This is like a birthday gift to me. Only the intermediary and advance mixers can understand this explanation because it defiles the law of attack and release in a traditional compressor. But the question is, are all attack and release time in every other limiter working as that of the pro-L2?
Very interesting! I'm a novice, but I've purchased and used it, using presets because I've no idea what I'm doing other than pushing the signal as high as possible before it sounds like crap. But I've understood your explanation perfectly and I'm glad I didn't pick up any false understanding first!
Amazing, just yesterday i was questioning attack time on this limiter. I thought, if its not set at the fastest attack, how is it going to limit??? Then this randomly comes across my youtube algorithm. Thanks YT, and panaroma
Haha... I have a way of putting things; but I love and respect Dan Worrall too much; I'll leave the technical side of things to those who respect it best ;) Not some boomerang weilding aussie who ride their kangaroo to work ;)
@@panorama_mastering hey man... we don't all think in anti-aliasing-oversampling-harmonic-distortion-%£*^%@@... The black & white answer is better sometimes. Hell; I'd take the Kangaroo!
I love it. Great explanation, with a proper test to back up your findings. The article is fantastic as well. Will this change the way you dial in your L2? iirc, your method is to crank the gain to maximum, dial in your advanced settings, then return the gain to your intended volume. If I'm interpreting these findings correctly, adjusting the attack knob with the gain set louder than you intend the final master to be will allow more (longer) transients within the attack value you set. Then, when you reduce the gain to the volume you want your master to be, you'll be left with a gap: the difference in transient length between the peaks at your final volume and the peaks at the volume at which you dialed in the settings.
Thanks mate! And absolutely! The way I was setting my limiter previously was so far flawed it's not funny; With this consideration I'm going to spend some time with the knolwedge and consider how I can re-impliment my limiting practice with unit then come back to TH-cam with my findings!
@@panorama_mastering Two things immediately come to mind: Using two L2's in series, the first with "flatline' settings, the second with attack at zero, release to taste. That way you have a designated "pure" clipper followed by a designated "pure" limiter. That and/or getting really good at understanding what you're hearing when you flip L2 into "delta" mode (the little headphone icon in the bottom right menu)
Dang! That's crazy! I knew the attack and release didn't work like any thing else I'd seen but I never understood why. I was always confused why the attack knob would change the release time, and I learned to set it in such a way that it wouldn't have release time on transients. I think that ended up being almost the right thing lol, But now I can get it truly right! Thank you sir!
Damn you sir haha. I worked this out early last year, sort of by accident, but kept it quiet. I think you may be surprised how many engineers do actually know how it works but they keep it to themselves. I had numerous conversations last year about this with colleagues and they immediately go all "hush hush". It's all good though, it is better the information is out there for all to see. Now perhaps you should get in to how the multi stage attack and release section works on the Weiss DS1 MK3. Now that is fun. Blew my mind when I worked it out. Cheers for the video bro.
Haha ! Wonderful! No secrets in my tool box; and I'm happy to say I was completely wrong about it in the past :) There are some tabboo' "secrets" which I know and you may too; which I've been tempted to post here... but maybe another time buddy;
@@panorama_mastering I remember 12 years or so seeing a Full Sail or Pensado interview with Mike Dean where he gave some vague ass, snake oil answer to a question about loudness and mastering. As recently as 3 months ago I'm now running around with a PhD (literally), still Mastering more or less full time and he's still up to the same shit and hiding his chain. I'm not kicking around your channel because there's some hidden cheat code giving you your edge, I'm here cuz you put in more work and actively demonstrate more commitment, continual improvement and passion/excellence of craft than anyone with the exception of Dan Worrall who I'd consider one of your very few peers. Streaky's spilling the beans daily and I think the whole culture is better because of it. Because of the two of you, I've recently been smashing the ever living shit out of my mixes and guess what: They sound that last couple of percent better than they did when I was still aiming for -12LUFS. Much like Dan, I imagine you can pull a beautifully conservative, dynamic mix if you so choose but if we're working on stuff that knocks, turns out you're right: It ought to knock!! Looking forward to trying your parallel compression trick from a few weeks ago and thanks heaps from WA for all your efforts. I owe you an entire case of whatever you're drinking next time I'm over east. :)
the attack time is it's a time-based that triggers the release to change from the quick release to the slow (or whatever setting you have) release to kick in. right?
