Help with mic'ing/recording acoustic guitar

[spantini]

Mack Caster : You can build an AC Vent silencer very cheap and get professional results. Just make sure that the output for the hole on the silencer, is twice as large as the ac vent hole in the house. Standard registers are 4 x 12 so the out put needs to have two installed or one that is 8 x 24....

Thanks, Mack. My space is small. My AC/Heat is a wall unit which can be turned off easily for the time it takes to record. I was lazy this time as I was just looking to get good recording level on the guitar. Next time the A/C will be off and I'll measure the noise floor.

This is the style A/C-Heat unit I have (not my actual room):

 

[ecc83]

Attached is a tune my son recorded years ago (Sontronics STC-2 iirc) and the meter shows a section hitting about where you should be. Actually I had to tweak the level down a bit to show it properly! Bit of a tale there...

When we got our ZED10 it sat a 5mtr balanced cable run from the 'unbal' traff box next to the PC that contained the M-A2496 card. Dad had made up an attenuator to bring the DAW level into line with the (excellent) meters on the A&H. Son had spent TOO long with his Teac A3440 and did not like the wee waveform that -18dB average produces!

His original .wav recording does not crack, quite! But hits about neg 2 a couple of times! Mooosicians eh?

Dave.

 

[spantini]

Nice doodle!

Attached is a tune my son recorded years ago (Sontronics STC-2 iirc) and the meter shows a section hitting about where you should be. Actually I had to tweak the level down a bit to show it properly!...

Dave.

So.. I should be dropping the meter level even more...

Like this?

 

[Mack Caster]

Attached is a tune my son recorded years ago. Dave.

Your son is very talented!

Nice doodle! So.. I should be dropping the meter level even more...Like this?

Regardless of what you are feeding through your interface, you always want your signal to be between a -12dbs and -18dbs as it is being recorded, via your DAW or an external digital recorder. A -18dbs is hotter than a -24dbs by 6dbs as the closer you get to 0dbs from the negative side, (left), the louder the signal will be for both the input as well as the output.

 

[DaleVO]

"Originally Posted by bouldersoundguy "Generally speaking: 0dBVU = +4dBu = -18dBFS. In the analog world you (usually) aimed for 0dB on the meter. In digital you (usually) aim for -18."

Reapers internal loudness analyzer is off by -3dbs using the SWS Extension. Do you have any incite on why this is?

I recalibrated my Master meter Offset to "0 = -18". It has been many years, so I do not remember the Reaper default. Right click your Master meter to access the config.
[MENTION=196922]Mack Caster[/MENTION]: Now, I am wondering if this change affects the plug-in readings.
Dale

This screenie is from another VO artist's professionally mastered demo that I was using as a reference track:


And a quick capture of an audiobook that I recently recorded, after some sweetening:

 

[snow lizard]

Digital is not like tape. Tape has a fairly narrow sweet spot to aim for when setting levels. Digital is more forgiving as long as you aren't clipping the signal going in. There is no sweet spot.

As a guideline, I've always thought along the lines of +4dBu = 0dBVU = -18dBFS. These numbers are very close to Boulder's numbers. The difference doesn't matter as long as you aren't clipping. Also, these numbers are a reference only. Different converters can be calibrated in different ways and there is no standard.

For a signal with some transients happening like vocals, acoustic guitar or whatever, it's safe to print levels that don't peak over -12 dBFS. For something like drums you're probably okay if you get some peaks a hair hotter than that. For a steady state signal like ultra distorted electric guitar you're better off closer to -18 dBFS. The RMS levels will still be good that way.

If your levels are lower than that on the way into digital, it doesn't make a huge difference. Might actually be cleaner. You can crank the fader or add a gain plugin or something after the signal is printed to get the levels up if need be. Maximizing the loudness is the last step, after the mix is complete. If that raises the noise floor too much then you probably have other issues to deal with. Ambient isolation from HVAC, trains, parrots, vacuum cleaners, computer fans, etc. Electric isolation from cycle hum, inductance from audio lines crossing power lines, single coil pickups and flourescent lights, inadequate R/F shielding, bad cables etc.

A good digital system should have a much lower noise floor than tape.

The bottom line is, don't clip and you should be okay.

 

[Mack Caster]

I recalibrated my Master meter Offset to "0 = -18". It has been many years, so I do not remember the Reaper default. Right click your Master meter to access the config.
@Mack Caster : Now, I am wondering if this change affects the plug-in readings.
Dale

This screenie is from another VO artist's professionally mastered demo that I was using as a reference track:
View attachment 100779

And a quick capture of an audiobook that I recently recorded, after some sweetening:
View attachment 100781

As long as it is used for digital recording, I don't believe it does. I have calibrated all my meters nine ways to Sunday, and checked the rendered audio using Orban. They always come within a (+ or -) 0.2 dbs. It is only after I place the same audio file back into Reaper after it has been rendered, that the Loudness Checker in SWR shows the 3dbs difference. I also tried to go into the preferences and changed the control to show a +3db difference, but I got the same results.

