Plug in delay compensation

[kushan_ku]

Let's give this topic it's own thread.

Charles, do you always compensate for plug in delay or do you go by your ears?

It hasn't really been a great issue with me but I'm wondering if it's secretly making my mixes sound less glued together....

[Charles Dye]

kushan_ku,

Roger Nichols told me Steely Dan were asking him to move hi-hats by samples (What!!!!?)

Let me just say this. Steely Dan I'm not.

As scandalous as what I'm about to say is—here goes. For the most part, I do not compensate for plug in delays. There! I've said it! Are you happy now? Seriously though, I don't usually compensate for plug-in delays for a few reasons:

• I'm not anal-retentive. (Not that those who do are, and "not that there's anything wrong with that.")
• I can't be bothered. (If I had to slide a track every single time I added or yanked a plug it would drive me nuts + make my mixes take waaayyy longer.)
• I don't think it matters. (Ever look at a live drum kit and notice that the Kick on the BD track and the OH track are many multiple milliseconds apart—not samples.)

There's simply not enough time in the day for me to spend creative time doing this. Does this make me a bad person?

The one place I always compensate for plug delays is when stereo phase is at issue. Though I don't often process the left from the right with different plug-ins, but it has happened. I also will usually slide a track back to its original position if I bounce it via a bus.

So, kushan_ku I guess I'm with you on this topic, and it hasn't been a great issue for me either.

I'm quite curious what wrath this heinous admission might bring. And when automatic plug-in delay arrives I'll no longer have to feel guilty about this, but will it screw up the sound of my mixes? :eek:

[C.Lambrechts]

hmmmm ..... can't help but frown my eyebrows here Charles. AD/DA latency issues ok. I sometimes compensate and sometimes not. Depends on the musician I guess. But with plugins .... especially the way you use them .... meaning loads and loads and loads ..... hmmmm .....

I can understand very well the time consuming factor ... THE main reason why we've been asking for automatic compensation for years now.

Just curious, have you ever tried this in on one of your mixes (and I do mean a loaded with tracks and plugins session where you drive that HD8 of yours to the max. Not a 24 track session where bus summing isn't an issue, but like a 50 + tracks session) :

Bounce the mix without compensating and do another bounce with compensating for them. time adjuster, or more likely in your case where all the inserts are filled with plugins ... nudge. I know I know ... that's even more a hassle. But I'm just curious if you can hear a difference. Import the bounces into a new sessions to A/B them.

I have done it several times and on several kinds of music. (Rock / Pop / Dance). The day that I did it is the day I started crying at digi's door for automatic plugin delay compensation. To me it makes a WORLD of difference. Sure ... I agree that a mix can sound good without it .... that's not the issue .... but IMHO it sounds SOOOOOO much better when you do compensate. Especially on loaded sessions.
I do firmly believe that it improves the sound of the system.

Do I sell more records by doing it ... nope most certainly not ... and neither will you if you would do it .... but if that's the only thing that would matter I would change my job immediately.

[Curve Dominant]

You won't hear a peep of protest out of me, Charles.

How many samples are we talking about here?

Then, if you slide one track, it's out of sync with all the others...

And is the effect the same whether the plug is used as an insert, or on an aux channel? IOW: If I attach a send to my singer's vocal channel routed to an aux track with a D-Verb inserted, is this delay still an issue?

I dunno, man, but I'm the newbie here...

"Real" music is not in sync down to sample level. You may like that sound, it may sound different, but not necessarily better IMHO.

[bassmac]

When I think back to my days of analog recording... a room full of twisted cables, cheesy stomp boxes, lousy headphone monitoring, tons of bleed, and every musician on some sort of mind bending upper or downer... It's hard to believe how many people get their panties all bunched up over a few samples here and there. Those records had delay you could probably measure seconds... but they sounded great.

Unlike our K9 counterpart, we'll belive our eyes before we belive our ears. Since DAW's give us the choice of using both, (which is a good thing) we need to make an extra effort not to forget that our EARS are what matters most.

Two important things I learned about working on a DAW;

1) It can sound right - and look wrong.
2) It can look right - and sound wrong.

