How do you use a Multiband limiter/compression tool?

[Doug Young]

Originally Posted by rick-slo

Narrow Q frequency sweep: for example
https://music.tutsplus.com/tutorials...ing--cms-19832

Don't use it much (so don't get carried away) but when I do it's usually for something in the 4500 to 7500 hertz range (a harsh frequency). Narrow band (high Q) sweep boosting 6 or more decibels. Find something awful, then cut it there by a few decibels.

Right, this is the basic technique for using parametric EQ to fix problems - what I meant by "figure out what it is", I should have been more detailed there. Very effective technique, tho with the EQ tools I have, I can usually see the problem frequencies visually as well, so this is just a 3rd step to confirm - 1, hear a problem, 2, visually see where it is, 3, use the sweeping technique to confirm that boosting accentuates the problem, and cutting fixes it. With the dynamic EQ, you can adjust so that it only triggers when the problem occurs, and doesn't add a fixed EQ to the entire track.

[Vindellama]

Originally Posted by Doug Young

I'm not sure what the issue is - you didn't do anything, like load the file into a DAW? I thought it was mono when I first listened.



I don't know the tune, and that's why I wasn't sure it's what you wanted or not. Its a bit fatiguing to listen to - the song comes across as just pounding on that 4th string on the upbeat, but maybe that's the intent. It sounds like you're playing really hard, especially with the thumb. If you want that, or that's how the song goes, that's fine. If not, I'd work on playing the bass notes more lightly. You might get some small differences from mic placement, but I don't think it will radically change how this aspect sounds. A big part of playing fingerstyle is figuring out how to balance the sound and controlling which notes stand out.

I didn't... Is as it came out of the sd card.
Also... Is there any difference between setting -1 -3 or 0 db as the final "peak level" when using a limiter and maybe normalizing after?

I think the artist intended to be this way. Looking at this live performance he hits the 4th string even harder than me (way harder):


So far thx for all the info and help from everyone. I'll update the post when I decide to give another tune a try (experimenting with the placement and split mono to see if the stereo image stays the same)

[rick-slo]

Originally Posted by Vindellama

I think the artist intended to be this way. Looking at this live performance he hits the 4th string even harder than me (way harder):

The youtube is fine. He's hitting the thumb notes hard but also the hitting the melody notes hard so that you can hear them clearly. In your recording (for whatever the different reasons) they can barely be heard.

[Vindellama]

Originally Posted by rick-slo

The youtube is fine. He's hitting the thumb notes hard but also the hitting the melody notes hard so that you can hear them clearly. In your recording (for whatever the different reasons) they can barely be heard.

Probably because I had short nails when recording (and the struggle keeping the same volume with a thumbpick).

I'll pay more attention to it the next time.
I don't think I can go back to not using the thumbpick with all the joint problems.
So I'll try either longer finger nails to catch more on the strings or going back to using fingerpicks.

[Doug Young]

Originally Posted by Vindellama

Also... Is there any difference between setting -1 -3 or 0 db as the final "peak level" when using a limiter and maybe normalizing after?

These are very different things. A limiter sets a level above which it will simply cut the volume, so if you have a spike of loudness, it will clamp down at that point and turn down the volume, then turn it back up once the level goes below that threshold. Normalizing scans your file, finds the loudest peak, and then adds or subtracts gain from the *entire* track to make the peak meet the peak value you specify. Limiting reduces the overall dynamic range, normalizing just raises or lowers the entire volume.

Or maybe I misunderstood - you're suggesting doing both? Limiting to say -3 db then later raising it to -1 db? Whether that's different than setting the limiter to -1 would depend on your recording and whether it exceeds the threshold or not. If you have a limiter set to -3db and nothing in your recording exceeds -3 db, then the limiter would have no effect. If your material already hits 0db, then a -3db limiter setting would add 3db of compression, while a -1db setting would add 1db of compression.

[Brent Hahn]

I took a quick swipe at it. Tried to help even out the thumb and fingers a bit, dialed in some delay and reverb. No noise treatment -- I don't have that. Left in a little more of the low mids than Doug did.

The file is downloadable.

[Vindellama]

Originally Posted by Doug Young

Or maybe I misunderstood - you're suggesting doing both? Limiting to say -3 db then later raising it to -1 db? Whether that's different than setting the limiter to -1 would depend on your recording and whether it exceeds the threshold or not. If you have a limiter set to -3db and nothing in your recording exceeds -3 db, then the limiter would have no effect. If your material already hits 0db, then a -3db limiter setting would add 3db of compression, while a -1db setting would add 1db of compression.

I didn't speak properly.
Like setting the limiter to -3 and later normalizing to -1.
Is there an ideal value for both?
I remember reading somewhere that for different plataforms like cd vs streaming the ideal maximum peak/db and overall perceived volume are different.

[Doug Young]

Oh, I see, you just are wondering how loud your track should be. That's tricky, and I'm not an expert on this, Joe Hanna might give a better answer. But in general, other than not exceeding 0db (which is impossible, but you'd get clipping), the issue is more about perceived volume, not peak, which is what you are talking about with normalization. You could have a track that has a loud squeak, normalize that to 0db, and still have the rest of the track be very quiet.

Average levels, which is more like what we hear, as best I can tell, are more about trying to fit in to some convention. Like on TV, there are strict loudness rules these days, so that commercials don't end up being twice as loud as the program (or at least they shouldn't be). In music, it's a bit more varied, look up "loudness wars", to see a lot about the issue - everyone wants their record to be the loudest. In some ways, it's always a bit subjective, you wouldn't expect a Segovia CD to be the same volume as Metallica. On You Tube, it's going to be totally the wild west because so much of the content is amateur-created, so I don't think you can predict how you'll fit into other videos as far as level, unless You Tube does something to try to even videos out.

For the mix I did on yours, I used the TC Electronic LM2n metering system, and targeted the CD master recommendation of -15 LUFS in that tool. RX says the result has a peak of. -.41db and total RMS value of -12db or -13 LUFS (Looks like I overshot my goal of -15 a bit). This is a complex subject that I don't claim to fully understand, and that involves a lot longer discussion. Look at:

https://www.tcelectronic.com/brand/t...ness-explained

you might also look up Bob Katz's K-System for some more background.