R/E/P Community

Please login or register.

Login with username, password and session length
Advanced search  

Pages: 1 ... 3 4 [5]   Go Down

Author Topic: Is -0.0 DB too high for digital levels?  (Read 8593 times)

Bob Olhsson

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 3968
Re: Is -0.0 DB too high for digital levels?
« Reply #60 on: October 06, 2005, 09:35:44 am »

In my experience the artist's tour manager frequently has far more input about how and by who a CD will be mastered than anybody in A&R.

I've said this before, if you don't like what a CD sounds like, return it to the artist's manager and demand an undistorted copy. Anything less is about as effective as complaining about a funky burger bun to a restaurant's landlord rather than to your waiter.

bobkatz

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 2926
Re: Is -0.0 DB too high for digital levels?
« Reply #61 on: October 06, 2005, 07:19:19 pm »

Dialnorm does not control the absolute loudness the customer or listener listens at. No one can.The customer can and still does adjust his volume control. What dialnorm does is adjust the relative loudness between pieces.

We lose the fact that heavy metal sounds louder than Joan Baez (and I would miss that), but the benefits that you can finally put a smashed disc in the same CD changer as an open sounding, lower level disc far outweigh any perceived possible detriments.

And people can always turn off their dialnorm and I predict a cult of people who prefer having it off and others prefer having it on Smile

But we NEVER had the ability to make heavy metal sound louder than Joan Baez. There is no god in the sky regulating the absolute monitor gain that everyone uses like there is in the motion picture theatres. Hence, string quartets sound just as loud as symphonies, so quitcherbellyakin Brad.
Logged
There are two kinds of fools,
One says-this is old and therefore good.
The other says-this is new and therefore better."

No trees were killed in the sending of this message. However a large number of
electrons were terribly inconvenienced.

TotalSonic

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 3728
Re: Is -0.0 DB too high for digital levels?
« Reply #62 on: October 06, 2005, 09:03:13 pm »

This thread seems to be addressing a number of issues at once and they seem to be getting jumbled up.  So I am only to address one at a time.

First off:
DIALNORM:
Since I tried to find where BK detailed this in this thread and couldn't see anything -  I advise people to read:
http://etvcookbook.org/audio/dialnorm.html
http://www.tvtechnology.com/features/Inside-Audio/f_dm_insid e_audio-12.10.03.shtml
http://www.ccta.ca/CMFiles/Measuring_DialNorm_05190564PLR-51 92005-655.pdf

Now what I can not for the life of me figure out is how you make it so a Dialnorm decoder would work on consumer playback systems where you are setting relative levels between individual discs or tracks when the recommended "prescribed measurement period" found in the pdf I linked "is to be no less than a 24-hour period.  The ideal measurement period will be between 3 and 7 days whichever reflects most accurately the broadcast programming schedule." !!  While I can see the value of a Dialnorm type of decoder as a very conventient to the consumer option for some playback systems, such as background music being played back from commercial CD's in a reIsaurant, in nearly all other cases I see this instead opening up an entirely new can of worms and really backfiring.  What are you exactly proposing - look ahead decoders?  This seems ridiculous if someone has to wait a looonnnnggg time for their darned disc to start playing back.  Or are you proposing a new subcode header required in every master?  What about backwards/fowards compatibility and the fact that some mastering house will eventually figure out a hack around it - and then probably become the most popular place around??  Or are you proposing something that would change level between songs??  And now instead of the relative levels between tracks that ME's have so painstakingly set we instead have strange or unexpected level drops or raisings??  

Sorry - thanks, but no thanks.

Best regards,
Steve Berson

TotalSonic

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 3728
Re: Is -0.0 DB too high for digital levels?
« Reply #63 on: October 06, 2005, 10:17:08 pm »

bobkatz wrote on Fri, 07 October 2005 00:19


And people can always turn off their dialnorm and I predict a cult of people who prefer having it off and others prefer having it on Smile


What about the fact, judging by lots of history, that a lot of lower end consumer oriented equipment makes itself less expensive by leaving off a few options??  What if the damn player doesn't let you take it off?

Best regards,
Steve Berson

blueboy

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 538
Re: Is -0.0 DB too high for digital levels?
« Reply #64 on: October 07, 2005, 12:01:52 am »

I just found this on Dialnorm...

    http://www.hometheaterhifi.com/volume_7_2/feature-article-di alog-normalization-6-2000.html

It seems to do a better job of explaining what DialNorm (Dialogue Normalization) does.

A dialog normalization setting for DVD audio (AC3) is done during the transcoding stage from PCM to AC3 (most software encoders I've seen default to -27, but range from -1 to -31db).

So I would imagine it would be up to the ME to set the DialNorm level during mastering on each CD (or any succeeding audio format),if it were implemented for audio playback. What's to stop everyone from setting the DialNorm level to -31db (no DialNorm attenuation)?

I'm assuming that DialNorm would also require new decoders in all playback devices as it looks like it is tied to Dolby decoding for either DVD or Broadcast purposes.

I'm not sure if there has already been work done to perpare it for use in audio-only playback devices though.

Now that I'm reading more about this, I think they may be using DialNorm to turn "up" my TV commercials instead of down.  Smile

It would be cool if it could be made to work though....

