Hey, I just figured maybe the nerds could interpret my data.
 
http://www.soundclick.com/player/single_player.cfm?songid=13262425&q=hi
 
Good Morning Campers,
 
Sorry to be such a pest, but I am trying to hone my skills.  Please listen to this new mix and bash the living heaven out of it if you wish.  Here are the Amp Stats, according to AA6.
 
Possibly clipped samps.....300
T RMS....-13.95
Max RMS...-4.35
Min RMS....-79.30
Avg RMS...16.27
DNR....75
DNR Used...68.00
Loudness....-13.00
Perceived Loudness....11.7
LUFS....14.60
 
Can anyone reply if I am in the ballpark?  Or am I looking in a dark room?
The angle I want to explore is transient vs  amplitude.  Trying to make that connection between fast attack to resolution and slow attack to resolution.  There is a crossover there somewhere.  Still mulling it over.
 
Jesse Q. Screed

Given the genre, those stats are fine.
 
Except for the troubling number of overs. 300 can be inaudible or they could be distracting, depending on their distribution. I don't bother asking myself if they're audible or not, I just make sure there aren't any (like you, I use Audition to verify that).
 
Audition has a feature that lets you jump to the first suspected over, which lets you quickly zero in on where they're coming from. In all probability, those overs are occurring on the snare hits, in which case all you need is a limiter on the drum bus to keep them under control. You can verify this in AA6 by selecting a snare hit, which should be clearly visible in the waveform, and running amplitude statistics on just that small section. 

bitflipper
Given the genre, those stats are fine.
 
Except for the troubling number of overs. 300 can be inaudible or they could be distracting, depending on their distribution. I don't bother asking myself if they're audible or not, I just make sure there aren't any (like you, I use Audition to verify that).
 
Audition has a feature that lets you jump to the first suspected over, which lets you quickly zero in on where they're coming from. In all probability, those overs are occurring on the snare hits, in which case all you need is a limiter on the drum bus to keep them under control. You can verify this in AA6 by selecting a snare hit, which should be clearly visible in the waveform, and running amplitude statistics on just that small section. 

Thanks Bit, I appreciate the advice.  You are correct about the overs, they were all snare hits.  You say there is a function that allows you to skip to each one?  I need to figure out what that is so I can make it easier to edit.
 
Would it be better to catch those overs in Sonar before transporting the wave in AA?
 
Thanks
 
Jesse Q. Screed

I use AA3, so I can't tell you if the feature is the same in AA6, but in AA3 there is a button next to the "Possibly Clipped Samples" box that will take you to the next suspected over.
 
And yes, it's best to knock those problem peaks down at the source, ideally at the track level. It could be as simple as letting SONAR show you the maximum peak and then backing the Gain slider down by that amount.

Thanks Bit, you're a swell guy.  I think I know the thing you're talking about.
 
I also need to start sending AD2 out to all audio outputs to mix it better inside Sonar.  I suppose I could mix it inside AD2 but I'll take one GUI at a time.
 
Jesse Q. Screed

Jesse Screed
Hey, I just figured maybe the nerds could interpret my data.
 
http://www.soundclick.com/player/single_player.cfm?songid=13262425&q=hi
 
Good Morning Campers,
 
Sorry to be such a pest, but I am trying to hone my skills.  Please listen to this new mix and bash the living heaven out of it if you wish.  Here are the Amp Stats, according to AA6.
 
Possibly clipped samps.....300
T RMS....-13.95
Max RMS...-4.35
Min RMS....-79.30
Avg RMS...16.27
DNR....75
DNR Used...68.00
Loudness....-13.00
Perceived Loudness....11.7
LUFS....14.60
 
Can anyone reply if I am in the ballpark?  Or am I looking in a dark room?
The angle I want to explore is transient vs  amplitude.  Trying to make that connection between fast attack to resolution and slow attack to resolution.  There is a crossover there somewhere.  Still mulling it over.
 
Jesse Q. Screed

Hi Jesse,
 
Was the above achieved by scoping an mp3 or the actual wave file? If it was an mp3, go back and try the analysis with a wave file and see if it changes because it SHOULD. If that is a wave file, then yeah, the 300 possibly clipped samples definitely need to be looked at more closely.
 
That said, I didn't hear anything that sounded clipped on this end. I can say this, I get different numbers in Wave Lab than I do Adobe or Sony or even Studio One. Some editors have a resistance option (Wave Lab) that you can set to a certain threshold and if you set it too sensitive, it may give off false readings. I don't know anything about AA6, so you may want to see if it too has some sort of threshold control on how it reports.
 
