Dynamic Compression and the "Lay Person"

[avaserfi]

I thought this would interest a few people here. As we all know dynamic compression, almost always sucks. I decided to do a little study on the preferences of the "lay person" or non-audio nuts.

I took an extremely dynamic Steely Dan song and compressed it (just dynamics) and then perceptually level matched the original song and the compressed version and started testing on my reference headphones (chosen for their linearity in frequency response). Right now the study is in its pilot stage (the reason why I am not going into full details). I will say though so far even if the participants consistently fail an ABx test between the two files and hearing differences it is statistically likely that in a blind comparison the subjects do prefer the uncompressed file. Due to this being a pilot study the results aren't generalizable enough yet, but I am working on that.

Hopefully, next semester when I have more time to commit to this project I will get more subjects and perhaps even the recording companies will listen, but that is probably being idealistic.

Lastly, sorry if I have posted in the wrong forum, but music seemed the perfect place for this as its ruining our music as I am sure most of you know!

[ISLAND1000]

avaserfi wrote:

I decided to do a little study on the preferences of the "lay person" or non-audio nuts.

Why did you do that?

[avaserfi]

ISLAND1000 wrote:

Why did you do that?

Really, why?

Because the people who control the industry aren't the ones who buy niche products and specialty recordings that are of great quality. It is the lay person who goes to the local big box and picks up the CD on the top of the charts. Sadly, nearly all, if not all, of these discs are horribly compressed and the majority of people do not realize that this happens. So I figured I could open a few eyes and find out if people really prefer this compression or if they are just unknowingly accepting it.

In the end, if any change is going to happen it will not be because of us, the audio nuts, but because of the average consumer and no I don't think my single study will change much if anything, but it is a start in the right direction as I see it.

[ISLAND1000]

I see . . . . .

[rcarlton]

Good luck. I am interested in reading more.

[Josuah]

CARS. Unless people stop listening to analog radio and CDs in their cars, or on the train, bus, etc. this won't change.

BTW - Be careful you don't clip. Compression makes it much easier to clip your amp.

[avaserfi]

Josuah wrote:

CARS. Unless people stop listening to analog radio and CDs in their cars, or on the train, bus, etc. this won't change.

BTW - Be careful you don't clip. Compression makes it much easier to clip your amp.

Cars have no bearing on the situation. Radio stations and such can compress songs on their own for their own application and often times do (check out classical stations). This compression is not due to sound quality concerns but to allow for less costs involving sending the signal as less bandwidth is used. Also, people using proper ear buds rather than the iPod buds make all the difference.

As far as I understand compression does not make it easier to clip your amp. You might hear the clipping on the source, but why would having a compressed source cause any more clipping than another?

Either way in this case it doesn't matter as there is no need for the headphones I am using to be amplified and I have never experienced any audible clipping when testing all the equipment and sources.

[Josuah]

Didn't know that about radio stations. But the argument I've always heard for why compression is done is specifically for stuff like radio stations, where you want to make sure everything is loud, and loud as or louder than everything else.

If the PCM signal generated by the redbook audio on the CD is clipped, then it'll try to drive your amp with that signal, causing the amp to clip (not due to hitting the ceiling, so perhaps I should not have used clip), and then possibly damaging your speaker, no?

[avaserfi]

Josuah wrote:

Didn't know that about radio stations. But the argument I've always heard for why compression is done is specifically for stuff like radio stations, where you want to make sure everything is loud, and loud as or louder than everything else.

If the PCM signal generated by the redbook audio on the CD is clipped, then it'll try to drive your amp with that signal, causing the amp to clip (not due to hitting the ceiling, so perhaps I should not have used clip), and then possibly damaging your speaker, no?

Yeah, radio stations are more than willing to compress music themselves as it is a very simple process if you aren't worried about introducing other artifacts.

As far as a source clipping and it being transfered through the amp: I don't think this form of digital clipping is harmful to the speaker. From what I understand, and I admit a lack of knowledge in the area of amplification as currently my focus is speakers, these are two different phenomena due to the way an amplifier itself acts when it clips versus the way it acts sending a clipped signal. So a source clipping isn't detrimental to speakers while an amp clipping itself can be, but isn't always. Think about how much trouble most speakers would be in with this current compression trend in the industry if this really did damage them.

[thxgoon]

Ya a compressed signal is more demanding on an amp because it will ask for more continuous high output. A compressed signal lacks dynamics you would normally find so the amp is not likely to clip but will have a more even output which could creat tremendous heat if played loud for long times. It has been my understanding that compression is used in recording and at radio stations to create the 'biggest' and 'loudest' sound that they can. They basically set the master recording volume very close to reference and then compress the snot out of it so that it neither clips nor dips into quiet.

On a side note - this is similar to what they do with tv commercials. The max volume for commercials is actually regulated to be lower than that for the program material but advertisers compress the commercials and set them as loud as they are allowed so they catch your attention. That's why they always seem louder than the show you are watching.

[avaserfi]

Well, everything is said and done and the results are in!

Everything was completed with 45 participants - 27 males and 18 females between the ages of 18-56. In the end 86.7% of listeners preferred the uncompressed file (Average RMS 19dBfs) to the compressed file (Average RMS 11dBfs) this ends us up with a 95% confidence interval saying that between 76.8% to 96.6% of participants would prefer the uncompressed to the compressed in the general population.

Just to see if there were any variables involved with preference I compared age, sex, music ability, and music interest to see if any trends were apparent and there were none which shows that this preference was universal.

I also have a independent study next semester so I can dedicate far more time to expanding this experiment so more results will come.

Here is an image of the files the people listened to the first part is clearly uncompressed and the second is compressed. Thanks to Chris (WmAx) for setting the files up for me.

[phaseshift]

Excellent post-

Any chance you would be willing to share the spreadsheet with some of us? I am curious to see how it broke down.

Also, was the comentary of the test subjects recorded? It would be interesting to see what the perceptual differences were.

[avaserfi]

phaseshift wrote:

Excellent post-

Any chance you would be willing to share the spreadsheet with some of us? I am curious to see how it broke down.

I will try and compile some ranged data tonight and post it for everyone. Shouldn't be a problem.

phaseshift wrote:

Also, was the comentary of the test subjects recorded? It would be interesting to see what the perceptual differences were.

I didn't record any actual commentary, but after participants completed the test and filled out the questionnaire I talked to a few about it. Most thought the compressed version sounded like the brass was missing (I used a section I.G.Y by Fagan); some simply couldn't tell me why they liked the uncompressed more, but they did; some just said the uncompressed version flowed far better. The few people who chose the compressed version told me the brass was too accentuated for their tastes.

The next study will be more in depth and will hopefully answer more questions.

Edit:

Here is a quick rundown of the data gathered:

Males - 27
Females - 18
12 Considered themselves musicians (the rest did not)
39 Preferred uncompressed compared to 6 comparing compressed

25 participants ages ranged from 18-24
10 participants ages were from 25-35
10 participants were from 35+

If there is anything else you want to know please let me know.