Comparing Camera Systems for Feature Films

[Brian Drysdale]

Originally Posted by Sareesh Sudhakaran

Dynamic range is important. However, the differences in DR comparing an HDSLR to the F3 isn't that great to cover up bad lighting. Just experience speaking. One can screw up even an 18-stop DR with a poorly planned and executed shot.

People tend to use the higher dynamic range because the hiighlights tend not to clip so badly.

Yes, the 80% of performance is relatively cheap to achieve, the extra percentage points cost a lot more, You're comparing the recorded signal, but not taking in the compromise system that the 60D uses for extracting a video signal from the stills sensor. I've heard people comparing it favourably to the the 7D, but this is the first time I've heard it suggested as a better option to the F3.

"The Blair Witch Project" was supposed to have been shot with the film school cameras that the students were using, so they could've filmed with a mobile phone or a F3

[Charles Papert]

Sareesh, you are making a lot of definitive statements that are going off mark, I think. If you are talking about a $100,000 feature, chances are that unless you are fortunate enough to get very skilled people working for you for essentially no money, there will not be enough time, gear and crew to be able to perform the kind of workarounds you are suggesting to some of the issues presented by these cameras. Take a typical day exterior situation. Shooting closeups all day is relatively easy: it doesn't take excessive gear or difficulty to control the sun even at its worse in mid-day (although it does take a little bit of skill to produce natural looking closeups under artificial conditions). Once you move to wider shots, controlling contrast becomes a more complicated task. On my larger jobs, I think nothing of asking for a 20x20 ultrabounce or bleached muslin, but on a feature that size, I'd surely downgrade it in size due to limited personnel (and likely experience levels) to ensure that it is rigged safely. With a higher dynamic range camera like the F3 in S-log mode, I may not even need it all, and that is a significant consideration. For a job that has a lot of exteriors, one may be able to move that much faster with better results. And what's this about ND's being redundant with proper lighting? They are still a basic and essential tool for day exteriors, even more so as camera sensors become more sensitive. No getting around that one.

One other thing about working around the limitations of DSLR moire: I've been doing just that for the past eighteen months and I'm just about done with it. To have to schedule camera/wardrobe tests for every piece of clothing on every person that will appear in a project is too time-consuming and impractical (you have to check it at multiple focal lengths as moire may only appear at a specific distance) and we constantly get burned. Same thing with set design. It's extremely hard to predict what will be troublesome other than the most obvious patterns. Then there are situations where moire pops out on people's hair, or even eyebrows--yes, there are some post production chroma fixes available but as I've learned, very often "we'll fix it in post" means "we'll never actually get around to it"

Now, a common theme in all of this is what size and experience level crew you can get on a micro-budget feature and that is something that may be a variable depending on location. I suspect things are quite different in India than they are here and this may be the factor that is influencing many of the statements you have made that don't quite gel for me.

[Dylan Couper]

Not to continue the dogpile but...

That's an issue for the AF100, too, isn't it? The good thing about lenses is that they can be rented if required. Nobody said lenses were cheap. In fact, the lens I used in the chart is around $3,000.

When I said m4/3 cameras, that includes the AF100.
Good lenses aren't cheap, but a useful range of fast/quality glass for the m4/3 cameras is difficult and expensive enough that it would probably eliminate them from my list of "low budget feature cinema" cameras. (having said that, I own an AF100, as well as many DSLRs)

Also, I'm not sure how you "using" a $3000 lens in a text based chart is relevant, but it seems contradictory to me that you'd put $3000 into a lens that's only fractionally better than a $100 Nikon 50mm 1.4, while in the paragraph below telling people that the extra gains in quality is not discernable by the audience are not worth the money (which I agree with). I'm just sayin....

I mean best. I don't plan on spending a few extra thousand dollars just for 8-bit 4:2:0 video (which is the baseline for this chart). Considering all the compromises a low budget indie makes on average, the extra gain in 'quality' is hardly discernible under practical conditions (nor by its audience, let me add).

So you don't mean "best" but "best on a no-budget feature." I'd support your conclusion of the 600D there.

[Sareesh Sudhakaran]

Originally Posted by Brian Drysdale

I've heard people comparing it favourably to the the 7D, but this is the first time I've heard it suggested as a better option to the F3.

I never said it was better than the F3. In my chart, the F3 stands heads and shoulders above every other camera easily.

"The Blair Witch Project" was supposed to have been shot with the film school cameras that the students were using, so they could've filmed with a mobile phone or a F3

They couldn't have shot that film with a CMOS sensor (even the F3's) due to the jello-effect and other artifacts that would have come about in their way of shooting. They did use one digital camera, and it had a CCD sensor.

[Sareesh Sudhakaran]

Originally Posted by Charles Papert

Sareesh, you are making a lot of definitive statements that are going off mark, I think. If you are talking about a $100,000 feature, chances are that unless you are fortunate enough to get very skilled people working for you for essentially no money, there will not be enough time, gear and crew to be able to perform the kind of workarounds you are suggesting to some of the issues presented by these cameras.

Isn't that the case with any project? Without a decent crew nothing can be accomplished anyway. Nobody said low budget filmmaking was easy, or that it didn't entail making many compromises and sacrifices.

With a higher dynamic range camera like the F3 in S-log mode, I may not even need it all, and that is a significant consideration.