Now, close yer eyes and LISTEN to it... before you fudge and nudge it all up just to make it LOOK pretty.

[Charles Dye]

Chris,

Excellent suggestion. I will try it.

I'm curious, were the sessions you tried this on tracked mostly live with lots of open mics with bleed + so forth, and by nudging you were sort of getting all the bleed back in time? I can see how that might matter. But there is still the 3:1 rule, which means that in most cases phase is still not the issue; the bleed is potentially beyond phase and more like "ambience" or "space". You know what I mean? I really don't mix music like that, most everything I mix is all overdubbed.

The other point is that I usually use lot's o plugs on many of the tracks (+ often almost the same chain) so the delay I'm creating is kind of uniform across the session. I have tried it a few times on live drum kits and I have to admit I couldn't hear it. That's just me, but I will try again and listen very carefully (maybe I didn't want to hear it). Again, unless we are comparing the top + bottom snare mics the 3:1 rule comes into play and "ambience" bleed is more what is on the tracks rather than "phase inducing" bleed. Do you agree or not?

If I hear an improvement after doing the nudge test, I will simply incorporate the nudging into my final steps before printing. Is that when you do it, or do you do it every time you change an insert?

Another reason I don't usually do manual compensation (with Time Adjuster), is that it will eat into my DSP resources (+ on MIX systems time slots) for potentially more important sound shaping plugs, but I suppose at this point for me that is yet to be seen. I'll tell you how the test goes.

Who else has tried this test and has head an improvement? Of any kind.

Thanks.

I told you guys I would learn stuff here. This is good. Let's really bash this topic around this month. Is plug-in delay compensation inherently a good thing?

[Charles Dye]

Originally posted by bassmac
1) It can sound right - and look wrong.
2) It can look right - and sound wrong.

Two very true statements. I definitely agree.

[Ckevperry]

I have noticed times when plugin delay between the overheads and the rest of the drum kit killed a low end.

I am NOT one of those people who phase align the overheads to the close mics...just doesn't sound natural to me. But when I spend time placing the overheads for a great drum sound- I do concern myself to keep all the delays even in the drums so I am not screwing that up and "fighting myself." Delay in the digital world effectively "moves" the mics I so carefully set for great sound (phase coherency) in the real world.

[Charles Dye]

Kevin,

My overhead mics don't normally have any low end in them—I roll most of that off.

Regarding moving OH's to phase align them (I realize you don't do this, + this has nothing to do w/ plug delay comp), but I've never understood the point. What exactly would you phase align them to? The kick? Snare? Toms? If you were to phase align the OH's to the kick the OH's would still not be in phase with the snare + toms. And so on. Even worse, by sliding the OH's to be in phase with any of the other drums you will be placing the cymbals earlier in time than when the drummer actually hit them. And on top of that, they would now be "out of phase" with where the cymbals are in the kick, snare + tom mics. It seems like you would be constantly chasing your tail. It really doesn't make any sense to me.

Again the 3:1 rule seems to solve all of this.

A drum kit is a single instrument and over the last century has always been heard with all these "phase" issues built in. This is a part of it's established sound. I ain't messin' with that concept. Can you imagine individually miking the 88 hammers on a piano + attempting to phase align the players performance? You could attempt it with todays technology, but not me.

As well, my kick, snare, hat + toms will almost always have all the same plugs on them anyway. Sort of self auto plug-in delay compensation.

[groundcontrol]

I'm with Kevin on that. If I really nailed the kit's sound the way I wanted, I'm very cautious about changing the relationship between the mics. On the other hand, on tracks that had badly recorded drums the plugs delay and/or moving things around sometimes improved the sound. So the listen first rule is a good one...

I'll admit that on programmed tracks and/or completely o'dubbed ones it's somewhat a moot point. Especially on stiffely programmed tracks where everything has been quantized. Smearing that downbeat in time can actually be pretty musical sounding... yuktyy

I can't wait for automatic delay compensation though because I like multing so much comes mixtime... rollz

[kushan_ku]

Awesome I needed this thread.

I'll continue to let my ears rule, and I trust them. I don't think I have a delay prob except sometimes when I do bounces or sends to auxes with a bunch of stuff stacked up on it.

