Having the same RMS level doesnt mean the song will play at the same loudness. A song with an RMS of -8dB can sound lower
than a track with an RMS of -10dB.

Use your ears to have them sound the same level. Thats all you need to do. RMS alone will not do it

Cj

A simple way to do this would be as follows. 

I'd set up one song as the "REFERENCE SONG" for loudness. Place a wave of it in a new "Reference for volume" project and save it. 
Then.... after you export the song you are working on, as a wave. Load it into a second track in the reference project.  You can actually have that second track set up and linked together to the solo button on the first track. I have done this on occasion. Properly linked, the solo button toggles you back and forth soloing the tracks one at a time so you can easily compare the levels between the two. 

Simply press play and click back and forth using the solo button to compare the volume levels between the two tracks.  make compression and level adjustments to the second (new) track until the balance is close. then export it and save it. 

I think that as you work and become more accustom to Sonar, your tracks should be getting closer to each other in sound and volume levels as you progress. I know that mine are.  I load them into my MP3 player and it is a rare thing to need to adjust the levels on a finished tune. 

CJaysMusic


Having the same RMS level doesnt mean the song will play at the same loudness. A song with an RMS of -8dB can sound lower
than a track with an RMS of -10dB.

Use your ears to have them sound the same level. Thats all you need to do. RMS alone will not do it

Cj

I think this needs some explanation. I'm not to sure how a song at a higher volume level will sound lower in level than a song of a lower volume level. I suppose it is possible depending on a very wide difference on song type but even so RMS is a very good indication of loudness.

John

CJaysMusic


Having the same RMS level doesnt mean the song will play at the same loudness. A song with an RMS of -8dB can sound lower
than a track with an RMS of -10dB.

Use your ears to have them sound the same level. Thats all you need to do. RMS alone will not do it

Cj

I think this needs some explanation. I'm not to sure how a song at a higher volume level will sound lower in level than a song of a lower volume level. I suppose it is possible depending on a very wide difference on song type but even so RMS is a very good indication of loudness.

Hi John,
  
  there is a time factor involved - 1 100Hz square wave will sound louder at the same rms reference as a 100hz sine wave.  The square wave spends 100% of the time scale at 100%  +/- output.  The sine wave does not. 


 I have probably oversimplified that but it demonstrates the concept.


Cheers


Grif

You could use this (30 day trial) http://hindenburgsystems.com/products/hindenburg-journalist 30 day trial should be enough time to get all your tracks done.

It levels any audio automatically to a set "Loudness" using EBU compliant loudness. Then you could just use gain to push the volume up a bit more if needed. Great for mastering to get all tracks at a similar perceived loudness ;-)
 
Of course you could do it in Sonar with something like toneboosters EBULoudness Meter, just takes longer as you have play every song all the way through.

Having the same RMS level doesnt mean the song will play at the same loudness. A song with an RMS of -8dB can sound lower than a track with an RMS of -10dB.

Only if the songs had a significantly dissimilar spectral content, which probably isn't the case for the OP and his live backing tracks. Of course -8db at 3KHz is going to sound louder than -8db at 30Hz, but it's unlikely he's playing sine waves onstage.

I suppose it is possible depending on a very wide difference on song type but even so RMS is a very good indication of loudness.

Loudness is a physiological/psychological concept, only roughly related to the sound pressure level (spl=power) of the sound. It is most obviously related to the sensitivity of the human ear, where the frequency of the sound is a predominant factor:
+



Even after adjusting for frequency characteristics (which the rms of the waveform does not), there remains a large area of subjectivity. A sound presented by itself arising out of silence, will be perceived as much louder than a sound buried in a piece of music. This can partly be explained by the physiologic tension on the tensor tympani that occurs reflexively to dampen the movement of the ossicles and protect the hair cells from damage. But it is also a psychological effect, in that the sound is noticed more and hence becomes louder in the mind. And another strong pschophysiologic effect, auditory masking, is even more important. 



The same spl voiced by different instruments or sources is likely to have quite different loudness to a human mind, even when it reads the same on a meter. And the same sound when heard in the company of other sound will have even more pronounced variance from that predicted by spl.

Algorithms that produce a standard loudness so you will not have to adjust the gain on your MP3 player, mainly act as limiting compressors, and can do some considerable damage to what was intended to be a piece with a wide dynamic range. Human beings playing ensemble, adjust their  volumes to match a compatible subjective loudness. If you want to use robots in your performance, I agree with CJ, that for the best effect, you need to rehearse with them as though they were members of the band. Just running them though a computer program, or watching your meters will not provide optimum results. 

bitflipper

Having the same RMS level doesnt mean the song will play at the same loudness. A song with an RMS of -8dB can sound lower than a track with an RMS of -10dB.

Only if the songs had a significantly dissimilar spectral content, which probably isn't the case for the OP and his live backing tracks. Of course -8db at 3KHz is going to sound louder than -8db at 30Hz, but it's unlikely he's playing sine waves onstage.

2:22 in hehe


http://www.youtube.com/watch?v=IiP4BZ-1mM4


it probably is messed up with compression , compared to the cd version 

Maybe we are talking about very different things but the OP simply wants to have his MIDI based background songs playing at similar loudness so he can play live without having to adjust his volume knob all the time. 

We don't need a dissertation on audio loudness for him to achieve that.  

Here is what I do.
 
1. Make all the tracks the same format. I use .wav because .wav files are not companded like .mp3s, and so it is much easier to limit them equivalently.
 
2. Assemble all the .wav files into a useful group. I like to put them all in a folder, but sometimes group them into styles because similar styles can be treated with similar volume/loudness attenuation.
 
3. Play the files back and watch the meters. Even with the same gain settings, some will be louder, some will be quieter.
 
4. Raise the volume of the tracks that are too quiet.
 
5. Lower the volume of the tracks that are too loud.
 
6. Put a limiter (UAD LA2A is my choice) and use it to keep any tracks from clipping. The actual settings should be adjusted by ear, but even having this limiter on the tracks will make them sound better.
 
7. Now this is the most important part: make sure the instruments you are matching to the tracks have similar effects, and are appropriately matched to the backing tracks. You can spend a lot of time making the backing tracks the same volume, and then screw it up by having the live instruments too loud or quiet. That is why I always have a volume pedal in the chain and can use it to match live instrument volume to backing tracks. This also can help with different rooms and the way their acoustics impact your overall loudness.
 
8. Get the best in-ear monitors you can afford. They will make volume matching between live instruments and backing tracks a lot easier.
 
9. If you can afford it, hire a sound engineer to run the tracks through a mixer and adjust volume in the mixer.
 
Paul