sorting the Pan Law setting(s)

Was wondering if someone could point to an example of why one "Pan Law" configuration would be better than another. Different situations?

I currently have my Pan Law set at -3bd Compensated ... I'm very picky about how audio is panned and I listen very attentively when doing so. I'm quite happy with the setting, just wondering when it may be appropriate to have it set otherwise.

cheers!

It's a big misconception that the choice of a pan law or another affects the sound of your mix - it doesn't. I can only think of one exception to that rule, and I'll discuss that exception a bit further.

My honest opinion: don't overthink it. Most DAWs don't have a pan law setting, and most successful engineers don't know what a pan law is. The few that do have generally no idea what pan law their software uses. In other words, knowing which pan law you're using is of little interest to you the sound engineer so long as you use your ears. When using your ears, you'll automatically adjust/compensate for anything, so the behavior of the tool is of little importance (in that case).

Now if you really can't sleep at night, and just HAVE to know, here's how I answered that question earlier on this forum:

0dB is the default setting, it doesn't compensate anything, so the result is that if a signal is panned in the center, it stays at its nominal level (0dB of gain), and as you pan a signal to one side, the level drops (as you reach only one side, you only get one speaker playing your signal rather than two).

-3dB is a first attempt at smoothing out the level of a signal as you pan it, by gradually reducing the gain of the signal as you bring it back to center, ending up at -3dB when panned dead center. Not an ideal solution, but the advantage is that the level remains fairly constant as you pan the signal across the stereo field.

-3dB compensated is the best of both world: using the -3dB setting but adding 3dB of gain to the whole signal. So it's pretty much raising the gain as you pan to a side, so that the level remains constant as you pan across the stereo field, but not losing -3dB when panned center. The level is consistent as you pan, although you'll see the level rise on the meters as you pan to a side.

About the only time when you'll actually HEAR a difference between pan laws is if you automate the pan of an instrument, having it go from left to center to right to center to left, etc... if you do this, both "-3dB compensated" and "-3dB" will sound the same, but they will both sound different from "0 dB" where the signal will drop in volume as it reaches one side or the other.

In any other situation, you'll automatically rectify the pan/volume manually while listening, using your ears, as you mix anyway, so pan laws don't make any difference.

PS: Many think (wrongly) that because they take an existing mix, change the pan law setting and get a wider/narrower image, that the pan law setting affects the perception of width in a mix. Really all they're doing is changing the volume of all their instruments depending on their panning position. Of course the width of the mix changes when you do that. But had you mixed with another pan law setting, for a given desired result, you'd have put the volume faders at another position yourself anyway, and gotten the same result in the end.

Thanks so much, David.

How you described it is exactly how I was thinking of it. Groovy!

and ps... I am thoroughly enjoying your Logic Pro 9 book. I'm skipping around a good bit (I felt real comfortable with most of the first sections) and am having a blast. Great job, as usual!

Cheers!

Great explanation David, however there is one thing you left out.

In LP8, I haven't experimented in 9 with this yet so it might be fixed, if you record in a signal from outside of logic. Say your final mix after it goes through a analogue board or summing amp and you have your pan law set at -3db, when you bounce you will loose 3db in signal. Here's how this happens:

You mix using outboard gear and it sounds fine coming in through a aux or regular track. When you record in the audio file mix it comes in with full level. But sense the panning of the track is set to center and your pan law is -3db when you solo and bounce that track it is bounced 3 db lower. Sinse you spent all that time mixing it and you listened to it carefully coming in you don't check the newly recorded audio file because you just heard it coming in. When you send it to someone and they ask why its so quite. This took me months to figure out, mostly because the music editor I was sending too didn't say something right away but anyways its a rare thing but it happened to me.

This is just a FYI and like I said a very rare situation but worth mentioning.

PS: Many think (wrongly) that because they take an existing mix, change the pan law setting and get a wider/narrower image, that the pan law setting affects the perception of width in a mix. Really all they're doing is changing the volume of all their instruments depending on their panning position. Of course the width of the mix changes when you do that. But had you mixed with another pan law setting, for a given desired result, you'd have put the volume faders at another position yourself anyway, and gotten the same result in the end.

