It's not that un-linking shifts the low frequency energy, it just inevitably generates more difference signal than fully linked. ie: the transient link, by definition, is HF.
I tested my limiters and the A.O.M Invisible Limiter 2 proved to be the best for this task It almost did not affect the lower range I did in mono (about 150 hertz and below). Even with fully unlinked channels its impact on this area was much less than in other limiters. The other limiters I tested were: Sonible Smart Limit, Ozone 11 Maximizer, TDR Limiter 6.
The multiband limiter with the most settings for things like this currently is Limitless by DMG Audio, but here’s hoping this is a feature coming up - though I’d never use it. Lol
Completely agree. Hopefully this could be implemented in Pro-L3 if it ever manifests itself. Pro Tip: The absolute best plugin for making your own 100% phase coherent crossovers with full control over steepness and type is actually FabFilter Pro-MB. Simply set up as many bands as you want, then copy the plugin instances to each channel and hit 'solo' for each respective band. Boom, instant perfect crossovers with 3 different phase modes (though minimum and dynamic phase modes will be sort of similar but not quite identical). Anyhow, personally I hope Pro-L3 will be a multiband capable limiter, basically the band splitting taken from Pro-MB. That would be dope!
What about just using the normal mastering chain but put the low cut of the side signal afterwards and add the gain that got missing from the cut or add an additional limiter with linked channels. This way the main limiter can be unlinked and the rest is taking care of the sub area.
Am I correct in saying Pro Tools still has the limitation of sidechain inputs being mono only? I can see even the limiter you have unlinked is applying equal gain reduction on both sides because it’s reacting to a mono signal. This idea, among many other workflows, would work perfectly if avid would finally address this limitation that no other modern daw has.
Now that you point it out, damn. That's actually pretty bitter for an industry standard DAW. There's a workaround, however it includes using a third-party plugin, namely BlueCat Patchwork.
there's no point in having a stereo key input. you can either use the info from one of the L-R channels or if the info is totally different, you can just sum them into one. you won't hear it anyway, it's only going to be read by the plugin.
I think to truly set this up in Pro Tools you'd have to set up two mono auxes for the high band, so that the L and R sidechains could be assigned independently. Otherwise both mid and high band are effectively stereo linked because they're reacting to a mono sidechain send.
Only a few days ago while doing my restoration on a vintage analog console, i was going over the signal path around the buses. Noticing that in addition to the Stereo bus (and the other 16 mix buses) there is a 1 and 2 mono bus, i had thought there must be some specific purpose for including this functionality...
Someone has to say it... may as well be me! 😁 Unlinking the left/right does more damage to your vocal's phantom image than to the bass... so presuming only the low end is linked, it will heighten your awareness to the vocals shifting. If you want the L/R to be processed separately without messing with the phantom image try setting the linking to Trans 0 and Rel 100.
@@ZCBeats1 i think that was just to demo low-end stereo field changing (he uses brickwall in side channel to make low-end mono before the limiter) due to the unlinking.
No-one "deserves" to have their job taken over by AI. It's a high ask when you set the average human worker against silicon Valley tech bros/gals with phd's in stem and billions of dollars behind them.
man, that looked complicated. in bitwig you'd just put a multiband container into the effect chain and then put individual effects on those bands. now the reason why this is not a feature in limiters is probably because the main reason for using limiters is typically to limit the signal at the peaks, but with a multiband context it would really obscure the meaning of limiting. nevertheless i think your idea is pretty cool and I hope it will be considered by some developer who's good at making mastering limiters
Surely there must be a way to re-center the low-end "drift" by duplicating a few certain channels with some hard L/R panning, reversing the polarity and then summing them back together?
i see a problem: in Pro Tools external sidechains can only be MONO busses (you can even see in the video that gain reduction was symmetrical in the unlinked pro l)
Althoug its not a proper limiter, something like that can be made in ProMB. Ratio 100, band split where you want and advanced tab gives stereo linking options. Whats cool to me about this way also has a knee,so i utilize "Clipper Like" behavior to lean into the limiting. Im curious if u try this experiment what pros and cons u will come across to. Also after proMB i still have final TP limiter to catch those rough peaks,and sometimes push an extra db or two.
Could you create a video about release and attack times in parallel compression? For example 300-400 ms as default in DynOne sounds exotic, but it somehow works to blend and smear nicely! Then If I extend the release time from zero to 400ms then the squeezing effect disappears and the attack transient cuts through. This is a bit counterintuitive to me. A few approaches how you use DynOne would be appreciated. Btw. I'm on your course in EDMProd, gaining the precious knowledge from you :)
What’s the benifit of unlinking the limiting in the high end anyway? So far as I can tell your just choosing to get more loudness at the expense of messing with the stereo field in an unnatural way right?