And like you , Charles, I was thinking the same thing...I use alot of the same plugs of most of the channels so theoretically I'm building it in because everything is being slightly delayed.

And as far as the drums go, I haven't had any phaze problems. I'm stacking compressed snares back in, maybe a sound replaced snare. Kicks too, although I usually opt for the sampled kick so I don't have to go thru and endlessly tweek it to the orig kick.

Usually if anything I find the sound replaced stuff sometimes ahead of the original groove and I'll just quick throw a 3-5 ms. delay on it which is faster to me than moving a bunch of audio bits around.

[groundcontrol]

Originally posted by kushan_ku
Usually if anything I find the sound replaced stuff sometimes ahead of the original groove...

That's quite an understatement!

[Steve Smith]

I never have adjusted for indivual tracks ( ie: if I put an insert on the bass for example, I do not adjust it to match the kit again) but i am pretty paranoid when doing drum subgroups and the like, or when putting a plug on one mic of a ,multi miced anything... (ie, if I put a 57 and a 421 on the same amp, I would not want those relationships moving for the most part..)

I guess I am just as unsure as you charles.. however, if somethng starts to sound effed up, I check the delay first

I am also not doing Steely Dan. hell, I don't even play the CD in my studio

[DTM]

Hey Charles;
This is Dan, thanks 4 turning me onto your column. Overall, I would agree with your assessment of ignoring plugin delays. Most are too miniscule to bother with. Since getting HD, I've started using a couple plugins, Amplitude & Max DUY which have 2000 & 1000 samples each. That's too large to ignore, so I have been compensating for that. Talk to ya Monday.....

[Charles Dye]

Originally posted by Steve Smith
but i am pretty paranoid when...when putting a plug on one mic of a ,multi miced anything

Steve,

Definetely. When using plugs on a multi-miked identical sound source (like a gtr amp as you said) I would definetely use the same plugs to maintain phase, or time-adjuster. I include these types of tracks in the same category as stereo tracks, which I do the same with.

[Jules]

I feel it is very bad voodoo to have audio played back even at fractionaly different times to the way it was recorded. (INTENTIONALY EDITED PARTS EXCEPTED)

I am taking about live musician performed parts.

The "I cant hear it' crowd really need to bow out of the discussion IMHO. Deafness is hardly a strong position.

Anyhow, right now I have a whole season of mixing taking place, it's a major PITA futzing with Time Adjuster on my songs 11 + multi mic drum tracks every time I want to try something different on em. grudge

Sheesh! I only just discovered 'Multi Mono" for stereo tracks

I pray for Auto delay compensation ASFP

I just tweak the multi mic drum tracks to have the same delays, I dont go match up the rest of the songs tracks.. I got to try that Chris! Good idea..

[Greg Heimbecker]

In general I agree with the notion of "let your ears decide". With the type of work Charles is doing with almost everything overdubbed I think it's not a big deal. Shifting a groove in microseconds is fairly innocuous. I think it gets far trickier with acoustic ensembles playing together and/or with multiple feeds from one source(like mic & DI on an upright bass).

Another example I wrestle with regularly are multiple pairs on a large ensemble like orchestra for instance. Say I have an ortf pair for image focus and a spaced omni pair for ambience and low end "weight". As you shift one pair against the other sample by sample you can change the focus and balance of the whole freakin ensemble, mono compatability, image depth, apparent ambience & frequency balance. Add spot mics for soloists and it gets even more complicated. I typically shift one pair against the other til it gels nicely for me and move on. Nobody is going to pay me to sit around and obsess over it with the work I do!grggt

I find it present in the studio as well when I start mixing something that sounded killer as we were tracking but starts subtly degrading as plugins get added that by themselves sound fine but when combined with other leakage start to lose focus. I suspect subtly shifting comb filters as delay is added to some sources are a big cause of this. As you guys all know some delays are modest and others are fairly substantial (Ren EQ for instance). It's blatantly obvious if you mult a source like kick or snr for instance. If I do I'll try to use plugs on both that have the same delay.

Great thread on a subtle and tweaky damned topic!