Would it not help in effect to use the headphone feature and 1:1 to hear what the limiter is doing at the peak level to see how harsh the clipping/limiting is sonically (in that what you can hear)? The limiter has a similar feature to Kazrog's KClip 3 delta feature where it isolates the clipping only so you can hear what the clipping sounds like/how much is being clipped
I feel like folks from TDR figured out to clear this mess by having peak and rms visualy in front of you. This on Fabfilter... Was a bit of a confusion but it's nice to know that now 😁
@@panorama_mastering I'm suprised with quality of their stuff for the price... Not that im gonna complain. But they really made some classics at least in my opinion. I warmly recommend the to anyone who is starting from a blank slate as first "go to" mixing and mastering plugin set 😁😊
I had no idea of what the Attack and Release dials were doing, though, I was setting them in a way which was not far off what I would do now knowing what they are doing. I was setting the Attack around 650ms (12 O'clock) and release around 50ms (9 O'clock). But now knowing what is going on I know what to listen for when setting both dials. Thanks for the video Now what does the Channel Linking stage knobs do? Transient and Release knobs? I don't really understand their use, all I know is they relate to how the L/R react to transients, but is it in conjunction to the Attack and Release dials?
Super interesting, but your last experiment at about 6:47 muddies the water. At 6:54 you say "It's not quite slow enough". Why? If your previous description of how it works was rock solid, it most certainly would be. My guess (and it can only be a guess since FF isn't saying and I don't have the ability to test better than you have) is that there's a ratio involved. Like maybe the attack time function works how you describe, but it's not a 1 to 1 length of time being measured before the signal is treated as sustained.
You still have it quite wrong (sorry Jonathan Jetter if you happen to read this) but you are at least on the right track if in a quite significantly simplified sort of way.. kind of how somebody would describe how a car works by saying that it goes forward when you press the throttle pedal. The good thing is that you don't need to know much more than that but suffice to say that more is happening under the hood. Here is a slightly more complex explanation of what the attack knob does: You can think of the attack time as being the "threshold" between full on clipping (=instant release) and many stages of program dependent limiting. The only exception to this simplification is the 'Dynamic' algorithm that adds an additional "transient shaper" like system under the hood where the attack knob controls a bunch of extra parameters that exaggerate the transients vs the sustain portion of a sound before hitting the limiter + clipping portion of the plugin. There is a reason why I created so many presets for Pro-L2, they specifically explore the usable extremes of the controls. I always urge people to go through them and explore the full ranges of every single knob in Pro-L2 and do so on a multitude of different audio material. For instance the look-a-head knob is not at all a traditional look-a-head as one would think, especially not in the 'Modern' algorithm. It literally affects the attack, internal hold AND the release knobs of the plugin as well. Basically you can force Pro-L2 in the 'Modern' algorithm to not hit the clipping stage at all if you so desire.. or do the exact opposite and clip almost everything. It's worth mentioning that _literally all the knobs in the Pro-L and Pro-L2 plugins are what you could call macro controls_. Under the hood they affect up to several dozen of internal parameters that then in turn affect even more parameters dynamically, all in constant motion. The only controls that are completely self explanatory and exactly what they say on the tin are: Threshold and output level. All other knobs are highly program dependent (yes even the stereo link controls). It's a deceptively complex plugin under the hood and because it has so few controls it is highly recommended to explore the full range of motion for ALL controls and test all the different algorithms. It's literally like having 8 different limiters in one plugin. The algorithms completely change the internal structure and aren't just "small variations" on a theme as some people seem to assume.
