Elaborate production - input wanted

David W. Richardson

Chapelgrove Films
Okay, I want everyone's input on this. I'm especially keen to hear what actors think. This is going to take quite a bit of explaining, so bear with me.

I have a feature project in development called 'The Circle'. It's been in development for quite a long time because I can't figure out a way to shoot it without it costing a good chunk of money -- money that I don't have and can't seem to lay my hands on.

It's the story of 9 people who are conversing in an online video chat room when one of the group is suddenly attacked, stabbed and left for dead -- and the others have to figure out where she is located so they can send help before she bleeds to death.

Now this is a real-time story. What you'll see on the screen are 9 boxes -- the 'video chat windows' of each of the characters. The boxes are arranged in a checkerboard pattern, each with a different character in it. (Each character is sitting in his/her home or office, in front of their computer, chatting via webcam.) Every character is on-screen the entire time. There are NO CUTS.

And that's what presents the challenge.

My original plan was to build 9 small sets, located on a single soundstage. These sets would be arranged in a circle, so that everyone is facing each other. This way they could see and hear each other when performing the script. The director would sit in the middle of the circle.

Each set would have its own dressing, props and lighting. And each would have its own camera and microphone. Tape would roll on each camera, and in post we would composite all of the footages together to make the checkerboard "video chat room".

On the one hand, this would make shooting very fast. The script is performed like a play, in real-time. So we could shoot the whole movie a total of 3 times in just 2 days.

BUT this would also be very expensive. Not only would we need a soundstage and 9 constructed sets, we'd also need 9 cameras, 9 tripods, 9 mics, and 9 sets of lights. It adds up in a hurry!

So I've been trying to think outside the box to come up with ways to cut down on the expense. If we could shoot each person individually and cut them together in post, that would save a LOT of money -- because we'd only need one small room (re-dressed for each character), one camera, one mic, one lighting set. But since this is a real-time story and everyone is seen on-screen the whole time, there are NO CUTS. So getting 9 people's dialogue to time out right so it all fits together without cuts would be next to impossible.

Or would it???

You see, I've had this idea. (And this is what I want opinions on.) See if you think this scenario would work...

We get the whole cast in a room around a table. They 'perform' the script from start to finish, doing all the dialogue and actions just as if we were actually shooting the movie -- but all we're doing is recording the audio. We 'perform' this as many times as we need until we get the best take.

Now we setup a room dressed for Character #1. We make a version of the recorded audio that has Character #1's lines removed -- but everyone else's dialogue is intact. We bring in the actor for Character #1 and shoot his/her bit, using the audio recording of everyone else's dialogue for the actor to play off of. This puts his/her dialogue in the right places with the right timing.

We re-dress the room for Character #2. Make a version of the recorded audio that has Character #2's lines removed -- and repeat the process.

We do this for each of the other characters until we have all 9 of them shot. Then we take the footage into editing and put it all together in the checkerboard layout.

That should work, right? Technically, I mean. Are there potential problems I'm not thinking about? Granted, it would take longer to shoot this way. We'd probably have to devote an entire day to each actor's shoot. Still, the overall costs should be much much lower.

And you actors -- what do you think of this? I realize it's always best for actors to be able to play off of real live people. But do you think this is a workable option?

Any input would be greatly appreciated!
 
so this will be a full length, circa 90 minute movie? With no cuts for nine separate video streams? Ouch... For a 5 minute short film i can see it working technically fairly easy, but for a full length movie I can see it getting real difficult to make.

some thoughts:
- after the victim is killed, what appears on their video 'square'? - their empty set (could get boring 80 minutes later, unless plot-relevant stuff is also appearing sometimes), or blackness or static or a 'user disconnected' screen at some point after the death?
- staring at a 3x3 checkerboard of videostreams could be hard on a lot of viewers at that length, we're not used to it
- you'll definitely need some sneaky way of inserting cuts. The obvious one would be for actors to move in and out of frame now and then - with locked down cameras, such a cut won't be visible and also wouldn't be unnatural.
- And surely some of the others will have good reason to log off now and then (to find info, change the baby, answer the phone, whatever) - that will also give some visual variety and hidden cut potential

I'm not convinced that your first 'shoot all 9 simultaneously' idea would necessarily be so expensive - if you're simulating 9 people staring into their fixed webcams (ignoring the possibility that one or more has a laptop and so can move around with it), then surely you're talking sets maybe 2 metres wide each at most? And as webcams are fairly nasty video quality, you'd get away with filming via cheap video cameras or even actual webcams - once you stack them 3 x 3 onscreen in your editor, that will make up for the lower res of otherwise inferior cameras. Similarly lighting doesn't have to be flash at all. You do need good (rather than 'realistic') sound though.
So a decent sized room (big enough for 9 mini-sets and a director) with sound-absorbent walls would seem ok to me.

do you actually have a video-chatting application that can set up something roughly like what your film will show? I ask because the best way to film this would be with them literally looking at each other on the real video chatting screens in front of them, and not being able to see each other in the flesh at all - that way you'll get nice natural eye movement.

so, you'd need 9 mini-sets, computers, webcams/cameras, microphones, and an electronics/video geek to network the computers together.

