, Left, Right, ref: 0dBFS

Peak Amplitude: 1.09 dB 0.60 dB
Total RMS Amplitude: -8.81 dB -8.78 dB
Maximum RMS Amplitude: -5.22 dB -5.32 dB
Minimum RMS Amplitude: -16.37 dB -14.29 dB
Average RMS Amplitude: -9.04 dB -8.98 dB
Dynamic Range: 11.15 dB 8.98 dB
ITU-R BS.1770-2 Loudness: -9.53 LUFS

 

[RRod]

I stated such example earlier in this thread. I had mentioned that both the first movement of Beethoven's Symphony #5 on CD, and a typical '80s pop tune on CD, registered a DR=12 in Foobar. If you need visible proof of that I can add screen shots. Both CDs were original '80s releases by the way - not reissues or "remasters". Listening to both, that is, "using my ears", led me to assume that the Beethoven selection would naturally - logically - return a significantly higher DR value in the Foobar software than the pop-rock song.

How does one "misuse" Foobar2000 Dynamic range meter? You drag the songs you want to analyze into the Foobar interface, right-click on the songs you want to check, and select 'Dynamic Range Meter'. A snapshot is then produced and a text document of the DR results is saved in the specific folder the song resides in. Simple! No settings to fudge with, no code.

Like others have said, the DR meter isn't made for what you want to do with it. It's mainly designed to compare different masterings of the same material or at least the same genre. Throwing out the bottom 80% of RMS values and the highest peak doesn't really make for a method that's meant as an absolute dynamic range metric. Beethoven 5.I isn't a track I'd expect to come up as highly dynamic anyway; the second or third movement would probably register better, having longer soft stretches and some pretty big bangs (for the time).