that's a good safe area

and you want peaks WELL below clipping.

peaks at -12 to -10 would be preferable.

I am starting to get the impression sharpdion23 that you are sort of not getting it. There are two very important articles you should read slowly and digest please written by the man himself Bob Katz.

http://www.digido.com/articles-and-demos12/13-bob-katz/22-level-practices-part-1.html

and the second but very important part to this:

http://www.digido.com/how-to-make-better-recordings-part-2.html

There is some slight mis information here. It is important to realise that when working with the K system you are immediately into rms mode not peak mode. (And by peak I am referring to transient peaks that are at the start of a sound) And you are basically using a VU meter that specialises in showing you rms and not really showing peaks. You use your DAW metering for the peak information.

The idea is that you track and mix keeping all your rms levels constant and the same. (and they are at the chosen ref leval as well and with K system there are three ref levels to choose from) The peaks you actually do NOT have to worry about because of the headroom that is already built into the K system approach. But you do keep an eye on them for sure using your peak meters but only to make sure they do not hit the 0dB FS ceiling. 

If you keep your peak levels constant they won't  tell you anything about the rms part of the signal that exists below the peak hence you end up with varying rms levels all over the place. I believe a better way to go is to keeps rms levels constant and the peaks above them will vary but who cares! K system has brought analog metering into the digital world. (along with peak metering which we need now) That is what so good about it. We used analog metering for years and it actually worked perfectly and for much longer than any digital metering.    

Read these first and I think you will get a better understanding of the situation. Basically the K system metering approach is much better than what any sole peak meter can show. Peak metering on its own is inferior and that is why many have problems with track levels varying all over the place and clipping as well. When you work with the K system you end up with all your tracks/busses having constant rms levels (and varying peaks) and no clipping anywhere. Pretty good don't you think! You never have to add or subtract gain from any tracks because you have tracked them all correctly and at the right level. 

You can do all this even with a stock DAW and Sonar will do a good job of it. But when you install a K meter what you were seeing down at -14 or -20 (rms wise that is) is now up around 0 dB and you have got the 14 or 20 dB of headroom sitting above that now, very similar to an analog system don't you think.  It is easier to read up there and you can see more variation.  +3 dB and -3 dB changes are large on a VU meter. 

Setting record levels is as easy as anything. Just get the talent to do their thing and adjust your input gain so the VU just comes up nicely to 0dB VU mark. You are now recording the rms part of your signal now down at either -14 or -20. I find working and tracking at K-12 a bit harder as you are recording hotter so therefore with only 12 dB of headroom the chances are greater of big transients clipping and we don't want that. You need to get into using limiters on the way in more.  K-12 is a good K level to switch to say after a mixdown where you have got much more control of transients overall. In mastering you shift up another 6 or 7 dB higher!

With very transient signals like kicks, snares, cowbell etc the rms part is not going to help you, just rely as you normally would on peak metering only and make sure the loudest hit is well clear of 0dB FS (eg -6 dB is often fine) Once you start sending a lot of drums and percussion sounds to a buss though, the VU rms part of the sound will become much more readable. Big open toms and kicks with a lot of sustain will have high rms components anyway.

The other aspect to K system is the constant and ref sound pressure level in your control room and that is a very important part of it too.

This has not been mentioned yet but needs to. The other thing that is VERY important too is getting hold of some test calibration signals. Unless you calibrate your setup you are wasting your time. I wonder how many people around here have actually calibrated their setups. You may need another program to make the test tones.

OK you need two types of signals. Sine waves at say 400 Hz or 1000 Hz created at all three K system ref levels eg -12, -14 and -20. The very tops or (peaks) of the sinewaves need to be at this level (NOT the actual rms level which is 3 dB down from that. That is an electrical engineering thing and it does NOT apply here) They should be 30 seconds or so.

The test tones must be stereo NOT mono. Mono test signals will give you varied results due to pan laws and the like.

Other handy test signals to have is pink noise also at those levels. Good news Bob Katz has got the -20 dB pink noise test signal on his website available for download. Just add 6 and 8 dB to that to create the other two.

