The Acoustic Guitar Forum

Go Back   The Acoustic Guitar Forum > General Acoustic Guitar and Amplification Discussion > RECORD

Reply
 
Thread Tools
  #1  
Old 03-26-2010, 12:39 PM
Fran Guidry Fran Guidry is offline
Registered User
 
Join Date: May 2002
Location: Walnut Creek, CA
Posts: 3,712
Default Audio Monitor calibration - ping Joseph Hanna

As I've gotten more involved with video I've been idly wondering about the possibility of calibrating my recording monitors so I could do a better job of judging the levels on my YouTube clips.

Joseph, from reading your posts this is right in your line of work. Can you give us some insight into this process?

Fran
__________________
E ho`okani pila kakou ma Kaleponi
Slack Key in California - www.kaleponi.com
My YouTube clips
The Homebrewed Music Blog
Reply With Quote
  #2  
Old 03-26-2010, 03:21 PM
Joseph Hanna Joseph Hanna is offline
Registered User
 
Join Date: Feb 2005
Location: Belmont Shore, CA
Posts: 3,206
Default

Quote:
Originally Posted by Fran Guidry View Post
As I've gotten more involved with video I've been idly wondering about the possibility of calibrating my recording monitors so I could do a better job of judging the levels on my YouTube clips.

Joseph, from reading your posts this is right in your line of work. Can you give us some insight into this process?

Fran
I can give it a try Fran Let me get some clarifying info.

First; I'm not yet sure I know what you mean by "calibrating" your monitors? If you're talking about metering specs and in fact meters..yea I can help with that. If your talking about what happens when something is posted to YouTube I'm afraid I haven't the FOGGIEST notion as I've never posted anything to YouTube.

Second; I'd be glad to run down the process of how I get video from editors, how it gets into Pro Tools, how I get On-Line full rez video when audio is done and ultimately how audio is then posted to the full rez video by spec.

Before I start am I close to understanding your questions???
Reply With Quote
  #3  
Old 03-26-2010, 04:08 PM
Fran Guidry Fran Guidry is offline
Registered User
 
Join Date: May 2002
Location: Walnut Creek, CA
Posts: 3,712
Default

I'm specifically talking about getting meters and monitors in sync, so I can consistently judge volume of tracks. It took a while but I finally figured out that if I change level in the DAW and at the (software) mixer at the same time, I could never tell how the result would sound on some other system. So I started thinking about figuring out a standard level that would give me a useful reference.

If it sounds _this_ loud in my monitors it will sound _that_ loud when I listen to the final on my other computer or in the car.

When I started Googling around I found that there are plenty of standards - one for every occasion apparently. I think you mentioned at some point that if you deliver audio at too high a level it gets cut off by some downstream carrier and you look bad.

I've heard of "K System" or something like that, I've just read a bit on the 3dB forum, I'm curious what procedure you might follow if you installed a new monitoring system to ensure that your ears and your metering lined up properly.

It's never been a problem when I'm doing audio only. Clearly the answer there is to turn everything up as loud as possible <grin>. But video seems to play by different rules.

Fran
__________________
E ho`okani pila kakou ma Kaleponi
Slack Key in California - www.kaleponi.com
My YouTube clips
The Homebrewed Music Blog
Reply With Quote
  #4  
Old 03-26-2010, 06:04 PM
rick-slo's Avatar
rick-slo rick-slo is offline
Charter Member
 
Join Date: Nov 2004
Location: San Luis Obispo, CA
Posts: 17,172
Default

You want your monitors to match in decibels what the DAW meters say, at least at a certain listening volume anyway, say 84 decibels? Basically you need a steady test tone to play and a sound pressure level meter (eg try Radio Shack). Get the DAW meters reading 84 and set your monitor volume for a match using the sound pressure level meter at the location where your ears are going to be listening.
__________________
Derek Coombs
Youtube -> Website -> Music -> Tabs
Guitars by Mark Blanchard, Albert&Mueller, Paul Woolson, Collings, Composite Acoustics, and Derek Coombs

"Reality is that which when you stop believing in it, doesn't go away."