Some tunes might need the stereo field unlinked if there's a lot of diferant things going on on each side up top. It can really narrow the stereo field of a very widely arranged and panned piece. If you're gonna be digging in hard, then it can shift the stereo field in weird ways, but if you're only touching it lightly, it often doesn't, and it keeps that open top end feeling. It's kinda program dependant though, So I always mess about and see what works best, but quite often I find linking some things just chokes them off and makes them sound way narrower than they did before as even if a loud thing is playing on just the right, the left goes quieter too, and you loose some of that speration. It wont sound bad, it'll just loose a sense of space in certian senarios. You kinda have to play it by ear as some things totally can sound better in linked, especially very evenly distrabuted arrangments with almost the same thing happeneing on both sides but that already has a decent sense of space that doesn't need to breath independantly of each other (like double tracked guitars etc), but being able to not mess up your low end's positioning in the process for them times when you do want to unlink the top end is a HUGE bonus to be able to have.
You can gain a slight stereo width(which may be good in some cases) with by unlinking as well as slightly more loudness. Does this technique here do anything…..no it doesn’t.
@@DaftyBoi412 thank you for the detailed response mate I appreciate it 🙏 personally I’m rarely working on anything that has any mad panning of different elements so that scenario never crossed my mind!
Nice workaround Never used it as limiter but DMG Multiplicity allows per band link/unlink This being said i had similar stereo image issues with "low end focus" from Izotope (not shifting but altering the width because how it creates contrast between loud and quiet elements) and ultimately using it in M/S was the key. For stereo consistency i always kept my channels linked but will try out extensively M/S limiting with channels unlinked in the coming days. This being said i might also just update my mastering template with a phase coherent split (curious to listen how this could also affect the tone of a clipper)
Hi, great idea ! Thanks. I tried it with two instances of weiss ds1-MK3 same filter on both, one with highpsss, the other with lowpass.... the one for the lows is ganged, the high not. It's quite good, I guess... ?
Great concept! ... Studio one has a splitter built in every channel, even the master... How about Waves studio Rack .... You can split the frecuencies too...
That's really pukka. It was kinda what I was thinking when I said multiband limiting (not something I useually do, so am unfamiliar with the plethora of diferant options a plug in of that nature would have), but I was assuming you'd be able to set each band's stereo linking in my "imaginary" multiband limiter. It sounds like that's not a thing, and you basically just made what I assumed the plug ins would already have! Really cool idea!
I thought I'd add, The day after your last vid I was thinking and I worked out the issues with M/S too, it's basically gonna act similar to stereo linked for all the side info when you do that, because the diferance is one channel, and that is everything that makes the left and the right unique, all being compressed together with that one side chain signal. It's not identicle to stereo linking (as all the mid info is not there, nor in the S side chain), but it's way closer to it than unlinked, and that almost defeats the point of what we were trying to achieve.
@@PeterPan-lu7cw yeah, sure. But you can then sabe a metaplugin present and be done with it :) You can even have one in which you can switch compressors :D
@@DirkParker well it's not just about splitting the frequency, the limiters must also have the stereo linking feature. I don't know if waves ones do, and you are limited to waves plugins.
@@DirkParker It's also about how you split the freqencies and put them back together. Conventually it's done with linear phase filters, but this can introduce preringing (which can be especially noticable when altering the low end). I'm assuming Nick is phase rotating after because he's using normal filters, and I'm guessing this is to avoid pre-ringing (I could be mistaken here, but that's what I took from it).
Really bizarre take at the end there, although I assume you’re being mostly facetious. As if a properly trained AI mastering model in a year or two would have any problem fixing a problem this simple.
It's not that un-linking shifts the low frequency energy, it just inevitably generates more difference signal than fully linked. ie: the transient link, by definition, is HF.
Good point.
this is what SideminderME2 does. it's basically a correlation limiter if you set the advanced tab and ratio to 100:1
I tested my limiters and the A.O.M Invisible Limiter 2 proved to be the best for this task
It almost did not affect the lower range I did in mono (about 150 hertz and below). Even with fully unlinked channels its impact on this area was much less than in other limiters.
The other limiters I tested were: Sonible Smart Limit, Ozone 11 Maximizer, TDR Limiter 6.
The multiband limiter with the most settings for things like this currently is Limitless by DMG Audio, but here’s hoping this is a feature coming up - though I’d never use it. Lol
Love Limitless
We need a Lorenzo Limiter
I second this!
Completely agree. Hopefully this could be implemented in Pro-L3 if it ever manifests itself. Pro Tip: The absolute best plugin for making your own 100% phase coherent crossovers with full control over steepness and type is actually FabFilter Pro-MB. Simply set up as many bands as you want, then copy the plugin instances to each channel and hit 'solo' for each respective band. Boom, instant perfect crossovers with 3 different phase modes (though minimum and dynamic phase modes will be sort of similar but not quite identical).