Now if I want to render to comply with the AES, I just set the leveler in SWR to a -15LUFS and when I check it in Orban, it is at a dead -18LUFS. I think, and I may be wrong, the meter offset on the master is used if you are coming into Reaper after using Analog gear. Maybe someone can confirm or deny this. Sorry if I don't make any sense but all I do is record digital.

I am sure that all the meters in Reaper are set to dbs as I have checked them with several calibrated sine waves. I can upload them if you need them. As far as the audio books go, the ACX requires you to be between a -18 and -23 dbs using the RMS measuring system and I have a few master buss plugin to get me where I need to be.

Now, regardless of what you may hear, good AIs are calibrated to operate at their best when your input level is between a -12dbs and a -18dbs. If you only paid $30.00 for the ones being praised on e bay these days, this may not be true. One of the members of this community wrote an article explaining why this is. I linked to his article, long before I became a member of this community to dispel all the false information that is on the www. Did I just take their word for it?

No, I sent an e-mail to Tascam concerning my US 16x8 Audio Face along with a link to @massive s article and asked them if they were correct. I wish I had kept their e-mail response as all it said was: "Yes, they are 100% correct". Now, I don't know them guys, I have never done business with them guys and I have had very little interaction with them in this community, but if they have an opinion on something that deals with audio, they go to the head of the class.

Now, this is how the article starts out. Heading: PROPER AUDIO RECORDING LEVELS. "You're probably tracking too hot. Here's why...This entry was viewed over 110,000 times in 2008, nearly 300,000 times in 2009, picked up by and published by Pro Sound Web in 2011 -- And is *still* waiting for tweaking... And I feel a bit guilty about it because (as of mid-2012), the e-mails just keep coming in". You can read it here.

 

[bouldersoundguy]

As a guideline, I've always thought along the lines of +4dBu = 0dBVU = -18dBFS.

The +4dBu value for line level is correct. I had it wrong and have changed it to match your numbers.

 

[DaleVO]

As long as it is used for digital recording, I don't believe it does.

To avoid confusing the OP discussion about recording levels, I failed to mention that my screenies were captured from playback of mastered or compressed signals, not raw recorded input levels. Like the advice given, I too target -18 to -12dB recorded input levels. Then, I use my ears and process the signal, to ACX-acceptable levels.
After reading your post about the consistent -3LUFS difference, on your rendered clips, I began wondering if the SWS loudness analyzer reading could be affected by the Master recalibration. You sparked some experimentation for me...
Dale

 

[Mack Caster]

After reading your post about the consistent -3LUFS difference, on your rendered clips, I began wondering if the SWS loudness analyzer reading could be affected by the Master recalibration. You sparked some experimentation for me...Dale

That's funny cause that's what your question to me did! So before I made my reply back to you, I opened Reaper, reset the master meter by 3dbs. I than went through my process to see if that made a difference. It did not. That led me to the conclusion that the meters just registered dbs and as long as it was a digital signal. it did not matter to any thing connected with the SWR.

Now with that being said, you may have very well lead me to the answer I have been searching for, close to a year now. I do not have time to mess with it now, but the pic below may hold the key! This was due to your simple question, I could not answer. Check out the settings in the drop down menu. If you play with the settings and it improves, please let me know and I will do the same. The one marked Reaper 2.3, may be where my 3db offset is coming from. THANK YOU!

View attachment 100784

 

[spantini]

I experimented. I added a track. I added a Small Full Meter to it. Using the input gain at the interface, I set the mic's level at just below -18 (peaked at -18) on that meter when strumming acoustic guitar chords. I then put the guitar down, and watched the meter...

With the mic just laying in my padded box, the meter showed small yellow signals at the -60 to -50(?) levels in a "quiet" room. As the mic lay there picking up some noise, I slowly reduced the gain at the interface until the yellow signal at the track meter disappeared.

I then recorded a few bars of this "silence". I then ran it through VST: ReaFir (FFT EQ+Dynamics...) to check the floor.

Here are two screen captures. I think the loudest part may be from me handling the mouse to start the Record process.

 

[spantini]

Another experiment (to reduce recorded noise I can't block yet).

I reduced the recording level of my acoustic guitar to -22dBs. Any higher and room/building noises appear on the meter in the yellow range. I don't have a stand or boom to get closer placement, I'm at about 12-inches on this last recording.

Would it be better to record at -18dBs to -12dBs for optimal signal to noise ratio... or... would recording lower at -22dBs (as I have done to reduce the room noise) be a better compromise?

