IPB

Welcome Guest ( Log In | Register )

 
Reply to this topicStart new topic
Imperfections of MP3s exposed when played out loud on good PA?, "Funktion One system will eat 192kbps MP3 for breakfast"
rgtb
post Sep 26 2012, 14:29
Post #1





Group: Members
Posts: 25
Joined: 26-September 12
Member No.: 103446



I regularly read a DJ blog called DJTechTools. It has an entry today entitled "A DJ's Guide to Audio Files and Bitrates."

They state "Many DJs don't notice a big difference in file type and bitrate until they're blasting the track on a high-end club system and find the track just isn't hitting hard enough." They also claim that "Funktion One systems [...] will eat 192kbps MP3s for breakfast." (Note: Funktion One makes expensive PA speakers which are considered top-notch by many DJs and promoters.)

These type of stories are quite common in the DJ world. Is there any reason to believe they are true? That is, is it conceivable that medium-bitrate lossy is transparent on a home system but not transparent when played on a PA at high volume? If so, why?
Go to the top of the page
+Quote Post
pdq
post Sep 26 2012, 14:54
Post #2





Group: Members
Posts: 3305
Joined: 1-September 05
From: SE Pennsylvania
Member No.: 24233



It is possible for a faulty sound system to expose artifacts that are inaudible on properly working hardware. Codec testing and tuning is generally performed using good quality speakers or, preferably, headphones. Issues with non-flat frequency reponse or high levels of distortion negate that testing.

I'm not claiming that this explains or confirms the stories, just that it might be a possibility.
Go to the top of the page
+Quote Post
dhromed
post Sep 26 2012, 15:07
Post #3





Group: Members
Posts: 1244
Joined: 16-February 08
From: NL
Member No.: 51347



The article is a weird mix of bits of sound advice and complete misinformed or irrational nonsense.

Lossy audio compression should not be compared to jpeg compression like that.
The idea that listening to music in a club during a party will reveal artifacts better than a proper home environment is bunk, regardless of the hardware.
Dithering has nothing to do with resampling an image. (that is, not in the way it's put in the article)

On the other hand, there's the recommendation of a blind testing tool, and the correct statement that you can't magically upconvert an audio file. These are good things.

I'm not sure whether this article should be criticized for being so noisy and all over the place with its almost-technical explanations, or commended for being a step up from the usual 100% tech superstition.

This post has been edited by dhromed: Sep 26 2012, 15:08
Go to the top of the page
+Quote Post
shadowking
post Sep 26 2012, 15:38
Post #4





Group: Members
Posts: 1523
Joined: 31-January 04
Member No.: 11664



Home speakers positioned 5 ~ 6 ft apart , away from walls give more precision and detail. In such cases good speakers can be revealing a bit like headphones .. But most people esp in public will have them wide apart in weird configurations to form abnormal bass and exaggerated sound stage. They become much more 'imprecise' and would probably mask a lot of artifacts. The reverse may also be true but unproven as far as i know.

I think "will eat 192 k mp3" is total BS and 320k will feel archaic??, frequencies breathing ?? hes talking spectrograms. I mean this is total garbage cause those same people never did proper abx tests 10 yrs ago and are still clueless.

This post has been edited by shadowking: Sep 26 2012, 15:51


--------------------
Wavpack -b450x1
Go to the top of the page
+Quote Post
zima
post Sep 26 2012, 15:50
Post #5





Group: Members
Posts: 128
Joined: 3-July 03
From: Pomerania
Member No.: 7541



QUOTE (dhromed @ Sep 26 2012, 16:07) *
The idea that listening to music in a club during a party will reveal artifacts better than a proper home environment is bunk, regardless of the hardware.

Oh yeah, club parties... I can think of some indispensable elements of those, effects of which could potentially influence (not necessarily "improve" - just make it different) hearing:
http://www.hydrogenaudio.org/forums/index....hl=psychoactive
http://www.hydrogenaudio.org/forums/index....hl=psychoactive

This post has been edited by zima: Sep 26 2012, 15:51


--------------------
http://last.fm/user/zima
Go to the top of the page
+Quote Post
benski
post Sep 26 2012, 15:52
Post #6


Winamp Developer


Group: Developer
Posts: 669
Joined: 17-July 05
From: Ashburn, VA
Member No.: 23375



"improper" room and speaker configurations can unmask artifacts that were assumed by the encoder to be masked. In theory, EQ can do the same thing. I think this is an interesting area for further testing. Blind testing of MP3 vs WAV with relatively extreme EQ settings would be easy for HydrogenAudio to set up. It would also give us an idea if "bad" listening environments might unmask artifacts from transform-based codecs.
Go to the top of the page
+Quote Post
greynol
post Sep 26 2012, 15:53
Post #7





Group: Super Moderator
Posts: 10000
Joined: 1-April 04
From: San Francisco
Member No.: 13167



Regarding revealing artifacts, there are two things at play, one is frequency response, the other is unwanted noise. Flat frequency response makes artifacts more difficult to hear, not easier to hear. Additional noise, on the other hand works the other way, by providing additional masking beyond the noise introduced by the lossy compression itself.