Anyhow, personally I hope Pro-L3 will be a multiband capable limiter, basically the band splitting taken from Pro-MB. That would be dope!
What about just using the normal mastering chain but put the low cut of the side signal afterwards and add the gain that got missing from the cut or add an additional limiter with linked channels. This way the main limiter can be unlinked and the rest is taking care of the sub area.
Am I correct in saying Pro Tools still has the limitation of sidechain inputs being mono only? I can see even the limiter you have unlinked is applying equal gain reduction on both sides because it’s reacting to a mono signal. This idea, among many other workflows, would work perfectly if avid would finally address this limitation that no other modern daw has.
Now that you point it out, damn. That's actually pretty bitter for an industry standard DAW. There's a workaround, however it includes using a third-party plugin, namely BlueCat Patchwork.
I was coming to the comments to post something like this.
+1, if sidechain is mono, there's no point to make this turnaround, cool but useless.. :(
there's no point in having a stereo key input. you can either use the info from one of the L-R channels or if the info is totally different, you can just sum them into one. you won't hear it anyway, it's only going to be read by the plugin.
I think to truly set this up in Pro Tools you'd have to set up two mono auxes for the high band, so that the L and R sidechains could be assigned independently. Otherwise both mid and high band are effectively stereo linked because they're reacting to a mono sidechain send.
Only a few days ago while doing my restoration on a vintage analog console, i was going over the signal path around the buses. Noticing that in addition to the Stereo bus (and the other 16 mix buses) there is a 1 and 2 mono bus, i had thought there must be some specific purpose for including this functionality...
I love the concept of this i am gonna try this tonight!
Will be easy to do on REAPER
Mad respect for editing the shout out! I'm glad you were able to find a unique solution to this!
Could you not Do the same, by using Waves StudioRack with Multiband Mode and put L3 or Ozone Miximizer on both Chanels?
Someone has to say it... may as well be me! 😁 Unlinking the left/right does more damage to your vocal's phantom image than to the bass... so presuming only the low end is linked, it will heighten your awareness to the vocals shifting. If you want the L/R to be processed separately without messing with the phantom image try setting the linking to Trans 0 and Rel 100.
What does the phase look like
Exactly what I was thinking I saw the brickwall filter and went 😧
@@ZCBeats1 i think that was just to demo low-end stereo field changing (he uses brickwall in side channel to make low-end mono before the limiter) due to the unlinking.
No-one "deserves" to have their job taken over by AI. It's a high ask when you set the average human worker against silicon Valley tech bros/gals with phd's in stem and billions of dollars behind them.
man, that looked complicated. in bitwig you'd just put a multiband container into the effect chain and then put individual effects on those bands.
now the reason why this is not a feature in limiters is probably because the main reason for using limiters is typically to limit the signal at the peaks, but with a multiband context it would really obscure the meaning of limiting.
nevertheless i think your idea is pretty cool and I hope it will be considered by some developer who's good at making mastering limiters
You can probably do that with Melda MXXX. Or in Reaper :D.
Reaper has a stock limiter where in the title it says stereo unlinked
Surely there must be a way to re-center the low-end "drift" by duplicating a few certain channels with some hard L/R panning, reversing the polarity and then summing them back together?
i see a problem: in Pro Tools external sidechains can only be MONO busses (you can even see in the video that gain reduction was symmetrical in the unlinked pro l)
Althoug its not a proper limiter, something like that can be made in ProMB.
Ratio 100, band split where you want and advanced tab gives stereo linking options.
Whats cool to me about this way also has a knee,so i utilize "Clipper Like" behavior to lean into the limiting.
Im curious if u try this experiment what pros and cons u will come across to.
Also after proMB i still have final TP limiter to catch those rough peaks,and sometimes push an extra db or two.
Won’t this introduce some serious phase issues especially using a brickwall filter?
I think thers a stock limiter in reaper ther was two of the same one said stereo unlinked
Could you create a video about release and attack times in parallel compression? For example 300-400 ms as default in DynOne sounds exotic, but it somehow works to blend and smear nicely!
Then If I extend the release time from zero to 400ms then the squeezing effect disappears and the attack transient cuts through. This is a bit counterintuitive to me. A few approaches how you use DynOne would be appreciated. Btw. I'm on your course in EDMProd, gaining the precious knowledge from you :)
I just unlink them and call it a day. if u can't tell the difference they wont either but... good to know...
I’d use pro L if it wasn’t sounding that bad. Always get a choked feeling out of it. Never liked how it muffles the snare
What’s the benifit of unlinking the limiting in the high end anyway?
So far as I can tell your just choosing to get more loudness at the expense of messing with the stereo field in an unnatural way right?
My initial instinct is to just always have the limiting be channel linked and not ever have to worry about stereo field issues.