The sine waves should be used first. You choose a ref level eg K -14 and then park the K-14 test tone on a stereo track and play it. You now insert a K meter on that track and set it to calibrate also to K -14 then the meter should read 0 dB VU. Good thing to do also is observe what the Sonar meters are showing you as well while the K -14 test tone is playing.

You can take it further as well and measure output voltages from your soundcard too but that is another story.

If you want I will upload all three K system test tones onto my Soundcloud and make them downloadable so you can get them easily. 

The pink noise signals are good for calibrating your monitor levels. A sinewave tone is not. You only have to move your head 3 inches and the SPL level at the meter will change a lot! After doing the sinewaves you put the pink noise waves on that track and select the correct one eg K-14. You will notice the signal does not quite reach 0dB VU but that is fine, that is the nature of the signal. You can set each monitor speaker one at a time by positioning a SPL meter right where you head is and aim it directly at the speaker and set your room level for 83 dB SPL. Do the same for the right speaker. When both speakers are on the SPL will be about 85 dB SPL which is about right. (Some say 85 dB SPL is loud, it is not. It may sound that way for the tone and the pink noise but music will sound a little quieter in general. SPL meter must be set for C weighting. 85 is a beautiful level.) I have got a permanent SPL meter rigged up and showing me the SPL level all the time. It is very nice to have around. It is so easy to creep up your monitor levels before you know it you are enjoying the music at 95 or 100 dB SPL! 

The good thing about working at 85 dB and K -20 is that the music can go up to 105 dB SPL if it needs to! You could also make your K -20 level produce 90 dB SPL in the room which means the sound level can go up to a whopping 110 dB SPL. And looking at the other end too the digital noise floor even at K-20 is still 70 or more dB down (for 16 Bit) so you are not going to hear any noise. Who said digital was not dynamic. I Love it!

Sorry sharpdion23 I did not answer your question about where to do the metering. Pretty straight forward actually. On tracks insert the meter in the FX bin first and just make sure you echo the signal through or monitor the incoming input signal and you will see it straight away. If you are recording with effects, put the meter last after all the effects so you can read effect levels on the way in. But most of us I think don't record with effects or at least I don't much. So the meter is usually the only thing on a track.

On buses put the meter last if there any effects on the buses. You want to meter last so the meter sees everything on the effects buses. K system metering also helps you set accurate levels going into and out of plug-ins too. (without any clipping)

On your master buss same thing, make the meter last so if you are mixing into a compressor for example you can just tweak the makeup gain to bring the VU level back up to the 0 dB mark.

My real VU's are sitting on the stereo buss permanently but that is cool because as soon as I solo any tracks or buses I am reading them directly anyway even with a meter at the end of the signal chain.

The BlueCat meter is nice but I like the VST's that are doing real VU's now. I have been comparing the ballistics of the real VU's to the VST VU's and believe me they are good. The Klanghelm meters give you lots of control over the ballistics so you can fine tune nicely. Be careful here because you can screw the meter up too! You need different test signals for adjusting meter ballistics. They are special pulses certain time distances apart.

I am going away for a few days but I will try and get the test tones up on my Soundcloud. I will add the link here when they are up.

Rus W


^ Now, we're getting to the nitty-gritty - Mastering; however, I shall let my sensai explain that to you. Oh, Danny Boy ... (hums tune) 

LOL Rus! I'm sorry I missed this post....I'm crackin' up over here. Ya know, I been hearing about "the pipes the pipes are calling" for 45 years now....one day I opened up my window and yelled "WHAT? STOP CALLING ME FOR THE LOVE OF GOD!" Hahahaha! Even funnier...I got this dog across the street from me. Little thing. When it barks it goes "herf herf herf herf herf herf herf" over and over and never breathes. I was trying to sleep...nice morning, windows open...it started "herfing" and I went ballistic and yelled out the window "Bark like a real dog you @#$%^&*(^%$##!!!!!!" What was funnier...the neighbors who were outside with their mouths wide open that started laughing hysterically after. LOL! :)
 