David is so right! And I might add that achieving a wider "sounding" mix is something that can also be accomplished in the mastering process by separating the mix into mid and side channels and changing those levels in relation to each other. This can be accomplished with Logic's "Direction Mixer" if used subtly as a mastering tool to reduce the level of the mono (center) signal.

Steve H

But sense the panning of the track is set to center and your pan law is -3db when you solo and bounce that track it is bounced 3 db lower.

Use the -3dB compensated setting and this won't happen.

David's summary is correct and accurate. All DAWs follow a pan law, and many allow you to change it.

One thing however - it's most common to have some kind of pan law that reduces the centre signals in comparison to when panned to the extremes, by various amounts. (For instance, ProTools uses a -4.5dB pan law, I believe).

Until Logic 7, Logic used to have one pan law, which was 0dB - ie, no drop in the centre. This was, I believe, a contributory factor into why many people found mixing in Logic less than satisfactory, in getting mixes with depth (and this is, of course, down to their mixing ability) compared to other systems.

Switching to a -3dB or -3dB comp pan law when it was introduced in LP7 in many cases was a bit of a revelation for these people - not because it instantly made their existing mixes better in one click (for reasons David points out), but because people seemed to find it *easier* to create mixes with depth, than when using the 0dB no compensation setting.

I showed this to many people back then and almost without fail people were amazed at the differences for them in getting better mixes.

In the end, it's down to user ability, but it definitely seems that some compensation results in an easier mix experience, and is also why most mixers, software and hardware, do not use the 0dB setting.

Just FYI

beej you are the pan guy today

thx for bumping this thread!

Just noticed.... there is a tick-box under the options for your Pan Law Settings in Logic 9 -- the box is regarding the option to apply -3db Pan Law to Stereo Balancers.

http://www.nielbrooks.com/images/box1.jpg

According to the manual, all old (pre Logic 9) Projects will have this box automatically ticked when opened in Logic 9. So, I'm assuming that in Logic 8/7, the Stereo Balancers were set to -3db Pan Law by default when you chose to use Pan Law. Interesting.... I'd never noticed that little box until this evening.

http://www.nielbrooks.com/images/manual.jpg

So, I'm assuming that in Logic 8/7, the Stereo Balancers were set to -3db Pan Law by default when you chose to use Pan Law. Interesting.... I'd never noticed that little box until this evening.

There's a thread discussing this a few posts down.

No - previous versions of Logic did not conform to your pan law at all when balancing stereo tracks - because stereo tracks cannot be panned, only have the relative balance of each channel adjusted.

The new option let's stereo tracks follow the pan law as if they were panned.

Ahh, I didn't see the other thread - thanks.

Totally bizarre that Logic defaults to ticking the box when opening older projects if the Pan Law didn't work as discussed. Their clearly noted "for compatibility reasons" phrase is exactly wrong in this case - it should be unticked "for compatibility reasons" if older versions ignored the Pan Law Comp on stereo tracks. (and yes, I understand the difference b/t mono and stereo tracks in logic with regards to the "pan" / "balance" knob.)

Many think (wrongly) that because they take an existing mix, change the pan law setting and get a wider/narrower image, that the pan law setting affects the perception of width in a mix. Really all they're doing is changing the volume of all their instruments depending on their panning position. Of course the width of the mix changes when you do that. But had you mixed with another pan law setting, for a given desired result, you'd have put the volume faders at another position yourself anyway, and gotten the same result in the end.

Hi David. Sorry if I'm being slow on the uptake here. Before I knew anything about Pan Law settings, I mixed a song with Pan law at -3db. I know need to export/bounce the stems for use in a live performance, and thanks to this thread, I've realised that the exported tracks are 3db lower due to my pan law settings. The quickest way to fix this seems to be to change the settings to -3db COMPENSATED, which means that the tracks are exported at the same volume as they are in the mix. But am I right in thinking that this will change the mix? If so, it's not a great option because I've already spent a lot of time getting the original mix of the lets call it "cd version" to sound how I wanted.