Some tunes might need the stereo field unlinked if there's a lot of diferant things going on on each side up top. It can really narrow the stereo field of a very widely arranged and panned piece. If you're gonna be digging in hard, then it can shift the stereo field in weird ways, but if you're only touching it lightly, it often doesn't, and it keeps that open top end feeling. It's kinda program dependant though, So I always mess about and see what works best, but quite often I find linking some things just chokes them off and makes them sound way narrower than they did before as even if a loud thing is playing on just the right, the left goes quieter too, and you loose some of that speration. It wont sound bad, it'll just loose a sense of space in certian senarios.
You kinda have to play it by ear as some things totally can sound better in linked, especially very evenly distrabuted arrangments with almost the same thing happeneing on both sides but that already has a decent sense of space that doesn't need to breath independantly of each other (like double tracked guitars etc), but being able to not mess up your low end's positioning in the process for them times when you do want to unlink the top end is a HUGE bonus to be able to have.
You can gain a slight stereo width(which may be good in some cases) with by unlinking as well as slightly more loudness. Does this technique here do anything…..no it doesn’t.
@@Limit5482 This vid is adressing the wandering low end in the stereo field issue when using unlinked stereo limiters talked about in a past video.
@@DaftyBoi412 thank you for the detailed response mate I appreciate it 🙏 personally I’m rarely working on anything that has any mad panning of different elements so that scenario never crossed my mind!
Nice workaround
Never used it as limiter but DMG Multiplicity allows per band link/unlink
This being said i had similar stereo image issues with "low end focus" from Izotope (not shifting but altering the width because how it creates contrast between loud and quiet elements) and ultimately using it in M/S was the key.
For stereo consistency i always kept my channels linked but will try out extensively M/S limiting with channels unlinked in the coming days.
This being said i might also just update my mastering template with a phase coherent split (curious to listen how this could also affect the tone of a clipper)
Limitless
Hi, great idea ! Thanks. I tried it with two instances of weiss ds1-MK3 same filter on both, one with highpsss, the other with lowpass.... the one for the lows is ganged, the high not. It's quite good, I guess... ?
Great concept! ... Studio one has a splitter built in every channel, even the master... How about Waves studio Rack .... You can split the frecuencies too...
That's really pukka. It was kinda what I was thinking when I said multiband limiting (not something I useually do, so am unfamiliar with the plethora of diferant options a plug in of that nature would have), but I was assuming you'd be able to set each band's stereo linking in my "imaginary" multiband limiter. It sounds like that's not a thing, and you basically just made what I assumed the plug ins would already have! Really cool idea!
I thought I'd add, The day after your last vid I was thinking and I worked out the issues with M/S too, it's basically gonna act similar to stereo linked for all the side info when you do that, because the diferance is one channel, and that is everything that makes the left and the right unique, all being compressed together with that one side chain signal. It's not identicle to stereo linking (as all the mid info is not there, nor in the S side chain), but it's way closer to it than unlinked, and that almost defeats the point of what we were trying to achieve.
Love those nerdy videos :)
Does it make a difference when summed in mono ?
No, since the drifting of the low frequency should maintain its phase coherence, which means no difference when summed in mono.
@@bakerlefdaoui6801 That's what I though, tho L2 is also clipping but saturation is phase coherent too.
You can’t link plugins in studio one. Lame
bx true limiter peak has channel link
Wonder when that limiter will be released :)
Sorry for starting all this. But after the last video I was thinking the same of setting this up as no limiter has this option (yet). Thanks!
Pro-L…..3? 🤔
Thats f**ing genius. Damn 🤯
Can't this be done in Metaplugin? To avoid all those channels and rounting and just have an insert.
Yes, but you still need two limiters to do different linking settings.
@@PeterPan-lu7cw yeah, sure. But you can then sabe a metaplugin present and be done with it :) You can even have one in which you can switch compressors :D
The waves plugin rack can do this too. Idk what it’s called but you can split by freq and put two limiters. Easy.
@@DirkParker well it's not just about splitting the frequency, the limiters must also have the stereo linking feature. I don't know if waves ones do, and you are limited to waves plugins.
@@DirkParker It's also about how you split the freqencies and put them back together. Conventually it's done with linear phase filters, but this can introduce preringing (which can be especially noticable when altering the low end). I'm assuming Nick is phase rotating after because he's using normal filters, and I'm guessing this is to avoid pre-ringing (I could be mistaken here, but that's what I took from it).
Waves Vitamin
U got a mistake on your title
Than might be one
@@Durkhead it might be two
Think outside the box - OR be replaced by a robot. LOL. Nice one.
Really bizarre take at the end there, although I assume you’re being mostly facetious. As if a properly trained AI mastering model in a year or two would have any problem fixing a problem this simple.
If AI can take your job, it should, is an awesome take