Jump to content
John Blankenship

5D Shoot Workflow

Recommended Posts

These days, if they hear anything, and it did not include syncing up the quality audio, they're happy.... A bit sad...

There is the crux of the problem, they can hear "something" on the camera while looking at the image. Before the cameras could record audio (Frezzi single system news cameras excepted) if anybody wanted to hear anything along with the picture they had to sync up the sound --- and this sound was always recorded by a SOUND person with appropriate sound recording equipment. I am confident that in the near future for most jobs we will return to double system methodology and a key component in this is that the people responsible for the image will be camera people and the people responsible for the sound will be sound people. One thing that may change, though, will be that it may be a single data storage repository that accepts the data generated by the two (double) systems, image and sound, but this data storage will NOT be the imaging device (camera).

Share this post


Link to post
Share on other sites

The reason to send audio to a 5D is that the syncing of the double system sound goes MUCH faster if the editor can sync audio-to-audio waveforms, instead of audio-to-picture.  Audio analysis apps like PluralEyes work much better comparing and syncing waveforms because the two waveforms they have to work with are nearly identical if the camera got a feed.  As with many popular prosumer cameras, syncing via TC is not an option--the camera's TC is not settable or outputable in any way.  In addition, if you have not sent a scratch mix to the camera then the director has no way of looking at a playback w/ sound of what he or she's shot--the 5D playback is crude but it does work.  I have less than no interest in clients using the sound I transmit to the camera--I usually send a mono mix of everything via a cheap wireless and leave the camera gain in auto, but having the audio sync go faster and easier due to the camera getting line feed actually helps sell the concept of double system sound to producers who may have no experience with it, or who are unhappy with the extra expense.  The people who actually want to use the 5D onboard audio for their project aren't likely to hire any of us to record it anyhow.

Philip Perkins

Share this post


Link to post
Share on other sites
Guest BobD

Agree with your post Phillip and Jeff...

  I was speaking more in terms of the RED where TC can be entered...  The 5D shoots I do the same, cheap Senn Wireless mounted to the hot shoe, mono mix for the editor.... works well for that function....  If interview style, I might choose to use the on board mic for the job...

Share this post


Link to post
Share on other sites

  Funny to see the Zax receivers and trans. on sale on the WTS page for $6000...  Why should we put out that kind of money for redundant audio, then get beat up on rates!!!  I'll tell you why... Too lazy to sync it, and we're just supposed pony up gear to be used for free.... or listen to static from production when they hear there is actually a charge for that gear...  duhhhhh...

Bob,

Most of the people that use the Zaxcom Stereo TX & RXs are not using them to send redundant audio but as the only audio. 

Andy

Share this post


Link to post
Share on other sites

So just to clarify, PluralEyes can easily match up the mono scratch mix sent to camera with, let's say, a 4-track poly .wav file (of ISO tracks) from my recorder?

I haven't used the software yet, but am interested!

Share this post


Link to post
Share on other sites
Guest BobD

Bob,

Most of the people that use the Zaxcom Stereo TX & RXs are not using them to send redundant audio but as the only audio. 

Andy

Point well taken, but still redundant to me...  Still like what I have control of... I will always think of "the other audio" as redundant....

It's only really redundant on the cameras with the poor signal processing...  Kinda what I was driving at...

Thanks Andy...

Share this post


Link to post
Share on other sites

So just to clarify, PluralEyes can easily match up the mono scratch mix sent to camera with, let's say, a 4-track poly .wav file (of ISO tracks) from my recorder?

I haven't used the software yet, but am interested!

Yes.  Manual sync is much faster too.

Philip Perkins

Share this post


Link to post
Share on other sites

Hi, all.

I've had recent experience with both the 5D and 7D and used a slate and my 744.  No complaints from me, especially about having to not be in the way of the camera and operator.  This is like going back to business as I remember it!

Last summer I turned down a 'job' working a feature for first timers that had a 5D. They wanted me to just feed to the camera and not be able to monitor what I was 'recording'.  We did not work together, especially because they wanted to pay me $179 for 21 days! 

Found out later that their sound did, indeed, suck.  We are now in negotiations for a film to be shot on the Dread. I want to do double system but they are not wanting to pay for it. They are trying to talk me into letting them provide me with a boom man.  They must know someone that has 'held a boom'. 

Digitally yours, Mike

Share this post


Link to post
Share on other sites

Share this post


Link to post
Share on other sites

I noticed alot of people griping about the issue of double system and why can't they just sync in post.  While I completely agree that a double system is generally the best way and the only way if shooting with a 5D or 7D, and I guess apparently the 1D now, it's seem quite obvious that why post doesn't want to do it is because they don't know how too.  The AE's that are being hired on jobs are not being taught how to do the job of syncing up the audio and are so use to having the audio recorded on the camera.  The larger part of shoots on the 5D or the like are low-budget and hiring a person with that kind of experience in post is generally not within their budget.

