Dugan AutoMix Experience - Sound Devices 633
Вставка
- Опубліковано 12 вер 2024
- Dugan Automix, Three Sanken CS-M1 mics in a pretty live room with close proximity to three talkers, Sound Devices 633 and a bit of fun primarily listening and comparing. Listening samples include normal one at a time conversation, simultaneous talkers, dynamic speech and signal to background noise level comparisons.
Nice demonstration, Michael! Dugan Automix always strikes me as a subtle but very effective effect in these types of situations.
Thanks Curtis, really useful for long corporate round table stuff as well!
Love how much the ambience is attenuated in active mode. Sounds great!
Thanks for watching Madhu V!
@@MichaelWynneCAS Thanks again, and for helping me zone in on the CS-M1 for a short shotgun mic.
@@MichaelWynneCAS Would love to see a comparison with 3 LAVs at the same distances as they can be much trickier.
Definitely the best automix I've used to date. Good stuff.
Thank you Shelby Watson, I completely agree. Even having used it for some time, I was stunned with these results!
Thanks for the video Michael. Been running Dugan on my 633 for years on corporate etc. but only just came by this video after watching some of your Cantar quick tips. Do you have the Aaton mix plugin installed on your Cantar(s)? How does it compare? RE Bag drop scenario (scripted with hidden lav mics) did you find clothes noise to affect the pusefulness of Dugan? - Does/would the ability to apply a threshold (SD automix) help in such circumstances? TIA
Really great questions leonsardine. Yes any noise lav can affect Dugan's performance. I have used Aaton mix on my X3, and it performed beautifully. Very not noticeable and helpful. Was a rapid fire dialogue scene with lots going on, made a big difference!
I've never seen a demo of the Dugan Automix before so correct me if I'm wrong... It appears that if it detects no signal in a mic, It's a circle with empty center. If it detects a signal, it's watching the input and puts a dot inside the circle to indicate that it's either playing the level just a tad or is standing by to play the mic. If it's gaining up on the channel, inside the circle starts filling up and if it's all the way up, the circle is solid. Is that, more or less, the way it is?
It sounds like when the Automix is on, the noise floor is lower which also drops the echo quite a bit. I'm assuming that's because when two of you are overlapping, you wouldn't hear the noise floor anyway but with only one mic open, you hear between the breaths and pauses. The result is a cleaner mix. Without Automix turned on, you're hearing the open room which sounds more like it would if I were sitting in the room listening to you. That's not better for recording dialogue in the mix though.
I noticed a few things I want your thoughts on Michael:
1) I know you set the gain correctly on all three of you but because Patrick has a higher voice, his voice gets buried under yours when overlapping because your voice is deeper and stronger than his. The levels were matched which is technically correct but your voice is more present because it's more powerful. In this circumstance, would it not be better to add 1 or 2 more dB of gain to Patrick so he's hotter in the mix? I know in the past you've told me the levels have to be matched for Automix to work correctly but I'm wondering if this is a rare circumstance it could be helpful? I imagine a female voice would totally vanish underneath your voice if level matched. I know this is the nature of acoustics and not something the mics, mixer or Automix is doing to the tracks. Now that I look again, Patrick's mic is higher than yours and Sals which "might" play a part in that? I don't know... your thoughts?
2) If you EQ the tracks, I know that only affects the mix track and not the ISOs. Does the Automix perform it's mix before EQ or after? That could make a difference, right? I'm just assuming that all three mics were flat and not EQed, right?
3) You and I have talked about how much UA-cam goofs up our pristine BWF sound when uploaded so this question is for about your BWF tracks... Many times when any of you guys are speaking, it's picking up in at least one nearby mic like your voice may pop into Sal's or Patrick's and Sal's may pop up in yours... I'm saying this based on my assumption that the more the circle fills, the Automix is using it more in the mix. If you ISO a track, is there bleed occuring between your mics? It could be because of your low camera angle but it appears that your mics are high enough overhead that they could be picking up the next voice over and if that's the case, it could phase. Am I over thinking this or is Automix sometimes playing two mics when only one person is speaking? Do you think this could (even if not here) cause phase cancellation? That would be my concern if playing two booms in close proximity and using Automix.
Thank you for this video and in advance for your knowledge and answer.
Allen if I could give this the comment of the year award, I would. Very good questions and observations! Also Thank you for numbering them. First thing to understand before I answer each questions is the Dugan automixing method which I've first pasted here from the SD688 manual.
Page 60 Sound Devices 688 Manual.
"Dugan’s system operates on a very elegant principle: each individual input channel is attenuated by an amount (in dB) equal to the difference (in dB) between that channel’s level and the sum of all channel levels. The gain of all channels is adjusted immediately and continuously based on what each channel is receiving. The neat aspect of this mathematical construct is that the total gain through the system never changes-always stays at 0dB. The core concept of this mixer is totally automatic, so there are no thresholds or other adjustments to set. With Dugan automixing, multi-mic speech applications are mixed with smooth automatic cross-fades for clear, optimized audio recordings A circle icon is displayed next to the meter for each active automixed input. The amount of purple filling the circles indicates the level of automix attenuation"
1. You are correct that Patricks mic is slightly higher then the others this was unfortunately not on purpose, also you'll notice his mix fader is slightly below unity gain which is also not on purpose. I caught this in post and was like ( sh*t ) Had Patricks mic been placed more optimally like Sal's and mine, the result would have been a better sonic result for Patrick. Interestingly enough though, it doesn't seem to negatively affect Dugan Automixer as you can see by what's being reduced and what your hearing. However is output level is also slightly lower due to the fader position. But not because of anything Dugan automixer is doing which is working perfectly for his input.
