My Avg volume is louder than Dave Mathews song BUT perceptively lower. Why?

[dblock]

I doing a comparison with a Dave Mathews song and a few bars of music I composed and mixed just for my experiment.
Mathews song avg volume is -12.5db, Peak volume 0, max rms is -5.89
My song avg volume is -9.26db, Peak volume 0, max rms is -4.29

Seems mine is technically louder but not perceptibly louder.
Do you guys have any basic suggestions what I can do to improve the perceived volume of mine?
Also are there any software auto correction eq's that might help?
Thanks

[Cellotron]

It's due to two things:
frequency (spectral) content
transient shape (the sharpness or smoothness of the initial attacks)

Tracks with a greater amount of mid to upper mid frequency content - along with sharper transients - will sound perceptibly louder even when the peak and RMS meters are showing the same exact levels - relative to tracks with less forward upper mids and with smoother transients.

Best regards,
Steve Berson

[Cheebs Goat]

Originally Posted by dblock

Seems mine is technically louder but not perceptibly louder.
Do you guys have any basic suggestions what I can do to improve the perceived volume of mine?

Less low EQ and more high midrange EQ can sound louder. Different arrangements and mix choices can seem louder as well...

Also are there any software auto correction eq's that might help?
Thanks

Probably not. And if there were, it wouldn't be a good idea.

And why are you trying to do this anyway? If you make EQ and mix choices for loudness instead of making those choices for the best sound, you end up with something shrill and harsh. If something shrill and harsh is coming out of a speaker, the listener is going to keep the volume low. So it won't matter how loud the computer analysis thingamajigger says it is. In the real world, it is quiet.

Make it sound good so the listener turns it up. Then it will be loud for real.

[dblock]

Originally Posted by Cheebs Goat

Less low EQ and more high midrange EQ can sound louder. Different arrangements and mix choices can seem louder as well...
Probably not. And if there were, it wouldn't be a good idea.

And why are you trying to do this anyway? If you make EQ and mix choices for loudness instead of making those choices for the best sound, you end up with something shrill and harsh. If something shrill and harsh is coming out of a speaker, the listener is going to keep the volume low. So it won't matter how loud the computer analysis thingamajigger says it is. In the real world, it is quiet.

Make it sound good so the listener turns it up. Then it will be loud for real.

Thanks for your input.
I'm making eq mix choices to be loud "and" good. I really don't think I necessarily have to choose one or the other. Just looking for ways to improve the odds that I can get both. I'm sure others can.

[dblock]

Originally Posted by Cellotron

It's due to two things:
frequency (spectral) content
transient shape (the sharpness or smoothness of the initial attacks)

Tracks with a greater amount of mid to upper mid frequency content - along with sharper transients - will sound perceptibly louder even when the peak and RMS meters are showing the same exact levels - relative to tracks with less forward upper mids and with smoother transients.

Best regards,
Steve Berson

Thanks Steve. That's a start.

[Cheebs Goat]

Originally Posted by dblock

Thanks for your input.
I'm making eq mix choices to be loud "and" good. I really don't think I necessarily have to choose one or the other. Just looking for ways to improve the odds that I can get both. I'm sure others can.

You can get both. Very easily. Just go 100% in the direction of "sounds the best". The listener will turn it up as loud as he wants. I don't know why people have any concern for volume at all on a recording (outside of the relationships between different parts of the same body of work).

The data on the CD has almost no relationship to the force with which the speakers push air. It would in a world where the listener/DJ/radio station is forbidden to touch the volume knob. That world doesn't exist and never will.

Look at what really makes a song loud. How powerful is the listener's amp? How large are his speakers? You have no control over that...but... How catchy is the song? How much does the production move people? How badly does the listener want to turn it up and feel the music? That stuff determines how hard the air gets pushed.

An infectious song with an aggressive production and no ragged shrill edges will come out of the speakers loud because the listener wants it to come out of the speakers loud.



Funny thing about human ears: We think loud sounds better. Always. Right up to the threshold of pain. So bands constantly make the mistake of A/B'ing a loud production against a soft one without compensating for the volume difference. When a twist of the volume knob is added to the softer production and a comparison is made without a volume difference clouding judgment, people can see how confined and small the "loud" production sounds in comparison. It is the opposite of "large and kicking ass".

[Table Of Tone]

It really starts with what's happening in front of the microphones and how well it's recorded!

If the above is all in order, the recording will still keep it's balls, all the way through the mixing and mastering, and the RMS doesn't need to be at an insane level.

[bkuijt]

Music that's played and sung effortlessly can also sound much louder then a whole full band shouting and being compressed to death.

