Vultures

Another possible problem: if every PP1 student is now wandering ABC Pool in search of users to comment on, are we unintentionally going to end up spamming our helpless user population? I’ve gone around to a few users and commented on their works, suggested that they consider submitting something to My Tribe, etc.. but I’m now thinking that if even one other RMIT student has given similar advice to this user, the “My Tribe Message” is somewhat cheapened. Correct me if I’m wrong, but it seems that the only solution would be to systematically check each of the user’s works for comments to make sure there are no My Tribe related ones.. and even then, the user may have been contacted via private message.
Maybe this problem hasn’t even arisen, but I don’t know. In my head, we’ve all become hungry vultures scouring the internet for users and, by extension, PP1 marks. It’s a harsh world I guess.

Adversity

So we finally shot our Screen Production segment. On Anzac day. At 6am. In a wasteland.

I’m happy with how it went. We had a great cast and crew.. I don’t think it would’ve been achievable otherwise. The odds were definitely stacked against us on this day.. I’ll list a few of our problems.

Phantom gear
I can’t explain it. Our equipment was possessed by demons. I set up the camera two nights before at Stevie’s house. Everything worked fine. But on sunday morning, the aperture ring was set to auto, the white balance was set to 25,000K, and no settings or details were displayed in the viewfinder at all. And apparently no one touched the camera after I’d set it all up…
I managed to fix all of these problems within ten minutes (the hard part was getting details displayed on the viewfinder again.. there’s a flick-switch on the top-front of the DSR, attached to the viewfinder, that toggles displays on and off). But in post, I discovered yet another camera problem.. the back focus was out. Not enough to be visible in the viewfinder, but it’s quite significant in our wide shots.. damn. I know, the theory is that you meticulously check that all of your gear works before shooting.. but I stupidly assumed that since the backfocus is always locked, it would’ve been calibrated by the techs and stayed in the right place. stupid stupid.
Anyway. Our audio gear was even more possessed than our camera gear. Every 15 minutes, the mixer would die a horrible death.. all sound became heavily distorted, then the VU meter stopped working. To fix this problem, we needed to put in fresh batteries.. which kept the mixer alive for a short while.. before it died again.. and so on.
Also, the radio mics may have been set to similar frequencies.. or else it was just another mixer problem.. we couldn’t get rid of this high pitched hiss.. probably should’ve looked at the radio mic frequencies, but we were working against the clock and i didn’t want to further mess with what was already half-working.

The Great Outdoors
I always thought that this would be our biggest problem on this shoot… we had an army of C-stands, blacks and cutters standing by, but nature was fairly nice to us. It was windy, but we were able shelter our lapel mics under layers of clothes, so audio is very good considering the harsh conditions. Lighting continuity was a pain at times.. we’d get the perfect cloud come over, then miss it because of a debate over something trivial, then have to wait 20 minutes for more cloud.. etc. But Emma J and Stevie kept track of which shots we needed when it was cloudy and which we needed in harsh sunlight.. so each scene of our film has quite consistent lighting, which is what we were aiming for.
Rain hit us late in the day, minutes after we’d finished up a scene, so that was good timing.. we took a break for 40 minutes and then shot our final scene with plenty of light left.. all good.

The Terrain
The only thing worse than shooting outdoors is shooting outdoors in a wasteland. Our camera positioning was limited by the lack of level ground and the abundance of rocks and thistles. Setting and resetting the jib in this location was especially difficult, though having Jim Elson onboard as camera assist was a big help in this respect.

And yea. Other than all the things mentioned above, everything came together well.
If I could go back and do it again however, here’s what I’d change:

Less jib.
At the start of the day, I thought that we’d be able to keep the camera on the jib for 90% of our shots. We could hold it steady for “locked off” shots and have the added flexibility of being able to move the camera without resetting the tripod position. However.. I didn’t factor the strength of the wind into this equation. No matter how hard you’re trying to keep the jib still in 35km/h wind, you’ll see considerable movement in your mids and tight shots. Unfortunately it took me a little while to figure this out, and some of our starting cutaways aren’t as stable as I’d like (I think we’ll just slow down the framerate in post where necessary).