While I'm writing, it's just occurred to me that, just for the purpose of exporting the tracks for use in my other Live project, I could change the mix to -3db compensated, export, and then revert the pan law settings to -3db so that my original "cd mix" project isn't changed. Am I correct in thinking that this way, the tracks exported to my live project would be exactly the same volume as my original project.

I hope this is clear and apologies if I'm doing something obviously wrong.

But am I right in thinking that this will change the mix?

It should not change your mix unless you start distorting an I/O plug-in or some 3rd party 24 bit plug-in or change the way a threshold-plug-in processes your file - for example let's say you had 8 channels of drums outputing to a bus and a compressor on that bus, changing your pan law means the compressor will no longer see the same level, therefore its threshold will have to be readjusted as well.

I'm reading "Mixing Secrets" by Mike Senior. He says, panning to extremes leads to a perception of a 3dB drop in those panned instruments when listening in mono on a radio for example p. 127

In pan law "-3 compensated", is the 3dB mono panned center equal to stereo at center, both at 0dB? and the 3dB mono at extremes is at 0dB whilst the stereo is at +3dB? In other words, the perceived loudness graph in mono is a straight line resting on 0dB but the stereo is a "smiling" curve?

I'm reading "Mixing Secrets" by Mike Senior. He says, panning to extremes leads to a perception of a 3dB drop in those panned instruments when listening in mono on a radio for example

You'll also get that perception when listening in stereo anywhere! (try it for yourself)

In pan law "-3 compensated", is the 3dB mono panned center equal to stereo at center, both at 0dB? and the 3dB mono at extremes is at 0dB whilst the stereo is at +3dB? In other words, the perceived loudness graph in mono is a straight line resting on 0dB but the stereo is a "smiling" curve?

Reading this made my head ache.

It's really simple: -3 dB pan law means you're applying -3 dB of gain to the signal when the pan is in the center, and raising the gain back up to 0 dB as the pan knob goes toward one side. If you decide to do the same with stereo balancer, then.. well the same will happen with stereo balancers: -3 dB when the balancer is in the center, progressively going to 0 dB when it's hard left or hard right.

Now -3 dB compensated means you compensate for the -3 dB of gain by applying +3 dB of gain. So if you want, you can re-read the previous paragraph and replace "-3 dB of gain" with "0 dB of gain", and "0 dB" with "+3 dB".

HTH?

PS: insert a test oscillator on your audio track, and a metering plug-in on your Stereo Out. Play with the pan and look at your levels. Try the different pan laws. It should help you realize exactly what's going on.

Hello all,

I've been invited to this discussion by Alex, primarily as this nullifies any copyright issues related to the attached illustrations.

The explanation below applies to two-channel stereo.

Acoustic Summing of Two Sources (Speakers)

Imagine yourself in front of two speakers, with the signal panned hard left. The left speaker will output the signal in a specific level, while the right speaker will be silent. Say we measure the level with a dBSPL meter and the meter reads 80 dBSPL (all obviously coming from the left speaker).

Now to pan that signal to the (phantom) center, we need to send the signal at the same level to both the left and right speakers. If the left speaker still outputs the same level as before (80 dBSPL) and so is the right speaker (80 dBSPL), the measured level for someone bang on between the two speakers will be 83 dBSPL - a 3dB increase compared to when the signal was panned hard left.

The 0dB Pan Law

This case, which is essentially the 0dB pan law, pose some problem on engineers - because as you pan across towards the center your signal gets louder (and so you may need to adjust the fader of the track).