Just remember, try to keep your master files no louder than -0.3 if you can help it before you convert to MP3. I keep mine between -0.3 to -1dB and never have any issues.
 
Sometimes when software reports clipped samples, it can be transients from snare drums, kick drums lack of compression or lack of automation. I'm actually recording your song as it is playing through SoundClick just to see the wave file, and it doesn't look abnormal to me. I'm waiting for it to finish so I can get some stats on it. But I primarily did this so I could physically see the file. This is what I see recording it as a wave file.
 
                               Left         Right
Min Sample Value:    -32598    -32596
Max Sample Value:    32597    32589
Peak Amplitude:    -.05 dB    -.05 dB
Possibly Clipped:    0    0
DC Offset:    0     0
Minimum RMS Power:    -106.8 dB    -113.33 dB
Maximum RMS Power:    -5.93 dB    -6.1 dB
Average RMS Power:    -15.12 dB    -15.6 dB
Total RMS Power:    -14.29 dB    -14.74 dB
Actual Bit Depth:    16 Bits    16 Bits

Using RMS Window of 50 ms


Nothing bad there in my opinion. The -6 Max RMS are due to some pretty strong peaks, lack of compression and you're in need of some fader automation as I mentioned above. Now granted, I captured this in real time playing off of Sound Click so it's not as accurate as me seeing/using your actual wave file. I really think that's a pretty accurate reading though because I can't see that -0.5 consistency on both sides happening by coincidence.
 
Now Studio One reports something else and claims your peaks are at 0dB yet the meters never actually hit 0dB. Wave Lab is around the same as Adobe with only your Max RMS being louder at -3dB...which is really loud even for a peak to be there. So though nothing sticks out to me as being horribly wrong, I'd concentrate on proper compression on each instrument in your tracks as well as fader automation to control your mix further. If the -0.5 is correct, and nothing ever touches 0dB, you're fine. Hope some of this helps. :)
 
-Danny

Danny Danzi

Jesse Screed
Hey, I just figured maybe the nerds could interpret my data.
 
http://www.soundclick.com/player/single_player.cfm?songid=13262425&q=hi
 
Good Morning Campers,
 
Sorry to be such a pest, but I am trying to hone my skills.  Please listen to this new mix and bash the living heaven out of it if you wish.  Here are the Amp Stats, according to AA6.
 
Possibly clipped samps.....300
T RMS....-13.95
Max RMS...-4.35
Min RMS....-79.30
Avg RMS...16.27
DNR....75
DNR Used...68.00
Loudness....-13.00
Perceived Loudness....11.7
LUFS....14.60
 
Can anyone reply if I am in the ballpark?  Or am I looking in a dark room?
The angle I want to explore is transient vs  amplitude.  Trying to make that connection between fast attack to resolution and slow attack to resolution.  There is a crossover there somewhere.  Still mulling it over.
 
Jesse Q. Screed

Hi Jesse,
 
Was the above achieved by scoping an mp3 or the actual wave file? If it was an mp3, go back and try the analysis with a wave file and see if it changes because it SHOULD. If that is a wave file, then yeah, the 300 possibly clipped samples definitely need to be looked at more closely.
 
That said, I didn't hear anything that sounded clipped on this end. I can say this, I get different numbers in Wave Lab than I do Adobe or Sony or even Studio One. Some editors have a resistance option (Wave Lab) that you can set to a certain threshold and if you set it too sensitive, it may give off false readings. I don't know anything about AA6, so you may want to see if it too has some sort of threshold control on how it reports.
 
Just remember, try to keep your master files no louder than -0.3 if you can help it before you convert to MP3. I keep mine between -0.3 to -1dB and never have any issues.
 
Sometimes when software reports clipped samples, it can be transients from snare drums, kick drums lack of compression or lack of automation. I'm actually recording your song as it is playing through SoundClick just to see the wave file, and it doesn't look abnormal to me. I'm waiting for it to finish so I can get some stats on it. But I primarily did this so I could physically see the file. This is what I see recording it as a wave file.
 