More depth of field.
I’m a big fan of shallow DOF.. and this back-focus problem has shown me that this obsession is becoming unhealthy. In the wide shots, I could have easily stopped down the lens and overcome our back focus problem.. I was using the strongest ND filter on the DSR, plus increased shutter speeds.. so there was plenty of light available to close down the iris. Unfortunately, I kept shooting everything at around F/2, because that’s what I wanted for the tight shots. In the future, I may make a policy of going for deep focus in my wide shots because there’s no reason to shoot a wide shot at F/2 anyway, and stopping down will protect against any more back focus issues (I’ll be checking the back focus religiously before shooting now as well…)

More testing of equipment
This shoot was obviously a sign from the Gods.. a warning.. an 11th commandment — “Thou shalt do a freaking equipment check and never trust RMIT gear”. And I’d be stupid to ignore such warnings. It’s good that we’ve gotten these problems out of the way prior to PP2 filming.

Anyway that’s enough lamenting and self-criticising. I enjoyed the day, and I’m very pleased with the way our footage looks. In particular, I’m proud of the shots in which I made the actors dance with the jib arm. They’re fairly wide angle shots with macro focus, as the actor needs to be quite close in order to hold onto the jib arm.. it’s quite a disorienting effect.

That’s probably enough for now.. so.. thanks to everyone who put this monster-shoot together. we had Sam giving us great sound despite horrible mixer problems, Emma J keeping us all on track as 1st AD, Stevie keeping our film cohesive and guiding our actors as director, Emma H doing god knows what and disappearing mysteriously to do something amazingly helpful that no one else realised had to be done (ie producing), Jim being a great help to me as camera assist and taking awesome stills to enhance our street-cred, Mal driving us (and our gear) all over the place, Sara Stone covering the set in blood, and our brilliant cast of actors — Peter Flaherty, Steve Sheeran and Kara Bert — enduring the elements, dancing with cameras and generally giving great performances.. was awesome. Thanks everyone.. next time we’ll shoot the film in focus. I promise.

Macroscopic

We’re shooting soon. And as you may recall, my plan was to create disorienting, distorted visuals by shooting as wide as possible on the 9-155mm Fujinon lens. Panavision’s DOF calculator led me to believe that at 9mm, my focus would still be reasonably shallow with the subject 60cm away, 45cm away, etc. But today, while stopping in at Stevie’s house for a beer, I opened up the DSR case to try this theory out first hand. And it’s extremely lucky that I did so.
It turns out that even with the macro ring, at 9mm, I had to be less than 10cm away from a subject to achieve reasonably shallow focus. On one hand, it’s impressive that I can focus the DSR at objects so close to the lens.. but obviously, it’s hardly a viable way to shoot a film.

I was about to conclude that we’d need to shoot the film at much longer focal lengths and then add fisheye filters (or something) to the footage in post to make it disorienting.. when I stumbled onto something. I noticed that the macro ring has a much greater effect on DOF when the lens is set at longer focal lengths.
“Wait a minute Josh. If you use a longer focal length, you’ll need to move your camera backwards — and with this increase in subject distance, you’ll be able to focus normally without using the macro ring at all.”
Yes, this is true. But I don’t *have* to focus using the focus ring..

This sounds ridiculous, but what seems like an option is to wrongly set the focus ring and then correct the focus using the macro ring, while using a reasonably long focal length (40mm.. 80mm.. somewhere in there). And when using this “technique” on subjects that are fairly close (around 1 metre away), DOF becomes amazingly shallow.

“But Josh, you’re not going to get barrel distortion when using the camera at between 40 and 80mm. What happened to using 9mm??”
To be honest, the barrel distortion I managed to get at 9mm was fairly negligible, the depth of field was too great and the subject distance had to be microscopic in order to get a reasonable-looking tight shot.. so.. “The wide closeup” isn’t all that feasible. It’s disappointing, but I’m now too interested in this new macro-possibility to care all that much. What this new use of the macro ring means, if my informal testing is correct, is that we can create shallow depth of field on much shorter focal lengths than I previously thought possible with a 2/3″ camera.