The following illustration show the perceived loudness for the 0dB pan law for both stereo and mono. We'll ignore mono for now.

http://www.mixingaudio.com/webassets/logicprohelp/0dB-Pan-Law.gif

The 3dB Pan Law

To solve the 3dB boost center issue, the idea was to introduce a 3dB drop for signals panned center (and a proportional ratio for anything between the center and the extremes). So now as we pan across, the level does not drop when we listen in stereo, as show in the following illustration:

http://www.mixingaudio.com/webassets/logicprohelp/_3dB-Pan-Law.gif

The 3dB Compensated Pan Law

To the best of my knowledge, this unique feature of Logic has to do with Logic's history. If I'm not mistaken (would be nice to know I am), before pan laws where introduced in Logic it had the 0dB law. Then the -3dB low was added. But if you had a mix that was done before panning laws were introduced (ie, 0dB pan law), and you now change to the -3dB law, all instruments panned center would drop by 3dB.

To combat this, the -3dB compensated law was introduces, which, as David explained, simply applies a 3dB boost to the whole signal regardless of the pan position. This means that center panned signals won't drop in level, although anything panned off center will go up, with the extremes up by 3 dB.

The problem with the 3dB compensated low is that if you take a mono signal, have it panned center, then bounce to a stereo file and convert it to a mono file, you'd get 3 dB increase. Generally speaking, not a good thing.

The -4.5dB Pan Law

I have mentioned at the start that if two speakers output at the same level there will be a perceived increase of 3dB. But to be perfectly scientific about it, this will only happen under very specific conditions - namely that the measurement is taken exactly 1 meter from each speaker and that the room is an anechoic one, ie, no reverb. If real life (ie, in real rooms) the increase is more than that, and is also frequency dependant (low frequencies will experience further increase than high ones).

So based on this and through experimentation it can be shown that by applying 4.5 dB attenuation for centrally-panned signals, you get a fairly even level as you pan across.

The graph for the -4.5 pan law is a bit misleading. While it shows a perceived dip of 1.5 dB for the center, in real life one would perceive something that may actually sound like no dip at all (like the flat line of the -3dB line).

http://www.mixingaudio.com/webassets/logicprohelp/_45dB-Pan-Law.gif

However, if so far things weren't complicated enough, here's another thing to consider - most masking happens in the center of mixes as typically vocals, kick snares and basses are all panned there. Also, due to the curved nature of the two-channel stereo image, signals panned center appear more blurry and distant (and thus less defined). So while when soloing an instrument a pan across the stereo panorama may sound even in level - it won't when the whole mix is played, when instruments panned center will mask what you are panning across. The -3dB law can actually achieve that.

What the -4.5dB pan law is definitely good at is that theoretically it provides the best compromise between stereo and mono panning (the maximum error is 1.5 dB for either stereo or mono, where with other pan lows it is 3 - 6 dB). But this is important mainly in live broadcast where both mono and stereo has to be considered.

If you look at all the illustrations above, the dashed line show the level behaviour in mono. That is, when the signals are summed internally and not acoustically (as they leave the speakers). This is due to the fact that internally (either analog voltage or digital sample values) the summation of two identical signals (the left and right channels if the signal is panned center) result in 6 db Increase, and that's how the dashed lines were produced. For mono-critical applications the -6dB provide even pan across in mono, but a -3dB dip when listening in stereo.

So Which One is Best?

It's all about compromise and it depends on:

• Your room.
  
• The nature of music you are working on.

But most people would argue that -3dB is most appropriate for music, while -4.5 is most appropriate for film sound.

Truth to be told, a proper DAW should really give you the option with 0.5 dB steps (so -3dB, -3.5dB, -4dB, -4.5dB) and let users experiment with these different settings in their rooms and with the music they typically working on.

But given how hard it is to understand this technical issue, it is not surprising that in the past Pro Tools only offered -4.5dB pan low - simple and works in most cases.

Thank you Roey for the detailed explanation. I believe this will prove to be very helpful in explaining the pan laws.

Very kind of you.

Hello all,

 

The problem with the 3dB compensated low is that if you take a mono signal, have it panned center, then bounce to a stereo file and convert it to a mono file, you'd get 3 dB increase. Generally speaking, not a good thing.

