Jump to content
Misha

Limiter while recording vocals.

Recommended Posts

Like others have said, it would be better to record at lower levels. I realise you said that this causes issues during mixing in regards to room noise. For that I would recommend investing in something like Izotope's RX7 or Steinberg's Spectral Layers 6. Both products will allow you to isolate room noise from the vocals. Although note that often entirely removing room noise does not sound as natural as simply lowering the amplitude of the extraneous/problematic sounds.

Share this post


Link to post
Share on other sites

I'll make a video. Yes, I'm saying you can use that compressor to impact what you hear through headphones or speakers.

Share this post


Link to post
Share on other sites

Gswitz, 

you do not have to do that. I think I have an idea how you want to route this. That was my fundamental mistake,  when I started this thread.... In most probable scenario, the software limiter will react to something post A/D conversion. Meaning, if clipping occurs before conversion it is still "game over" and software limiter would not save the day.

 

I think in what I want, only outboard will accomplish...or built in DSP in interface, which unfortunately I do not have. 

Share this post


Link to post
Share on other sites

Just to clarify, the OP is seeking a limiter before the audio is committed to disk. Anything added after the A/D converter will not affect that. 

Share this post


Link to post
Share on other sites

mettelus,

Yes, I was a bit naive when started this thread, thinking it is possible to use software limiter to curb peaks in live scenario - "before the audio is committed to disk"

Most of my clipping occur before A/D.   So if I got this correctly either outboard limiter/compressor or DSP in audio interface would help to reduce it.

Share this post


Link to post
Share on other sites

It is essential to apply compression before the signal hits the ADC.

Audio only gets converted to floating-point data after the interface, which works exclusively with signed integers. That means that instead of clipping you get inversion, e.g. an overflow that turns a positive number into a negative number. This doesn't sound like analog clipping at all. It is particularly nasty-sounding and noticeable even at low levels.

There are only two options for preventing this: analog limiting prior to the interface, or recording at a low-enough level that clipping can't happen. The latter is indeed practicable -- if you have a quiet microphone, a quiet room, and acoustical treatments to mitigate boundary reflections and room resonances. 

  • Like 1

Share this post


Link to post
Share on other sites

I would just like to say that if Andrew Huang can record vocals on a frozen windy lake, that there has to be a better solution than aiming to clip and letting an analog compressor munch up the peaks.  (11:42 shows the signal from the vocal mic.)

 

Share this post


Link to post
Share on other sites

bitflipper,  also if a interface has built in DSP... which is probably the most clean way to do it.

 

Royal Yaksman, I had a fairly simple question and I believe I got the answer. The video... Cool guy doing cool things to be cool... that is cool, but it does not relate to my  particular question.  I have heard a drunk opera singer in abandoned tunnel that had more talents than the guy in your video.  My question is not about  "talents" or vocal techniques. It is specifically about using a Limiter/Compressor. And I do not "aim" to clip :)

P.S. Is that a compressor on the table of the guy in the video at his studio?  :)

Share this post


Link to post
Share on other sites
10 minutes ago, Misha said:

bitflipper,  also if a interface has built in DSP... which is probably the most clean way to do it.

 

Royal Yaksman, I had a fairly simple question and I believe I got the answer. The video... Cool guy doing cool things to be cool... that is cool, but it does not relate to my  particular question.  I have heard a drunk opera singer in abandoned tunnel that had more talents than the guy in your video.  My question is not about  "talents" or vocal techniques. It is specifically about using a Limiter/Compressor. And I do not "aim" to clip :)

P.S. Is that a compressor on the table of the guy in the video at his studio?  :)

The point of me posting that video has nothing to do with cool. Quite frankly I thought the whole thing was ridiculously unnecessary. Though I am sure even the artist is aware of that. But I posted it because there is no way in hell the sound in your room could even come close to the wind on that lake and yet he got a fairly clean signal. I know that you did not specifically request technique tips. But for sure whilst you can compress on the way in, there will be less you can do with the captured audio in the mixing process as a result.

Lastly, you say that you do not aim to clip, but are intent on continuing to employ a practice that you know will result in clipping?...🤪

Share this post


Link to post
Share on other sites

I remain undeterred. 

There may be reasons not to use cakewalk for live performances, but inability to compress or limit isn't one of them.

Any compressor can be overdriven, before or after adc. 

If you are wanting to print the compressed track instead of the initial input you can using patch points and aux tracks. This is really no different than using the rme compressor except the round trip to the interface.

To save clipping you have to properly set your gain. Once done, everything else you ask for you can have. 

Failing to properly use preamp gain will not be avoided by having external gear. I find external gear is subject to all the same gain staging criteria as internal gear.

You can absolutely use cakewalk to create beautiful live mixes with eq, compression and limiting. To do so, you must set inputs low enough to avoid clipping. You can still maximize output though, targeting your favorite crest factor.

