Magic sends not so magical anymore :/

Discuss music production with Ableton Live.
Aweeeezy
Posts: 35
Joined: Mon Apr 13, 2009 1:23 pm

Magic sends not so magical anymore :/

Post by Aweeeezy » Wed Mar 14, 2012 2:21 am

So you know how in order to phase out the dry signal of a send effect in Ableton, you have to create two return tracks? The first return has an inverted utility on it and is sent to the second return, which has the FX chain loaded in...

Well I'm not sure what happened, but my sends stopped phasing out the dry signal. My dry signal(s) comes from a group(s) with a number of tracks inside of it. If I ungroup the tracks and send them, instead of the group, to the return track, the dry signal is properly phased out.

I was able to fix this problem in 2 of my 3 groups by ungrouping and regrouping the tracks inside, however, the last group will not work properly and I can't figure out what's different about this one compared to the other two.

I've tried closing and reopening Ableton, and I've ungrouped and regrouped the tracks at least 10 times to no effect.

Any ideas?

madlab
Posts: 1389
Joined: Fri May 02, 2003 6:38 am
Location: France

Re: Magic sends not so magical anymore :/

Post by madlab » Wed Mar 14, 2012 8:47 am

Could be related to some variable latency voodoo.
Aboard from V. 1
MBP M1 Pro 2021 - 16 Go RAM - Monterey 12.6.3
MBP 2.5 Ghz I7 16 Go SSD OSX 10.14
iPad + Mira+ TouchOsc
RME FF UC Live 11.3.21 M4L Max 8
Band : https://elastocat.org/
Madlab sound unit / objects, guitar, electronics / end_of_transmission

Nokatus
Posts: 1068
Joined: Fri Jul 01, 2005 7:06 am

Re: Magic sends not so magical anymore :/

Post by Nokatus » Wed Mar 14, 2012 10:31 am

Instead of using two sends, try using an audio effect rack on one return channel, and create the parallel signal chains inside that. The other one contains the Utility phase inverter for the dry signal, and the other one contains the wet effect.

beatmunga
Posts: 728
Joined: Sat Dec 03, 2011 5:20 pm

Re: Magic sends not so magical anymore :/

Post by beatmunga » Wed Mar 14, 2012 11:01 am

'Scuse my ignorance, but why not just have one return on 100% wet signal?
mendeldrive wrote:NOBODY designs their own sounds... There is ZERO point in reinventing the wheel.

Nokatus
Posts: 1068
Joined: Fri Jul 01, 2005 7:06 am

Re: Magic sends not so magical anymore :/

Post by Nokatus » Wed Mar 14, 2012 11:07 am

beatmunga wrote:'Scuse my ignorance, but why not just have one return on 100% wet signal?
Because the amplitude of the dry signal in the overall mix doesn't lower in proportion to the amount you send to the return channel. When you want this behavior, you need to sum an inverted version of the dry signal along with the mix, controlled by the amount you send to a return channel at any given moment. The most convenient way to do this on the fly is the one described above (when it works).

beatmunga
Posts: 728
Joined: Sat Dec 03, 2011 5:20 pm

Re: Magic sends not so magical anymore :/

Post by beatmunga » Wed Mar 14, 2012 11:35 am

Nokatus wrote:
beatmunga wrote:'Scuse my ignorance, but why not just have one return on 100% wet signal?
Because the amplitude of the dry signal in the overall mix doesn't lower in proportion to the amount you send to the return channel. When you want this behavior, you need to sum an inverted version of the dry signal along with the mix, controlled by the amount you send to a return channel at any given moment. The most convenient way to do this on the fly is the one described above (when it works).
Got it... so you can go from 100% dry to 100% wet using sends instead of inserts

Didn't know about this - nice trick!
mendeldrive wrote:NOBODY designs their own sounds... There is ZERO point in reinventing the wheel.