 

 

Thanks for your detailed explanation. I've no problem with the actual mix as really, in the end, it's a question of playing it by ear, literally. But I'm still having difficulty knowing the best setting to export/bounce my stems then. I want to export the stems for a live performance, so that it sounds as close to the original mix as possible. I've read conflicting info from different sources as to volume loss/gain when exporting and (re) converting files to mono. So my question is, which setting will give the closest db match when exporting/converting files?

Thanks very much

In other words, the perceived loudness graph in mono is a straight line resting on 0dB but the stereo is a "smiling" curve?

Reading this made my head ache.

Me too when I wrote it. It's better to explain a graph by using a graph. I was going to post a few from another great book' date=' Mixing Audio by Roey Izhaki. I asked permission and he offered an even better solution! Thanks for your post Roey.

For fun, I made a mono vs. stereo graph depicting Logic's "-3dB compensated" setting. It is the perceived levels caused by the two effects Roey talked about:

1) +3 dB boost at center from the physical effect of acoustic summing and

2) the internal halving of L and R when generating a mono signal.

Also added the +3 dB "compensation" of course

Let me know if the graph is incorrect.

Alex

Hello ,

first of all I want to say "thanks" for all the infos and explanations in this thread . Very interesting, indeed.

Unfortunately , I'm now starting to think about it. And thinking is a bad thing. Leads to over-thinking, which is even worse.

In general I'm wondering , if working with PanLaw Settings can have a negative influence on the Stereofield.

Maybe this thought is completely dumb and results from my total misunderstanding of the whole thing.

So, here's the challenge, time is running:

can someone stop me from doing dangerous thinking and rescue me with an answer to the following questions :

1) Do Pan Law settings influence internal Audio Recordings in Logic ?

Lets say you have several Sub-Mixes / Sub-Groups in a Logic session,

be it for Song Production ( Sub-Mix Drums / Sub-Mix Backing Vocals / etc. with Compressors /EQ's etc)

or for Orchestral Arrangements ( Stems for Strings , Wood, Brass, Vocals, etc. ).

( All Instruments are Virtual Instruments and Mono AUDIO Recording Tracks.)

When you then route such a Sub-Mix to an STEREO AUDIO Track and record it while one

of the Pan-Law Settings is selected ( i.e. -3dB compensated ), is the

recorded Stereo AUDIO Track affected by the PanLaw Setting ?

( So , we don't Bounce to the outside world but stay completely inside Logic .)

Or does the PanLaw Setting only affects a bounced MasterTrack, when a Mix is bounced through the Master Output ?

2) Issues with applying PanLaw Settings several times on a recording ?

a)

Based on question 1) lets assume that a selected PanLaw Setting ( i.e. -3dB compensated )

actually does affect an internal Stereo Recording .

What happens if I then finally bounce my MasterTrack , still with a PanLaw setting activated ?

Does this mean that I apply the PanLaw setting twice ?

Which means , the PanLaw Setting is applied for the first time when I internally record the Sub-Mix to a Stereo Track

and then a second time when I finally bounce the complete Mix .

b)

Lets say we have several Stereo Tracks (Stems / SubMixes) , some of them bounced with , and some without certain Pan-Law settings.

We now import these Stereotracks into a Mastering Session to do a Final Mix.

Again , when I choose now a particular PanLaw Setting for this Master Session , can this screw up the Stereo field

when I work with Stems each of them bounced to disk before with different PanLaw settings ( or without activated PanLaw Setting ).

c)

I'm working with VEPRO5 to host nearly all my samples and Synths outside Logic.

In its Preferences VEPRO5 one can select various PanLaw settings ( 0dB ; -3dB ; -4.5 dB ; -6dB ).

What does this mean in combination with the host sequencer :

What can happen when I choose different PanLaw settings in VEPRO and Logic ?

Thanks in advance for any insight / infos on this.

Best,

Gerd

In general I'm wondering , if working with PanLaw Settings can have a negative influence on the Stereofield.

