Two questions regarding spaced pairs and compression
 

Would appreciate any feedback here.

First topic - Read a UAD blog on solo acoustic mic placements. In spaced pair suggestions, they used an SDC for the front mic (pointed at the neck junction) and an LDC for the mic pointed at the bridge section of the lower bout. Would they use the LDC because the cardioid pattern would be larger (than the SDC) to cover the larger area of the lower body of the guitar? Last night I did some recording using a KM184 for the front and a Rode NT1A for the body and it sounded quite nice.

Second topic - Compression used specifically for solo acoustic guitar recording. From what I have gathered from articles and videos, you should not need it if you get a good raw recording. The last few weeks I’ve been using two KM184’s for recording and played around with mixing the tracks through a Focusrite compressor plug-in and the effects seem to be very subtle at best. If the recording is relatively balanced, it seems like compression is not really needed.

Thanks again!

Since the desired sound is subjective there is not really a "standardized" industry method . Renowned engineers have used different mic'ing methods to successfully achieve the sounds they are after.

Pairs of the same exact mic models in both SDC and LDC configurations. ,,,,pairs of the same type (SDC or LDC) but different models or mixed brands . And also using pairs of different types and or brands , one SDC and one LDC. All of these different methods have been used to make great sounding recordings. In other words the "rule" is there are no rules ,,,, only results

"Compression"
First the phrase "should not need" is subjective and ambiguous at best
A more accurate statement would be ...
Is compression "required" to a good recording ? NO
Is compression "desirable" that is totally subjective to the individual .
In compression "subtle" ( or judicious) use,, can often be exactly what you might want,,, and usually yields a more natural effect of moving the sound forward in the sound field , while not sounding "pumped" or artificial.

Consider that the notion that, in the "good old days" the engineers often did not add EQ or Compression (especially on solo instrument recordings) , is based on largely on an ambiguity.. Because recording to Analog Tape itself, provides subtle forms of both EQ and compression type results , regardless of whether it is also added or not, from the recording console or outboard gear.

Excerpt from article on analog tape

"While it’s still extremely common to see analog preamps and tube mics in a modern studio setting, tape machines are a somewhat rarer bird. While you'll still find them in many studios, the industry-wide shift to working in the box has made bypassing tape a somewhat more feasible practice. Because of this, many engineers feel that modern music is missing something.
Recording to tape gives music a certain character. It rolls off the high-end and adds a small bump in the lows, similar to an equalizer. It rounds off transient peaks, creating a subtle form of compression. Perhaps most importantly, it adds a unique form of saturation, primarily boosting what are known as 3rd order harmonics.

3rd order harmonics mimic the frequency response of a signal two octaves above the source material. For instance, if you recorded a low E on a bass guitar, the root frequency would be 40Hz. 2nd order harmonics, which are typically added by tubes, boost one octave above the source (80Hz). 3rd order harmonics are primarily added by analog tape, and boost two octaves above the source (120Hz). This helps signals “cut through” a mix by adding higher frequencies in a musical way, instead of simply boosting the top band on an EQ. As with all good things, too much can cause negative effects.

Well, there are many, more serious recording folks here that I'm sure will weigh in, but the choice of mics and their placement is probably 80% or more dependent on the player, the instrument, the mics, and maybe the space, though not so much in a closer mic'd setup. There are no rules.

Me, if I don't have 2 SDCs or a single mic, I've used the LDC as the one off the neck join and an SDC at the body because I use it more as a LF control than something to capture the instrument sound. But, TBH, a couple good mics placed well are not going to sound that different, though, back to ❡1, "it all depends."

Compression, well, I use it all the time because the performer (cough) usually does not have perfect dynamic control, or even if they do, the guitar is sitting in a mix, perhaps just a vocal. It's a rare performer that can balance their guitar playing and vocal performance to the point an audio *recording* doesn't show some improvement with a little compression, especially if you are trying to target a specific kind of loudness and dynamics. Just my experience, though.

When I use 2 mics, I will have those bussed to an aux where the compression is applied. Threshold is the control that is most useful, and in the case of a mix, probably some side-chaining. Again, my own fumblings...

I like to use a little compression to make the recording sound a bit fuller. I try to use plugins that add a little warmth also. But the mic placement makes the most difference to my ears. Things like compression and reverb and just icing on the cake after the fact, so to speak.

Given my application is solo instrumental fingerstyle acoustic, what has been frustrating in searching out advice on learning about applying compression (and understanding compression parameters) is that most of what I find is in the context of making the guitar sit properly in a mix with other instruments and/or a vocalist. One of the articles that made some sense to me was looking at compression where a solo instrumental player is using percussive techniques as the percussion can create transients that benefit from compression to smooth out the recording.

