Skip to main content ↑ ↓

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: Bob Katz - Loudness: War & Peace (Read 16221 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Bob Katz - Loudness: War & Peace

Reply #25 –
Quote from: cliveb on
As long as people's primary mode of listening remains via mobile devices in noisy environments, compression is a requirement.

I have to challenge this. There are actually two types of dynamics to be considered.

Dramatic music with long-term variation in loudness will need to be compressed so that you can hear the quiet bits.

I don't think punchy recordings with good short-term dynamics need to be compressed to be heard in noisy environments. Don't take my word for it. Find some loud music mastered at reasonable levels in the early 1990s (e.g. Jane's Addiction, U2, RHCP, Public Enemy) and take it out for a drive or walk.

Bob Katz - Loudness: War & Peace

Reply #26 –
Quote from: googlebot on
Meta-data which lets the producer somewhat control the amount and type of client side compression would be the way to go.

Agreed. A producer's job, in essence, is to make certain decisions regarding the material, the manner in which it's recorded and the manner in which it's mastered, in alignment with that producer's 'vision'. While consumer control is a Very Good Thing, one should also take into consideration how an artist wants their content to be experienced. Letting a producer exercise some control over playback while providing consumers with a way to override that, if desired, seems like the right move to me.

Producers still get the control they crave, but the listener can always take the reigns if he wants to experience the content differently, so everybody wins.

Bob Katz - Loudness: War & Peace

Reply #27 –
Quote from: Ron Jones on
Quote from: googlebot on
Meta-data which lets the producer somewhat control the amount and type of client side compression would be the way to go.

Agreed. A producer's job, in essence, is to make certain decisions regarding the material, the manner in which it's recorded and the manner in which it's mastered, in alignment with that producer's 'vision'. While consumer control is a Very Good Thing, one should also take into consideration how an artist wants their content to be experienced. Letting a producer exercise some control over playback while providing consumers with a way to override that, if desired, seems like the right move to me.

Producers still get the control they crave, but the listener can always take the reigns if he wants to experience the content differently, so everybody wins.

I'm not sure I fully understand. Are people saying that the compression metadata is the defined way that compression should be applied? Or that the consumer may override this with their own compression settings if they wish?

If the former, then that's not good enough. Different listening environments (eg. planes, trains & automobiles [great movie BTW]) have different compression requirements.

If the latter, then what's the point of having the metadata in the first place? Presumably the uncompressed version is the artist's/producer's vision of how it's supposed to be - so what does the "official" compression metadata define? Are these two supposed to be akin to the director's cut and studio release of a movie?

Bob Katz - Loudness: War & Peace

Reply #28 –
On DAB, the DRC data is 1-band only (i.e. not multiband compression), but many radios let you choose:
2) apply it as-sent
1) apply it half as strong, or
0) not apply it at all.

However, since it's single band, and most stations want that multi-band compressor sound, it's rarely used.


I don't think any music producer is going to trust compression to the end device. Radio stations just might, if forced to. Apart from classical, some jazz, and some speech+drama, most content is already easily audible in the car, and pop music with-dynamics would still be easily audible.

Beatles (or most other pop) tracks mastered in the 1960s work fine in the car today (except for the stereo effect). This is a non-problem IMO.

for classical, 1-band DRC, already available on DAB, is fine for in-car listening. Though the UK commercial station Classic FM already applies bucket loads more of compression than is "needed" in-car - it wants to make the classical recordings sound more like pop recordings, and uses multiband compression to do it. That's a choice, not a necessity.

Cheers,
David.

Bob Katz - Loudness: War & Peace

Reply #29 –
Quote from: 2Bdecided on
I don't think any music producer is going to trust compression to the end device. Radio stations just might, if forced to. Apart from classical, some jazz, and some speech+drama, most content is already easily audible in the car, and pop music with-dynamics would still be easily audible.

This might apply to hypercompressed chart music (not that I listen to it).

But I think you either have a very quiet car (do you drive something like a Bentley/Mercedes/Lexus, perhaps?  ), or you don't listen to much quality pop/rock music.

Try listening to something like Roger Waters' Amused to Death in my Ford Fiesta or my wife's Focus, and you'll find that at least 50% of it is inaudible without some sort of added compression or AGC.

Quote from: 2Bdecided on
Beatles (or most other pop) tracks mastered in the 1960s work fine in the car today (except for the stereo effect). This is a non-problem IMO.

Same argument as above holds for the White Album.

Bob Katz - Loudness: War & Peace

Reply #30 –
Quote from: cliveb on
But I think you either have a very quiet car (do you drive something like a Bentley/Mercedes/Lexus, perhaps?  ), or you don't listen to much quality pop/rock music.
I probably listen to rubbish, have it on far too loud (so any quiet bits are still audible), and don't mind turning the volume down for brief "extra loud" bits, manually.

In truth, I suspect I self-censor what gets played in the car, avoiding the issue.


Still, I think hypercompression neither helps nor hinders in-car listening.