[tld]

One situation that hasn't been mentioned here is when some of the plug(s) that are causing delay are in use while tracking. That's often my situation when I'm recording my own material...for example, direct guitar with Amp Farm. The performer will tend to compensate slightly, even if they aren't aware of it, to keep in the groove. In that situation, I would think that delay compensation after the fact could do more harm than good.

Tom

[bassmac]

Originally posted by Jules
The "I cant hear it' crowd really need to bow out of the discussion IMHO. Deafness is hardly a strong position.

Huh?

Are you really saying people who can't hear a tiny little delay caused by a plugin shouldn't take part in the discussion?

... or did I misunderstand that?

[Charles Dye]

That in fact is the discussion.

[kushan_ku]

rollz rollz

[Jax]

Originally posted by tld
The performer will tend to compensate slightly, even if they aren't aware of it, to keep in the groove. In that situation, I would think that delay compensation after the fact could do more harm than good.

Tom

This leads back to the main way to deal with this issue: use your ears. Don't compensate just because you can, or because you think you're supposed to.

Regarding those who don't hear a difference either way, then you have nothing to worry about. If it sounds like it works, and the people who are paying you don't think it sounds off, it probably doesn't matter. Not to mention that many musicians would be hard pressed to explain exactly what it is about a sound that isn't right, if it's related to plug-in delay.

The question is: Can you add or restore any more magic by P-I delay compensating? If you can, then it counts for something. So, use you your ears.

That said, I tend to plug-compensate for everything across the board. I do find that adding a load of different plug-ins alters the subtleties of timing that the musicians should get credit for, be that for better or worse.

[Jules]

Latency to coin a phrase of Jay Kahrs - Sucks!

It's a gold plated PITA for live drums and simply a weird vibe as a concept.

Eradicating it is surely favorite.

However DAW manufacturers could, I suppose, be encouraged to enable a 'legacy' option in the preferences where all tracks would play back at fractionally different times due to plug ins.

That would keep everybody happy!

[Jax]

Hey, this wasn't an issue when analog was all there was! Musically, plug-in delay sounds unnatural... but only when it's noticeable. The noticeable factor isn't as obvious to consider as it could be, apparently.

[kushan_ku]

I guess it must boil down to tastes and types of music.

With the common exception being that it should be used on aux's / mults, etc. or plugs that have an extra ordinary delay.

And the uncommon exception (had to say that Jules! )

I'll play around with some of the suggestions here, seeing that I have plenty o' power to add the adjuster for giggles and grins (gotta love HD3 - although I've been maxing it freequently lately!! ..time for another process card)

I'll let ya'll know what I think.

j_ho

[C.Lambrechts]

Originally posted by Charles Dye
Chris,

Excellent suggestion. I will try it.

can't wait to hear the result

Originally posted by Charles Dye
I'm curious, were the sessions you tried this on tracked mostly live with lots of open mics with bleed + so forth, and by nudging you were sort of getting all the bleed back in time? I can see how that might matter. But there is still the 3:1 rule, which means that in most cases phase is still not the issue; the bleed is potentially beyond phase and more like "ambience" or "space". You know what I mean? I really don't mix music like that, most everything I mix is all overdubbed.

nope, like you 95% of the stuff I do is overdubbed. Nowadays I tend to play back the dubs, take care of potential ad/da latency by nudging 100 samples back in time. comp everything and start the mix.

Originally posted by Charles Dye
The other point is that I usually use lot's o plugs on many of the tracks (+ often almost the same chain) so the delay I'm creating is kind of uniform across the session. I have tried it a few times on live drum kits and I have to admit I couldn't hear it. That's just me, but I will try again and listen very carefully (maybe I didn't want to hear it). Again, unless we are comparing the top + bottom snare mics the 3:1 rule comes into play and "ambience" bleed is more what is on the tracks rather than "phase inducing" bleed. Do you agree or not?

hmmm ... somehow while thinking about your way of mixing I thought of that today. thinking that if of course most tracks get the same plugin chain. The delay created by them will not be an issue. That's logical. My mixes will have all kinds of plugins spread all over them. Some tracks get the full load ... others get nothing. Back when I compared it the difference was the most obvious in a 50+ track mix loaded with plugins. I had a mix 6 + 2 dsp cards back then and dsp was almost maxed out so that should give you an idea. But I'm sure you take things way beyond that limit in the way you work so ....
In comparing them I felt 2 major things :

1. Most importantly : in the time adjusted mix things definately tightened up rock solid where as in the non-adjusted mix it sounded less tight. Let me make clear that the non-adjusted one sounded tight too. Of course it did. The differences are subtle and if you would make me listen to both an hour apart from each other I would not be able to tell which was which. But in a blind A/B I picked the adjusted one 8 out of 10 times. Concentrating on "feel".