I’ve got a “heavy” thumb that I am always trying to tame and perhaps compression could be used to smooth it out in the mix rather than EQ’ing?

My attitude about compression for solo acoustic is set the ratio light (3 to 1, maybe 4 to 1) then bring up input gain until you just start to hear the effect, then back it off just a tad. Too much and it brings out finger squeaks, nail noise, etc. Also, be aware of your gain staging. If the output is too hot, bring down the output gain.

Yes, a lot of compression has to do with making things fit in a mix, or "gluing" the mix together.

You can use it to fix things, too, though be careful. I briefly used a compressor with my electric playing but quickly decided it was making me lazy about some technique. Now, there are styles where it's practically required (e.g. "chicken pickin'), but as a matter of course, it's best reserved in solo playing for those things that just stick out a bit much, and not worth a retake or punch-in may not be worth the trouble (or possible).

Something like that is probably where I'd either use a side chain filter or dynamic EQ, possibly with automation. (Feels like this was discussed recently here.) I.e., you only want the EQ to be applied when it occurs on notes of a certain (excess) dynamic, and so you can set up an EQ and just automate it on/off in those parts, or if there are a lot, use a compressor, but filter it for the EQ range where your thumb is causing problems, adjusting threshold to just hit the dynamic you wnat squashed a bit, or, employ one of those newer, dynamic EQ plugins (which are kind of EQ filtered compressors in an EQ kind of UI wrapper, IMO).

You may not want any "makeup gain" if you are using the compressor to strictly target a few notes/frequencies, at least not in the filtered compressor/dynamic EQ scenario.

Videos like this from Michael Watts have been driving my recording priorities:

https://www.youtube.com/watch?v=lQV2...uEGuZH&index=1

and

https://www.youtube.com/watch?v=7G7-...uEGuZH&index=3

He's focused completely on the front end of the production (player, guitar, mics, and room).

I'll try this to see if I hear a difference. Thanks for the feedback!

Hope the Edwinson is moving along well.

Check into multiband compression. You can divide the signal into frequency ranges and set compression differently for each range. (Similar to side-chain compression, but more versatile.)

It can be amazing how effective a small bit of compression in a narrow range can be in tamping down problems.

I will, thanks for the suggestion.

I know I've told this story here before: There was an article in one of the recording magazines a few years back, one of those "20 sure-fire techniques for recording acoustic guitar" things. I read it with great interest, all kinds of ideas I hadn't thought of. Got to the end and the "famous recording engineer" who was writing says "All these techniques are sure to work unless you're doing something totally crazy, out in left field, that no one ever, ever does, like recording solo instrumental guitar". Sigh... Tosses article... You do have to be careful about the context :-)

I view compression as something I'd do in a mastering step, and it needs to be done very carefully for the kind of music I play, or else it will be obvious. I don't use it to fix or balance out a recording. That's best done with mic placement and fingers. That said, I do find dynamic EQ useful, which is basically an EQ side-chained to a compressor. You can set it so that when a certain frequency triggers a threshold, you can cut (or boost) that frequency. You set it up like an EQ, but with that additional threshold feature. Very handy for occasional boomy notes, maybe a raspy high note that you only hit now and then. Even tho it's leveraging compressor technology, the idea is more "EQ that only kicks in when you need it".

In the mastering stage, I do typically use some simple compression/limiting. Ozone's been my go to limiter for a while, but lately I've been using Limitless. In either case, we're talking *very* little compression. I typically use it to bring the track up to release level and that's it. In other words, I'm trying to find the point just before compression starts, with perhaps an occasional overshoot. Ozone has a nice display that shows you when limiting is kicking in. Here's an example. That line is showing when compression is applied. So here, it basically isn't, except for that short little divot on one note a few seconds in.

Attachment 50314

I also sometimes like the sound of an LA2 compressor on a track. Again, adjusted to where the meters don't move, so in theory, "no" compression. In reality there's maybe a 1/2 db or something happening, and the LA2 just seems to add a nice color and smoothness.

On your mics question - personally, I tend to use identical mics most of the time. I can see in theory that using different mics could be useful, LD vs SDC, or even condenser and ribbon. I've been able to play with those ideas easily lately, since I got the Townsend L22 modeling mic, which lets me record stereo and swap virtual mics around after the fact while listening. You can certainly get some different colors both by choosing different mics and by choosing different pairings. On the Christmas CD I recorded last year, I used a LD/Ribbon mic pairing on a few tracks, and liked the way that came out. But other times, I've thought the result seemed a bit odd and matching mics seemed more natural sounding.