Hmm, hypothetically this could even be filmed in a real internet cafe, dressed with different backgrounds to look like 9 locations?
 
Thanks for the response, clang. First let me address your point...

- after the victim is stabbed (not "killed" necessarily), she is out of the view of her webcam (lying on the floor) so we continue to see her empty set. We do hear her coughing and wheezing occasionally, letting the others know she's still alive but unable to communicate.

- I know staring at the checkerboard will be completely out of the norm for the audience. I'm hoping they'll get so engrossed in the people and the tension of the situation that they will soon forget that, however. It's a risk. For what it's worth, the overall run time will only be about 70 minutes -- I think we'd be hard pressed to keep the audience for 90 minutes. And it will be comprised of a total of 3 scenes. In the first scene, there is only 1 person and 1 set, which will fill the whole screen. The 2nd scene is the bulk of the movie -- with the 9 squares/people -- and runs 60 minutes. The final 5-minute scene is the same 9 squares/people, but takes place a few weeks after the main part of the movie.

- I think if all 9 characters move out of view of their cams and back in, even once or twice, that's going to look odd. I much prefer to work this out with the characters glued to their computers.

- This group of people meets online once a week in this chat room. Nobody will log off to do other things -- especially not after the one girl is attacked. They are extremely focused on trying to save her life.

Now, on to your thoughts about the possibility of shooting them all at once. This is actually the way I'd truly prefer to do it, if at all possible. I had planned on using consumer quality, single-chip camcorders for this, which can be had for $200. But these cameras generally have no external mic jack, so it would be necessary to have some sort of audio recorder for each person, plus a QUALITY microphone for each. I could use a recorder like the Zoom H2. It has 2 mic inputs, so one recorder could cover two actors' mics. Still, that would be at least $1800 for cameras, $950 for the Zooms, plus the cost of 9 quality microphones -- at least $200 apiece. And tripods and mic stands.

I don't think using actual webcams is feasible. The ones I have experience with don't give me good, fluid motion capture. There is always some degree of hesitancy with them. And each would have to be connected to a PC of some kind, recording to the computer's hard drive. Add 9 PC's and monitors to the budget. And add all that PC fan noise to the audio problems. I'd much rather avoid all of that.

That also precludes the possibility of setting up any kind of software to display all 9 webcam feeds on monitors in front of each actor.

You're right that the sets would be fairly small, but they would still require materials and construction, and that ain't cheap. Shooting one person at a time, I could use actual rooms in actual homes, which would save money.

I had hoped I could find a company willing to do a product placement deal with us. B&H Photo and Video, for instance, who could provide us the camera and audio gear we need in exchange for a prominent 'banner ad' on the simulated 'video chat room' website -- an ad that would be visible for virtually the whole movie. After shooting, we could return the gear to them and they could sell it in their 'used' department. But I haven't been able to interest any company in that kind of a deal.

I do appreciate the ideas. And believe me, if a way can be found to shoot this thing with everyone in one place at the same time, that's the way I want to do it! I wish I were better at digging up investors -- all this project really needs is a decent budget.

Thanks!
 
Wow,

You are going to need pretty amazing actors for this to even be possible. You might want to consider splitting it up between the webcams and a standard hollywood style film making. 70 minutes of 9 small screens with 9 character voices overlapping is a logistical nightmare. Not to mention the presentation. Maybe it can fly on a 70 foot screen. But on a small television or a computer screen? That is going to be insane with the information.

I don't think an audience would be able to take staring at the same 9 characters, essentially not moving for 70minutes. Hell, I can't even carry on a 70minute conversation online and have updates that doesn't require a minute to five minute wait time.

I just can't foresee the conversation continuing for 70minutes. Also, the best way to go about that might be to record the dialog of each character - assuming we are working scripted in one location at one time. And have the other dialogues play. But to physically do it 9 at a time - you would need sound proof rooms so we don't hear everyone else talking at the same time.