“But Josh, for most purposes, this ultra-shallow DOF will be unusable — an actor would have to stand completely still because you can’t pull focus using a macro ring.”

Alright — first off, as we’re shooting outdoors, I can always stop down the lens to achieve whatever DOF I want (within reason). And second, I don’t forsee this film requiring a great deal of focus pulling.. we can work around this problem.

Anyway. This is all quite exciting. But I’m slightly dubious… I’ve never heard of anyone defocusing and compensating with the macro ring in order to enhance their shallow DOF. Maybe it’s such a ridiculously bad idea that it’s never even discussed. But maybe it’s that most people shooting on a DSR-450 aren’t overly concerned with shallow depth of field.. and because when you’re shooting on 35mm, you don’t have to resort to such drastic measures to achieve shallow DOF anyway.
As long as optical quality remains sufficient when using macro, I can’t forsee any significant drawbacks to using this “method” I’ve come up with.. and in any case, I’ll find out the hard way soon enough.

Bloody Sunday

This sunday is looking pretty bleak, and there’s no backup plan. According to the Bureau of Meteorology, our outdoor shooting day will feature the following: “Partly cloudy. Shower or two clearing. Min 12 Max 18″ with “Winds west to southwesterly averaging up to 35 km/h decreasing below 20 km/h during the afternoon”. Right. So that’s basically the worst possible scenario, apart from torrential rain or cyclones. The cold weather will cause our camera batteries to die faster, the “partly cloudy” will make lighting continuity extremely difficult, and the 35km/h winds will wreck our audio.. this will be interesting.

The wide closeup

We had a production meeting for “The Tip” earlier this week where we finally met our screenwriter. He’s somehow wound up writing scripts at RMIT after being a director at the ABC for 20 years… crazy. Anyway, he suggested that to further convey a sense of “wrongness” in our film, wide-angle distortion in our tight shots would be a good idea.

It’s something that I haven’t tried before.. I generally fall into the trap of using a longer focal length and keeping my camera at wide-shot distance from the subject for CUs. But if I set the focal length at 9mm rather than 85mm on our 2/3″ 3CCD camera, we’d have more of a chance of getting barrel distortions and slightly warped faces in our tight shots. It’s not hard.. just move the camera forwards. A lot.

Generally, on the DSR-450, I set the 17x zoom lens at ~85mm for closeup shots.. the subject is maybe.. 4m away (the distance is a guess.. which is problematic, but I’ll work with this assumption). At F/2, this setup will give me a DOF of 18.22cm.
Now.. say I want to get the same frame with the focal length set at 9mm… I’ll have to get pretty damn close.. less than a metre away I’d wager.
So.. with the subject 60cm away, using a 9mm focal length at F2, my DOF is 40.58cm. Not quite what I had when using 85mm, but still shallow, and vaguely usable (there’s room for the actor to move.. somewhat).
And as subject distance decreases, the DOF falls away rapidly. At 45cm (again with 9mm lens at F2), DOF becomes 21.7cm, similar to what I had with an 85mm lens 4 metres away.
At 30cm, DOF becomes an unusable 9.26cm… it might look fairly twisted, but the focus would be extremely unforgiving..

Now the next question is: how close can I focus on a DSR-450? Well I’ve searched on google for lens-information. It turns out that our DSRs at uni are equipped with Fujinon A17x9B RM lenses.. you’d think they could’ve come up with a nicer name. Anyway, the M.O.D (minimum object distance) on these lenses is 90cm.. most likely not close enough for my purposes. But there is still hope — the lens has a macro focus. So no matter what the “M.O.D” is on the DSR, the macro should let me in as close as I want to go, especially at 9mm focal length.

So my current plan is to use the macro focus and get closeup shots using a 9mm focal length, or thereabouts. For locked off shots, this should be fine. But there’s a dance sequence, and we’ve hired a jib arm for it… this should be fun. I’ll have to see on the day, whether it’s viable to use a jib so close to actors (most likely I’ll need to zoom in a bit and move the camera back to avoid any unfortunate accidents.. but I’ll still try it at fairly wide angles just to see what works)

I’m sure that all this has been done before, thousands of times. It’s hardly groundbreaking. But it’s new to me, and that’s good enough.

vs Nature

We’re shooting in a wasteland.

