IPB

Welcome Guest ( Log In | Register )

The Loudness War: Dynamic Range over the years, An analysis of the DR database
bug80
post Feb 19 2013, 00:05
Post #1





Group: Members
Posts: 398
Joined: 23-January 05
From: The Netherlands
Member No.: 19254





I thought it would be interesting to see how the Dynamic Range (Album DR) has developed over the years, and if this shows a clear picture about the development of the Loudness War. In order to check this, I made a dump of the DR database. I choose to focus on the period 1985-2012. The figure above shows the mean DR-value per year (the error bars represent the standard error). It is clear that, overall, the Dynamic Range of albums has decreased in this period from approximately 13 dB to the 8-9 dB range. It is also interesting to see that the minimum mean value is reached in 2008, the year that Death Magnetic was released. In 2009, we see an increase of +/- 0.5 dB (a double-sided t-test shows that this difference is statistically significant at the 0.01 level). Maybe this has to do with the awareness created by this album with respect to the Loudness War?

Furthermore, it seems that the Dynamic Range does not further decrease over the last four years. Has the Loudness War come to an end?

Finally, the figure below shows the number of albums per year that were analyzed. In total, 14,680 albums were included in the analysis.



This post has been edited by db1989: Feb 19 2013, 10:12
Reason for edit: as per request in post #11
Go to the top of the page
+Quote Post
 
Start new topic
Replies
guerrillero
post Feb 19 2013, 11:17
Post #2





Group: Members
Posts: 3
Joined: 19-February 13
Member No.: 106742



the problem i have with this data is the following. the submissions to dr.loudness-war.info are probably biased due to self-selection. i expect that the site is frequented primarily by very tech-savvy users (i.e., "geeks" or "nerds"). there is no reason to believe that group's musical taste to be identical to the entire population. it's also conceivable that users elect to submit albums not randomly, but are more likely to submit releases that stand out in one way or another.

not saying the stats are entirely meaningless, but they should be taken with a grain of salt, just as you should take studies in experimental psychology in which all subjects were undergrads in psychology with a grain of salt.

This post has been edited by guerrillero: Feb 19 2013, 11:18
Go to the top of the page
+Quote Post
bug80
post Feb 19 2013, 12:41
Post #3





Group: Members
Posts: 398
Joined: 23-January 05
From: The Netherlands
Member No.: 19254



QUOTE (guerrillero @ Feb 19 2013, 11:17) *
the problem i have with this data is the following. the submissions to dr.loudness-war.info are probably biased due to self-selection. i expect that the site is frequented primarily by very tech-savvy users (i.e., "geeks" or "nerds"). there is no reason to believe that group's musical taste to be identical to the entire population. it's also conceivable that users elect to submit albums not randomly, but are more likely to submit releases that stand out in one way or another.

Although I had this concern as well, I don't think it is that much of a problem. Note that almost 15,000 albums were included, I have sufficient reason to believe that there is enough 'spread' in genre, dynamic range, etc.
Go to the top of the page
+Quote Post
guerrillero
post Feb 19 2013, 14:16
Post #4





Group: Members
Posts: 3
Joined: 19-February 13
Member No.: 106742



QUOTE (bug80 @ Feb 19 2013, 12:41) *
Although I had this concern as well, I don't think it is that much of a problem. Note that almost 15,000 albums were included, I have sufficient reason to believe that there is enough 'spread' in genre, dynamic range, etc.

if self-selection is going on, a large sample size does not help you. stats can and often will be biased even asymptotically. the representation of certain genres may never approach their population counterparts. so, if certain genres are under- (or overrepresented) and those genres had a different path of DR, your stats are misleading. and i can very easily imagine that certain genres (such as classical, folk music, country) which are quantitatively important in the population as a whole are not too much included in the database you got.
Go to the top of the page
+Quote Post
bug80
post Feb 19 2013, 15:03
Post #5





Group: Members
Posts: 398
Joined: 23-January 05
From: The Netherlands
Member No.: 19254



QUOTE (guerrillero @ Feb 19 2013, 14:16) *
QUOTE (bug80 @ Feb 19 2013, 12:41) *
Although I had this concern as well, I don't think it is that much of a problem. Note that almost 15,000 albums were included, I have sufficient reason to believe that there is enough 'spread' in genre, dynamic range, etc.

if self-selection is going on, a large sample size does not help you. stats can and often will be biased even asymptotically. the representation of certain genres may never approach their population counterparts. so, if certain genres are under- (or overrepresented) and those genres had a different path of DR, your stats are misleading. and i can very easily imagine that certain genres (such as classical, folk music, country) which are quantitatively important in the population as a whole are not too much included in the database you got.

