Audio for Dance Recital and the Alesis FW Multimix 8

Dave Stern · February 2nd, 2009, 06:40 AM

Quote:

Originally Posted by Ron Evans

....Turning frame sync off allows the audio to be dragged so that they can be line up independent of the video and by expanding or contracting one of the tracks can be made to sync up precisely independent of the video track. ...

Ron - not to move this thread OT, but quick question - how to you turn off the snapping of the audio to the video frame? I did have that problem and just couldn't find it...

Ron Evans · February 2nd, 2009, 08:10 AM

Dave it's under Options>Quantize to frames, snapping is also under the options menu. The process I use is to export audio from the synced up video in Edius into Vegas. Set the main camera audio as the master, expand timeline so that I can see the waveform in detail, line up the waveforms of the remaining audio tracks at the beginning( you will have to shorten the beginnings to give room to move tracks around then expand back when set)) and scroll to the end of the track and expand or contract track to line up the waveforms. With quantize to frames off the audio tracks will move smoothly with the mouse in very fine increaments. Then do an audio playback check to see if it sounds good using the mixer to set levels etc. With my older cheap DV cameras they would not hold sync all the way through so I usually had to adjust a few times( I am really picky!!!) HDV and AVCHD are better at maintaining their relative position. In our case the Zoom H4 or H2 is set at stage level, the FX1's use shot gun mics and the SR11 shoots in 5.1. The main centre FX1 is used as reference audio( SR11 is in the same position), rarely use the second FX1 audio and the Zoom and SR11 audio provide stereo ambiance in the mix.

Ron Evans

Bill Warshaw · February 2nd, 2009, 05:30 PM

Hi,

I use the MOTU 8pre Firewire audio interface with my primary camera (XH-G1) and record to a small, portable, USB HDD hooked up to my Laptop using Vegas 8 Pro. The G1 has timecode out, and the MOTU happily syncs to SMPTE timecode in one of the 8 input channels (a nice feature of the MOTU's). There is no significant drift over 1hr long recordings. My secondary camera (XH-A1) is simply genlocked with the G1. This is a more expensive setup (G1 & 8pre) but works really well. Recording using my laptop (plugged into AC) has never been an issue to date.

One trick (safety net) that I use is to record a submix of the important elements out to the primary & secondary cameras from the 8pre. I also tell Vegas to record this submix to the HDD. This makes lining up the clips in Vegas very easy, and once aligned I just delete/mute the camera audio tracks and the reference track from the HDD.

On one occasion, I made a mistake in setup and the MOTU wasn't looking for sync from the audio input - rather it was running off it's internal clock. Having the submixes on both camera tapes and the HDD really paid off: I aligned the camera clips as usual and then stretched all the MOTU recorded audio tracks using the reference submix from the MOTU as a guide so it matched the camera audio. They should look exactly the same when sync'd. Took a couple of minutes and sounded fine. I guess I spent the extra $'s for the G1 for nothing <g>.

I'm not sure if the FW unit you're looking at has the capability to record a mix your sending out to the camera back to the computer along with the individual tracks, but if so it will really help to get things in sync.

/BILLW

Steve House · February 3rd, 2009, 04:26 AM

Quote:

Originally Posted by Bill Warshaw

Hi,

I use the MOTU 8pre Firewire audio interface with my primary camera (XH-G1) and record to a small, portable, USB HDD hooked up to my Laptop using Vegas 8 Pro. The G1 has timecode out, and the MOTU happily syncs to SMPTE timecode in one of the 8 input channels (a nice feature of the MOTU's). There is no significant drift over 1hr long recordings. My secondary camera (XH-A1) is simply genlocked with the G1. This is a more expensive setup (G1 & 8pre) but works really well. Recording using my laptop (plugged into AC) has never been an issue to date.

One trick (safety net) that I use is to record a submix of the important elements out to the primary & secondary cameras from the 8pre. I also tell Vegas to record this submix to the HDD. This makes lining up the clips in Vegas very easy, and once aligned I just delete/mute the camera audio tracks and the reference track from the HDD.