Woods hands pick by eye and ear
Made to one with pride and love
To be that we hold so dear
A voice from heavens above
Reply With Quote
  #5  
Old 03-26-2010, 06:20 PM
Joseph Hanna Joseph Hanna is offline
Registered User
 
Join Date: Feb 2005
Location: Belmont Shore, CA
Posts: 3,206
Default

Sure,

This may get involved so we can take things in bits and pieces.

My industry, by and large, uses the dBfs scale. dBfs is a scale designed specifically for digital monitoring relative to a full scale where digital clip is represented by 0 dBfs. Therefore all usable audio is represented by numbers below 0 dBfs. In other words negative numbers.

To put things in general perspective before we get to deep it's vital to understand that every network and more specifically every show that we do ALWAYS comes complete with a spec sheet. Some even get slightly comical but by and large it's (the spec sheet) written with industry standards in mind and outlines what peak levels must be contained to and even sometimes suggesting where they'd like average levels to live. It could of course get as specific as to indicate where average levels for dialog should be as they pertain to audience reaction ect, ect. Most if not all shows spec that peak levels will not exceed -20 dBfs.

The point in all of this is we work to predetermined levels. It of course sounds as easy as not letting your peaks get out of hand but believe me there's a mountain of land mines in that mind set.

Before we move on as to how you can convert this info to your home studio..... does all of the above stuff make sense?
Reply With Quote
  #6  
Old 03-27-2010, 12:16 AM
Fran Guidry Fran Guidry is offline
Registered User
 
Join Date: May 2002
Location: Walnut Creek, CA
Posts: 3,712
Default

Quote:
Originally Posted by Joseph Hanna View Post
Sure,

This may get involved so we can take things in bits and pieces.

My industry, by and large, uses the dBfs scale. dBfs is a scale designed specifically for digital monitoring relative to a full scale where digital clip is represented by 0 dBfs. Therefore all usable audio is represented by numbers below 0 dBfs. In other words negative numbers.

To put things in general perspective before we get to deep it's vital to understand that every network and more specifically every show that we do ALWAYS comes complete with a spec sheet. Some even get slightly comical but by and large it's (the spec sheet) written with industry standards in mind and outlines what peak levels must be contained to and even sometimes suggesting where they'd like average levels to live. It could of course get as specific as to indicate where average levels for dialog should be as they pertain to audience reaction ect, ect. Most if not all shows spec that peak levels will not exceed -20 dBfs.

The point in all of this is we work to predetermined levels. It of course sounds as easy as not letting your peaks get out of hand but believe me there's a mountain of land mines in that mind set.

Before we move on as to how you can convert this info to your home studio..... does all of the above stuff make sense?
Right, 0 dBFS, all bits on, is the basic reference for digital audio. There can be no positive values, the FS stands for Full Scale. There can only be bits turned off for lower values. There can be no part of any waveform expressed in digital form that exceeds 0 dBFS.

So digital audio level is described as a negative value, with the area around -18 dBFS roughly equivalent to 0 dBVU depending on this and that.

Fran

Fran
__________________
E ho`okani pila kakou ma Kaleponi
Slack Key in California - www.kaleponi.com
My YouTube clips
The Homebrewed Music Blog
Reply With Quote
  #7  
Old 03-27-2010, 08:39 AM
Joseph Hanna Joseph Hanna is offline
Registered User
 
Join Date: Feb 2005
Location: Belmont Shore, CA
Posts: 3,206
Default

So assuming you're going to either purchase or perhaps you already have a meter and assuming you're gonna do most of your work in digital I'd recommend the AES -20 dBsf devices.

As I mentioned in another post we actually got so busy a couple of summers back my company bought all the audio guys Dorrough meters so we could mix at home. They are VERY expensive but if you plan to be in audio I can't personally imagine doing anything without my meters, it's a solid long term investment. I think mine are the little 280d series. A BIG NOTE HERE is that I find the Dorroughs the best at displaying peak vs rms levels clearly even better than the crazy DK meters at work.