Nokatus
Posts: 1068
Joined: Fri Jul 01, 2005 7:06 am

Re: Magic sends not so magical anymore :/

Post by Nokatus » Wed Mar 14, 2012 11:44 am

beatmunga wrote:Got it... so you can go from 100% dry to 100% wet using sends instead of inserts

Didn't know about this - nice trick!
Yeah, that's the idea :)

There are other ways to have sends behave this way, just not quite as convenient. There's the "Sends Only" option in the channel output section so that you could, for example, keep one send for completely dry signals and have otherwise 100% wet send effects, and then on a given mixer channel just adjust/automate the dry level independently.

The main goal of these methods is to be able to go from a completely (or almost) dry signal to a completely (or almost) wet send signal smoothly, in a live situation or in an automated mixing setting. The less parameters you have to adjust to accomplish that, the better.

Anyway, to the original poster: if you can't get the inverted signal to align, try the method above. It requires you to control the dry signal level separately, but you can get the same smooth and proportional transition in effect levels this way. Not as convenient, though.

JuanSOLO
Posts: 3236
Joined: Sun Jul 15, 2007 8:21 am
Location: Shreveport LA, sometimes Dallas/Ft Worth TX

Re: Magic sends not so magical anymore :/

Post by JuanSOLO » Wed Mar 14, 2012 2:23 pm

Nokatus wrote:There's the "Sends Only" option in the channel output section so that you could, for example, keep one send for completely dry signals and have otherwise 100% wet send effects,
I prefer this method. However, my sends are set up as toggles on a controller so just one button press toggles between clean return or fx return.
I think a lot of people dont even know Track have the 'Sends Only'

salatspinatra
Posts: 235
Joined: Fri Jan 21, 2005 5:17 pm
Location: NYC

Re: Magic sends not so magical anymore :/

Post by salatspinatra » Wed Mar 14, 2012 3:41 pm

Aweeeezy: see if you haven't changed your levels on the polarity reversing channel or any of the return tracks you're playing with. If they're anything but unity gain, all bets are off for canceling the original signal. And once you fiddle with it, I'm really not sure how one can get the canceling back other than to stop playback, reconstruct the devices and levels. Is it because the gain on the Utility itself has not been compensated? Not sure why this is, but its a precursor to the trick. This is unfortunate for me, as I had hoped to keep my faders at -12db and not tempt my brick wall. Anyone find a way around this?

Also, you might want to look into the order of your return tracks. I put my Utility trick track farthest to the right. There's an issue with sample delays and the internal summing-mixer of ableton as to why the ordering of tracks will illicit various behavior, some desired some not. Listen, I'm not a tech expert-I consider it like good diet advice-skip the scientific papers and tell me that I should eat more broccoli, get it? But should you find these explanations on the boards, by all means I'd welcome your reporting back here and explaining them to us.

Nokatus: your parallel chain rack sounds like an elegant approach: I'm always happy to find ways to lean up my tracks, and I wonder if it would solve either one of the above constraints. One thing that it does better than a dedicated Return track is assuring that you'll have some signal out at any point in time. With the dedicated return track approach, you need at least one send on the Utility Return track at any time delivering full signal. To me, this might mean keeping a signal processor on at all times even if it isn't in foresight of using the magic send trick creating my chosen effect, or having an empty chain delivering unaffected signal. Otherwise, turn off all sends from the return track and get awkward silence. With your rack approach, you don't have this worry.

BUT, I think you loose some flexibility in how you midi assign your output with full effect, don't you? With the dedicated return track, I can hit the corresponding send on my audio/midi tracks and know that they're going through magic send (mind you, marginal levels have an underwhelming behavior, so I tend to give this trick over to a button rather than a fader.) With the rack approach, each of those same track's signal is serving double duty. I then need a separate control to change the chain on each return. I haven't played with the midi assignments for your approach enough to see how that would be more fun for the performer, what you'd do if on occasion you wanted to send partial signal to the return tracks in a traditional sense, or if a single control assignment could make up for this formality. I hope that we all give this a go and report back with our preferred approach.