Ok, sorry about that....you asked for me...here I am. Another novel, but a pretty good one I think you will enjoy. :)
 
To be honest, I have tried as hard as possible but cannot hear the difference between a properly encoded 320 kb mp3 and a 16/44 wave. If one exists, my ears can't pick up on it at all. What I SEE is 13k on up completely removed from an mp3 where I don't see that in a wave file of the same song. I can hear up to about 16/17 k these days via hearing test...anything after that though, and I'm dead. But to be honest, I STILL don't hear the high end as being "removed" or "missing" nor do I hear a difference that in reality, should be pretty drastic compared to what I SEE when I compare a wave and an mp3 of the same song. So to be honest....yeah, I definitely think we can burn 320 kb mp3's to CD and no one would know the difference. Would I do it? No because even though I can't hear a difference, we know that an mp3 is not as pure as the wave file.
 
It's like dithering....I can't hear a difference with or without dithering. But I use it because I'm supposed to and I don't notice it messing anything up either. If I heard stuff within a dithered file that I didn't like, I wouldn't care who said "you should dither"....I'd not do it.
 
Mastering and over-all levels compared to major label releases: For DIY'ers, your best bet is to NOT try to get the levels reached by major labels and mastering engineers that know how to master things loud. The reason being, you don't just take any mix and make it loud. There are lots of things that need to be considered before you can do this and literally compete with a major.
 
Getting the volume they get is easy...making it sound clear is the problem most home studio guys have. When a major sends a project to a mastering engineer and his job is to make it sound loud while remaining clear, he takes a different approach to the pre-mastering stage than he would if he was allowed to just master the tune to a decent volume. The editing part of a pre-master is where you set the tune up to be loud. However, garbage in, garbage out. Firstly, it needs to be mixed right. That means you do NOT take on the role of the ME and try to master within your mix.
 
Those who do this and are comparing their mixes to MASTERED material are failing here. We add way too much at the mix stage...so when you master, you've already buried yourself. When a major label submits a mix to be mastered, it does NOT sound like the mastered version you end up with. The mix is balanced 90 % of the time. Meaning, there is no excessive meat in the low end, there is no congestive mids, there are no harsh highs. AND...unless instructed by the producer or the label, they do not squash the living heck out of the master bus at this point. Some do...but then the ME's hands are tied. 
 
Next, the tune can't have loads of spikes in it at the mix stage. Remember, you will only be as loud as your biggest spike. A limiter can tame it, but it is still the biggest volume boost  that will gauge how loud things will be. This is where manual leveling of the audio comes into play. Let's take a look at a client mix to further illustrate what I'm talking about.
 

 
In this example, I've circled the biggest spike in the mix. (meant to circle the part above it as well in the other side of the clip becuase you would adjust both sides equally, not just one...sorry) A major label would never release something that looks like this, so if you guys that are going for loud mixes have spikes like you see here, don't even bother going for "loud and clean" because it ain't gonna happen.
 
The peak/spike you see circled will be the loudest point this song can go no matter what limiter I use. It's capped right there. Sure, we can make it loud, but it's not going to get "commercial" loud without sounding horrible. In a situation like this, you have two choices.
 
1. You master as loud as the tune can handle without sacrificing quality and accept that it ain't major label loud.
 
2. You manually level the audio or send it to a dope like me to do it for you. LOL!
 
Now let's take a look at what that spike really looks like when it's exploded.
 

 
What you see is what makes up the first spike I circled. When we zoom in like this, if we hi-lite the clip at the spike in say something like Adobe Audition or Rx Advanced, when you press your space bar, you are going to hear nothing more but a pop or a click. So we can literally adjust this spike hump at a time until it goes away. All it is, is a transient of a snare drum that was not taken care of at the mix stage.
 
But this little devil here can be the reason you lose volume when trying to go for a loud mastered volume. When you have enough of them, your mix degrades and this will affect the final outcome. The key is to automate more within your mix whenever possible. But at times, the better choice for certain instruments due to their transient delivery, would be to compress a bit more. The problem most home studio guys are faced with is, they over-use the compression/limiting side of things and they do not know how to use a compressor or limiter properly to begin with.
 