That said and as many have mentioned here there is a push towards software metering solutions. I had also mentioned earlier that we had tried the software metering package by Inspector XL and it was too slow on the LE systems. When you're mixing dialog phrases one has to know exactly when a problematic peak is rearing it's ugly head. If that peak show up on the meter 200 ms late, it's simply worthless.

They may be (software meters) MUCH better these days and Dorrough apparently does offer software solutions but I don't know first hand how they work in application.

The next step This will be something very similar to ric-slow's idea. That is understanding that for music there is no spec to follow. At least not at the home/project music studio level . If you follow the Katz article on metering he uses some assumptions for sure but they're not only good assumptions but when it comes to "a reference" ya gotta start somewhere. I'm sure you've read this but here is a section from Katz's article and it should be next in the process.

In 1996, we measured that monitor gain, and found it to be 6 dB less than the film-standard for most of the pop music we were mastering. To calibrate a monitor to the film-standard, play a standardized pink noise calibration signal whose amplitude is -20 dB FS RMS, on one channel (loudspeaker) at a time. Adjust the monitor gain to yield 83 dB SPL using a meter with C-weighted, slow response. Call this gain 0 dB, the reference, and you will find the pop-music "standard" monitor gain at 6 dB below this reference.

Understanding how to properly setup your monitoring/metering is only the first step in a long chain of things. The application of metering is another beast all together.

So far so good?
Reply With Quote
  #8  
Old 03-27-2010, 09:30 AM
Bob Womack's Avatar
Bob Womack Bob Womack is offline
Guitar Gourmet
 
Join Date: Dec 2000
Location: Between Clever and Stupid
Posts: 26,991
Default

Good stuff, Joseph. It's fun to have another professional who is willing to explain things on the forum!

Bob
__________________
"It is said, 'Go not to the elves for counsel for they will say both no and yes.' "
Frodo Baggins to Gildor Inglorion, The Fellowship of the Ring

THE MUSICIAN'S ROOM (my website)
Reply With Quote
  #9  
Old 03-27-2010, 02:35 PM
Joseph Hanna Joseph Hanna is offline
Registered User
 
Join Date: Feb 2005
Location: Belmont Shore, CA
Posts: 3,206
Default

Thanks Bob! Always up for sharing good information.

I want to add a bit to this thread/process for the benefit of those that are following. That's probably just Fran, Bob and myself but who cares.

It's important to distinguish that here to fore we have discussed the very tangible issue of metering. That is physically looking at a meter and making decisions based on what story that meter tells.

I blurred the lines slightly by then introducing Katz's methodology as to setting up physical output levels. Essentially how loud you should listen and where that ends up from a metering perspective and ultimately a mastering perspective. One could then conceivably calibrate the physical meters (especially for home recordist) to reflect Katz's ideas and 'to hell with" any spec sheet. Ah oh... we're then coming dangerously close to the K-metering idea. You'd at least always have a visual idea as to where your levels were in "reference" to where you want them to go.

There are boatloads of reading available on why one should mix somewhere around the 83-84 dB mark and that's fairly easy to do without an expensive meter. A $40.00 Radio Shack dB meter will measure summed outputs but as I said earlier knowing the total output (particularly as a snapshot) holds a minimal amount of useful information.

In all of this we still then have not discussed the levels as they pertain to laying off the product. (burning cd's) and how that relates to peak vs rms and most importantly perceived audio levels.
Reply With Quote
  #10  
Old 03-27-2010, 03:05 PM
Fran Guidry Fran Guidry is offline
Registered User
 
Join Date: May 2002
Location: Walnut Creek, CA
Posts: 3,712
Default

Actually, Doug Young (who's also following this thread, so add one to the count) clued me to a Gearslutz link in which Bob Katz talks about monitor calibration and K system. And that's really what I'm thinking about these days, much more than metering. The stuff I do is simple, has no time constraints, no spec sheets, and will never be broadcast <grin>.