Quote
Same argument as above holds for the White Album.
I guess I must turn tracks up and down more than I realise.

Cheers,
David.

Bob Katz - Loudness: War & Peace

Reply #31 –
Quote from: 2Bdecided on
I don't think any music producer is going to trust compression to the end device.


Film producers are doing this for a long time now on DVDs with AC3 and it works great.

Bob Katz - Loudness: War & Peace

Reply #32 –
Quote from: cliveb on
I'm not sure I fully understand. Are people saying that the compression metadata is the defined way that compression should be applied? Or that the consumer may override this with their own compression settings if they wish?

Essentially both. If you can define the nature of the compression in metadata such that there can exist a 'reference' level, that 'reference' level could be what the artist or producer intends regardless of listening environment. Some sort of button, switch or software setting on the playback device would allow listeners to override that, either eliminating additional DRC altogether or selecting from a range of DRC presets (or something along those lines). The default setting on the override would probably be at the discretion of the device's manufacturer.

If you're listening in the car, for example, the head unit in the car would have its own built-in DRC preset which would apply fairly significant compression, and would probably vary the compression level based on the speed of the car. There might also be a button on the head unit that enables the standard 'as the artist intended' DRC as contained within the file's metadata.

Quote from: cliveb on
Presumably the uncompressed version is the artist's/producer's vision of how it's supposed to be...

A presumption based on what information? I'm not sure I follow.

The director's cut/studio release analogy is fitting, though, yes. A director's vision for a film is often at odds with the studio's vision for a commercial release, and that may also be the case with respect to artists/producers and player-controlled DRC.

Bob Katz - Loudness: War & Peace

Reply #33 –
Quote from: googlebot on
Quote from: 2Bdecided on
I don't think any music producer is going to trust compression to the end device.

Film producers are doing this for a long time now on DVDs with AC3 and it works great.

That's additional compression. They still use in-studio dynamic processing on the music used in movies, including hyper compression when they want that sound.

There are three compression modes in AC3:
0) = off (available on hi-end 5.1 receivers)
1) = normal (default for stereo outputs)
2) = heavy (default for RF modulation)

Obviously 0) won't touch the music, but you can also drop a normally compressed piece of pop music into a film soundtrack such that 1) won't touch it either. A 2000s style pop master probably won't get touched by 2) either.

Outside of the pop world, I agree it could work well. But outside of the pop world, the abuses are less already.


I agree with the core proposal. My point is that I think some producers will still hyper-compress, and almost all will still compress. Almost none will release the entirely compression-free version.

Have you ever heard compression-free pop music? I have, and it sounds very strange to my ears.

Cheers,
David.

Bob Katz - Loudness: War & Peace

Reply #34 –
Quote from: 2Bdecided on
That's additional compression. They still use in-studio dynamic processing on the music used in movies, including hyper compression when they want that sound.

There are three compression modes in AC3:
0) = off (available on hi-end 5.1 receivers)



I presume this is something other than the common DRC 'off/low/medium/high' option that can be applied to any kind of signal, including AC3.  If it is, I can't say I've ever seen this feature offered or even written about -- can you point to a make/model that has it?  Does it go by any other common name?

Bob Katz - Loudness: War & Peace

Reply #35 –
Quote from: krabapple on
I presume this is something other than the common DRC 'off/low/medium/high' option that can be applied to any kind of signal, including AC3.  If it is, I can't say I've ever seen this feature offered or even written about -- can you point to a make/model that has it?  Does it go by any other common name?
AC-3 compression is driven by metadata in the AC-3 file - so the compression characteristics are defined at the time of AC-3 encoding. There are several choices, all carefully documented by Dolby (see page 7). There's also the full AC-3 spec for some light bedtime reading.

Various DVD players I've seen have a DRC menu option that internally selects between the various AC-3 modes - I had a UK Philips one that did this (I tested it with various AC-3 files to verify), but I can't remember the model number. I've never owned a 5.1 amp so can't comment - it's too long since I tested any to remember details.

Cheers,
David.

Bob Katz - Loudness: War & Peace

Reply #36 –
Is it possible to design a practical set of parameters that describe the desired method of dynamic processing sufficient for producers to actually use such a system for the next 10 or 20 years? Clearly, simple "Compression on"/"Compression off" flags don't have the flexibility to do what producers desire and will therefore be effectively bypassed by simply turning it off and doing processing in the studio. I can't imagine a in-device dynamic processor that is flexible enough while being dirt-cheap enough to be standardized and used for any long time-period.

I imagine that having a flexible filter-bank-like structure where the (subsampled) instantaneous gain is directly encoded along with the uncompressed/less compressed reference is the only way to allow the producers to have "their own" sound using "their own" processing for standing out on radio while still allowing the musician/producer to convey how they would like it to sound for their own pleasure, or for "retargeting" the processing for different environments/playback equipment.

I realize that this is just a twist/bandwidth reduction on transmitting/distributing two independent (but highly correlated) streams: one "audiophile", one "loudness maxximized".

-k