I also feel that if I'm asked to do something like mix to the 5D or the RED and I can make it work then I will.  I've never had a problem with the RED's audio and have always run a second recorder anyway.  Treated every shoot as if I was only doing double system, but just happened to have that connection to camera as well.  For the 5D I've never had production request anything be sent to the camera, but have always tried if the budget allowed.  Usually just a scratch track and nothing more though.

J Hemmerlin

Share this post


Link to post
Share on other sites

I noticed alot of people griping about the issue of double system and why can't they just sync in post.  While I completely agree that a double system is generally the best way and the only way if shooting with a 5D or 7D, and I guess apparently the 1D now, it's seem quite obvious that why post doesn't want to do it is because they don't know how too.  The AE's that are being hired on jobs are not being taught how to do the job of syncing up the audio and are so use to having the audio recorded on the camera.  The larger part of shoots on the 5D or the like are low-budget and hiring a person with that kind of experience in post is generally not within their budget.

I also feel that if I'm asked to do something like mix to the 5D or the RED and I can make it work then I will.  I've never had a problem with the RED's audio and have always run a second recorder anyway.  Treated every shoot as if I was only doing double system, but just happened to have that connection to camera as well.  For the 5D I've never had production request anything be sent to the camera, but have always tried if the budget allowed.  Usually just a scratch track and nothing more though.

J Hemmerlin

Right on bro.  Syncing audio in FCP was something I figured out how to do in about 2 minutes w/o the manuals.  A monkey could do it.  They know this--it is more about the time and rote work involved--no comment on that.  Pluraleyes costs $150 and will work on really old FCP versions (back to v5) which can run on very slow (cheap) computers, so.....?  I generally send a mono sum w/ the camera AGC on--they can hear everything in a playback that way, but there will be no question that the double system sound sounds better.  Attaching two RX or even an RX as large as an SR w/ a battery to the 5D hotshoe will not make you any friends in the cam dept..  Adding all that gak to one of the recent 5D "Battlestar" type rigs will make you even less popular....  I say keep it simple.

Philip Perkins

Share this post


Link to post
Share on other sites

If pluralEyes is being used for sync in post, I have found the easiest method for dodging the cam dept. and achieving the same "scratch" results is to have each camera equipped with it's own mic using it's included mount and "mic level in".  During my last 5 week 3 cam shoot (1d, 5d, 7d), PluralEyes had no problems sync each camera's reference sound to my recorder's sound.  No TC, no wireless feeds to cameras, no getting in the way of the cam dept.

Personally, I more comfortable working with a TC lock-it box scenario than relying on this software to do it's job correctly, but it's 2 for 2 now and everyone is happy.

Share this post


Link to post
Share on other sites

What did you do in instances when the camera mics could not pick up the audio?  Long shots, outside with camera inside, etc.  I guess at that point a cable or something is in order. 

Testiing a 5D next week.  If the client likes it,  then a serious lengthy shoot this fall is in the offing. 

thanks...

Share this post


Link to post
Share on other sites

I often use a G2 wireless to send a scratch track to camera I feed it via 1/8in out of my 442 simple easy. I would NEVER cable to a 5D at all  i always try to clap sticks so if it is a long shot or one without enough audio for pluraleyes they can just use the slate

Share this post


Link to post
Share on other sites

What did you do in instances when the camera mics could not pick up the audio?  Long shots, outside with camera inside, etc.  I guess at that point a cable or something is in order. 

Testiing a 5D next week.  If the client likes it,  then a serious lengthy shoot this fall is in the offing. 

thanks...

The answer for these scenarios is Slate.  I mean all it takes is just a few seconds of unique sound (like during the slate) for the program to identify what's what.  The mics on the cameras always picked up something as long as they were on. Their auto gain pumping everything to max seemed to be enough for PluralEyes to interpret/match my off camera recordings to the camera's.  This shoot actually had a fairly novice editor that had no issue getting the sync together regardless of the type of shot. 

Share this post


Link to post
Share on other sites

I work with  usually two 5D's on a regular basis. One of my clients has bought two of them and he uses them for most of his shoots. On other jobs they pop up on a regular basis. I have only had positive results putting sound into 5D's. I don't see it as my concern whether my clients use the sound that I am putting on the camera through a Lectrosonics 100 receiver or them syncing the files from my 744T. The client who owns the two 5D's, has even stopped taking the files from the 744T because he does not use them. This won't stop me recording them, but I just can't understand why everyone is so strident about this camera. Its just another job for me.