Why do we do this? Why do we further complicate the already challenging process of creating a short film by placing ourselves in the hands of the infamous Mother Nature? Despite all this, I am looking forward to this shoot. It’s ambitious.. most likely too ambitious. But it’s the first outdoor shoot that I’ve had any control over, and I’ve worked on several naive outdoor projects that have gone horribly wrong.. this is a chance to rectify all the mistakes I’ve seen in the past. But at the end of the day, if nature wants to fuck us over, that’s what will happen.

I’m not sure who will win this battle, but we will come as prepared as possible.

I want an army of C-stands, shot bags, blacks and cutters. Maybe.. 8 C-stands, 10 shot bags.. if that’s at all possible. It will be epic. We can shield our audio from excessive wind noise, we can create shaded areas for the sake of lighting continuity.. I’ve found it quite useful in the past, on our last nightmare outdoor shoot (see previous posts).

We’ll probably need wireless lapel mics, since wind and traffic noise can be terrible out in industrial areas (though we’re yet to do a location scout). This then means that we need clothing for our actors that will conceal said lapel mics without creating excess scratching noise.. that could be difficult. But I don’t think a boom mic will cut it in these conditions.. we’ll find out soon.

Obviously we’ll need millions of umbrellas, tarps and garbage bags in the event of showers (it always rains when you’re filming. Australia’s drought and Australia’s struggling film industry are probably linked in some way..)

Our lighting will consist of a reflector or two, and that’s about it. I’d really like a giant frame of grid or diffusion.. but I don’t think RMIT has any. I guess that this film is meant to be harsh and post-apocalyptic looking.. so clipped highlights won’t be horribly detrimental.. but still. I wish we had some grid, just in case.

Then there’s the question of what we actually shoot on. Ideally, we hire out a DSR-450 from the techs.. but because this film is made for an elective subject, TV1 students have priority over us. We’ll lose our booking to TV1 students doing a test shoot, let alone an actual shoot.. so we’re in limbo. I wish I could get a straight “yes” or “no” out of RMIT, because it’s hard to plan without knowing what you’re shooting on. The other alternative is shooting on a 5D mkII, but if we want one we’d need to book it right now. RMIT has two 5Ds, and they sell like hotcakes. Also we’d have to endure great disapproval from various techs who are not fans of HDSLR shooting (can’t say I blame them.. I feel so dirty even asking them about using the 5D..). Worst cased scenario is we shoot on an HVR-Z7p.. or a Z5, or a Z1.. ugh. I will hate my life if this happens.. such a small, noisy sensor, such a wide, slow lens.. such tiny buttons.. (that’s probably enough bitching, I’ve already said this all before..)

No matter what camera we end up shooting with, dynamic range will be a big problem. Of course that’s part of the post-apocalyptic aesthetic, but I’d rather create that effect in post rather than work with clipped highlights right from the start..

And then, just to make things even more difficult, we’re thinking of hiring RMIT’s jib arm for the day.. there’s a dancing sequence at the end of the film, which sounds like a good opportunity for some strange camera movements.. I’m all for it. But it’s one more thing to set up. We’ll need to rock up at dawn for this shoot, with our location locked down and a detailed weather report. Then set up the army of C-stands, all the audio gear, camera, jib, art department set dressings, actor’s makeup… It’s going to need to happen fast, like a secret military exercise. Sounds good. But when you’re working outside, there’s never such thing as a guarantee.

5D TV

I’m shooting a short film. It will be shown on television (C31). That’s pretty much guaranteed.