I used the towel-wrapped box (again) to place the mic into, this time I positioned it at the 12th fret and angled it more towards the hole. I got a slightly brighter sound, but it also seemed to be smoother - especially after a little EQ. I normalized the track and that greatly increased volume levels without bringing much of the noise with it. But my ears aren't sensitive to that yet, so...

I know I should be striving for perfection in the noise area, but right now my only option is to build an upright screen enclosure using heavy towels - and this AT2020 XLR mic is so sensitive I'm not sure that will help much.. maybe cut the noise by half. That's something, anyway - right?

 

[Mack Caster]

Since you are aware of the room conditions, you can try several locations to see where it sounds the best. Sometimes 12 to 18 inches left or right, front or back can make a great improvement. I am talking about moving your gee-tar. As far as recording at a lower input level, if you lower it to say a -30dbs. We assume that since the mic is at a lower level, it will not pick up the bg noise. Now, I am going to catch hell for saying this but I have done my own test and I know what I know.

You mic hears things just as you do. Even if it is just a dog barking off in the distance. When you raise the volume to around a -13 or -14dbs, you are going from a -30 dbs to a -13dbs and your noise floor has no choice but to come right up with it. So, IMHO, I would rather have the recommended input for my AI and try and deal with the noise in post, since I will be dealing with it anyway.

You can do this same test. Just set up your mic and record at different input levels, being as quite as you can. After you bring all the files up to the proper levels, slap a VU meter on the file and see what the db level is. The key is to be as quite as you can and try to duplicate each setting for each file. It is one of the first things I have new podcasters or ACX narrators do, so I can get a sense of their recording environment.

 

[bouldersoundguy]

Now, I am going to catch hell for saying this...

As well you should.

 

[spantini]

...

... Now, I am going to catch hell for saying this but I have done my own test and I know what I know.

You mic hears things just as you do. Even if it is just a dog barking off in the distance. When you raise the volume to around a -13 or -14dbs, you are going from a -30 dbs to a -13dbs and your noise floor has no choice but to come right up with it. So, IMHO, I would rather have the recommended input for my AI and try and deal with the noise in post, since I will be dealing with it anyway.

...

That's what I was looking for. I actually don't recognize the noise when I hear it in loud passages, so it doesn't bother me. I'll stick with the -18 to -12dBs recording levels and go from there.

Thanks, Mack.

___________________________

After reading your post I had a thought.. this AT2020 has a cardioid polar pattern which is supposed to cut down on the sensitivity at the rear and sides - concentrating on the front. My "thought" is.. maybe turn the mic to face into my towel-lined box, leaving only the rear and sides exposed to room/building noises.

That will probably reduce the input signal, but what-the-hey!, as the Indians say. I'm gonna try that just for the hell of it and we'll see..

And then I'll give those varied tests you mentioned a try.

 

[spantini]

As well you should.

 

[Mack Caster]

Have you considered recording in a closet? Sounds silly but you never know. It should help with the bg noise as well.

 

[spantini]

I am currently leasing an efficiency which has virtually no closet space. Next year I hope to be in larger quarters with a dedicated music room - and less noise. Years ago, I set up a live recording space in a concrete basement using very large Persian rugs hung from the joists to the floor - then two layers of pile carpeting on the floor. No studio, but produced some decent close-mic'd recordings of our jams. I could sure use that space now

Maybe I can make a booth out of a couple sheets of plywood and some pile carpeting. If I hinge it the right way it could be collapsible. Piano hinges, I think.

See, now you've got me thinking. There goes the budget..

 

[ecc83]

I am currently leasing an efficiency which has virtually no closet space. Next year I hope to be in larger quarters with a dedicated music room - and less noise. Years ago, I set up a live recording space in a concrete basement using very large Persian rugs hung from the joists to the floor - then two layers of pile carpeting on the floor. No studio, but produced some decent close-mic'd recordings of our jams. I could sure use that space now

Maybe I can make a booth out of a couple sheets of plywood and some pile carpeting. If I hinge it the right way it could be collapsible. Piano hinges, I think.

See, now you've got me thinking. There goes the budget..

You would do better IMHO to make a 'booth' out of 1.25" plastic waste pipe and cover with old carpet, duvets and /or blankets. A decent size made in ply is going to be firkin' heavy.

Dave.

 

[Mack Caster]

I agree with Dave, [MENTION=89697]ecc83[/MENTION] 100%. Keep in mind you don't want to block everything. The pic below is so simple to do and you can add "Tee" in on the top rail to throw a quilt over the top if needed. You can build smaller ones for just vocals or to place behind your mics. Keep one side open and you can use elect zip ties from Lowes or any hardware store as the hinges.

You can build one for under $30.00. When not in use, remove blankets, fold flat, lean against the wall or slide under bed. When you move, simple to take with ya. ;-)

View attachment 100801