No. A pan law setting is just what governs how much gain is applied to an audio signal as you pan that signal to one side of the stereo field. Since you use your ears when you mix, if panning a signal resulted in that instrument being now too loud or too soft, you would just reach for that channel strip's volume fader to re-adjust. So the pan law setting in itself doesn't affect your mix at all, it only affects your work flow as you're mixing (so for example if you've been using one pan law and realize that you keep having to raise the volume back every time you pan something to one side, then you might want to switch to another pan law to avoid having to do that - but in the end your mix will be the same independently of each pan law you're using.

So to answer your questions:

1) No.

2) a) The question is based on a wrong assumption.

2) b) Nothing can be "screwed up" by a pan law setting.

3) c) The audio signal generated by VEPro is first routed through VEPro's pan (and processed according to VEPro's pan law), then the audio signal is routed through any audio FX plug-ins on the channel strip, then routed through Logic's volume and then pan (and processed according to Logic's pan law).

I hope that helps, but really the most fundamental thing you have to understand is that a pan law setting doesn't affect your mix anymore than using 4 inch volume faders vs using 6 inch volume faders: they just change the way you're mixing, they don't change the mix itself.

In general I'm wondering , if working with PanLaw Settings can have a negative influence on the Stereofield.

No. A pan law setting is just what governs how much gain is applied to an audio signal as you pan that signal to one side of the stereo field. Since you use your ears when you mix, if panning a signal resulted in that instrument being now too loud or too soft, you would just reach for that channel strip's volume fader to re-adjust. So the pan law setting in itself doesn't affect your mix at all, it only affects your work flow as you're mixing (so for example if you've been using one pan law and realize that you keep having to raise the volume back every time you pan something to one side, then you might want to switch to another pan law to avoid having to do that - but in the end your mix will be the same independently of each pan law you're using.

 

So to answer your questions:

1) No.

2) a) The question is based on a wrong assumption.

2) b) Nothing can be "screwed up" by a pan law setting.

3) c) The audio signal generated by VEPro is first routed through VEPro's pan (and processed according to VEPro's pan law), then the audio signal is routed through any audio FX plug-ins on the channel strip, then routed through Logic's volume and then pan (and processed according to Logic's pan law).

 

I hope that helps, but really the most fundamental thing you have to understand is that a pan law setting doesn't affect your mix anymore than using 4 inch volume faders vs using 6 inch volume faders: they just change the way you're mixing, they don't change the mix itself.

David ,

thanks so much for your quick reply and for taking the time to explain all this !

Much appreciated .

Best wishes

Gerd

so my question to you is this :

the pan option pan law ompensation you to keep active or not ???

The Pan-Law settings in Logic just mean this:

- if you have set the input gain in your interface correctly to not exceed 0db:

panning the signal will overdrive Logics channel gain if you set this setting to any other than 0db.

Very funny, honey........

On 4/8/2013 at 8:11 PM, Izhaki said:

Hello all,

I've been invited to this discussion by Alex, primarily as this nullifies any copyright issues related to the attached illustrations.

The explanation below applies to two-channel stereo.

Acoustic Summing of Two Sources (Speakers)

Imagine yourself in front of two speakers, with the signal panned hard left. The left speaker will output the signal in a specific level, while the right speaker will be silent. Say we measure the level with a dBSPL meter and the meter reads 80 dBSPL (all obviously coming from the left speaker).

Now to pan that signal to the (phantom) center, we need to send the signal at the same level to both the left and right speakers. If the left speaker still outputs the same level as before (80 dBSPL) and so is the right speaker (80 dBSPL), the measured level for someone bang on between the two speakers will be 83 dBSPL - a 3dB increase compared to when the signal was panned hard left.

The 0dB Pan Law

This case, which is essentially the 0dB pan law, pose some problem on engineers - because as you pan across towards the center your signal gets louder (and so you may need to adjust the fader of the track).