Regards,

JL
Logged
"Only he who attempts the absurd can achieve the impossible." ~ Manuel Onamuno

bobkatz

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 2926
Re: Is -0.0 DB too high for digital levels?
« Reply #65 on: October 07, 2005, 12:40:04 am »

blueboy rote on Fri, 07 October 2005 00:01




So I would imagine it would be up to the ME to set the DialNorm level during mastering on each CD (or any succeeding audio format),if it were implemented for audio playback. What's to stop everyone from setting the DialNorm level to -1db?




I thought it was an absolute value of attenuation, but not to quibble.

Cheating will not be allowed if this thing is not going to turn into a fiasco. Someone has to create a standardized meter and an unambiguous definition and I assume someone at the plants has to be able to confirm or reject the dialnorm setting.

It sounds complicated and scary, but I see no alternative if we are looking to have sound quality in the future media.

Quote:



I'm assuming that DialNorm would also require new decoders in all playback devices as it looks like it is tied to Dolby decoding for either DVD or Broadcast purposes.




Well, "coding" is not necessary to implement dialnorm. All you need is metadata. This can work with a non-compressed format as well as a data compressed format. There's even enough room in the current CD subcode for dialnorm, but it's too late to implement a standard.

One thing that would have to happen is that all new players for the new medium (e.g. "blue ray") would have to have metadata decoders in their hardware, implement the metadata (dialnorm) in their D/A converters and if external converters are used, metadata will be required to be communicated and dialnorm as a standard in the DAC.

Since there is a new medium, there is room for the new rule....


It would be VERY cool if it could be made to work.

BK
Logged
There are two kinds of fools,
One says-this is old and therefore good.
The other says-this is new and therefore better."

No trees were killed in the sending of this message. However a large number of
electrons were terribly inconvenienced.

TotalSonic

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 3728
Re: Is -0.0 DB too high for digital levels?
« Reply #66 on: October 07, 2005, 12:53:11 am »

Bluray vs. DVD-HD is looking like it's headed to the same fate as SACD vs. DVD-A right now.  I'm willing to bet that Bluray won't be around in 10 years.  

Personally I think placing any kind of "hope" in a dialnorm metadata header on discs is beyond foolish.  If it is implemented I am willing to bet that it will be implemented poorly and that it will cause a lot more problems than it will solve.  I can think of hundreds of releases that have true drama and effect that arises from the radically different relative levels of their tracks (i.e. Rachel's "The Sea & The Bells" where there is that one track in the middle of the CD where it is painful screehing loud as heck after a bunch of mellow quieter tracks) that will get completely screwed up - and that ton of people will never hear things the way the artist intended them because they never bother or realized how to take their dialnorm off.  

Best regards,
Steve Berson

bblackwood

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 7036
Re: Is -0.0 DB too high for digital levels?
« Reply #67 on: October 07, 2005, 01:42:50 am »

As everything else, dialnorm seems to be another Utopia, unattainable in the real world. How does one measure it?

Instead of impugning people trying to create art, let's struggle to make people appreciate what we have now...
Logged
Brad Blackwood
euphonic masters

blueboy

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 538
Re: Is -0.0 DB too high for digital levels?
« Reply #68 on: October 07, 2005, 02:58:45 am »

I should stress that my only experience with DialNorm is from the creation of AC3 audio for DVD's.

I don't know how much different its application in broadcast or uncompressed audio would potentially be, or what might be required to fully implement it successfully.

I just thought the article would clear up some confusion about the general premise behind DialNorm.

Dictating to people how loud to make or play their audio doesn't make sense to me. But if DialNorm could be turned into a "step up" feature that offered the benefit of evening out playback levels, many people may see that as a good thing.


Regards,

JL
Logged
"Only he who attempts the absurd can achieve the impossible." ~ Manuel Onamuno

bobkatz

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 2926
Re: Is -0.0 DB too high for digital levels?
« Reply #69 on: October 11, 2005, 02:57:40 pm »

TotalSonic wrote on Fri, 07 October 2005 00:53

Bluray vs. DVD-HD is looking like it's headed to the same fate as SACD vs. DVD-A right now.  I'm willing to bet that Bluray won't be around in 10 years.  

Personally I think placing any kind of "hope" in a dialnorm metadata header on discs is beyond foolish.  If it is implemented I am willing to bet that it will be implemented poorly and that it will cause a lot more problems than it will solve.  I can think of hundreds of releases that have true drama and effect that arises from the radically different relative levels of their tracks (i.e. Rachel's "The Sea & The Bells" where there is that one track in the middle of the CD where it is painful screehing loud as heck after a bunch of mellow quieter tracks) that will get completely screwed up - and that ton of people will never hear things the way the artist intended them because they never bother or realized how to take their dialnorm off.  

Best regards,
Steve Berson


Steve. Dialnorm will not affect the integrity of the track to track levels within a single work. A single dialnorm level will be assigned to the entire work. You won't have to worry about whether an artist knows how to take it off or not.

Regardless, Blu-Ray has been etched in silicon already and the only formats on there that have dialnorm are AC3 and MLP. Well, well, well..  But there is no standard to define how dialnorm is measured, it is a producer's prerogative, and with that, there might as well not be dialnorm at all. The stereo versions on both of these high res formats are destined to have a volume war and the same consequences as the audio CD within, let's say, 5 years. Unless A WORKBLE, PROPERLY PLANNED, UNAMBIGUOUS AND EFFECTIVE dialnorm becomes a standard.


BK
Logged
There are two kinds of fools,
One says-this is old and therefore good.
The other says-this is new and therefore better."

No trees were killed in the sending of this message. However a large number of
electrons were terribly inconvenienced.
Pages: 1 ... 3 4 [5]   Go Up