Great video and I had no idea even though I use L2 on every mix. I just fiddle with attack/release until it sounds right but didn’t really understand what I was doing. What would be useful for a follow up video would be to test how this actually affects the mix in practice. I genuinely just turn the knobs until it sounds “better” but will now see if I can marry up what I’m hearing with how (I now know) the L2 works.
While I understand now how it functions, I’m not sure I can apply that practically without just using my ears as a guide anyway. While that’s always the best idea anyway, what would you say is the practical purpose and use case for this feature?
Typically how I'm using hard clippers pre-limiting is to help great a consistent peak level to feed the release stage; But I do this with a seperate plugin all together where I can control the gain staging seperate to the limiter; This is more-so just for the knowledge of what you do and how it interacts with the sound when you do it;
I heard a lot by the past that modern digital limiters were actually half-clippers, half-compressors. I finally understood, thank you !!
This is nuts. This channel is such a gem
Thanks mate!!
This is like a birthday gift to me. Only the intermediary and advance mixers can understand this explanation because it defiles the law of attack and release in a traditional compressor. But the question is, are all attack and release time in every other limiter working as that of the pro-L2?
Wonderful vidddddddd🎉
Glad you liked it!! Thanks for watching!
Very interesting! I'm a novice, but I've purchased and used it, using presets because I've no idea what I'm doing other than pushing the signal as high as possible before it sounds like crap. But I've understood your explanation perfectly and I'm glad I didn't pick up any false understanding first!
Glad it was helpful! Welcome aboard!!!
Amazing, just yesterday i was questioning attack time on this limiter. I thought, if its not set at the fastest attack, how is it going to limit???
Then this randomly comes across my youtube algorithm. Thanks YT, and panaroma
“Come in deal with the peak and fuck off…” is the best description ever 😂 Fabfilter should have you on board as the anti-Dan Worrall!
Haha... I have a way of putting things; but I love and respect Dan Worrall too much; I'll leave the technical side of things to those who respect it best ;) Not some boomerang weilding aussie who ride their kangaroo to work ;)
@@panorama_mastering hey man... we don't all think in anti-aliasing-oversampling-harmonic-distortion-%£*^%@@... The black & white answer is better sometimes. Hell; I'd take the Kangaroo!
I love it. Great explanation, with a proper test to back up your findings. The article is fantastic as well.
Will this change the way you dial in your L2? iirc, your method is to crank the gain to maximum, dial in your advanced settings, then return the gain to your intended volume.
If I'm interpreting these findings correctly, adjusting the attack knob with the gain set louder than you intend the final master to be will allow more (longer) transients within the attack value you set. Then, when you reduce the gain to the volume you want your master to be, you'll be left with a gap: the difference in transient length between the peaks at your final volume and the peaks at the volume at which you dialed in the settings.
Thanks mate! And absolutely! The way I was setting my limiter previously was so far flawed it's not funny;
With this consideration I'm going to spend some time with the knolwedge and consider how I can re-impliment my limiting practice with unit then come back to TH-cam with my findings!
@@panorama_mastering Two things immediately come to mind: Using two L2's in series, the first with "flatline' settings, the second with attack at zero, release to taste. That way you have a designated "pure" clipper followed by a designated "pure" limiter.
That and/or getting really good at understanding what you're hearing when you flip L2 into "delta" mode (the little headphone icon in the bottom right menu)
Yeap; I think there's a myriad of ways this will unleash to use this limiter! So damn happy I was sent this article!
Here is your good 👍. I already knew this, but you are the first youtuber I see spreading the right info on attack and release for the Pro L 🎉
Thanks mate!
Dang! That's crazy! I knew the attack and release didn't work like any thing else I'd seen but I never understood why. I was always confused why the attack knob would change the release time, and I learned to set it in such a way that it wouldn't have release time on transients. I think that ended up being almost the right thing lol, But now I can get it truly right!
Thank you sir!
I know! My reaction was the same when I found out!!!!
so what kind of gain reduction "shape" are we aiming for? (the red)?