I hand edit each and every spike in very mix that comes through my doors. By the time I get done with the wave file, it looks beautiful and sounds beautiful as it is not only leveled for these spikes, I manually level it like riding a fader as the final editing stage before the song really gets mastered.
 
If you have a smooth mix that remains dynamic before you master it, the final master should blow your doors off. Just remember, handling peaks/spikes will not do it alone. The mix has to have all the right stuff in the eq area as well as the instruments being compressed properly.
 
So after the manual leveling phase of my pre master here....I end up with a wave file that doesn't have any huge spikes like what you see in the pics. This removes the blatant caps I would have in this mix. When I put a limiter on the mix, there isn't a cap on how loud I can go. I mean there will still be a cap, but the cap will be due to how the mix is mixed as well as how I master it. It won't be because there are spikes stopping me from making it as loud as it COULD be, understand?
 
From here, if I wanted to compete with Metallica, I could do that without a problem. However, when you start hitting the -4 dB RMS range, no mix will sound good in my opinion no matter who masters it. There is just no reason for anything to be so loud, it is not only degraded....it's completely destroyed and no longer sounds like music. When I have a mix that is nice and tight without spikes, I treat it like a luxury car. Sure...I could drop a Corvette engine in it and make it loud and proud, but because of how I handled things, a nice 305 cubic inch 8-cylinder motor or even a 6 cynlinder would be perfect. We don't need to super-charge everything we do just because we have the power and the means to do so.
 
It's like driving your car....you know it will do 100 or faster, but you don't exercise that option often unless there is a reason. The same with our mixes. You can literally go louder and stay cleaner than a super loud mix if you go about it right just by turning up your volume knob. A super loud mix caps off at a certain volume level where volume is replaced with distortion. A clean master with a little volume will go right past it and keep on going until you max out the volume knob while remaining clear. I'll take clean, clear and control over the volume knob over super loud, distorted and non-dynamic.
 
That said....there HAVE been a few recordings I've heard where they are super loud that don't seem to distort when you turn them up. This is rare though. You'll even see them clip your meters at times yet you don't HEAR the clipping. Again I use another car scenario....mix/master for the tune and use a volume level that compliments the tune you're working on just like you "drive as fast as you can control" a vehicle. Hope this helps and sheds a little light onto the subject. :)
 
-Danny  

seems like several of us keep repeating the same information over and over, take the highlights and run with it.



an interesting side note:

that clip that danny shows, the spike...

THIS is one of the reasons why you have to really watch your tracking.

if you allow PEAKS to sneak thru, at that moment it peaks, it will cause a digital distortion that, even if it's ultra fast, it will cause your mix to sound grainy, even as you mix it down.

this is why i suggested outboard compressors and limiters, and running conservative tracking levels.
once you have hit that ceiling going in, you are stuck with it.

a lot of folks get confused about the ability to use limiters and compressors AFTER the fact (after the signal has already hit the convertors) and that's all i'm trying to clear up.

there are a newer batch of preamps that have built in limiters and compressors, even if they are software, as long as they are upstream of the convertors, they are fair game.
whether they sound good or not, is another discussion.



irregardless of "K" metering and any of that, if you follow conservative rules, you'll always be safe, and splitting hairs about headroom is really pointless.

@ Danny:

 I'm glad you see the humor in that and come on your dog was only kidding! You've heard that for 45 years? That's an immense amount of pressure of me then! haha!

I have noticed (and I'm guilty of this, too) is that most DIY'ers that are confused are mislead by the terms Master Effects with Global Effects. These two are not the same thing. They think Master Effects a la Master Volume

"Do this to the Master, everything is affected." It's not wrong, but what is wrong is not realizing that this is still mixing.

"I wanna take away some highs or lows on the overall track." This is still in the mixing stage. Mastering is more intricate, but doesn't equate to mixing being easy. I also think that when a someone says "Here's my 'Master Track' - s/he really means "Globally Mixed" track; however, DAWs misleads user with a term such as Master/Mastering effects.