As much as the idea of product placement is a good one - it still presents the issue of wether or not this thing can be distributed.

You might want to re-look at the idea. I mean, I guess it could work. I just don't know how well it will translate for the audience.
 
Okay, I knew it would come to this. Let me try to say this without sounding harsh or ungrateful...

I'm not looking for advice as to whether or not the concept of the project is a good one. I believe in the project. I believe in the vision I have for the project. It won't win an Oscar. It won't make a zillion dollars. And there will be plenty of people who just won't be able to get into it for one reason or another. But I believe it WILL find an audience that will enjoy and appreciate it. And it WILL earn a profit for the investors. It may even develop a cult following.

So I'm not looking for input on whether or not to DO it the way I want it done. I'm looking for advice on the best way to EXECUTE the shoot so as to achieve the result I'm looking for. Not that I don't appreciate the concern, I really do. But I've already settled that aspect of it in my mind. I believe I can make it work. What I need now is advice on the mechanics of making it happen.

I hope that doesn't offend anybody. :)
 
I can see where you're coming from believing in your project, but just as another suggestion, watch TimeCode as that's 4 continuous takes with 4 boxes and I got confused even with that.

Back to the technical problem though, the answer seems pretty obvious to me. Get the 9 sets to record all at the same time, and simply rent all the equipment. Drive a few hours to a big city if need be, but buying 9 setups is completely ridiculous especially if you're just filming a few days (as would be the case if you film everything at once). You can find cheap rentable cameras with mic-inputs you just have to look harder (or find filmmaking friends who you can borrow cameras from). Or you can rent a multi-track recorder many of which can record 8 tracks simultaneously.

Good luck.
 
Is it script-based?

Instead of filming everything all at once, maybe get everyone together for a table-reading, audio-record a good version of that, then film each actor individually as they react and act to the table-read audio.
 
I can see where you're coming from believing in your project, but just as another suggestion, watch TimeCode as that's 4 continuous takes with 4 boxes and I got confused even with that.

Back to the technical problem though, the answer seems pretty obvious to me. Get the 9 sets to record all at the same time, and simply rent all the equipment. Drive a few hours to a big city if need be, but buying 9 setups is completely ridiculous especially if you're just filming a few days (as would be the case if you film everything at once). You can find cheap rentable cameras with mic-inputs you just have to look harder (or find filmmaking friends who you can borrow cameras from). Or you can rent a multi-track recorder many of which can record 8 tracks simultaneously.

Good luck.

I have seen 'Time Code' and agree it's confusing as heck -- but that's because you have four separate things going on at the same time. This is more like 9 people in a room having a conversation together. So it should be a whole lot less confusing than 'Time Code' was.

I live a stone's throw from Charlotte, NC. The only rental gear available here is the pro stuff -- like DVX100's and such. And I doubt there are 9 rental cameras in the whole county. Even if there were, it would cost more to rent those than it would to buy the cheap consumer-level cameras outright.

I considered a mixer or recorder, but finding a decent one with 9 inputs has been a challenge. And again, cost is a factor.

I wish I had enough local filmmaker friends to dredge up that many cameras and mics. :)

I actually considered making owning (and bringing) a MiniDV camcorder one of the requirements for actors to be cast in the movie, but that didn't really seem fair.

If I had the money to buy the gear up-front, then I could recover most of it after the shoot by selling the gear. And that's a possibility. Wish my credit cards weren't all maxed out. :(
 
Is it script-based?

Instead of filming everything all at once, maybe get everyone together for a table-reading, audio-record a good version of that, then film each actor individually as they react and act to the table-read audio.

Yes, it is script-based. And that's exactly what I was asking about -- whether doing an audio-record of a table reading could be used to help set the timing of dialogue when shooting each person individually. It would be a very involved way of accomplishing the shoot, but it may be the only approach that fits my pocketbook.
 
Sounds like a great idea. I remember a few years ago there was a news story in the UK of a girl was trapped off shore in Australia and for some reason her phone worked once and she was able to call her Mum in England who then raised the alarm to the coast guard in Australia.

As a point of reference check out Timecode by Mike Figgis. It is similar is as much as it plays over four screens in real-time with no cuts.

You'll be able to get away with an awful lot in terms of little "jump" cuts to catch up with time...as something draws your eye to someone talking on screen you can probably cut one of the screens to catch up (or speed ramp slightly every now and again).

So if it is well rehearsed and you have someone on clock I don't see why recording one at a time wouldn't work. You would probably want to have the other actors there to read-in their parts.

