It's best to be cautious with levels during recording and sort out the details in post anyway, so what you're doing is probably close to the ideal workflow. Compressors take some skill to use optimally. Until you have that skill you're as likely to cause problems as solve them. Compression in the digital realm is at least as good as an affordable analog compressor, and it has the added benefit of undo so you can experiment without risk.

On top of that, some amount of manual editing may be a good idea, instead of or in addition to compression.

If you're going to spend money, get a portable audio recorder and sync up the audio and video in post. Cameras are rarely good audio recorders. A decent audio recorder will have better noise specs and better control. I use a Zoom H5.

What software are you using to edit your videos?

Thanks for the reply! I'm using Vegas Pro to edit the videos. And I record from the main outs of the mixer ZED10FX to the XLR inputs on the Canon XA10 camcorder
I think it would be extremely difficult o use an audio recorder because the camera is starting and stopping frequently.

Do you think it's better to record to the camera with lower then optimal levels and continue to just normalize the audio track in Vegas?

If you absolutely must use the camcorder, yes, it's best to avoid overloading the audio circuits or clipping the converters. Also, defeat any automatic gain control. You can control things better in post. The main downside is the audio performance of the camera itself, which may show itself with more conservative input levels.

There's a reason they use a clapper board in film and video. It provides a clear cue for synchronizing separate audio and video recordings. Simply clapping your hands in the video frame will do the same thing. And you can slate each take with a spoken "song x, take y" followed by the clap. Since the camcorder will also have audio, it's not as necessary to have a clapper board with the take info written on it, just find the clap in the cam and audio recorder waveforms and line them up.

Actually, syncing with audio is easy enough that I don't bother with claps (even though I have a clapper board). It's syncing multiple video angles that's more challenging. I've gotten good enough that it's not a obstacle, but I used to take still shots with a flash between each song to facilitate the process.

I also use Vegas Pro 14.

From my teaching years in college - the one audio tool that was virtually impossible to teach and demonstrate were compressors. The changes to the sound are subtle, until they suddenly become savage, and loads of people just could not hear what a skilled user can. If you use a compressor in a way that controls your levels, it sounds horrible. Guitars naturally sustain - the compressor makes this worse and you will hear the noise floor going up and down, a kind of nasty pumping action that really sounds dreadful. You simply need to manage record levels properly so the loudest doesn't distort and then any level changes will follow the player. If it's an instructional video and the watcher cannot replicate what they hear, they will not last. If the subject in the video twangs a note and it goes doooooooooooiiiiiiiiiinnnnnnnnnnng and the students guitar goes dong, they won't work it out. I personally NEVER compress a recording, I often add compression on replay, but most cameras with manual record levels cope well. Most cameras with auto recording levels mess it up anyway - so in the edit you spend time matching audio levels. This is normal, and you do not normalise, or change the dynamics. If the player has pp written on the music, and you bring it up to 'normal', then he plays the ff section and you bring it down, you ruined the music. Levels need to be appropriate and matched and considered, but never compressed, normalised and limited. Soap box mode off. Sorry.

So I take it you would not recommend this pedal? https://www.boss.info/us/products/cp-1x/
Based on the description, first 30 seconds into the video, it sounds like it does exactly what I need.

Personally - guitar pedals are for guitars, and I leave them for guitarists who need them - I'm happy hanging a mic on a cab, or DI'ing them - but for the guitarist they are an effect, rather than a recording processing tool. Only you can tell if it's right for you? Just not something I'm knowledgeable about.

When we record video, the guitarist adjusts his volume level frequently, and sometimes plays soft, sometimes loud. It doesn't sound good on the recording when he plays soft, or with a light touch. That's why I thought a compressor pedal might be helpful to even everything out for more consistent sound during the video recording. Plus, the guitar itself, in my opinion, can sound rather small and mousy, not a big fat sound.

Paul seems to be taking a different tack than the OP. Paul is advocating a classical purist approach while the OP seems to think that technology is a substitute for talent.

this is a performance issue, not a technical one. garbage in, garbage out .... i have to ask, if a person can't play and present themselves correctly in the first place, why are they being recorded?

you could try recording a pass and then play it back and ask them, "See what happens when you do that?" ........