Not placing blame on them. I just see how this can skew what it means to Master which is why many experienced (or gaining experience, such as we), say that Mastering isn't activating plugins. What you're doing when you do that is still mixing, but globally.

However, The Mastered Track is often used because Globally Mixed is a mouthful. lol Again, I think this is what DIY's really mean when the term master and all its forms is used. Just my .2 cents on how I look at it.

Now, it could be that I'm admitting ineptitude in this area, but I'd rather someone who knew what s/he was doing do this instead of me.

In terms of globally mixing, I think I am getting the hang of it.

Hey Jeff,

I am actually not suggesting a volume envelope at all.  I am asking if that is what Danny meant because I don't know.  When Danny said that he "hand edits each spike individually"; I was trying to actually understand how he does this.

that was my whole point about 'conservative' recording levels.

MakeShift


Excellent post Danny, I am soaking it all in.  One question, so when you say you manually reduce these spikes, are you saying that you create a volume envelope, click in some nodes, and then literally pull those spikes down in volume?

On the dog story, I think every neighborhood has one of those dogs that just, for what ever reason, can't bark like a real dog.  That was funny and I laughed pretty good.....been there.

Mike, thanks! I made you (and anyone else that wants to see it) a 12 minute video explaining a few things. You'll totally get where I'm coming from when you see it. It's rendering now....as soon as it's done I'll post up a link. :)
 
-Danny

That's an excellent video Danny- very interesting.

I was wondering when you record a project do you use this technique track by track or just over the finished wave?


Totally agree Sonar should have this sort of capability! Cubendo sort of does but I am one of the dopey guys you mention at the start who just grabs a big chunk and lowers it indiscriminately.  now I know better!

Thanks again for the cool video.

I must say that was very informative and I agree with your "hand-written" approach. Not saying it's the best - most certainly not the worst, however, Mastering is an intricate art and details definitely matter and the hand-written approach to me seems the safest method.

I'm sad to say that upon watching you've garnered a new nickname. But I'll leave to you: Inmate or Warden? :)

I do have a question:

When you mentioned in the vid about the snare causing the pop (once you isolated it), generally, how do you get mixes to be mastered because if there's alot going on in the song (despite your trained ears), how do you isolate a particular sound/instrument - especially if you don't get the mix ready to be mastered as individual tracks?

I ask because I've seen videos where the ME will just have a unmastered track comparing it to after s/he is done with it.

This isn't like where something is obvious (too much low/high end, etc). I do understand about the clarity of the mix, but how do you deal with a cake that's already baked and go about adding the sprinkles? Or would you prefer getting the layers "store-bought" if you know what I mean? (No pun intended with said analogy) Or would you want both as to have a reference with all the individual tracks included? This isn't me asking for the ME to mix or remix (production) the track, but I do want him/her to see everything that is going on. Would you agree or disagree with this?

I hope this question(s) isn't confusing. My apologies if it is.

PS: I'm sad to say that upon watching you've garnered a new nickname. But I'll leave to you: Inmate or Warden? :)

Rus W


I must say that was very informative and I agree with your "hand-written" approach. Not saying it's the best - most certainly not the worst, however, Mastering is an intricate art and details definitely matter and the hand-written approach to me seems the safest method.

I'm sad to say that upon watching you've garnered a new nickname. But I'll leave to you: Inmate or Warden? :)

I do have a question:

When you mentioned in the vid about the snare causing the pop (once you isolated it), generally, how do you get mixes to be mastered because if there's alot going on in the song (despite your trained ears), how do you isolate a particular sound/instrument - especially if you don't get the mix ready to be mastered as individual tracks?

I ask because I've seen videos where the ME will just have a unmastered track comparing it to after s/he is done with it.

This isn't like where something is obvious (too much low/high end, etc). I do understand about the clarity of the mix, but how do you deal with a cake that's already baked and go about adding the sprinkles? Or would you prefer getting the layers "store-bought" if you know what I mean? (No pun intended with said analogy) Or would you want both as to have a reference with all the individual tracks included? This isn't me asking for the ME to mix or remix (production) the track, but I do want him/her to see everything that is going on. Would you agree or disagree with this?