                               Left         Right
Min Sample Value:    -32598    -32596
Max Sample Value:    32597    32589
Peak Amplitude:    -.05 dB    -.05 dB
Possibly Clipped:    0    0
DC Offset:    0     0
Minimum RMS Power:    -106.8 dB    -113.33 dB
Maximum RMS Power:    -5.93 dB    -6.1 dB
Average RMS Power:    -15.12 dB    -15.6 dB
Total RMS Power:    -14.29 dB    -14.74 dB
Actual Bit Depth:    16 Bits    16 Bits

Using RMS Window of 50 ms


Nothing bad there in my opinion. The -6 Max RMS are due to some pretty strong peaks, lack of compression and you're in need of some fader automation as I mentioned above. Now granted, I captured this in real time playing off of Sound Click so it's not as accurate as me seeing/using your actual wave file. I really think that's a pretty accurate reading though because I can't see that -0.5 consistency on both sides happening by coincidence.
 
Now Studio One reports something else and claims your peaks are at 0dB yet the meters never actually hit 0dB. Wave Lab is around the same as Adobe with only your Max RMS being louder at -3dB...which is really loud even for a peak to be there. So though nothing sticks out to me as being horribly wrong, I'd concentrate on proper compression on each instrument in your tracks as well as fader automation to control your mix further. If the -0.5 is correct, and nothing ever touches 0dB, you're fine. Hope some of this helps. :)
 
-Danny

Hello Danny, thanks for taking the time to have a look.  Having both Bit and you take a look gives me a little confidence that I'm heading in the right direction.
 
I did the stats on a wav file, not an mp3.  I originally had over 1000 peaks, so I deleted the first wav and went back into sonar and remixed the snare because that appeared to be where the issue was. I only lowered the volume and did not add any compression or limiting.
 
The next wav I imported into AA only had 400 peaks so I did go in an correct a passel of peaks in AA that were taller on the scale.  I got tired of that tedious exercise, not because I didn't want to correct the peaks, but because I thought they were so fast that they would be of no real concern.  However, I do know that I don't want to have them in the first place.
 
Inside sonar as I mixed I tried to avoid using any compression on individual tracks because I wanted to keep it as real as possible.  There wasn't any compression or limit on the master bus or eq, just a stereo widener (which I'm not sure was a good idea, but I'm still experimenting.)
 
I executed some rudimentary volume automation on most of the individual tracks, where it was obvious I didn't capture the full sound, or didn't have good mic technique.  But I do feel like I could have been more exacting in the automation.  I considered dropping each of the tracks into AA and working on the individual waves there because it just seems to be easier....is that something that might be a good idea?
 
I guess that I'm trying to get a good enough mix so that I don't need any compression or limiting on the master bus.  I would like to avoid EQ too, but I'm still learning about my room and mics, and anyway, eq is like salt and pepper to food.
 
Thanks for the advice.  I really do appreciate it.
 
Jesse Q. Screed
 
 
 
 

Jesse Screed
 
Hello Danny, thanks for taking the time to have a look.  Having both Bit and you take a look gives me a little confidence that I'm heading in the right direction.
 
I did the stats on a wav file, not an mp3.  I originally had over 1000 peaks, so I deleted the first wav and went back into sonar and remixed the snare because that appeared to be where the issue was. I only lowered the volume and did not add any compression or limiting.
 
The next wav I imported into AA only had 400 peaks so I did go in an correct a passel of peaks in AA that were taller on the scale.  I got tired of that tedious exercise, not because I didn't want to correct the peaks, but because I thought they were so fast that they would be of no real concern.  However, I do know that I don't want to have them in the first place.
 
Inside sonar as I mixed I tried to avoid using any compression on individual tracks because I wanted to keep it as real as possible.  There wasn't any compression or limit on the master bus or eq, just a stereo widener (which I'm not sure was a good idea, but I'm still experimenting.)
 
I executed some rudimentary volume automation on most of the individual tracks, where it was obvious I didn't capture the full sound, or didn't have good mic technique.  But I do feel like I could have been more exacting in the automation.  I considered dropping each of the tracks into AA and working on the individual waves there because it just seems to be easier....is that something that might be a good idea?
 
I guess that I'm trying to get a good enough mix so that I don't need any compression or limiting on the master bus.  I would like to avoid EQ too, but I'm still learning about my room and mics, and anyway, eq is like salt and pepper to food.
 
Thanks for the advice.  I really do appreciate it.
 
Jesse Q. Screed

Hi Jesse,
 
And you're very welcome. I'm curious, why do you not want to compress the individual tracks? You don't have to squash them, you just want to condition them so that nothing steps on another instrument or lashes out. You're sort of defeating the purpose of engineering with the method you are trying to apply. Don't get me wrong, I'm not trying to tell you what to do on your music, but you don't have to hyper compress like people are doing in the loudness wars. You are literally making way more work for yourself than you need to be doing, honest.
 