On one occasion, I made a mistake in setup and the MOTU wasn't looking for sync from the audio input - rather it was running off it's internal clock. Having the submixes on both camera tapes and the HDD really paid off: I aligned the camera clips as usual and then stretched all the MOTU recorded audio tracks using the reference submix from the MOTU as a guide so it matched the camera audio. They should look exactly the same when sync'd. Took a couple of minutes and sounded fine. I guess I spent the extra $'s for the G1 for nothing <g>.

I'm not sure if the FW unit you're looking at has the capability to record a mix your sending out to the camera back to the computer along with the individual tracks, but if so it will really help to get things in sync.

/BILLW

How are you genlocking the XH-A1 to the G1? I don't recall the A1 having an input for external sync?

Bill Warshaw · February 3rd, 2009, 09:26 AM

Sorry, other way around. BNC video out from the A1 to the BNC Genlock In on the G1.

Jay Massengill · February 3rd, 2009, 03:29 PM

Going back to your original post regarding the mics you hope to use and their setup locations. I don't think the side positioned shotguns will give you the quality of audio that you're hoping. In addition to their likely pickup of noticeable off-axis coloration, the timing differential will likely be very noticeable between these two mics and difficult to adjust since the on-stage sources will be constantly moving.
I think your track isolation list is good, but I would substitute a stereo pair of cardioids or hypercardioids in X/Y configuration rather than the side positioned shotguns.
I would also think about substituting a boundary layer mic on the stage lip for your central Sennheiser dynamic, or mount the Senn dynamic as close to the stage as you can. (You may need to shift your ambient stereo pair forward in time to match the extremely short delay of this closely placed source when you edit.
With my setup, the central mic gives the isolation of the stage sounds due to its closer proximity while the stereo pair adds ambience and some stereo imaging to augment the direct music and wireless feeds.
For all your acoustic sources, remember that distance equals time. Roughly 3 NTSC video frames per 100 feet if you're recording at 30 frames per second. (2.4 frames per 100 feet at 24fps). Even at very short distances, with music and tap, any delay between sources at equal volumes can be very noticeable. This delay will also come into play when using your on-camera audio as a line-up guide, even though you hope not to use the on-camera audio as a source.

Bryan Daugherty · February 3rd, 2009, 07:44 PM

Jay, thank you for your input. I think either I have not been clear with my plans for using the mics or I am not understanding your post. Let me preface by saying that for this particular job, I have to use the mics i have and do not have the budget to add additional mics this yr. The main question would be deployment. That said, I am not planning to just drop the mics in to the mix based on their location but rather am trying to capture sound based on the environment that can be mixed to match the video. I am not hugely proficient in sound so my terminology may be off a little, so I ask your patience if I don't say things correctly.

On the shotgun mikes, I want to deploy them in a manner so that they record the sounds close to them such as the tap shoes, and other stage noise but away from the speakers so they pick up less of the room mix. I want each one to pick up the sounds as heard from that part of the room. So that when I cut to the CU cam (cam 2 on the right of the stage) and we see the dancers further from the camera to stage left i would spin up the tap on the rt shotgun mic so that the sound seems further away. It would not be on the right speaker channel but more distributed across all front channels, as that op zooms in to get a tighter shot, i would bring in more of the left shotgun sound that is closer to the dancers actual position, etc. In other words, i am not wanting to use the mics to capture speaker channels but rather points of view. When i cut back to the wide shot i would pan the best sound to the dancers position.

I get what you are saying about distance and sound and appreciate that I will need to sync my audio to video to account for what (when) the mic hears vs what (when) the camera sees, my sync worry is not so much sync but sync drift. If i sync the sound at the beginning of the clip will it still sync at the end of the clip? The consensus seems to be no (or maybe), and that I will have to manually adjust via stretch or shrink.

The production makes occasional use of the orchestra pit so i won't have any mics at stage edge but rather at the edge of the pit. (The last thing i want to be responsible for is a dancer tripping on a cable I run.)

Thanks for your input and i look forward to learning more from you fine folk.