Aweeeezy
Posts: 35
Joined: Mon Apr 13, 2009 1:23 pm

Re: Magic sends not so magical anymore :/

Post by Aweeeezy » Wed Mar 14, 2012 7:16 pm

I have to go to work soon, but I will try out some of the suggestions posted here and let you know what happens...in the mean time, here is my Live Pack. Take a look at it and see if you can find anything fishy.

http://dl.dropbox.com/u/10059848/The%20Rig.alp

Here's my repost in glitchhopforums.

http://glitchhopforum.com/glitch-hop-mu ... tml#p68391

Aweeeezy
Posts: 35
Joined: Mon Apr 13, 2009 1:23 pm

Re: Magic sends not so magical anymore :/

Post by Aweeeezy » Thu Mar 15, 2012 8:54 am

Nokatus wrote:Instead of using two sends, try using an audio effect rack on one return channel, and create the parallel signal chains inside that. The other one contains the Utility phase inverter for the dry signal, and the other one contains the wet effect.
So two chains in one audio effect rack...the first chain has an invert utility on it and the second chain has the FX in it?

How do I send the audio through an invert utility and external out the inverted audio to the korg pad?

ikeaboy
Posts: 1685
Joined: Sat Jun 05, 2004 9:38 pm
Location: Ireland

Re: Magic sends not so magical anymore :/

Post by ikeaboy » Thu Mar 15, 2012 11:09 am

This happened me before aswell but the set was too complex to figure out what was causing the slight signal delay from the sends which ruined the phase cancelling. From reading your post it looks like grouping is the offender. I also remeber reading somewhere that each send is 1 ms delayed compared with track outputs and this accumulates i.e. 2ms for send 2 etc... I'm not sure if this is still the case

braduro
Posts: 238
Joined: Wed Aug 31, 2011 12:03 am

Re: Magic sends not so magical anymore :/

Post by braduro » Thu Mar 15, 2012 4:50 pm

ikeaboy wrote:I'm not sure if this is still the case
Yes, indeed, this is the prevailing and intended case. There's a good explanation from Robert Henke on why this is if you follow the original magic send threads.

matthews
Posts: 388
Joined: Mon Jan 10, 2011 11:18 pm

Re: Magic sends not so magical anymore :/

Post by matthews » Thu Mar 15, 2012 6:13 pm

Nokatus wrote: Because the amplitude of the dry signal in the overall mix doesn't lower in proportion to the amount you send to the return channel. When you want this behavior, you need to sum an inverted version of the dry signal along with the mix, controlled by the amount you send to a return channel at any given moment. The most convenient way to do this on the fly is the one described above (when it works).
Any chance you can detail how this is setup? I've been trying to figure out the best way to achieve this type of behavior for a while now, but haven't had any luck...

Aweeeezy
Posts: 35
Joined: Mon Apr 13, 2009 1:23 pm

Re: Magic sends not so magical anymore :/

Post by Aweeeezy » Thu Mar 15, 2012 7:11 pm

matthews wrote:Any chance you can detail how this is setup? I've been trying to figure out the best way to achieve this type of behavior for a while now, but haven't had any luck...
http://www.anthonyarroyodotcom.com/thea ... n-ableton/
ikeaboy wrote:From reading your post it looks like grouping is the offender.
I'm not ruling this out, but if this is the case, then why does choosing a different output device in Ableton temporarily solve the problem? Any why does only the one group behave this way while the other group's dry signal is properly phased out when sent to any of the return tracks?

I think it's just some kinda bug relating to my interface...I ordered a second Audio DJ 8 that should be arriving early next week. I'm hoping that combing the two Audio 8s into one aggregated device and choosing that device as my output will solve the problem. We'll see!...I hate waiting.

Post Reply