However, I cannot guarantee that this short film will be any good. It will be shown as part of a larger project.. a 6-part TV series on Channel 31, called “The Temperature’s Rising“. As such, it must to some extent address the issue of climate change.. though it may be as abstract and non-literal as we want. The main problem I’m currently facing is that we’re waiting on a mysterious group of screenwriters to come up with 6 different short film scripts, one for each episode. We’ve got to start shooting in around 3-4 weeks, and I have no idea what the script will be, or which production crew gets which script.. it’s a bit of a nightmare. I want to make something visually impressive and reasonably interesting, but it sounds like these writers are intent on turning the short film segments into small sketch-comedy routines.. *shudder*…

Hopefully there’ll be *one* passsable script that we can steal, I should find out later in the week. Until then, there’s not much I can do for this project, despite the looming deadlines. There are 65 students working on this TV series, and everyone’s kept in the dark. Tres frustrating.

Anyway, despite not knowing anything, I can speculate…
I know that I’m a “location DP” for one of the short film segments. I’ll be working with a location director and producer, and all other crew will be outsourced. I want to shoot on a DSR-450. It’s great to operate, it’s got a 2/3″ sensor and a reasonably long lens (9-155mm, f1.7 i think). But this may not be possible. 2nd year media students will be borrowing DSRs in the coming weeks for TV1, and they get the priority over Screen Production students. We’ll probably end up shooting on something horrible like a Z1. I figure that if worst comes to worst and we’re stuck with a Z1 (or worse, a Z5), we should opt for shooting on the Canon 5D mk2.

“Oh god!”
“Is he serious?”

I can assure you, I’m not a fan of HDSLRs.. I’m quite dubious and I realise that there are many logistical problems that come with shooting on a 5D. But I’m not getting stuck with a Z1. I’m not getting stuck with a 1/3″ sensor and a 4.5-54mm lens that stops down as you zoom.. no. At least shooting on the 5D would be a learning experience.. I hear so many whisperings about shooting on this camera that I may as well give it a shot and make up my own mind. I realise it’s not ideal, but if we don’t have access to a DSR450.. it’s the next best option.
What I’d need is a follow-focus kit, or at the very least a baseplate so that I could attach it to a fluid-head tripod. Hopefully that’s viable. We’d need to post-sync sound with solid state recorders and slating, which may be good practice for later shoots on 16mm or RED. I’m quite worried about how I’m going to focus accurately, given that my usual “crash zoom, focus, zoom out” method won’t work for still camera lenses… hopefully the push-auto function will be sufficiently accurate, and hopefully we can get away without millions of focus pulls. We’d probably need a data wrangler onset to ingest and wipe the CF cards for reuse, depending on how many CF cards we have access to. And then there’s the lenses.. RMIT has a 50mm F1.2, L-series lens which should be fairly awesome. And then there’s the kit zoom lens.. can’t remember the specs for that. (maybe a 24-105mm L series f/4 zoom? not sure)
In any case, the fast 50mm and the kit zoom would most likely be flexible enough for our shoot, and it would be a good chance to get a sense of full-frame 35mm shallow depth of field.. good practice before shooting on RED (which isn’t quite as shallow). If I can convince the techs to let me shoot with it (it’s a polarising issue down in the tech’s office.. lots of different opinions), I’ll deal with the consequences and hopefully end up with something that looks half-decent.
What happens in post?
We could try cutting in native h.264.. probably that’s a horrible idea, but we’ll see. It’s easy enough to transcode to ProRes, that may be best for all involved. And while I’ve been bitching about h.264 compression on HDSLRs, I noticed in the manual that the 5D captures video at 330MB per minute. That works out to 5.5MB per second, whereas HDV captures at around 3MB per second. And after cutting a large project in HDV, I can say that for the most part, it looks decent. Not amazing.. but passable. Of course, HDV is only compressing 1440×1080 images to 3MB/s, whereas the 5D is compressing full 1920×1080 images to 5.5MB/s, so the compression ratio may be just as harsh as HDV. But now you’ve got to account for the fact that HDV uses MPEG-2 compression, and that according to Apple, h.264 is much smarter than MPEG-2, achieving similar results with half the data rate.
I’m skeptical, but assuming this is true, the 5D’s h.264 @ 5.5MB/s may be equivalent to 11MB/s of HDV compression, which isn’t nearly as harsh as I was imagining. I’ll find out in post I guess.