Damn you sir haha. I worked this out early last year, sort of by accident, but kept it quiet. I think you may be surprised how many engineers do actually know how it works but they keep it to themselves. I had numerous conversations last year about this with colleagues and they immediately go all "hush hush". It's all good though, it is better the information is out there for all to see. Now perhaps you should get in to how the multi stage attack and release section works on the Weiss DS1 MK3. Now that is fun. Blew my mind when I worked it out. Cheers for the video bro.
Haha ! Wonderful! No secrets in my tool box; and I'm happy to say I was completely wrong about it in the past :)
There are some tabboo' "secrets" which I know and you may too; which I've been tempted to post here... but maybe another time buddy;
@@panorama_mastering I remember 12 years or so seeing a Full Sail or Pensado interview with Mike Dean where he gave some vague ass, snake oil answer to a question about loudness and mastering. As recently as 3 months ago I'm now running around with a PhD (literally), still Mastering more or less full time and he's still up to the same shit and hiding his chain.
I'm not kicking around your channel because there's some hidden cheat code giving you your edge, I'm here cuz you put in more work and actively demonstrate more commitment, continual improvement and passion/excellence of craft than anyone with the exception of Dan Worrall who I'd consider one of your very few peers.
Streaky's spilling the beans daily and I think the whole culture is better because of it. Because of the two of you, I've recently been smashing the ever living shit out of my mixes and guess what: They sound that last couple of percent better than they did when I was still aiming for -12LUFS. Much like Dan, I imagine you can pull a beautifully conservative, dynamic mix if you so choose but if we're working on stuff that knocks, turns out you're right: It ought to knock!!
Looking forward to trying your parallel compression trick from a few weeks ago and thanks heaps from WA for all your efforts. I owe you an entire case of whatever you're drinking next time I'm over east. :)
the attack time is it's a time-based that triggers the release to change from the quick release to the slow (or whatever setting you have) release to kick in. right?
Would it not help in effect to use the headphone feature and 1:1 to hear what the limiter is doing at the peak level to see how harsh the clipping/limiting is sonically (in that what you can hear)? The limiter has a similar feature to Kazrog's KClip 3 delta feature where it isolates the clipping only so you can hear what the clipping sounds like/how much is being clipped
Great video. I've also been pretending ignorance is bliss when it comes to L2.
I know, right?
Gotta get that 0dB LUFS with perfect sound quality
Thanks for this.
I feel like folks from TDR figured out to clear this mess by having peak and rms visualy in front of you.
This on Fabfilter... Was a bit of a confusion but it's nice to know that now 😁
Bang on! Fabien is a genius (From TDR) if you check out my video on TP limiters he's one of the brains behind a lot of the research I pulled together;
@@panorama_mastering I'm suprised with quality of their stuff for the price... Not that im gonna complain.
But they really made some classics at least in my opinion. I warmly recommend the to anyone who is starting from a blank slate as first "go to" mixing and mastering plugin set 😁😊
so what should we do to get louder masters ? I didn't understand
The secret is the two stages to the attack and release; the attack being a hard-clipper stage
This was amazing! thank you
You're so welcome!
I get it, but I don't get it
EXACTLY! If this aspect of the plug-in was properly documented by FabFilter it would open up a whole new world to the way MANY approach using it;
So where should we set the attack time? Now I regret watching this, as I'm totally confused.
I had no idea of what the Attack and Release dials were doing, though, I was setting them in a way which was not far off what I would do now knowing what they are doing. I was setting the Attack around 650ms (12 O'clock) and release around 50ms (9 O'clock).
But now knowing what is going on I know what to listen for when setting both dials. Thanks for the video
Now what does the Channel Linking stage knobs do? Transient and Release knobs? I don't really understand their use, all I know is they relate to how the L/R react to transients, but is it in conjunction to the Attack and Release dials?