Believe it or not, my methods are similar to yours. The stuff I record is recording ready without much eq needed. However, if I do not eq the instruments to be out of each others way (cutting a frequency on an instrument because I boosted it on a like-instrument) or high pass/low pass, my mix will definitely not be as audible as it could be. Throw in no compression on anything, and I'm going to wind up with something very inconsistent and some of the instruments will lash out....and not in a good way.
 
Keeping things natural is great. But you can still eq and compress. As long as the instruments change for the better and work in the mix, you're not really changing how natural they are. You are making them a little more consistent and working as a team in the mix. Eq will help as well as compression done correctly without being over-done.
 
If you don't use any compression at all, you're going to be sitting there for hours doing automation. To me, it's not worth it when you can use a light compression and then a little bit of automation on top. As long as you don't go crazy with anything, you won't lose the natural sound. Things will sound a little tighter, less cluttered and way more cohesive.
 
But like I say, I'm honestly not trying to tell you how to mix your song. I'm telling you what will make it sound better (in my opinion) as well as what will help you with those peaks that are turning into possibly clipped samples. Snare drums are always going to have peaks because they are hard transient instruments. You can compress a little if need be or use a limiter as to not go over a certain dB if you choose without making the sound suffer. It's all in how you process. You're not losing the natural sound though. You'd just be controlling it for the song and making everything sound better.
 
There are light compression settings and basic eq things you can do that will help immensely without losing the natural sound of the instrument. As long as you don't squash your tracks with compression/limiting and eq to the point where the instrument no longer sounds like what you recorded, you should be fine. Do what you feel is best to get your music across to others...I'm just trying to help you maintain cohesiveness so things aren't all over the place within the mix. Good luck! :)
 
-Danny

Hey Danny, thanks again for the response. 
 
I really don't have a reason not to compress individual tracks.  I think what is happening to me is that the pendulum is swinging back and forth.  When I first started using sonar I was like a kid in a candy store.  I saw the pro channel and thought "hey, look at all the cool goodies, let's turn them all on, and feel like a pro engineer!"  Of course it is all subjective, and I thought my mixes sounded ok, but I realized they were too (insert adjective here,) and not exactly what I really wanted
 
Now I think I have swung a little to far to the other side.  Use as little of the pro channel and other effects as possible.  rely on good recording and performance techniques, use well thought out automation, and sprinkle in a little eq where deemed necessary.
 
It is time for me to settle into the middle ground.  I am starting to realize that in some instances there are more than one way to accomplish things.  I could automate the volume on a snare, or i could compress it, or I could pull or boost some freqs on it.  All of those may be useful, but I'm learning that not all of them are correct in relation to the music as a whole.
 
Trust me on this, I don't get the impression that you are telling me how I should mix my music.   I feel you are sharing  your experiences with the tools and tricks you have gleaned from so many years of professional use.
 
So now it is back to the desk, (when I get the time,) and experiment around with what you have shared. 
 
In the end of it all, I want to get to where I am going.  It is a process, and I wasn't endowed with any innate gift to do this stuff, so for me it is a matter of brute force trial and error.  That is why I shared my stats with the tech forum, I thought I had an idea I was on to something, but wanted some input on just what was going on.  In turn, that led to more dialogue about eq and compression.  So you and Bit have helped me make a step in the direction towards greater knowledge and understanding of the art of mixing.
 
Right now I'm imaging an analogy.  It's kind of like wine.  You can get the $2 bottle, and you will surely get the buzz you wish, with all the side effects.  Or you can spend a little more, and have the buzz along with some of the other nuances that come along with a well crafted wine.  Initially, I react to music on a visceral level, does it make me feel anything? Now as I think about it more, I am trying to figure out how to bring out the little things in a piece of music that not only make you want to dance and squirm, but also make you take a second look, because it shines like a brand new dime.
 
Any other comment or tips are appreciated.
Jesse Q. Screed

jesse,
you SHOULD have a reason to compress and/or limit that snare!!
 
the idea with mixing,
is that it is YOUR job to make each individual track behave.
 
either by eq, compression, simple track volume and/or panning, or all of the above.
 
and it depends on the style of music,
and the way each track was recorded.
 
if for example, you recorded a guy hitting a snare by himself, in a room, with one mic,
and the mic is close,
it will respond being recorded completely differently, and playback differently, than if the mic was 5 feet away.
 
 
YOU have to decide how dynamic something needs to be,
to be mixed properly.
 
that's the trickiest part of mixing, in my opinion,
being able to make THE RIGHT CALL.