Image006.jpg

Share this post


Link to post
Share on other sites

the problem with h264 in editing is that it's an insanely processor intensive codec.  ProTools simply will NOT work with it and keep sync of any kind..even on my 8 Core monster tower.  I always get 422 ProRes Proxy files for audio post.

Since I'm a newbie, I asked the guys at Juiced Link for some advice when sending from my 702T to the JL231 to the 5D'..so this is what I did...

"The typical setting on the 5D is to manually throttle back its gain all the way (1 click above off), then adjust the CX231 to the peaks are splashing just past -12dBFS on the 5D meter."

Share this post


Link to post
Share on other sites

You should check your configuration

I edit 3-6 video channels AVCHD Canon 5d files with no problem on 6 core AMD with 8 Gigs of ram.

Adobe premiere has a native support for avchd

the problem with h264 in editing is that it's an insanely processor intensive codec.  ProTools simply will NOT work with it and keep sync of any kind..even on my 8 Core monster tower.  I always get 422 ProRes Proxy files for audio post.

Since I'm a newbie, I asked the guys at Juiced Link for some advice when sending from my 702T to the JL231 to the 5D'..so this is what I did...

"The typical setting on the 5D is to manually throttle back its gain all the way (1 click above off), then adjust the CX231 to the peaks are splashing just past -12dBFS on the 5D meter."

Share this post


Link to post
Share on other sites

ahh..thanks for the heads up. 

I should have clarified that I was specifically referring to ProTools not playing nice with the h264 files during audio post, and, at least on the Mac side that I've seen, the 422 ProRes Proxy files I usually get from Final Cut seem to work great.

Share this post


Link to post
Share on other sites

Sorry not a protools user.

ahh..thanks for the heads up. 

I should have clarified that I was specifically referring to ProTools not playing nice with the h264 files during audio post, and, at least on the Mac side that I've seen, the 422 ProRes Proxy files I usually get from Final Cut seem to work great.

Share this post


Link to post
Share on other sites

I agree with Christian on his sentiments about the 5D (and its little brother, the 7D) -- very cranky for audio. The lack of audio monitoring is a killer.

I just did a 5D project over the weekend, and we dodged a couple of bullets only because I insisted on checking the takes before we moved on to a different scene. On two occasions, the cameraman had bumped the mike-in plug, resulting in several minutes of static and ground-loop him. Quite a mess -- but we caught it, it got redone, and life moves on.

I hate working with equipment I can't trust.

--Marc W.

Share this post


Link to post
Share on other sites

Read all the 5D workflow posts and I have a question. Assume the following : 5D with new software, camera in 24P, recording at 23.97 fps, as stated in new 5D specs. What do I set TC in sound recorder at?

On the surface it would seem to me 23.97. However, having done a bunch of reading re: the Panasonic 900 cameras, FC, and RED, it looks to me like ALL video device's frame rates are either 29.97 or 59.94, based on the line frequency of either the 60Hz or 50Hz environments they were designed to operate in, regardless of whether the camera is placed into 24p mode. Have I got this wrong?

How about with the RED in 24P?

I've had several post people tell me they can deal with any time code to fit their needs in post. BS, or no?

Thanks.

Share this post


Link to post
Share on other sites

Read all the 5D workflow posts and I have a question. Assume the following : 5D with new software, camera in 24P, recording at 23.97 fps, as stated in new 5D specs. What do I set TC in sound recorder at?

On the surface it would seem to me 23.97. However, having done a bunch of reading re: the Panasonic 900 cameras, FC, and RED, it looks to me like ALL video device's frame rates are either 29.97 or 59.94, based on the line frequency of either the 60Hz or 50Hz environments they were designed to operate in, regardless of whether the camera is placed into 24p mode. Have I got this wrong?

How about with the RED in 24P?

I've had several post people tell me they can deal with any time code to fit their needs in post. BS, or no?

The reason for pull-up or pull-down is for going to, or from, film when its running at a true 24fps.  So, let's remove that from the equation for now and just talk about 23.976 and 29.97.

Good news:  The post people pretty much told you correctly -- it's not too big a deal (again, unless there's a speed change involved, and then it becomes a big deal).

It used to be that most projects were edited in a 29.97 timeline -- even those things shot on 23.976, so always using 29.97 timecode made sense (again, we're not discussing any speed changes to or from film -- that's a different discussion). 

As technology has improved, and NLEs have matured, even many projects that may end up being broadcast at 29.97 are now both shot and edited at 23.976.

So, typically, I will try to find out what timeline post will be editing on and match that.  Lacking that information, just match what the camera is doing and you'll usually be okay.

Most Red projects I've worked on have been at 23.976 even though people may call it 24.  The camera can do either, so just check and see which it's doing.

John B., CAS

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now


×