So many questions are coming out of this video across platforms; I think it might call for a completely in depth look at the plug-in for TH-cam!
free puma commercial 😛
Do you even puma?
thank you a lot
My pleasure! Enjoy!
Super interesting, but your last experiment at about 6:47 muddies the water. At 6:54 you say "It's not quite slow enough". Why? If your previous description of how it works was rock solid, it most certainly would be. My guess (and it can only be a guess since FF isn't saying and I don't have the ability to test better than you have) is that there's a ratio involved. Like maybe the attack time function works how you describe, but it's not a 1 to 1 length of time being measured before the signal is treated as sustained.
Definitely a good theory here!
Damn, this is already my favourite thing to use and I don't know how to use it.
It's wild hey! Thanks for watching!
You still have it quite wrong (sorry Jonathan Jetter if you happen to read this) but you are at least on the right track if in a quite significantly simplified sort of way.. kind of how somebody would describe how a car works by saying that it goes forward when you press the throttle pedal. The good thing is that you don't need to know much more than that but suffice to say that more is happening under the hood.
Here is a slightly more complex explanation of what the attack knob does: You can think of the attack time as being the "threshold" between full on clipping (=instant release) and many stages of program dependent limiting. The only exception to this simplification is the 'Dynamic' algorithm that adds an additional "transient shaper" like system under the hood where the attack knob controls a bunch of extra parameters that exaggerate the transients vs the sustain portion of a sound before hitting the limiter + clipping portion of the plugin.
There is a reason why I created so many presets for Pro-L2, they specifically explore the usable extremes of the controls. I always urge people to go through them and explore the full ranges of every single knob in Pro-L2 and do so on a multitude of different audio material. For instance the look-a-head knob is not at all a traditional look-a-head as one would think, especially not in the 'Modern' algorithm. It literally affects the attack, internal hold AND the release knobs of the plugin as well. Basically you can force Pro-L2 in the 'Modern' algorithm to not hit the clipping stage at all if you so desire.. or do the exact opposite and clip almost everything.
It's worth mentioning that _literally all the knobs in the Pro-L and Pro-L2 plugins are what you could call macro controls_. Under the hood they affect up to several dozen of internal parameters that then in turn affect even more parameters dynamically, all in constant motion. The only controls that are completely self explanatory and exactly what they say on the tin are: Threshold and output level. All other knobs are highly program dependent (yes even the stereo link controls). It's a deceptively complex plugin under the hood and because it has so few controls it is highly recommended to explore the full range of motion for ALL controls and test all the different algorithms. It's literally like having 8 different limiters in one plugin. The algorithms completely change the internal structure and aren't just "small variations" on a theme as some people seem to assume.
Wow; this is awesome; thank you Niklas for sharing; would you ever be interested in jumping on an interview with me about this?
I'm here cuz of Dee and Lou!
WELCOME! Thanks for joining the live stream yesterday!
Great video and I had no idea even though I use L2 on every mix. I just fiddle with attack/release until it sounds right but didn’t really understand what I was doing.
What would be useful for a follow up video would be to test how this actually affects the mix in practice. I genuinely just turn the knobs until it sounds “better” but will now see if I can marry up what I’m hearing with how (I now know) the L2 works.
Don't worry a very very comprehensive and thourough look through the L2 will be on it's way ;)
How about 0 attack and 0 release?
While I understand now how it functions, I’m not sure I can apply that practically without just using my ears as a guide anyway. While that’s always the best idea anyway, what would you say is the practical purpose and use case for this feature?
Typically how I'm using hard clippers pre-limiting is to help great a consistent peak level to feed the release stage; But I do this with a seperate plugin all together where I can control the gain staging seperate to the limiter;
This is more-so just for the knowledge of what you do and how it interacts with the sound when you do it;
How can this be a clipper when actual clippers aren’t time based?
You're right clippers aren't time based; but the ballistics in this limiter has a clipping stage;
So it’s not an „attack“ it’s more like a „window“
Correct a "clipping window"
they updated pro L, you can't replicate these results any more