These are just rules of thumb, of course. If I try to use just woofers to ABX a lossy file with an overly-liberal low-pass then I probably won't detect it. Similarly, if I drown out high frequencies with massive amounts of bass, it can serve to mask artifacts as well.

This post has been edited by greynol: Sep 26 2012, 16:00


--------------------
Your eyes cannot hear.
Go to the top of the page
+Quote Post
benski
post Sep 26 2012, 15:54
Post #8


Winamp Developer


Group: Developer
Posts: 669
Joined: 17-July 05
From: Ashburn, VA
Member No.: 23375



QUOTE (greynol @ Sep 26 2012, 10:53) *
Similarly, if I drown out high frequencies with massive amounts of bass, it can serve to mask artifacts as well.


Unless the artifacts were in the bass frequency region!
Go to the top of the page
+Quote Post
shadowking
post Sep 26 2012, 16:03
Post #9





Group: Members
Posts: 1523
Joined: 31-January 04
Member No.: 11664



What 'difference' may be unmasked will probably not sound anything like the disaster predicted. The ever increasing bitrates of today will probably provide some headroom to cover such cases. In the past 128 CBR this may have been an issue.

This post has been edited by shadowking: Sep 26 2012, 16:04


--------------------
Wavpack -b450x1
Go to the top of the page
+Quote Post
greynol
post Sep 26 2012, 16:05
Post #10





Group: Super Moderator
Posts: 10000
Joined: 1-April 04
From: San Francisco
Member No.: 13167



QUOTE (benski @ Sep 26 2012, 07:54) *
Unless the artifacts were in the bass frequency region!

That's the easiest region to encode, but point taken. Generally it's the exaggeration of the high end that can help to unmask artifacts, or emphasizing or de-emphasizing critical bands so the overall response is unnatural.

This post has been edited by greynol: Sep 26 2012, 16:08


--------------------
Your eyes cannot hear.
Go to the top of the page
+Quote Post
greynol
post Sep 26 2012, 16:14
Post #11





Group: Super Moderator
Posts: 10000
Joined: 1-April 04
From: San Francisco
Member No.: 13167



QUOTE (shadowking @ Sep 26 2012, 08:03) *
What 'difference' may be unmasked will probably not sound anything like the disaster predicted. The ever increasing bitrates of today will probably provide some headroom to cover such cases. In the past 128 CBR this may have been an issue.

Isn't smearing the true achilles heel of mp3 at those bitrates; and, won't poor speaker placement combined with poor room acoustics serve to make it less difficult to detect?

This post has been edited by greynol: Sep 26 2012, 16:29


--------------------
Your eyes cannot hear.
Go to the top of the page
+Quote Post
2Bdecided
post Sep 26 2012, 17:19
Post #12


ReplayGain developer


Group: Developer
Posts: 4945
Joined: 5-November 01
From: Yorkshire, UK
Member No.: 409



I think it's just another group of people who like to think that they and/or their equipment is special.

They are not, and it is not - at least, not in terms of revealing artefacts. The only exceptions are certain types of processing - extreme EQ has been mentioned, and vocal cut (or similar spatial processing) could be a killer.

However, EQ is often used to compensate for properties of the speaker or venue - so the use of EQ is to ensure that the sound which reaches the listener's ears is "flatter" than it would have been in that "coloured" environment. The other use (blasting more bass and treble) is pretty common in many people's homes + iPods, so maybe we can consider mp3 well tested in this area? Depends how many people listen to that much bass, and how much you distrust mp3 in handling bass. I've never noticed a problem.


What's very common is for people to DJ with lousy mp3s (or worse: 48kbps HE-AAC on their phone!) and then claim that the "superior" equipment makes the artefacts more audible - whereas those artefacts would have been painfully obvious on a $20 pair of IEMs.

Cheers,
David.
Go to the top of the page
+Quote Post
greynol
post Sep 26 2012, 17:28
Post #13





Group: Super Moderator
Posts: 10000
Joined: 1-April 04
From: San Francisco
Member No.: 13167



QUOTE (2Bdecided @ Sep 26 2012, 09:19) *
The other use (blasting more bass and treble) is pretty common in many people's homes + iPods, so maybe we can consider mp3 well tested in this area?

I think this generally applies to DJs as well!

QUOTE (2Bdecided @ Sep 26 2012, 09:19) *
distrust mp3 in handling bass

...is silly.


--------------------
Your eyes cannot hear.
Go to the top of the page
+Quote Post
[JAZ]
post Sep 26 2012, 19:35
Post #14





Group: Members
Posts: 1710
Joined: 24-June 02
From: Catalunya(Spain)
Member No.: 2383



The article in itself is good, in the sense that explains the problems that can appear in this scenario, and also the recommendation is logical (using lossless).

I wouldn't say that one has to blindly believe all what the article says (I've read it fast, so not paying attention to details), but a club/concert speaker system is not perfect ( crossovers, filters, horns, highs and woofers separated by several meters... ), and we all accept that DSP processing is not something that lossy encodings accept blindly.