[therealbigd]

this effect is one of the key differences between a good ME and a bloke with brickwall limiter.

[PatrikT]

Originally Posted by dblock

I doing a comparison with a Dave Mathews song and a few bars of music I composed and mixed just for my experiment.
Mathews song avg volume is -12.5db, Peak volume 0, max rms is -5.89
My song avg volume is -9.26db, Peak volume 0, max rms is -4.29

Seems mine is technically louder but not perceptibly louder.

What you should really reflect upon is that it obviously seems that RMS is pretty useless and still about everyone is talking about RMS or even try to set standards using it.

Think about it for 0.4 seconds. Robot ear just can't hear.

I guess some people seem to think that measuring speed by only reading the RPM meter while driving a car is a marvelous way of knowing how fast things go.


Best Regards
Patrik

[PatrikT]

Originally Posted by Cheebs Goat

The data on the CD has almost no relationship to the force with which the speakers push air.

You need to reboot.


Best Regards
Patrik

[Cheebs Goat]

Originally Posted by PatrikT

You need to reboot.


Best Regards
Patrik

I don't follow what you're saying.

But I stand by what I wrote.

Take a hypothetical digital song that is nothing but a full-scale square wave at 500 Hz. Now place it in some playback system and play it under some sort of conditions.

How loud is the sound you are hearing? Is it loud enough to hear over 10 people talking? Is it soft enough that you don't need ear plugs? Is it soft enough that you don't get instant hearing damage? Is it loud enough to even hear at all from the next room?

If the data on the disc correlates in any meaningful way to the playback volume, that question could have at least a ballpark answer.

[PatrikT]

Originally Posted by Cheebs Goat

Take a hypothetical digital song that is nothing but a full-scale square wave at 500 Hz. Now place it in some playback system and play it under some sort of conditions.

There are no such songs in the real world so why relate to that?

How loud is the sound you are hearing? Is it loud enough to hear over 10 people talking? Is it soft enough that you don't need ear plugs? Is it soft enough that you don't get instant hearing damage? Is it loud enough to even hear at all from the next room?

Who cares, 500 Hz of constant square wave is not even executed by Squarepusher.

If the data on the disc correlates in any meaningful way to the playback volume, that question could have at least a ballpark answer.

The data on the disc correlates to voltage. That's how things work.


Best Regards
Patrik

[Cheebs Goat]

Originally Posted by PatrikT

The data on the disc correlates to voltage. That's how things work.

Are you forgetting that amplifiers are involved or are you trolling?

OK, a 500 hz square wave is boring. Take the original CD master of Sgt. Pepper and play "When I'm 64" on some system under some conditions. 43 seconds into the song, tell me how many volts are going to the right speaker.

Again, if the data on the disc had any correlation to real-world amplified voltage, sound volume, air displacement, or any measure of loudness, you should be able to give that a ballpark answer.

You can't. Because it doesn't. So it makes no sense to push for volume by manipulating data on the disc.

[PatrikT]

Originally Posted by Cheebs Goat

Again, if the data on the disc had any correlation to real-world amplified voltage, sound volume, air displacement, or any measure of loudness, you should be able to give that a ballpark answer.

On my DAC and ADC I have volts on one side and data on the other side.

It must be a truly unique design.


Regards
Patrik

[Jesse Graffam]

Both of you guys know what each other is saying. Stop being silly. thumbsup

The OP should be happy with the answers so far anyways, and there's a ton of other topics already about the same subject.

[princeplanet]

Originally Posted by therealbigd

this effect is one of the key differences between a good ME and a bloke with brickwall limiter.

Yeah, those good ME's, what genuises, they don't just brickwall, they eq the low mid mud out before the brickwall! .... Look, just mix it right (with a real pro), then any idiot can bring it up tp -7dB RMS and make it loud, and good....

[Table Of Tone]

Originally Posted by princeplanet

Yeah, those good ME's, what genuises, they don't just brickwall, they eq the low mid mud out before the brickwall! .... Look, just mix it right (with a real pro), then any idiot can bring it up tp -7dB RMS and make it loud, and good....

That's actually wrong!
You need a good recording, a good mix eng and a good ME!

[Jesse Graffam]

Originally Posted by Table Of Tone

That's actually wrong!
You need a good recording, a good mix eng and a good ME!

Any idiot can do all of that.

[FBM]

Originally Posted by dblock

Also are there any software auto correction eq's that might help?
Thanks

http://www.spl.info/index.php?id=746&L=1

+

Waves | Plugins | L3 Multimaximizer