You'll also have plenty of options in post to change it, if you find that 9 on screen "screens" is too much for the eye...Think of the "24" effect where one screen becomes prominent and the others become smaller.


If you shoot HVX you have the chance to make a film that goes way beyond anything out there in terms of resolution...7680 x 3240!!!!

Look forward to hearing more about the project!!
 
You can sneak cuts in there by having an actor walk off camera (to go pee or something) and cut to a later time when they return. A static shot of a room is easy to cut when nothing changes.

I think a project like this really demands that you do it real-time, for the sake of the acting. If you shoot on a stage, you will have problems with isolation. The sound of one person screaming will bleed into the other sets' audio.

I would suggest webcams and the mics that they have. It will be more "real" that way. Afterall, it really only matters about the story and the acting, not the picture quality.
 
I don't think anybody's going to wander off to pee when a young girl's life is hanging by a thread, do you?

I thought about the sound bleed issue. But since all of the dialogue will be in mono and basically come out of the two front speakers, nobody's really going to be able to tell that the SCREAM from one mic was picked up on another. If I see it's a problem in post, I can just drop out the audio from the other tracks during the SCREAM. But I don't really think it will be an issue.

I do agree that doing it 'real-time' would be best from an acting standpoint. It's just so much more expensive that way.

And I don't want to use webcams and their crappy microphones. Picture and sound quality DO matter to me. There's real reality and there's 'movie reality'. I'm aiming for 'movie reality'. :)
 
Shoot it on the set of Hollywood Squares. :thumbsup:

Seriously, I'd shoot it all simultaneously. Charlotte is a massive town with a bigtime dealer (Platts) and probably plenty of DVXUsers with cameras nearby. I'd be shocked if we don't have at least 8 other DVXUser members nearby; I can think of one already who should be able to help out. David and I might even come out, and that'd be an HVX and two DVXs right there.

And, heck, they don't all have to be HVXs. If you're taking up 1/9 the screen space, you can use DVXs, you can use XHA1s, you can use HV20's. You could easily shoot the whole thing on 9 HV20s for that matter. If it's a static locked-down shot, you could shoot Aipteks. By the time you rescale down to 640x360, anything is going to look fine. At a 1080p frame size, each person's window would be 640x360.
 
Last edited:
Hi,

I think, if I get it right, you could shoot the thing with web cameras, which is how I would likely do it.

If your bent on doing the project, do it as a play, recording each of the work stations to the drives, by script, in the same room with web cameras and highly discreet microphones to isolate sound with the director and crew to monitor everything in real time.

Tic-tac-toe the cameras up on the time line in sync for output. I can't think anything more simple or less complicated to accomplish what I think I see your plan to be. If color is off a bit, so much the better. Same with sound and the rest - so long as it isn't boorishly bad.

Cost shouldn't be outrageous.

Of course, I might miss your vision.
 
As an FYI, Canon has put 1/8" mic jacks back on their new DV consumer cams, the ZR900 series. Those are listed at $249 on Canon's website, but they're probably going cheaper than that somewhere. Or everywhere. :D Still, it's not exactly cheap to buy nine...
 
one other thought - and I'm saying this without any idea of your script, so it may be completely impractical - but 9 characters is quite a few, and is causing most of the problems with the 'shoot everyone at once' idea. Would it be possible to rewrite to only have 4 characters? - or to be more precise, 4 at any one time, coz I think the concept still works if one or more actors logs off at various times and others log on (dramatically justified by them pretending to use software that only allows 4 simultaneous users - that could actually help the story, a person with vital info might be desparately trying to log on but can't because 4 others are already connected, and there can be background e-mailing or cellphoning going on.) Obviously, the same idea with just 4 screens would be a whole lot cheaper and easier...

But I totally understand if you have a finished script that absolutely needs exactly 9 'screens'. :)
 
Ok, so if you shoot it all parallel, then how many takes to get all 9 actors to do it right...I mean get it right at the same time?
 
I would find a recording studio, set up a green screen behind the actor (all at separate times), run an indiscreet monitor in the actors ear and have the group reading outside the room for each take.
 
good point hazelwould - given the static nature of the shots and the relative unimportance of the backgrounds, greenscreen would certainly be worth considering if David goes for the 'shoot consecutively' option

and your mention of monitoring also raises another issue - David, are these people all sitting at their computers wearing head-phone/microphone headets? That certainly gives some opportunities for letting them hear something other than what the audience will hear, e.g. live direction from the director
 
Back
Top