Jay Massengill · February 3rd, 2009, 11:35 PM

My prediction is the shotgun mics won't be able to pick up the isolated sound that you're imagining they can.
They will have leakage from the stage sounds on the opposite side, and also the PA and crowd. What will make this even worse is the leakage will not sound good generally but it will also have a varying time relationship as the dancers move. So it won't just be sync, and sync drift, but it will be the constantly shifting time relationship between the 3 mics as the people move on the stage that you'll have to adjust as your shot changes.
But you'll be the only one to judge whether it's working satisfactorily. I'd suggest that you not only do a full recording test during tech week but that you also actually edit a typical dance number and see if it's possible to achieve.
Unless the performance space has outstanding acoustics and you're using the highest quality shotgun mics, I think it will be very difficult to record and edit the way you've stated, mainly due to the nature of shotgun mics with reverberent indoor sources.
What exact model of shotgun and Sennheiser dynamic do you have?

Bryan Daugherty · February 4th, 2009, 12:10 AM

Jay thanks for the follow up. The acoustics are outstanding in this theatre, i was quite shocked with the clarity my shotgun recorded on camera last yr during the test run. Since my shotguns will be in use on camera, my second op is lending his mics to the project. We are meeting (weather permitting) later this week and I will try to get the model numbers from him but he just upgraded them recently and my memory is they were extremely high grade and quite expensive. He is the media director for a non-profit and they take their equipment very seriously. I am limited in equipment placement for this production and am making do, so is there a better way to do it with this same equipment?

EDIT:

Quote:

Originally Posted by Jay Massengill

...the constantly shifting time relationship between the 3 mics as the people move on the stage that you'll have to adjust as your shot changes....

Are you referring to doppler shift or something else here?

Steve House · February 4th, 2009, 04:41 AM

Quote:

Originally Posted by Bryan Daugherty

...Are you referring to doppler shift or something else here?

Sound travels at about 1 foot per millisecond. Let's say a dancer is on the right side of the stage so she is closer to the right-hand mic than the left-hand one by, lets say, 30 feet. That means the clack of a tap hitting the stage will arrive at the left mic 30ms later than it arrives at the right. 30ms is one full frame.

Jay Massengill · February 4th, 2009, 08:49 AM

That's good news about the performance space acoustics and the potential quality of the mics. However I would for safety use a stereo pair for your centrally located mic, even if you had to beg, borrow or steal either a matching Senn dynamic or a low cost matching pair of something like AT3031 cardioids.

You could then use this stereo pair in three different ways:
1. As two slightly directional mics to reinforce your stated editing goal (but without the large time shift and odd-sounding leakage of the shotguns);
2. Mixed to mono if needed;
3. Or most importantly as a normal stereo backup if it becomes obvious that editing with your shotguns is too time consuming or has too much leakage.

I think you have enough channels to do this: 2 for prepared music, 2 for centrally located stereo pair, 2 for your shotguns and 2 for isolated wireless feeds from the performers.

By using a stereo pair in the central position, you gain much more flexibility and safety without dramatically changing away from your original shotgun idea when you record.
After the fact you'll have the full 8 tracks and the time to determine what can be edited to best effect.

To add one thing to Steve's example, it's the "leakage" of the dancer on the right of the stage that will arrive at the left shotgun 30ms late. I think you were hoping for enough isolation from the shotguns that the far side of the stage wouldn't be a factor but I think that's impossible in this case.

February 3rd, 2009, 03:29 PM	#21
Jay Massengill Trustee Join Date: Jul 2003 Location: Burlington Posts: 1,976	Mics Going back to your original post regarding the mics you hope to use and their setup locations. I don't think the side positioned shotguns will give you the quality of audio that you're hoping. In addition to their likely pickup of noticeable off-axis coloration, the timing differential will likely be very noticeable between these two mics and difficult to adjust since the on-stage sources will be constantly moving. I think your track isolation list is good, but I would substitute a stereo pair of cardioids or hypercardioids in X/Y configuration rather than the side positioned shotguns. I would also think about substituting a boundary layer mic on the stage lip for your central Sennheiser dynamic, or mount the Senn dynamic as close to the stage as you can. (You may need to shift your ambient stereo pair forward in time to match the extremely short delay of this closely placed source when you edit. With my setup, the central mic gives the isolation of the stage sounds due to its closer proximity while the stereo pair adds ambience and some stereo imaging to augment the direct music and wireless feeds. For all your acoustic sources, remember that distance equals time. Roughly 3 NTSC video frames per 100 feet if you're recording at 30 frames per second. (2.4 frames per 100 feet at 24fps). Even at very short distances, with music and tap, any delay between sources at equal volumes can be very noticeable. This delay will also come into play when using your on-camera audio as a line-up guide, even though you hope not to use the on-camera audio as a source.

