Harvey - I need help with db! I am so lost.

  • Thread starter Thread starter wes480
  • Start date Start date
wes480

wes480

New member
Ok, I started off reading this article on db a while back....and, it made pretty good sense in genereal. Except for the thing where it says that 6db gain makes something twice as loud - and I have heard from pro audio stuff that 3db is the smallest difference we can hear (or thereabouts). So that means the smallest difference we can hear is 50% louder...which...I guess..anyways, that article is here.

http://www.prorec.com/prorec/articles.nsf/files/EA68A9018C905AFB8625675400514576

My problem is - now I am in a physics of sound class at school...and we were talking about db today. And...according to the prof, we can barely notice a 1db change...and, roughly 10 times intensity makes something twice as loud to the human ear - so I am just all confused.

If someone really knowledgeable on the subject wouldn't mind posting maybe a basic db primer? I just need to know the info I should trust.

Oh, and also....where did 0db maximum come from, in regards to digital? Did they do that just to separate it from analog?
 
Well, three dB would be a doubling of volume. and one dB is supposedly the smallest increment that a human can detect a change in (but in reality it's more like .1 dB that a human can detect a change).
 
Ah we just had a similar discussion in the cave recently!

+6db is a doubling of amplitude (e.g. voltage), not loudness. A +6db change correlates to a sound that is 1.52 times louder than the original. It takes a ~ +10db change before something sounds twice as loud.

The 0db normalized db scale simply comes from the fact that in the digital world, there ain't nothin higher than the maximum value of a particular data type. For instance, the maximum value that can be represented by a signed 16bit integer is 0111111111111111, or +32768. That's 0db. You can't go any higher.

I'll let harvey & skippy and the guys explain the real nitty gritty. I'm just killin some time :)

Slackmaster 2000
 
ok. so that article is wrong? it says this..(has the math there..as well)


+ 1 dB means about a 12% increase in SPL
+ 3 dB means about a 40% increase in SPL
+ 6dB means about twice as loud (200% of the original SPL)
+ 12 dB means about four times as loud (400% of the original SPL)
- 1 dB means about 90% of the original SPL
- 3 dB means about 70% of the original SPL
- 6 dB means about half of the original SPL
- 12 dB means about one-quarter of the original SPL

(http://www.prorec.com/prorec/articles.nsf/articles/3F50D03B51D16EF386256904007DFFC0)
 
Track, +3db is a doubling of power, not loudness or amplitude. It looks like this:

+3db = 2X power increase, 1.4X voltage, 1.23X Loudness
+6db = 4X power increase, 2X voltage, 1.52X Loudness

Slackmaster 2000
 
+6db is a doubling of amplitude....not loudness. - Slack

Ok, I think we are getting somewhere. Could you or anyone go into a little more detail on that? Becuase, I just see "db" and assume its all talking about the same thing...hell, why wouldn't it be? heh.
 
I won't tackle all of that, but...

I should be able to clear a few things up for you. The problem is that some of the info you're quoting is subjective. While it is true, technically, that adding 6 db to any sound will double it's power, the "fact" that we can only hear a 3db change is really kind of ridiculous. Is this a 3db increase in overall program material from 85 db, or a reduction of 3db of the acoustic guitar backing in a rock tune, or an increase of 3db in broadband white noise? It also depends on the accuity of the ears of the listener. When I started out years ago I probably would have had trouble hearing a 3 db change in a whole song. Now I can hear a 1/2 db increase in 450 Hz in a vocal track while it's in the mix (as long as I'm doing it to A/B). THAT part of the equation is all about experience and practice.

What your professor referred to is the more subjective part of the topic. The 1 db part is about right for the experienced ear. As to the other part, what people PERCEIVE as twice as loud varies from person to person, but an order of 10 times the intensity must be an error on his part or a misunderstanding on yours.

I believe the 0db digital convention was used simply because analog has no hard limit (more distortion is perceived as warmth up to a certain point), and digital does. It wouldn't make sense to make that digital limit 71.4 or 213, so 0 was used. Pass that and you have unusable, distorted, clipped audio.

Hope this helps.
 
Slackmaster2K said:
Track, +3db is a doubling of power, not loudness or amplitude. It looks like this:

+3db = 2X power increase, 1.4X voltage, 1.23X Loudness
+6db = 4X power increase, 2X voltage, 1.52X Loudness

Slackmaster 2000

Man you have it down. Gives me hope that maybe a couple days from now I will get a clue ;)
 
Well I'd hate to disagree with a prorec article, because I'm really new to this stuff, just like you. The prorec article talks loudness in terms of SPL, which is into the realm of "that's not something I'm too sure about."

Also, "power" and "voltage" are not the same thing. Voltage is a measure of potential, and power is measured in watts. By doubling the "power", you gain 3db, so a 20W amplifier should crank out 3db more than a 10W amplifier.

I'm going to step out so that somebody "in the know" can answer your question with more authority.

Slackmaster 2000
 
Last edited:
Just in case anyone was curious the 0db rating is where no amplification or cut is applied to the signal. Makes sense to put the digital ceiling at unity otherwise you would have to introduce further gain just to get a 'perfect' level. It's always better to cut than add gain.
 
Yes and no...

The human ear can discern very minute changes in dB with very specific individual sounds by themselves. The 3dB figure your professor was referring to is the level detectable in program material. This is a very important distinction that is often not mentioned when discussions of this type begin.

Personally I think our ears are more sensitive than that figure, but it works fine as a general rule.

Tom Cram
dbx Senior Technical Support
(801) 568-7530
tcram@dbxpro.com

"Jazz, pfffft...They just make it up as they go along."
-Homer Simpson-
 
there was a very good article in Sound on Sound a few months ago about dB and the various expressions - dBv, dBu, and dB, and how they correspond to voltage, power, and SPL.

If we're not all talking about the same dB expression, no one will understand anybody.
 
Well, the "revalation" was the difference between "Intensity" and "Sound Pressure Level." If the author is correct by saying that perceived loudness and SPL are the "same thing", as he seems to imply, then +6dbSPL is indeed twice as loud.

I'm still looking forward to somebody else jumping in with a clear cut explanation. Like you, I have always been told that it took +10db for something to be considered twice as loud.

Slackmaster 2000
 
hmm...i am getting it now.
gonna go have a chat with the prof. on thursday, should be able to get everything straight.
 
I typed for about half an hour, and just as I was getting ready to post my explanation, my laptop got the Blue Screen Of Death. Gone.

Oh, well: find that thread from a few days back, over in the Cave. Sorry about that, but I hhave to get on with work now- if time permits later, I'll try to reproduce that...

Meanwhile, try a google search on "Fletcher-Munson", sones, and phons...
 
Skippy, I did look up your post to my thread in the cave, which was very informative BTW, but it seems to contridict what the author of these articles at prorec says. He says that +6dbSPL is twice as loud, and you said that +10db is twice as loud. Perhaps it's all just a confusion of terms?

If you could sort of simply reiterate what you said in that thread, but apply it directly to whatever the heck is going on in this thread, then that would be GREAT! I learn a lot from your posts.

(BTW, the post in question is called "Let's talk decibels" and is ~2 weeks old)

Slackmaster 2000
 
Back
Top