It's possible to get an absolutely outstanding recording of a guitar or any other set of audio signals with very bad dynamic balance. Judiciously applied compression really reduces the cringe factor when listening to such a recording.

Do you have an example? Seems like a contradiction - a recording with bad dynamic balance seems like either a bad performance, a bad recording, or just a piece with both very quiet and very loud parts. In the latter case, a bit of volume automation would be what I'd go for. Compressing a very loud part enough to even it out with a very soft section, on solo acoustic guitar, is at least in danger of creating audible, pumping compression - tho of course it all depends on the degree.

One place people talk about using it - maybe this is what you mean - is for percussive guitar. Certainly you can use limiting to reduce peaks caused by hitting the guitar. But a while back when I interviewed Andy McKee for an article, one of the things that impressed me was his dynamic control - he sounded in person exactly like his records. His loud stuff (hitting the guitar) was relatively quiet. His "quiet" stuff - harmonics, tapped notes, etc were nearly as loud as picked notes. So on my recordings of him, I didn't have to do anything special to make him sound balanced. He did it with his fingers. Jon Gomm talks a lot about this same thing in one of his instruction videos. If the performance is right, you don't have to "fix it in the mix".

That’s what inspires me about you, Andy McKee and what I mentioned above about Michael Watts. You all seem to be in pursuit of a recording that requires little to no post processing because the front end is done with commitment to a really good signal chain (great mics, good room, good mic pre, and most importantly a great player who knows how to control his dynamics).

Well, it's mostly because that's what I find works. I've never had much luck fixing a bad recording with post-processing, at least not to the point that I'm truly happy with it.

I just "fixed" a really terrible recording for someone - a video demo done in a noisy, lively room, probably with phone. But all I could do was take it from unusable to pretty mediocre. There are always improvements one can make, but what I've observed over the years is that the really good recordings (speaking just of solo guitar here) weren't created by studio tricks. You can do things, a little EQ, a little compression, add some reverb, but it all works best if the raw recording sounds pretty good to start with.

The good news is that for solo instrumental guitar, a good quality recording chain isn't that hard to come by or that expensive (all relative...). Room acoustics are most people's biggest challenges, but even that is a lot simpler than for those recording more complex music. With close micing, many people's home environments work out fine for solo guitar.

Record solo guitar. Almost always use a matched pair of SD condenser mikes. When I have used a LD and SD combo
I have put the LD on the neck side and the SD on the body side because the neck side has more high frequency content
and I wanted reduce the amount of high frequency sizzle (in theory anyway). Compression I don't use though sometimes
use a volume envelope.

[QUOTE] There are lots of guitars with bad dynamic balance and oddly resonant notes.

I am no expert in the fine nuances here. It seems like automation vs compression is sort of an odd discussion when it's just a one track recording. Compression is often applied to one track and automation is often applied to the whole recording (in this case, one track).

[QUOTE=H165;6611457]Ah. The cure for that is a new guitar :-) I guess you can technically make a great (aka "accurate") recording of a bad sound, but for a good final result, you need the entire signal chain, starting with the guitar, to sound good.

I use automation for anything I'd like to have automated and reproducible, including one track. In many DAWs you can automate anything - you could even automate a variable amount of compression. I use it for volume, reverb, EQ, just about anything. One of the coolest tricks I've run across was in a recording session (solo guitar) I did that was produced by El McMeen and engineered by his regular engineer. There was a short lick where El felt like I had dragged the tempo - just a little ascending run of 4 or 5 notes. We were debating whether to re-record, when El suggested the engineer try using automation to reduce the volume by a few db starting with the 1st note, and then bring it back up by the end of the run. Presto! That fixed the "timing" issue, or at least the perception of it.

It sounds like your idea of compression and automation relates to multi-track recording. For me, with a single track, you're not trying to compress anything to get it to sit in a mix, so you basically don't do those sort of mixing things for 1-track solo guitar. I think of compression as a mastering step, and therefore applies to the "whole mix", which in this case means just one guitar. With a band mix, you might compress individual tracks, and then the mastering engineer might apply overall compression to the mix. With solo guitar, you just have the mastering step.

Automation is certainly useful in a full mix, where you want to bring things in and out. But it's also useful for single track recordings, where you want to have total control. If, for example, you have a good recording, but there's one note that you just hit too hard, it's usually better to bring down that note with automation than to apply heavy enough compression to bring it down. You're less likely to hear "pumping".

Well yes and no,
While compression and automation are definitely two different methods , either can be used to tame transient attack, on a solo guitar recording

Yes certainly automation can be applied to an entire recording BUT
As Doug noted and Rick mentioned , it can also be applied to any single section of a single track, and even to a single transient of very short duration on a single track. And as noted can applied to any single track parameter, for any portion of a track, or any plugin parameter for any portion of a track.