Kurt Foster, post: 457767, member: 7836 wrote: Paul seems to be taking a different tack than the OP. Paul is advocating a classical purist approach while the OP seems to think that technology is a substitute for talent.

this is a performance issue, not a technical one. garbage in, garbage out .... i have to ask, if a person can't play and present themselves correctly in the first place, why are they being recorded?

you could try recording a pass and then play it back and ask them, "See what happens when you do that?" ........

What Kurt says.^^^

Not sure Kurt? In my band, our guitarist who is a really good friend - but there is 35 years difference between us (many think I'm his Dad!) drives me nuts by doing the same thing - he plays very quietly, then plays very loudly. He's also a very good recordist but knows how he wants his instrument to sound. He would never let me add a compressor to his carefully programmed sound because it would change it. In this topic we're talking about changing a guitarists sound to make it record better? I suspect the problem is the recording technique - which I assume is simple direct to stereo - is the problem. If there are no other instruments, then why not record his processed output, complete with the flaws, and also record the dry feed with a DI before the processing - and then you could control this afterwards - perhaps re-creating his sound, but with you doing it with more control.

It's not what we're talking about but I've never known many musicians who could change their playing style to solve a technical problem. I've heard recent graduates go up to a performer and tell them that they're too far from the mic, and either need to move closer, or play louder, and I've seen that person (with a shelf full of gold awards and record credits) look them in the eye and say very quietly. "I've been playing this way for 40 years, I'm too old to change now". There were tears in my eyes.

Sorry I am not a pro with this stuff. The guitar is going into a DI box, then into a PreSonus BlueTube, and then into the ZED mixer. I'll just leave the set up as is and forget the compressor pedal since you don't seem to think it would help. I tend to think differently because when I add a similar effect in the video editor, it makes it sound better, but I'd rather record it sounding better than screwing around with it in the video editing program.

that's all fine and dandy if the results are acceptable Paul. i'm all for technique. when the guy in your band does it do you need to fix it in post? i guess no. i don't see how recording a direct to stereo technique could be the issue as long as theres sufficient level going to tape at the soft passages. to me that line of thought approaches the problem from the wrong direction. the answer to the problem is like that old joke, "Doc, it hurt's when i do this, ..... " "Then don't do that".

MC208, post: 457771, member: 49667 wrote: Sorry I am not a pro with this stuff. The guitar is going into a DI box, then into a PreSonus BlueTube, and then into the ZED mixer. I'll just leave the set up as is and forget the compressor pedal since you don't seem to think it would help. I tend to think differently because when I add a similar effect in the video editor, it makes it sound better, but I'd rather record it sounding better than screwing around with it in the video editing program.

actually i think that is the correct approach. all i have ever gleaned out of all my years of doing this stuff is it has to be right at the source.

Rather than trying to explain, listen to a sample. The first part is a chord solo, played with fingers, and the second part when the backing track comes in (not in this sample) is played with a pick.
This is the recorded sound, no normalization, nothing done to it in post.

[MEDIA=audio]https://recording.o…

Attached files sample.mp3 (4.4 MB) 

i think that sounds pretty good actually. good player, good tone. nothing distorted. dynamic range is a good thing.

MC208, post: 457774, member: 49667 wrote: Rather than trying to explain, listen to a sample. The first part is a chord solo, played with fingers, and the second part when the backing track comes in (not in this sample) is played with a pick.
This is the recorded sound, no normalization, nothing done to it in post.

[MEDIA=audio]https://recording.o…

That's exactly the kind of thing that needs intelligent manual adjustment (Edit: which might mean leaving it alone). It's not a case for a compressor.

can we hear it with the backing track but not normalized?

Yes, here is the guitar and backing track together. They're recorded on separate channels, left and right, but then I pan center on both when I bring them into the video editing software.

[MEDIA=audio]https://recording.o…

Attached files guitar and backing.mp3 (4.3 MB) 

sounds fine. i really like the guitar work.