But that doesn't mean I don't value insights into CD levels, peak vs rms, perceived levels. In fact, those are the things I'm hoping to understand better in an experiential way by having my playback system calibrated to a consistent level.

Fran
__________________
E ho`okani pila kakou ma Kaleponi
Slack Key in California - www.kaleponi.com
My YouTube clips
The Homebrewed Music Blog
Reply With Quote
  #11  
Old 03-27-2010, 03:26 PM
Joseph Hanna Joseph Hanna is offline
Registered User
 
Join Date: Feb 2005
Location: Belmont Shore, CA
Posts: 3,206
Default

Quote:
Originally Posted by Fran Guidry View Post
But that doesn't mean I don't value insights into CD levels, peak vs rms, perceived levels. In fact, those are the things I'm hoping to understand better in an experiential way by having my playback system calibrated to a consistent level.

Fran
Makes sense Fran. You still NEED some visual frame of reference as to level even if you don't have to adhere to a spec sheet. I dunno what software you're running but there must be some type of metering included yes??

In the end it all tells a story and although I know the current crop of young music engineers sneaker at metering, in truth a good meter does SO much more than just displaying peaks and rms.

Every system needs a meter and a practical idea of how to implement it..even if only peripherally!
Reply With Quote
  #12  
Old 03-27-2010, 06:11 PM
Bob Womack's Avatar
Bob Womack Bob Womack is offline
Guitar Gourmet
 
Join Date: Dec 2000
Location: Between Clever and Stupid
Posts: 26,991
Default

And that's where familiarity with your gear and with the parameters surrounding a particular genre' are so important. Back in '75, Tom Scholtz of Boston took a real time analyzer and watched material he wanted to emulate. Being a Polaroid engineer, he also shot Polaroid snapshots of the RTA and labeled them for the song they represented. They became "music mixing flashcards." Doing this, he taught himself the ropes of basic mixing. But he studied the genre and used his tools to understand much of the nature of the signals he was attempting to create.

Once Boston was hired by CBS records, they assigned veteran producer John Boyland to help him. John said he was extremely knowledgeable and just needed someone to show him the techniques to accomplish his goals. That's a high compliment from someone like Boyland.

Bob
__________________
"It is said, 'Go not to the elves for counsel for they will say both no and yes.' "
Frodo Baggins to Gildor Inglorion, The Fellowship of the Ring

THE MUSICIAN'S ROOM (my website)
Reply With Quote
  #13  
Old 03-27-2010, 07:32 PM
Pokiehat Pokiehat is offline
Registered User
 
Join Date: Jun 2009
Posts: 181
Default

If you adopt K-20 you need to either get the pink noise test file from Bob Katz's site or use a pink noise generator to produce the same thing at -20 dBFS RMS. Some software synthesizers like Linplug Albino are capable of doing this.

Your DAW RMS meter should be setup as follows:

RMS Scan Window = 500ms
+20dB offset

Pink noise should peak at 0dB when you are done. The reason why pink noise is used is because it has nearly guassian amplitude probability density that is inversely proportional to frequency (I think. I'm fuzzy on the principles). It just means that it is a very high bandwidth signal with equal power in all octave bands.

Next you need an SPL meter and you need to place it at the point where you normally sit when you are mixing. Turn on the left speaker and get a readout of the sound pressure level. Use your soundcard mixer to trim or boost the main outputs to the left speaker until the meter reads 83dB SPL. Turn off the left speaker and do the same for the right speaker.

If you are like me and think mixing for semi long periods at 83dB SPL wrecks your ears then calibrate the speakers to a lower SPL that you are comfortable with. The reason 83dB SPL is used becomes clear if you look at a fletcher/munson curve since the curve is flattest at around 80dB SPL. If you calibrate to a lower level you just need to be aware that you ears will according to Fletcher's sample group be less sensitive to changes in the amplitude of low bass and high treble.