I understand your points, and agree with them for a large part. However, I still think that self-selection is more likely to result in an "overall" bias than a "per-year" bias, in other words, it should not affect the trend(s) too much. I still believe the size of the database is a strong point; and if I browse through it I do not observe a clear bias towards low or high DR-values (this is not scientific at all, I realize that).

For completeness: here is the distribution (# albums per DR-value):



Now, we could debate if this looks like a valid distribution for a big population... I have no idea personally.
Go to the top of the page
+Quote Post
icstm
post Feb 28 2013, 15:55
Post #6





Group: Members
Posts: 121
Joined: 25-January 12
Member No.: 96698



is this dataset somewhere to download?

I would love to cut this by year and genre and see these distributions for each.
To me it is clear that we have lost DR and that this is a shame as live music is very much about DR.

This post has been edited by db1989: Feb 28 2013, 19:20
Reason for edit: A question this general in no way warrants a screen-high quotation of an old post, graph and all.
Go to the top of the page
+Quote Post

Posts in this topic
- bug80   The Loudness War: Dynamic Range over the years   Feb 19 2013, 00:05
- - extrabigmehdi   Interesting graph. QUOTE (bug80 @ Feb 18 201...   Feb 19 2013, 01:19
- - Fandango   Hm, I wonder how they managed to produce CDs in th...   Feb 19 2013, 02:06
|- - extrabigmehdi   QUOTE (Fandango @ Feb 19 2013, 01:06) Hm,...   Feb 19 2013, 02:59
|- - Porcus   QUOTE (extrabigmehdi @ Feb 19 2013, 02:59...   Feb 19 2013, 10:10
- - DigitalMan   The increasing sample size of albums analyzed in l...   Feb 19 2013, 03:58
|- - Kohlrabi   QUOTE (DigitalMan @ Feb 19 2013, 03:58) J...   Feb 19 2013, 08:20
|- - cliveb   QUOTE (Kohlrabi @ Feb 19 2013, 07:20) If ...   Feb 19 2013, 09:29
||- - bug80   QUOTE (cliveb @ Feb 19 2013, 09:29) QUOTE...   Feb 19 2013, 09:53
|||- - db1989   QUOTE (bug80 @ Feb 19 2013, 08:53) Howeve...   Feb 19 2013, 10:12
|||- - bug80   QUOTE (Porcus @ Feb 19 2013, 10:10) QUOTE...   Feb 19 2013, 10:22
||- - Kohlrabi   QUOTE (cliveb @ Feb 19 2013, 09:29) QUOTE...   Feb 19 2013, 12:58
|- - dhromed   QUOTE (Kohlrabi @ Feb 19 2013, 08:20) ...   Feb 19 2013, 10:07
||- - Porcus   QUOTE (dhromed @ Feb 19 2013, 10:07) QUOT...   Feb 19 2013, 10:28
||- - Jplus   QUOTE (dhromed @ Feb 19 2013, 10:07) QUOT...   Feb 19 2013, 13:06
|- - Kees de Visser   QUOTE (Kohlrabi @ Feb 19 2013, 08:20) Thi...   Feb 19 2013, 13:21
|- - Kohlrabi   QUOTE (Kees de Visser @ Feb 19 2013, 13:2...   Feb 19 2013, 14:21
|- - extrabigmehdi   QUOTE (Kees de Visser @ Feb 19 2013, 12:2...   Feb 19 2013, 14:42
|- - dhromed   QUOTE (extrabigmehdi @ Feb 19 2013, 14:42...   Feb 19 2013, 14:57
|- - guerrillero   QUOTE (dhromed @ Feb 19 2013, 14:57) QUOT...   Feb 19 2013, 15:12
|- - bug80   QUOTE (guerrillero @ Feb 19 2013, 15:12) ...   Feb 19 2013, 15:15
- - mzil   QUOTE (bug80 @ Feb 18 2013, 19:05) Furthe...   Feb 19 2013, 07:01