MC208, post: 457771, member: 49667 wrote: Sorry I am not a pro with this stuff. The guitar is going into a DI box, then into a PreSonus BlueTube, and then into the ZED mixer. I'll just leave the set up as is and forget the compressor pedal since you don't seem to think it would help. I tend to think differently because when I add a similar effect in the video editor, it makes it sound better, but I'd rather record it sounding better than screwing around with it in the video editing program.

Having done a few videos my self, I should say that I always mix my audio in my DAW and then import it to the my video editing software. The tools are of highest quality and easier to work with.
That said, there are a very different things about compressing in the way directly after the output of a guitar or after the preamp or post in the DAW.
First any compressing done on the way in couldn't be undone. Done before the preamp or after will give different results because of gain staging. Also I would never use a compressor guitar pedal after a preamp but before, as a effect would be acceptable if there is a reason. For exemple if an effect placed after the comp pedal needs a more EQ level to be fed to and produce more constant saturation or level reacting effect.
The compression done after the preamp would, in my opinion, need to be of higher quality or special sonic qualities because inbox comp are so more precise and noiseless these days. So I feel you need to have a reason to use an outboard compressor (sonicly, character etc..)

The performace you present is kind of classical music, maybe jazz and/or blues influenced.. What's mostly different in this style is an approach of fidelity of the playing which mostly offer more dynamic end results compare to modern music. That's why you had many comment saying it's just fine without compression.. Of course if your song goes to a mastering engineer, he might use a bit of compression to even out the peaks and obtain a level that stands up against other album of your genre. But at mix time, I'd be very carefull about compression.

In the end you are the judge of what you want to hear. being the performer and producer. if the sound you aim is more modern, feel free to compress the hell out of it if it sounds good to you... Nobody can make that decision for you. ;)

Thanks for the information and suggestions. Most of the videos produced are of an instructional nature, not really performance. But I guess I'll continue on without a compressor.

I'm looking at the idea of a limiter again for recording instructional videos. I'm not sure if it's the right piece of equipment but here's the problem. It's very easy for the guitar to spike depending on on hard the guitar player plays something, likewise, it can go very soft. And I find myself having to constantly monitor the levels, and constantly adjusting them. For the purpose of instruction (talking and demonstrating), I do like to normalize because it just gives everything a bigger presence. but when I get spikes in the audio, normalization doesn't work well. Would a limiter allow me to set it so that nothing gets recorded louder than say -10 db, and then in my NLE, I can normalize it to 0 db for a nice big full sound on the video?

You are recording digital? Then simply set the maximum so it doesn't clip and use the tons of dynamic range you have, then you sort it out in the edit, or post-production. Perhaps simply riding a fader, or maybe a compressor to retain at least some dynamic range. Limiters on record are pretty pointless - as you have the ability to record more dynamics that the playback systems can handle. Normalisation is a crude device to bring average levels up, as you say, till the highest peak reaches your normalisation maximum. Compression brings down the dynamic range, but in general - record every source separately synced, and sort out the balance when you are in a controlled environment. Compression, limiting and effects are bad news to use on the master source material. I never record via any form of processing - as it's often a case that you cannot undo the damage. A brick wall limiter might have a place if your recording is totally wild and uncontrolled.

paulears, post: 458335, member: 47782 wrote: Then simply set the maximum so it doesn't clip

How is that done? I'm using a ZED10FX mixer. I try to set the level so it doesn't go above what I want it to, but the guitar is so annoying because it constantly does. Never stays at a consistent level.
I don't know compressor, or limiter, but just something I could put in that would keep the recording from having spikes that I don't want. This isn't a music performance, it's an instructional video. I know hardly anything about this stuff, but I do know that when my recording levels are lower, and then I normalize in the software, it sounds infinitely better. When I try to record at higher levels to avoid normalizing, then it sounds worse.

You bash the guitar at the point where it peaks and back off a bit. My tascam interface distorts quite easily on the instrument inpout, so you hit hard, watch the red light and back it off so no matter what you do it can't peak . I presume you are panning voice hard left and guitar hard right then blending in the edit/PP?