February 3rd, 2009, 07:44 PM	#22
Bryan Daugherty Major Player Join Date: Jan 2009 Location: Lexington, Ky - USA Posts: 552	Jay, thank you for your input. I think either I have not been clear with my plans for using the mics or I am not understanding your post. Let me preface by saying that for this particular job, I have to use the mics i have and do not have the budget to add additional mics this yr. The main question would be deployment. That said, I am not planning to just drop the mics in to the mix based on their location but rather am trying to capture sound based on the environment that can be mixed to match the video. I am not hugely proficient in sound so my terminology may be off a little, so I ask your patience if I don't say things correctly. On the shotgun mikes, I want to deploy them in a manner so that they record the sounds close to them such as the tap shoes, and other stage noise but away from the speakers so they pick up less of the room mix. I want each one to pick up the sounds as heard from that part of the room. So that when I cut to the CU cam (cam 2 on the right of the stage) and we see the dancers further from the camera to stage left i would spin up the tap on the rt shotgun mic so that the sound seems further away. It would not be on the right speaker channel but more distributed across all front channels, as that op zooms in to get a tighter shot, i would bring in more of the left shotgun sound that is closer to the dancers actual position, etc. In other words, i am not wanting to use the mics to capture speaker channels but rather points of view. When i cut back to the wide shot i would pan the best sound to the dancers position. I get what you are saying about distance and sound and appreciate that I will need to sync my audio to video to account for what (when) the mic hears vs what (when) the camera sees, my sync worry is not so much sync but sync drift. If i sync the sound at the beginning of the clip will it still sync at the end of the clip? The consensus seems to be no (or maybe), and that I will have to manually adjust via stretch or shrink. The production makes occasional use of the orchestra pit so i won't have any mics at stage edge but rather at the edge of the pit. (The last thing i want to be responsible for is a dancer tripping on a cable I run.) Thanks for your input and i look forward to learning more from you fine folk. __________________ 3x-HD1000u - Ikan 8000HD- custom i7 PC - Vegas Pro 13 and 11 64 bit - Premiere Pro CS4 - and a whole mess of other equipment...

February 2nd, 2009, 08:10 AM	#17
Ron Evans Inner Circle Join Date: Mar 2003 Location: Ottawa, Ontario, Canada Posts: 4,220	Dave it's under Options>Quantize to frames, snapping is also under the options menu. The process I use is to export audio from the synced up video in Edius into Vegas. Set the main camera audio as the master, expand timeline so that I can see the waveform in detail, line up the waveforms of the remaining audio tracks at the beginning( you will have to shorten the beginnings to give room to move tracks around then expand back when set)) and scroll to the end of the track and expand or contract track to line up the waveforms. With quantize to frames off the audio tracks will move smoothly with the mouse in very fine increaments. Then do an audio playback check to see if it sounds good using the mixer to set levels etc. With my older cheap DV cameras they would not hold sync all the way through so I usually had to adjust a few times( I am really picky!!!) HDV and AVCHD are better at maintaining their relative position. In our case the Zoom H4 or H2 is set at stage level, the FX1's use shot gun mics and the SR11 shoots in 5.1. The main centre FX1 is used as reference audio( SR11 is in the same position), rarely use the second FX1 audio and the Zoom and SR11 audio provide stereo ambiance in the mix. Ron Evans