2. And this must be a "in my mind" thing bacause I cannot find a logical nore technical explanation for why this would be the case. I felt that the time adjusted mix helped the mix bus sound. And I still have that feeling now with HD. I know, it sounds silly .... With exactly the same settings I felt that the non adjusted mix lacked life. you know what I mean. It's so easy with ProTools to screw up every life that a recording has in it and so damn hard to keep that feeling in there. But it can be done and timeadjusting is one of the elements that helps me do that.

Ever since I start mixing with a timeadjuster plugin set on every track and aux. I definately don't change every track for every plugin I insert but like 2 or three times during a mix ... usually when I'm about to take a break, I'll grab the calculator and take 5 - 10 minutes to set the values right.

btw,you mentioned R. Nichols moving hihats with single samples .... ok ... I have to admit ... that's over the top for me too. I'm not that nitty gritty on things either. I'm just a strog believer that every single element is important in making a good recording. Good music as the main ingredient, good musicians the second, but being aware of everything that can possibly degrade those pefrormances technically is the third, .... delay compensation is a part of that for me.

And the fact for digi not coming up with a better solution then time adjuster for that is no longer an excuse. Protools has come a long way since I started out on my very first nubus system. They can't stop now that its finnaly getting ready to go really high end .... not this close to the line.

[Charles Dye]

Originally posted by Jules
However DAW manufacturers could, I suppose, be encouraged to enable a 'legacy' option in the preferences where all tracks would play back at fractionally different times due to plug ins.

Ten months ago I began conversations with a plug-in manufacturer about a new type of analog emulation plug that I thought they were the perfect company to design because they essentially had all the required technology available within their current plug-ins. It's currently in beta, and we are hoping to debut it at Winter NAMM. They gave me permission to pre-announce it here while I'm the guest moderator, as long as I wouldn't say the name of the manufacturer.

Here is the preliminary info (subject to change):

Product name: The Synchrolator

What it does: It is an analog 2" 24 track synchronizer emulator. It perfectly emulates all the analog inaccuracies of locking two or more analog 2" 24 track tape machines together.

Parameters include: tape machine manufacturer, tape speed, synchronizer manufacturer, and virtual slave machine number (up to 2 Slaves for MIX/5 Slaves for HD)*

How it works: You can insert it on any audio track (max. 24 tracks per virtual synchronizer) and those tracks will all act as if they are a single slave machine. Some features include Locked, a constant yet subtle speeding up and slowing down of the tracks, with slight changes in both pitch and time; SlaveStop, a buffer that holds the audio of the Slave tracks so they will stop after the Master machine has stopped; SlaveCatch, a high pitched garbled whirring sound when you hit play to simulate the Slave tracks catching up with the Master, and as the sound slows down it becomes the audio of the Slave tracks slowing to normal pitch and it then switches to the Locked effect; and SlaveStutter, when Synchrolator has been in use for a number of hours in a row it will start to react more erratically, and from time to time when you hit play the audio on the Slave tracks will stutter with a kind of low pitched hiccup and not lock-up. Once this happens, the only way to prevent SlaveStutter is to put Synchrolator into PowerOff mode for a half hour or so, and work with just your Master tracks. If SlaveStutter persists, at this point you have no choice but to go home for the night.

The Synchrolator makes working in Pro Tools sound and feel like a real 48 track lock-up.

*LE available Summer 2003

[kushan_ku]

grggt yuktyy grggt yuktyy

[Jules]

grggt

[Curve Dominant]

I just looked at my calander to make sure today wasn't April Fool's Day.

I will take a guess that the manufacturer of that plugin is Funk Logic.

I'm still waiting for their plugin version of the Palindrometer.