One example: I just finished mixing a vocal , where I applied a short duration tape echo plugin (to get a delay type effect) to the entire vocal track in parallel,,, and I also applied that same plugin directly on the vocal track with a longer single repeat on just a few words in that vocal. (so it's automated to be in "Bypass" except for just the words I want echo repeated. But as Doug mentioned I could have simply only use the one plugin on the vocal track itself, and automated the duration parameter to be short for entire track but change to longer for just the words I wanted repeated .

Thanks Derek, I sure like the KISS priorities!

My betters in recording achievement are talking here, so I should be quiet and listen, but I'm only going to the do the later. :)

I think I've tended to over-compress my recordings with just acoustic guitar instrumentation. It may be because I'm also mixing full band mixes and I'm used to "that sound." I also used to test my proposed masters in a car, and without compression a lot of the performance would be unheard. Now-a-days I try to test my masters on earbuds where I can get away with less compression.

It's not a solo guitar instrumental (which Doug Young was discussing) but as to an example of a compressed/limited "bad sound" making for an emotionally compelling recording I'd point to "Street Fighting Man" by the Rolling Stones where Keith Richards was insisting that his acoustic guitar should go through a cheapo portable cassette recorder preamp with its crude auto-level control.

Yeah, that's why context is important. A Stones recording is a whole different thing than a solo fingerstyle guitar instrumental, like Bob is asking about. Typical fingerstyle guitar has more in common with classical recording. I say typical because there are some fingerstyle players who are more like bands where various "low-fi" or extra expansive techniques make sense, but I think what Bob's trying to achieve is what you hear from the majority of fingerstyle players these days - a somewhat accurate replication of a good sounding acoustic guitar, which is very different from rock bands who are trying to create new textures and sounds.

[QUOTE=Doug Young;6611493]Of if you have one or two really soft good notes you can make them pop by bumping them up. :)
Using automation to say, just bring up just the tiny sliding sound between two notes, or maybe an unintended pinch harmonic, etc, in a track that is sparse is probably one of my favorite things in all of recording.

I agree. I've used my plague time to buckle down try to understand more about recording myself at home. And I certainly have gotten more from looking for Youtube tutorials by typing "recording solo jazz piano," or "classical..." When you type "acoustic guitar recording" just about all results give advice to -on everything from compression settings to mic choice- get the guitar to sit in an ensemble mix.. Unless you also type in "Doug Young." :up:

[QUOTE=min7b5;6611632]Yes, volume automation is one of the great things about modern DAWs. It's easy to zoom in down to to the note or even part of a note and change its volume or envelope. Easy to go down a rabbit hole there "re-pedaling" an entire performance, but gee it's nice when one's best performance still has one or two things that need a fix or there's an accidental note or dynamics effect that adds something.

I tend to use volume automation before compression on a lot of things. Certainly vocals. Sometimes guitar parts. MIDI parts are even more tempting as you change so much after performance. Sometimes with orchestral parts I feel like a very demanding conductor.

This is such a great forum! It's this kind of hands-on information and discussion that improves the amateur's (me) chances of doing something well. I have begun tinkering with the automation stuff on Garage Band just to understand the benefits of it in solo track dynamics. I'd never really thought of it as controlling very small sections...very useful.

Doug is spot on. I’ve reached a point in my journey with the guitar where my playing is starting to sound musical to others and myself. As I put in the time and effort (and joy) to learn new songs, I want to put down a good recording of it to keep for myself and to share with others before I move on. The solo fingerstylist does have the advantage that our recording work flow can be relatively simple if you have decent talent, a great guitar to work with, and a modest investment in hardware and software. I feel very privileged to receive advice from players like Doug, Wrighty, TBman, KevWind, Rick-Slo, and many others to get to a point where I am having some success and consistency in my recording efforts. This is such a great community to be a part of, I hope I can give back as much as I have received.

[QUOTE=min7b5;6611632]I need to figure how to do this on Reaper. I have plenty of takes where I have a few high or low notes that I would like to “treat” but instead I trash the recording and try again. That can be time consuming (and tiring at times).

[QUOTE=SprintBob;6612191]Regarding using a volume envelope you need to be aware that when you increase or decrease the volume of a particular note at a
particular time in the recording you will be doing the same with the other notes that were already ringing out on the guitar (even
true with a equalizer envelope due to all the overtones going on). Probably more noticeable (going on for a longer time) when using
a reverb. Listen carefully to tell whether that becomes an issue.