Edited by Gswitz

Share this post


Link to post
Share on other sites

 

 

Gswitz, Royal Yaksman

"set inputs low enough to avoid clipping."  If I experiment with different ranges it is almost impossible to find a "sweet spot" of gain.  

Read about "Caveman" approaches that couple of folks here (including me) are using. Different takes (gains) for different loudness. It works. 

Just want to know if analog limiter or compressor would squash some peaks that occasionally occur if I record at different ranges in a single take :)

 

 

 

 

Share this post


Link to post
Share on other sites

Always do a sound test and make sure that your noise floor is as low as possible. If that isn't possible, you will need an expander, and a noise reduction plugin. Whatever you do, DO NOT GO BEYOND 0dBFS. However, if you need a more consistent recording level, the PRO VLA II can be what you're looking for. it will set you back $299 tho.

 

Share this post


Link to post
Share on other sites

Gswitz,

Thank you for the effort! Per your suggestion, I did watch it twice :) some parts more...

I am familiar with Cakewalk at solid begginer/intermediate level.  You got your point across. Yes, limiter/compressor has activity at recording. My RADIO button was off. Still, I never used it in such manner (I use FX bins only at playback/ mixing),  It would not correct clipping on input, since, as was explained at the start of the topic, if (clipping)  occurs before A/D.  Just as you said it in the beginning, It would have to be an interface with built in DSP or external compressor.  Unfortunately, my interface (Arturia Audiofuse) does not have DSP onboard, but it has inserts for external FX to be mixed in before A/D conversion.

In any case, I thought that your video was put together well and should stay in Cakewalk archive. 

Thank you!

Misha.

 

Share this post


Link to post
Share on other sites

By 2x i meant you could watch it on double speed. I'm kinda a slow talker.

There really is no difference between using rme dsp and cakewalk except the time cost to go into the computer and back. Both are applied after initial digital conversion. Dsp stands for digital signal processing.

You can move your gear externally but you will face the same gain staging issues. Overdriven preamps and compressors distort.

If you were recording to tape it might be more important to use compressors before the tape because tape introduces a good amount of its own noise. But adcs these days are awfully quiet. 

For example, my bro and i thought to test the rme specs, but we don't have any mics quiet enough to do a valid test.

There are many good ways to do this. Nothing wrong with external gear. I have some myself. I took issue with you having to use external gear. I don't think you do. If you set levels correctly, i think you can do what you are describing in cakewalk.

Certainly, you can play and learn. You already have it. 😄

Share this post


Link to post
Share on other sites

Gswitz,

Thank you for nice and educational write.  I guess it comes down to finding a proper balance even in chaotic situations :) 

I never had an interface with built it DSP, now when you mentioned that it comes after the pre-amps seems to me  the limiter  would not work well as a live sound clipping agent 007.   Only if an outboard limiter that is placed first in the chain would.

Again, Thank you for your time!

Share this post


Link to post
Share on other sites

@Misha some folks like the sound of a hot preamp. If that is what you want you may have to attenuate it before digitizing. That is why the Rupert neve 5211 has two outputs for each channel... One louder and one softer... Otherwise the same. 

But if you don't want preamp distortion, you can use tons of preamps and just come in softer. A quieter signal will not clip. After digitizing, you can raise your level internally using compressors and limiters or channel gain. Whether you do this in the interface or in a daw is a difference of the round trip on the USB cable. Other than that 3 millisecond latency, there isn't really a difference. I suppose the code behind the effects might be different but that argument favors your daw choices which will be many.

So, there are a lot of good things about better interfaces and external hardware. Keep in mind, hardware is heavy. Currently, i leave my two channel compressor at home when i go to capture field recordings.

But i do use it often when making recordings at home. I pretty much always run through it when practicing.

I got a nice compressor before i got a nice interface. If i knew then what i know now, i would not have prioritized that compressor purchase.

My video i made for you does not set compression or limiting well at all. I was just demonstrating. If you want me to try to set up an in-the-box chain that might be useful for live performance, i will.

I'm sure you noted how handy the eq would be at finding and notching feedback in a live situation.

Anyway, it seems I'm mostly alone here. Everyone is waiting for the crazy guy to stop talking. Ha ha. Live stuff has to go well and you don't have time to mess around. This is a fair argument for limiting your choices. Keep it simple. That way you can't mess it up so badly.

Most people end up with an array they trust and understand clearly for live stuff. You get to pick what yours will be. Don't mistake free for unworthy.

Cakewalk is pretty great and might in the end be a better tool than thousands of dollars of hardware.

Edited by Gswitz
  • Thanks 1

Share this post


Link to post
Share on other sites

For live band vocals, i find use of expanders more valuable than compressors or limiters. When the singer isn't singing, don't send the room sound to the mains. This is especially important when the band stops.

Share this post


Link to post
Share on other sites

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now

×
×
  • Create New...