What would NOT be correct is to say that such speakers sound better and/or that playing at a higher volume makes hearing artifacts easy. (Just remember that you have to shout to people near you to talk. That shows how high an artifact should sound to be heard).

Go to the top of the page
+Quote Post
zima
post Sep 26 2012, 19:56
Post #15





Group: Members
Posts: 128
Joined: 3-July 03
From: Pomerania
Member No.: 7541



QUOTE (2Bdecided @ Sep 26 2012, 18:19) *
I think it's just another group of people who like to think that they and/or their equipment is special.
They are not, and it is not

Plus a group likely to have above average rates of hearing damage, with all the loud music exposure regularly going on?

This post has been edited by zima: Sep 26 2012, 19:56


--------------------
http://last.fm/user/zima
Go to the top of the page
+Quote Post
krabapple
post Sep 26 2012, 22:27
Post #16





Group: Members
Posts: 2159
Joined: 18-December 03
Member No.: 10538



now read this gem

http://www.djmag.com/news/detail/445


which ends with

QUOTE
It's worth noting lossy's underlying technology is founded on perceptual coding, which is based around a psycho-acoustic model of human hearing.

In plain English, the bits of audio that are messed with in a compressed audio file are based on an opinion of how humans hear, rather than on any fundamental laws of maths or physics.

In the case of MP3, a group of scientists in Germany (The Franhofer Institute) decided amongst themselves what to chuck out and what to keep, so it's probable the sound will be coloured to a degree and therefore may actually sound better than the original to some people.
Go to the top of the page
+Quote Post
Woodinville
post Sep 26 2012, 22:49
Post #17





Group: Members
Posts: 1402
Joined: 9-January 05
From: JJ's office.
Member No.: 18957



QUOTE (krabapple @ Sep 26 2012, 14:27) *
now read this gem

http://www.djmag.com/news/detail/445


which ends with

QUOTE
It's worth noting lossy's underlying technology is founded on perceptual coding, which is based around a psycho-acoustic model of human hearing.

In plain English, the bits of audio that are messed with in a compressed audio file are based on an opinion of how humans hear, rather than on any fundamental laws of maths or physics.

In the case of MP3, a group of scientists in Germany (The Franhofer Institute) decided amongst themselves what to chuck out and what to keep, so it's probable the sound will be coloured to a degree and therefore may actually sound better than the original to some people.



Wow, the author of that quote is ignorant and proud, I guess. If he would classify cochlear dynamics as an "opinion", I guess he's not into physics, either.


--------------------
-----
J. D. (jj) Johnston
Go to the top of the page
+Quote Post
Woodinville
post Sep 26 2012, 22:51
Post #18





Group: Members
Posts: 1402
Joined: 9-January 05
From: JJ's office.
Member No.: 18957



I should note that a rising high end, put into a sound system that is used to play music to people with nasty hearing damage, may very well expose some artifacts on MP3.

So it's not out of the question, but the misinformation quoted is insane.


--------------------
-----
J. D. (jj) Johnston
Go to the top of the page
+Quote Post
rgtb
post Sep 26 2012, 23:47
Post #19





Group: Members
Posts: 25
Joined: 26-September 12
Member No.: 103446



thanks for all the insights.

This post has been edited by rgtb: Sep 26 2012, 23:54
Go to the top of the page
+Quote Post
DVDdoug
post Sep 27 2012, 17:50
Post #20





Group: Members
Posts: 2441
Joined: 24-August 07
From: Silicon Valley
Member No.: 46454



ohmy.gif I'm thinking the just the opposite... I think if I could plug my iPod (MP3s) into a big top-notch PA system (like The Rolling Stones or Lady Gaga uses), I'll bet it would sound amazing and everybody would be impressed with the loudness, clarity, and bass you can feel. I could tell 'em I'm playing 24-bit lossless high-definition audio, and I think they'd believe me.

I think the 1st time I heard MP3s, it was on a DJ system. There was "something wrong" with the highs, and I wasn't impressed. My conclusion was that MP3s were lousy... Obviously, this was quite a while ago and I don't remember exactly what was wrong with the sound. These could have been low bitrate MP3s, or maybe the speakers had piezo horns (which can sometimes have odd-sounding highs), or maybe he was over-driving the tweeters, or something like that.
Go to the top of the page
+Quote Post
Porcus
post Sep 27 2012, 20:24
Post #21





Group: Members
Posts: 1779
Joined: 30-November 06
Member No.: 38207



I for one was surprised that a simple beat loop posted here drove LAME -V0 past the 300 kb/s mark and was still ABXable, but I would still be surprised if the that compression would turn out significant in a study that properly controlled for the “my DJ set emptied the dance floor and I didn't get laid at the end of the night, what went wrong?” effect tongue.gif

This post has been edited by Porcus: Sep 27 2012, 20:24


--------------------
One day in the Year of the Fox came a time remembered well
Go to the top of the page
+Quote Post

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: 20th April 2014 - 14:54