2. If the EQ is set pre fade such a high pass or selected in the menu it will be before the Dugan automixer. If it is post fade EQ I'm not exactly sure. That would be a question for Sound Devices. However this unit is a 633 and does not have post fade EQ options available.
3. If your were to ISO a track there would be no bleed from other tracks because only one mic is feeding each ISO per channel. What your seeing in the other tracks is normal Dugan automix operation. In fact since the mic's are closer in proximity your seeing more gain reduction on those channel outputs before it hits the mix. The reason for this is not because the Dugan automixer knows what's happening, it's just continuously doing the math.
It really is all quite brilliant to me Allen! And the results are very good and I look forward to expanding on this in future workflows on set. I'd really like to see a weighting option, groups and one touch bypass/activation on the 6 series recorders like you find on hardware and live console integration. More can be found on that online from user manuals which I recommend. Waves has a software version to you can download a demo for free for..
@@MichaelWynneCAS It's really impressive and I appreciate the clarification. I thought Patrick's trim and gain wasn't exactly right but know how much of a perfectionist you are so assumed I forgot a hidden trim independent of the knob itself. I'm glad to see you are indeed human. ;-)
Thank you for the full explanation with your detailed response including the Sound Devices excerpt. I look forward to any follow ups you release.
Maybe it's a given, but what I like about the Automix on the 6 series, is that it's stereo. So it's essentially 2 discrete automixers. I use it on 2-person corporate chats where I'll put 2 booms into one automix on the left and the 2 lavs into the other automix on the right. I prefer Dugan to the MixAssist option. MixAssist is bit too noticeable, sounds more like a gate/ducky.
Hub Location Sound! It wasn’t a given for me so I had to go test this really quick and your exactly right. This is excellent, thank you for the insight!
HLS, you're definitely right about MixAssist having a noticeable gate, but what I don't love about Dugan is that voices tend to sound less full IMO because its algorithm focuses on keeping the background equal, so often you're still hearing a subject on their mic as well as a bit on another person's mic. It'd be great if Dugan had even the slightest amount of customization to it like how MixAssist allows you to set the attenuation threshold. I've found both extremely useful in different circumstances though.
@@blaine691 I would welcome that customization - I specifically was thinking just yesterday, it would likely be useful to the algorithm if we could tell it which mics neighbor other mics. Even if it's just a checked box for "numerical order mode". I suppose this could get hairy if the distances vary, but if you have 8 people sitting in chairs evenly spaced side by side across a stage, if mic 3 knew that it was next to mic 2 and mic 4, it might duck those more aggressively to help with what you're describing.
@@blaine691 I prefer MixAssist because of what you mention and because MA does a better job than Dugan managing Last Mic On background ambience.
Great demo even for a French sound recordist ! Can't wait trying active mix on SD 833
Thanks Joel! It’s a useful tool
@@MichaelWynneCAS I was wondering if I can use automix on by ex Left output ONLY ? That should give me the opportunity to send a Left audio WITH Automix and a Right audio WITHOUT Automix . it looks like impossible, again on SD 833 ...
@@joelflescher7634 yes essentially their are two Dugan auto mixes on the 6 series. One for the Left and one for the right, on the 8 series you have bus options as well.
@@MichaelWynneCAS Thank's a lot Michael ! Best
Great demo! How are you liking the cs-m1 for indoor dialogue versus the usual suspects?
Archetype Guitars thank you! I’ve used it in some tough interiors on a stage where we were dealing with some noise from some power ballasts above on scaffolds and it was quite helpful. I think it’s a good sounding mic and also a problem solver in tight spaces
@@MichaelWynneCAS - another excellent problem solver in tight spaces is the new Schoeps CMC1 with the 41 capsule. It's a little beast!
@@nycvideoauditions Agreed, I've used the CMC 41 one time in a controlled environment and it really was magic.
Hello ! i really like your videos very much ! you said it only affects the LR mix, so in this configuration all 3 channels are routet to a mid position. will ducan engage if i pan left left right ? or right right left ? how does the auto mix behave if 1 have a boom on channel 1, and 2 wireless on channels 2 and 3 . is panning even necessary anymore ?
Hi Tobias Gerlach, You are correct that Dugan auto-mix only affects the LR, L or R mix. What’s really cool is since launching the video I learned that both L&R essentially have there own Dugan auto mixer which are not being affected by the other. So in essence it’s two auto mixers in one on the 633. So panning absolutely matters, and could possibly be used for weighting which is a Dugan auto mixer feature found in some more advance live consoles.
Nice demo ! Would it be possible to hear it from myself on a computer and perhaps to be able to download the 3 mono channels and test it with a few other softwares to compare the results ?
Again thanks for the demo it is enlightening !
HI sapym2408, I typicaly clear source files for these vids to save space on drives. So I don't have those available. Good idea though for testing.
Might sound like a silly question. But I'm trying to learn more about sound. Is there a version of this you can use with tracks in post?
Great question Effin A! Waves audio has a plug in version that will work in a DAW. If memory serves they also have a free trial period or demo
for most of their plugins. It also includes some cool features not in the 633 such as weighting and grouping.
@@MichaelWynneCAS Thanks for your help.
I don't ear any differences between on and off ?
Hi Voc0der, thanks for the comment. Have you tried listening with headphones. It may not be as clear on computer speakers or a mobile device
@@MichaelWynneCAS Yes i have, and same feeling. I can't make any differences. I'm not a pro, maybe my ears are broken, but if you need to be trained for that, then this plug don't do so much, and at the end of the production line, comon people like me will not make any differences. Am i the only one ? It would be interesting to do a blind test.
The most noticeable difference occurs when only one person speaks in this reverberant room and the other two mics are attenuated down by about 20 dB, you don't hear all the room reverb that you do hear when automix is off.
@@Voc0der you may not notice but without you would get a headache after 15 min of listening