Yes that's correct, voice and guitar and panned left and right when recording, there' also a piano that's panned in the same direction as the voice

The piano could confuse things as you'll lose the separation?

I panned them that way because typically the piano isn't going to be playing when the guitarist is talking

I went ahead and picked up a DBX 166A unit off eBay and connected it to the main outs from my mixer, and to the inputs on my camcorder. Can anyone recommend a really dumbed down tutorial on how to use the unit to accomplish my needs? I'll use one channel for voice and the other channel for the guitar. Main goal is to prevent spikes in audio, so I think that's where the peak limiter feature will be used (just one knob, I can handle that one!). Someone on my video forum told me not to bother with the expander/gate feature. So that leaves the Compressor feature to figure out.

It's going to take some practice to get it right. The settings are going to depend on the specifics of the signals in question and the desired output results.

You're correct, the peak limiter will deal with keeping peaks under control. The compressor section is slower. It doesn't have attack and release controls like many compressors but it has "content dependent" automatic attack and release. So at least that's less for you to think about. I would use "over easy" for the voice and probably not use it for the guitar. It creates a relatively gradual transition from 1:1 ratio to whatever you set. Beyond that you'll have to experiment to get the right ratio and threshold for the source material. You need to be able to hear the results in real time and take some care to set it correctly, which is why I would lean toward doing it in post.

I would avoid setting the limiter on the 166 series to anything other than maximum (+20dBu). It's the analogue equivalent of going over full scale on a digital system (a pair of diodes connected to a +/- static voltage set by the limiter knob), and despite the soothing talk in the manual about soft limits, it sounds just as bad.

If you want to use the dbx166A to reduce peak excursions, set a high compressor ratio with fast attack and release times, and then carefully adjust the threshold control downwards starting from maximum until you reach a setting that you think suits the programme material. Then notch it back again by a couple of dB.

A warning about the 166 series: the contacts on the push switches go dirty very quickly, resulting in the acoustic operation jumping all over the place during use. To correct this, I would unplug the mains, take the top cover off and give the switches a good soaking in a decent-quality contact cleaner while operating them vigorously. Concentrate on one switch at a time, and let the unit dry out well before re-applying power.

You may have to remove the LED board to get proper access to the switches, but it's not difficult to do. I often perform switch cleaning on 166XLs that I encounter, but for some reason have never met a 166A that needed instant attention.

The 166A doesn't have attack and release controls.

bouldersound, post: 458703, member: 38959 wrote: The 166A doesn't have attack and release controls.

There is none on the LA2a either and no one as ever complained ;)

pcrecord, post: 458708, member: 46460 wrote: There is none on the LA2a either and no one as ever complained ;)

Right, but I'm responding to Boswell's suggestion to set the attack and release controls.

It must be the original 166 and the 166xl that have the attack/release controls, where the "cut down" 166A does not. It's the xl I see the most of. Equipping them with BB op amps in the signal path and replacing the original VCAs with That devices turns them into remarkably respectable units for the money.

The original black 166 also lacks attack and release. The 266 varieties always had them.

Boswell, post: 458702, member: 29034 wrote: If you want to use the dbx166A to reduce peak excursions, set a high compressor ratio with fast attack and release times, and then carefully adjust the threshold control downwards starting from maximum until you reach a setting that you think suits the programme material. Then notch it back again by a couple of dB.

So you're saying NOT to use the peak limiter feature to catch peaking, but instead use the compressor feature? I messed around with the unit a little bit the other day testing the noise gate feature to see if it got rid of some hum we're experiencing. It did, so I ordered a dedicated noise gate pedal to take care of that. I think it will be better than using the gate in the 166A. Anyways, I will try your suggestion with the compressor, by following your instructions and see how it goes.

OK, but Boulder correctly pointed out that the 166A doesn't have the same compression attack and release time controls as the 166xl that I was basing my comments on - it just has a fast/slow button. Worth a try, though, at least in the fast mode (MODE button out).