|- - bug80   QUOTE (Fandango @ Feb 19 2013, 02:06) Hm,...   Feb 19 2013, 08:10
- - LithosZA   QUOTE There are plnty of other possible reasons fo...   Feb 19 2013, 09:46
- - shadowking   I have to sadly agree: We lost - at least with ...   Feb 19 2013, 10:04
|- - 2Bdecided   QUOTE (shadowking @ Feb 19 2013, 09:04) A...   Feb 19 2013, 10:33
- - Engelsstaub   I would honestly settle for the solid DR 9 average...   Feb 19 2013, 10:43
- - guerrillero   the problem i have with this data is the following...   Feb 19 2013, 11:17
|- - bug80   QUOTE (guerrillero @ Feb 19 2013, 11:17) ...   Feb 19 2013, 12:41
|- - guerrillero   QUOTE (bug80 @ Feb 19 2013, 12:41) Althou...   Feb 19 2013, 14:16
|- - bug80   QUOTE (guerrillero @ Feb 19 2013, 14:16) ...   Feb 19 2013, 15:03
|- - icstm   is this dataset somewhere to download? I would lo...   Feb 28 2013, 15:55
|- - bug80   QUOTE (icstm @ Feb 28 2013, 15:55) is thi...   Mar 11 2013, 09:10
- - Porcus   Re the selection issue: I am not so concerned abou...   Feb 19 2013, 14:15
|- - BFG   It's interesting that Muslimgauze has an album...   May 20 2013, 19:21
|- - delusionn   QUOTE (BFG @ May 20 2013, 14:21) It's...   May 20 2013, 19:26
- - rick.hughes   I have the impression that a lot of people buy mos...   Feb 19 2013, 14:42
- - extrabigmehdi   Well I believe in fragmentation of market, because...   Feb 19 2013, 16:03
- - hawkeye_p   Well at least Krautrock was invented way longer th...   Feb 19 2013, 16:07
- - probedb   Prog rock, industrial, stoner etc were around over...   Feb 19 2013, 16:07
|- - extrabigmehdi   QUOTE (probedb @ Feb 19 2013, 15:07) Prog...   Feb 19 2013, 16:19
- - greynol   Don't tell Kansas and Yes fans that! Also...   Feb 19 2013, 16:33
- - Porcus   Obviously there's a greater chance of knowing ...   Feb 19 2013, 16:44
- - 2Bdecided   I suspect people's assumptions about music sal...   Feb 19 2013, 18:07
- - ianshepherd   For those who are interested, I'm compiling a ...   Feb 19 2013, 21:22
- - icstm   As you have so expertly done teh work can I just a...   Mar 11 2013, 19:54
- - bernhold   I always wondered what the reason of the loudness ...   May 4 2013, 21:15
- - db1989   I won’t claim to speak for our industry-minded mem...   May 4 2013, 21:22
- - smok3   Is there a link/explanation to what exactly DR is ...   May 5 2013, 14:21
- - soundping   Maybe hearing loss inspired the loudness wars?   May 5 2013, 17:19
- - Gecko   I've always pictured a room full of record-lab...   May 5 2013, 19:26
|- - 2Bdecided   QUOTE (Gecko @ May 5 2013, 19:26) I'v...   May 6 2013, 19:43
- - audiofilaid   I'm reading "Appetite For Destuction: The...   May 15 2013, 23:20
- - delusionn   I just did a data visualization and a background p...   May 20 2013, 18:29
- - AliceWonder   I can't comment on why sales have declined but...   Mar 23 2014, 22:05


Reply to this topicStart new topic
1 User(s) are reading this topic (1 Guests and 0 Anonymous Users)
0 Members:

 



RSS Lo-Fi Version Time is now: 30th July 2014 - 16:13