I hope this question(s) isn't confusing. My apologies if it is.

PS: I'm sad to say that upon watching you've garnered a new nickname. But I'll leave to you: Inmate or Warden? :)

Everyone will sort of pick up their own ways of doing things. As long as they work for a person, it's all good no matter how goofy. :)
 
As for your question, the answer is...you're limited. However, each frequency you deal with, represents a syllable. Upon isolating these syllables or exeperimenting with them, if you will, they reveal instruments in how they are heard....or in some cases, sort of covered up if they lash out too much.
 
As for the cake that may be too baked, this is what listening sessions are for and why I give advice before I master a thing. I simply won't master something that is so baked it may sound bad, may be too compressed, may be way too colored to where nothing I do can make it better.
 
See Rus, the thing with me is I don't NEED the money as much as I WANT to make a difference. If someone sends me something that I cannot make a difference on, I tell them straight up that I don't feel I can make what they've sent better. And if I tried, I'd be taking their money for the sake of a pay check. Can I make it better? Sure I can! But...does the price I charge justify the RIGHT amount of better? Here's a scenario to explain it better.
 
You and I do a shoot-out with the same guitar. You record it on your end with a $4000 mic pre, I run my mic into some little DI box gizmo that cost $150. We post our results and it turns out the people that voted, liked your guitar sound better. It was close though and I got quite a few votes, but you won over the majority. When everyone seriously analyzes what we presented with our test, we find that it's also subjective no matter who won. We also come to the conclusion that the difference in your recording is about 3% for the better. One has to question: "can we justify paying $4k for a 3% difference?" To some, yeah...to others...no one will know as far as common listeners go.
 
My point? I can't live with making a 3% or less difference on someone's song and charge what I charge....and trust me, I'm more than fair price wise. I know sometimes I come off like a hot head that may seem heartless at times...but that's only when someone makes me mad. I'd never be heartless or that desperate to take on a job for the sake of money even if I was ready to lose my house. So the client either takes my advice and tries to fix the mix, they send me stems (which makes the procedure cost a little more) and I fix them that way, or I kindly walk from the job. So I never get to the "too baked" stage. :)
 
I'd like to share something else with you to further illustrate some of the stuff we've been talking about if you don't mind? A few years ago, Michael Wagener (famous rock producer from the 80's) shared a piece of an un-mastered song on Gear Slutz that he worked on for Ozzy called "Perry Mason". He of course didn't get the gig, but what he shared had more value to me than "wow Michael, that was great....you should have gotten the gig!" I was one of the one's that praised him and jumped on that train as well because it was great to see him there and I am a fan of his work.
 
However, I kept that file he shared and years later, a lightbulb went off as I got deeper and deeper into recording and mastering. When I say all the time that "major label guys do not share mixes that sound like mastered material" this is EXACTLY what I'm talking about. When you hear this, listen to what it DOESN'T have.
 
Every piece that I have heard from a major that was un-mastered, has these flat qualities. Flat qualities in a GOOD way though because they leave room for the ME do perform his magic. There are no low end nasties, no excessive mids, no harsh highs, no compressors pumping and breathing, the file doesn't look like a square box and it doesn't have loads of spikes all over the place.
 
It's balanced the way a mix should be. Now mind you, what he shared might have been doctored up a bit by him as most guys don't really expose themselves completely naked...especially on that forum. But this is definitely a good representation of what a mix should sound like in my opinion because of what it DOESN'T have in it YET.
 
https://dl.dropbox.com/u/4909348/pmfade.mp3
 
See how open it is to be sculpted into something really killer? Try it yourself. Throw it in Sonar or the DAW of your choice and eq it to your specs, add a limiter or some compression and then turn them all off and hear the differences. A mix like this is an ME's dream come true. Of course this mix is a little hotter than we'd like to have delivered to us, but you'll notice if you run some numbers on it, it's neither perfect nor is it stretching anything other than in 192kb mp3 format, the numbers are exaggerated. In wave form, they'd look a lot better.
 