The following illustration show the perceived loudness for the 0dB pan law for both stereo and mono. We'll ignore mono for now.

http://www.mixingaudio.com/webassets/logicprohelp/0dB-Pan-Law.gif

The 3dB Pan Law

To solve the 3dB boost center issue, the idea was to introduce a 3dB drop for signals panned center (and a proportional ratio for anything between the center and the extremes). So now as we pan across, the level does not drop when we listen in stereo, as show in the following illustration:

http://www.mixingaudio.com/webassets/logicprohelp/_3dB-Pan-Law.gif

The 3dB Compensated Pan Law

To the best of my knowledge, this unique feature of Logic has to do with Logic's history. If I'm not mistaken (would be nice to know I am), before pan laws where introduced in Logic it had the 0dB law. Then the -3dB low was added. But if you had a mix that was done before panning laws were introduced (ie, 0dB pan law), and you now change to the -3dB law, all instruments panned center would drop by 3dB.

To combat this, the -3dB compensated law was introduces, which, as David explained, simply applies a 3dB boost to the whole signal regardless of the pan position. This means that center panned signals won't drop in level, although anything panned off center will go up, with the extremes up by 3 dB.

The problem with the 3dB compensated low is that if you take a mono signal, have it panned center, then bounce to a stereo file and convert it to a mono file, you'd get 3 dB increase. Generally speaking, not a good thing.

The -4.5dB Pan Law

I have mentioned at the start that if two speakers output at the same level there will be a perceived increase of 3dB. But to be perfectly scientific about it, this will only happen under very specific conditions - namely that the measurement is taken exactly 1 meter from each speaker and that the room is an anechoic one, ie, no reverb. If real life (ie, in real rooms) the increase is more than that, and is also frequency dependant (low frequencies will experience further increase than high ones).

So based on this and through experimentation it can be shown that by applying 4.5 dB attenuation for centrally-panned signals, you get a fairly even level as you pan across.

The graph for the -4.5 pan law is a bit misleading. While it shows a perceived dip of 1.5 dB for the center, in real life one would perceive something that may actually sound like no dip at all (like the flat line of the -3dB line).

http://www.mixingaudio.com/webassets/logicprohelp/_45dB-Pan-Law.gif

However, if so far things weren't complicated enough, here's another thing to consider - most masking happens in the center of mixes as typically vocals, kick snares and basses are all panned there. Also, due to the curved nature of the two-channel stereo image, signals panned center appear more blurry and distant (and thus less defined). So while when soloing an instrument a pan across the stereo panorama may sound even in level - it won't when the whole mix is played, when instruments panned center will mask what you are panning across. The -3dB law can actually achieve that.

What the -4.5dB pan law is definitely good at is that theoretically it provides the best compromise between stereo and mono panning (the maximum error is 1.5 dB for either stereo or mono, where with other pan lows it is 3 - 6 dB). But this is important mainly in live broadcast where both mono and stereo has to be considered.

If you look at all the illustrations above, the dashed line show the level behaviour in mono. That is, when the signals are summed internally and not acoustically (as they leave the speakers). This is due to the fact that internally (either analog voltage or digital sample values) the summation of two identical signals (the left and right channels if the signal is panned center) result in 6 db Increase, and that's how the dashed lines were produced. For mono-critical applications the -6dB provide even pan across in mono, but a -3dB dip when listening in stereo.

So Which One is Best?

It's all about compromise and it depends on:

• Your room.
• The nature of music you are working on.

But most people would argue that -3dB is most appropriate for music, while -4.5 is most appropriate for film sound.

Truth to be told, a proper DAW should really give you the option with 0.5 dB steps (so -3dB, -3.5dB, -4dB, -4.5dB) and let users experiment with these different settings in their rooms and with the music they typically working on.

But given how hard it is to understand this technical issue, it is not surprising that in the past Pro Tools only offered -4.5dB pan low - simple and works in most cases.

Hello! Is this just -4.5db or -4.5db compensated?

17 hours ago, cainnamusx said:

Hello! Is this just -4.5db or -4.5db compensated?

Yes! These two options have been added to Logic Pro since that discussion, along with the -6 dB options.

51 minutes ago, David Nahmani said:

Yes! These two options have been added to Logic Pro since that discussion, along with the -6 dB options.

I was wondering with one he was referring to using.