One of the advantages of the system is that if you use it regularly you begin to strongly correlate volume with what you hear more than what you see on a meter and its easier to mix without looking at the meters so much. Also you know when something is too loud because it cooks your ears although thats not a good way to figure that out. There are other advantages in the sense that its close to numerous industry standard references when working in both analogue and digital production environments. Its common these days to move between the two or have hybrid software setups with external hardware running into the line/mic inputs designed around +4 dBu. By having the same reference levels throughout you don't go overloading your signal processing and pegging all your meters which shortens all your channel faders since the operable range of the fader from silence to 'oh god my ears, clipping GAAAH' is like 1 cm.

Last edited by Pokiehat; 03-27-2010 at 08:42 PM.
Reply With Quote
  #14  
Old 03-27-2010, 08:14 PM
Doug Young's Avatar
Doug Young Doug Young is offline
Charter Member
 
Join Date: Apr 2005
Location: Mountain View, CA
Posts: 9,879
Default

Well, I think Pokie answered the question I had about all this, which I think goes back to Fran's original question. What level should you put your monitors at? Sounds like a matter of taste, tho being aware of Fletcher Munson is important.

To me 85db seems way too loud to listen to for any length of time, and for the kind of music I record (and Fran, too) - solo fingerstyle guitar, 85db seems inappropriately loud for the genre. Unlike you guys who record a variety of stuff, Fran and I both do 99.999% one thing - solo acoustic instrumentals. It makes metering and so on pretty simple. If I record so that my peaks leave some headroom, even just a db or 2, in dbfs-land, my average levels will land right around -20dbfs, just naturally. In that case, monitoring level seems a matter of taste unless I'm EQing or self-mastering, in which case fletcher/munson comes into play, and I need to be careful.

Does this seem correct, or am I missing something?

I do use lots of meters, and have spent ages poring over every analysis tool I can find to understand phase, stereo spread, frequency response curves, etc, for the type of music I play, so I can pretty much look at the meters and tell you if a solo guitar track was recorded X/Y, spaced pairs, has lots of reverb, etc, etc. That's been more helpful to me for mic placement than anything else, not such a big deal for monitoring. Some meters are useful for EQ, but mostly I don't use that much EQ, I just get a good raw sound from the mic setup, set the right levels, which I can see both from software meters and the meters on my A/D converters, (which seem to be calibrated with each other), and I'm done. Life's pretty simple when all you do is record solo guitar :-)
Reply With Quote
  #15  
Old 03-27-2010, 09:02 PM
Pokiehat Pokiehat is offline
Registered User
 
Join Date: Jun 2009
Posts: 181
Default

Firstly the human ear isn't linear across the entire audio frequency band. Your speakers aren't either and neither are your mics. Then there are certain electronic components with decidedly non linear characteristics and then theres the influence of your room.

Your hearing will also degrade over the span of your life with perception of high audio frequency typically being the first to go. And we all know that very low bass is felt more than heard at high amplitude so it figures that when you turn down the amplitude, its easier to miss it when you can't feel it anymore. That may explain why fletcher munson looks the way it does.

For various reasons it doesn't make sense to obsess over flat graphs and you should simply focus on making your recording environment comfortable and safe. So you should monitor at a level thats comfortable even if it is quieter. Quieter is safer over the same exposure time anyway and balls to Fletcher if you ever find yourself torturing your ears for the sake of a graph. Just don't take it all too literally and it wont turn you into a robot statmonkey producer automaton. On the other hand, do see the logic in the system and reap the benefits where applicable as long as it doesn't get in the way of your enjoyment and creativity. Thats the real important bit.
Reply With Quote
Reply

  The Acoustic Guitar Forum > General Acoustic Guitar and Amplification Discussion > RECORD

Thread Tools





All times are GMT -6. The time now is 03:48 AM.


Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2024, vBulletin Solutions Inc.
Copyright ©2000 - 2022, The Acoustic Guitar Forum
vB Ad Management by =RedTyger=