But you'll notice...this doesn't rattle with low end...you can hear it and feel it a bit, but it's not making your bile ducts quake. It's got a nice mid range sound to it that is not congested or trying to over-emphasize "warmth" which can create "boxy sounding" mixes that are lifeless. It has a bit of sparkle in the high end...part of that due to Zakk's pinch harmonics needing to lash out and cut you like a knife as a part of his guitar style.
 
Anyway, this has made for some really great discussion. Thanks to sharpdion for letting us talk in depth without yelling at us for thread hi-jacking. :) Hope this answers your question, Rus. :) Hmmm....Warden sounds a bit better than inmate...though Phil Z dubbed me "The Daniac" a few years back. LOL! Ah I'm easy...call me whatever ya want as long as ya like me when ya say it. :)
 
-Danny

I think I will just call you Danny... I like that the best.

Dude, you are unreal, I did n't expect a video, but this was great.  I learned a lot from that video. Mostly, that I am not ready for that kind of mastering.  What I learned is that I need to still work on the tracking part of recording; working to keep the spikes that you have illustrated in check.  I know Bat started chiming in on the use of light compression early in this thread and I was n't really getting it until I saw the illustration.

So, point learned... I need to invest in a hardware compressor to know a little edge off on the way in.  Get the tracks to be as good as possible, before I even worry about mastering, or having some one like yourself, do some real mastering for me. 

I need to take a listen to this second file as well.  I am sure that is going to bring a whole new set of revelations to me.

I can't tell you how much I appreciate you sharing your wisdom, Danny.

No problem Mike, my pleasure man. As for the hardware comp...no, don't dare spend a bundle. I'll share a little secret with you...but shhhh...don't tell anyone. :) Since we're really (at leas I'm not) not using this particular compressor as a coloration comp, you can buy anything really. At my house, I use a Behringer Multi-Com which is 4 comps in one rack space and super cheap.

Of course we all know that though Behringer puts out decent stuff (ok, some of it sucks) it's perfect for what we want to do here...which is just lightly condition the signal going in, that's all. Now, other guys like to use the pricey hardware comps due to how they literally color the sound as well as condition it going in. That decision is up to you if you want to go that route, and there will still be a few cheaper comps out there that can do that too. Heck you can get away with an Alesis or an ART or a Samson....anything that compresses lightly, doesn't give you noise or hiss and just gives you a decent signal going to disc.

There are some really cool options out there without spending loads of cash. I DO have some really nice hardware comps that I use at my studio, but even there....I never smash them enough to where it matters if I used something pricey or something cheap. As long as my signal is nice going in without spikes and there is no noise or hiss, it will work. Now keep in mind...like I said before, you also want to pay attention to your performance. The reason being, it's all too easy to play or sing a certain way, and then keep jumping on your hardware compressor more and more to stop peaks.

That is NOT what you want to do. Peaks are ok to a small extent, we just don't want them looking like nasty spikes. If you pay attention to your performance and "work" your mic on a vocal or alter a picking style on an acoustic or change picks etc...you stop working the compressor as hard. See, we don't want to hear the effects of the compressor so this is why your performance, mic position, execution, proper leveling and all that goes with it is important.

Here's a quick way to think about it. Say I'm going to play a guitar track. I want my levels going in to be -6dB because that's just what works for me the best. So I'll play the part I'm going to record while watching my meters. I try to get it so that the hardest thing I play hits -6dB. From there, I'll compress just a little taking out maybe -1 to -2dB of gain tops to where the comp kicks in on its meter showing me that on the hardest/loudest part I play. Same with vocals. Never more than -2dB of gain taken out with the compressor and I use a ratio of like 1.5:1 on my guitars...-5 to -8 on the threshold...maybe a 3:1 on a vocal or a bass guitar...it all depends what I'm going for as far as the conditioning part of the compressor goes as well as the person performing. But whatever you do, you don't want to use this to the extreme because it's destructive. If you find out later that the track seems to sound restricted sounding or you hear a pump in the track, you'll need to recut it. So that's why you want to be extra careful here and use the comp to condition more than to actually compress. Make more sense now? :)

-Danny