February 2nd, 2009, 05:30 PM	#18
Bill Warshaw Regular Crew Join Date: Nov 2007 Location: Brea, CA Posts: 37	Hi, I use the MOTU 8pre Firewire audio interface with my primary camera (XH-G1) and record to a small, portable, USB HDD hooked up to my Laptop using Vegas 8 Pro. The G1 has timecode out, and the MOTU happily syncs to SMPTE timecode in one of the 8 input channels (a nice feature of the MOTU's). There is no significant drift over 1hr long recordings. My secondary camera (XH-A1) is simply genlocked with the G1. This is a more expensive setup (G1 & 8pre) but works really well. Recording using my laptop (plugged into AC) has never been an issue to date. One trick (safety net) that I use is to record a submix of the important elements out to the primary & secondary cameras from the 8pre. I also tell Vegas to record this submix to the HDD. This makes lining up the clips in Vegas very easy, and once aligned I just delete/mute the camera audio tracks and the reference track from the HDD. On one occasion, I made a mistake in setup and the MOTU wasn't looking for sync from the audio input - rather it was running off it's internal clock. Having the submixes on both camera tapes and the HDD really paid off: I aligned the camera clips as usual and then stretched all the MOTU recorded audio tracks using the reference submix from the MOTU as a guide so it matched the camera audio. They should look exactly the same when sync'd. Took a couple of minutes and sounded fine. I guess I spent the extra $'s for the G1 for nothing <g>. I'm not sure if the FW unit you're looking at has the capability to record a mix your sending out to the camera back to the computer along with the individual tracks, but if so it will really help to get things in sync. /BILLW

February 3rd, 2009, 09:26 AM	#20
Bill Warshaw Regular Crew Join Date: Nov 2007 Location: Brea, CA Posts: 37	Sorry, other way around. BNC video out from the A1 to the BNC Genlock In on the G1.

February 3rd, 2009, 11:35 PM	#23
Jay Massengill Trustee Join Date: Jul 2003 Location: Burlington Posts: 1,976	My prediction is the shotgun mics won't be able to pick up the isolated sound that you're imagining they can. They will have leakage from the stage sounds on the opposite side, and also the PA and crowd. What will make this even worse is the leakage will not sound good generally but it will also have a varying time relationship as the dancers move. So it won't just be sync, and sync drift, but it will be the constantly shifting time relationship between the 3 mics as the people move on the stage that you'll have to adjust as your shot changes. But you'll be the only one to judge whether it's working satisfactorily. I'd suggest that you not only do a full recording test during tech week but that you also actually edit a typical dance number and see if it's possible to achieve. Unless the performance space has outstanding acoustics and you're using the highest quality shotgun mics, I think it will be very difficult to record and edit the way you've stated, mainly due to the nature of shotgun mics with reverberent indoor sources. What exact model of shotgun and Sennheiser dynamic do you have?

February 4th, 2009, 08:49 AM	#26
Jay Massengill Trustee Join Date: Jul 2003 Location: Burlington Posts: 1,976	That's good news about the performance space acoustics and the potential quality of the mics. However I would for safety use a stereo pair for your centrally located mic, even if you had to beg, borrow or steal either a matching Senn dynamic or a low cost matching pair of something like AT3031 cardioids. You could then use this stereo pair in three different ways: 1. As two slightly directional mics to reinforce your stated editing goal (but without the large time shift and odd-sounding leakage of the shotguns); 2. Mixed to mono if needed; 3. Or most importantly as a normal stereo backup if it becomes obvious that editing with your shotguns is too time consuming or has too much leakage. I think you have enough channels to do this: 2 for prepared music, 2 for centrally located stereo pair, 2 for your shotguns and 2 for isolated wireless feeds from the performers. By using a stereo pair in the central position, you gain much more flexibility and safety without dramatically changing away from your original shotgun idea when you record. After the fact you'll have the full 8 tracks and the time to determine what can be edited to best effect. To add one thing to Steve's example, it's the "leakage" of the dancer on the right of the stage that will arrive at the left shotgun 30ms late. I think you were hoping for enough isolation from the shotguns that the far side of the stage wouldn't be a factor but I think that's impossible in this case.