F3 with the S-log upgrade is beyond the reach of most people (and out of the budget I was considering). Also, the hardware required on set to read that data means extra crew and more time. If I really had to shell out that much, I'd rather shoot in 16mm.

And what's this about ND's being redundant with proper lighting? They are still a basic and essential tool for day exteriors, even more so as camera sensors become more sensitive. No getting around that one.

A well-planned feature won't need more than one ND filter at most. It isn't hard or time consuming to screw one on, and it is not the case that they are permanently screwed on a lens anyway. Plus, the rigs I have mentioned have matte boxes for this specific purpose.

Also, it's funny how people always use the ND filter argument. Are the ND filter options (2 on the F3 and 3 on the AF100) exactly what will be needed on a particular shoot? If so, why can't manufacturers agree on the best stops for them? These in-built ND filters are really great when shooting documentaries or ENG work, but on a feature, the DP should know better.

One other thing about working around the limitations of DSLR moire: I've been doing just that for the past eighteen months and I'm just about done with it.

A little bit of moire is an acceptable risk. Like Hitchcock said: If the story and characters are interesting enough, nobody is going to notice the crew member in a white coat in the background. I can understand how on big budgets moire is unacceptable, but the low budget filmmaker has fewer choices. Even the biggest blockbusters goof up.

Now, a common theme in all of this is what size and experience level crew you can get on a micro-budget feature and that is something that may be a variable depending on location. I suspect things are quite different in India than they are here and this may be the factor that is influencing many of the statements you have made that don't quite gel for me.

I have worked with American crews in India, too. A good, experienced crew is a godsend anywhere. In fact, without a good crew, it is better to not start a project. Actually, I haven't made any statements with regards to crew, only to equipment.

I can add one thing though - lighting in India is much cheaper than in the US. I had a mini lighting and grip truck and a generator for around $80 dollars a day (in 2008). I used 2 4K HMIs, 5 1K Babys, a few 1K Ultras, 20x20 & 10x10 silks, plus dolly with track, stands, etc. It came with eight crew members (grips, one electrician and driver) for $10 each per 8-hour shift. Not bad, eh?

[Sareesh Sudhakaran]

Originally Posted by Dylan Couper

Also, I'm not sure how you "using" a $3000 lens in a text based chart is relevant, but it seems contradictory to me that you'd put $3000 into a lens that's only fractionally better than a $100 Nikon 50mm 1.4, while in the paragraph below telling people that the extra gains in quality is not discernable by the audience are not worth the money (which I agree with). I'm just sayin....

I'm not an expert on lenses. I chose the F-mount because it is the most universal mount (that can be adapted to almost every camera made). I use Nikon primes with my 550D. I chose Zeiss so that nobody would complain! Oh well...

So you don't mean "best" but "best on a no-budget feature." I'd support your conclusion of the 600D there.

That's all I've meant all along. For a feature, this is what I'd do:

$0 to $100,000 - The 600D - Buy (if possible buy two bodies). Rent the lenses unless the costs are the same. Avoid major post processing. Get the best stereo sound mix possible, but keep options open for a Dolby 5.1. 99.9% of these films will not see a theatrical release or break even.

$100,000 to $1,000,000 - Rent a RED ONE or shoot in 16mm (the latter if less post processing is required). Best possible theatrical and stereo sound mix. 99% of these films will not see a theatrical release, but can make money off television and video.

$1,000,000+ - Shoot 35mm (or risk not getting distributed) or RED ALEXA (for major post processing) - renting only. Without a theatrical release, most of these ventures are doomed.

[Brian Drysdale]

At a $100,000 budget I wouldn't use a 60D, as Charles points out moire is a real issue. You mightn't even notice the moire when you're shooting, but you can get stung with it during post The FS100 (or AF100) fits into this budget, you don't have the the same problems and you can always sell the camera afterwards or just rent.

All 8 bit 4:2:0 isn't equal, even using DVCam the pictures from a DSR 500 or DSR 450 are better than those from a PD150.

BTW There are $100,000 features that have been shot on 35mm film. You can get theatrical distribution on a film shot with a RED ONE so the $1M limitation doesn't apply, a number of high budget films have been shot with it. The Alexa is made by Arri and I suspect RED would argue that for big screen the RED ONE offers better resolution. There's no set path, especially at these low budgets ($1M is still low budget for a feature film).

[Sareesh Sudhakaran]

Originally Posted by Brian Drysdale

BTW There are $100,000 features that have been shot on 35mm film. .... There's no set path, especially at these low budgets ($1M is still low budget for a feature film).

You're absolutely right. I should have been clearer. A low budget indie ($0 to $30,000) has a higher percentage going towards its camera department than a well-balanced feature would have. Sometimes it is higher than 50% (as the case was with my feature).

Assuming the same movie had two budgets - one at 100K and the other at a 1 million - one cannot drastically reduce the expenses going into makeup, costumes, set design, etc, without losing production values. Yet, low budget filmmakers are always under pressure to do this. In the olden days, cameras, too, couldn't be compromised on - you either shot in 16mm or 35mm. If you couldn't afford 16mm, you couldn't shoot a movie for theatrical distribution. In today's age, it's possible to shoot a feature on a 600D and if one lights within the dynamic range of the camera, the result will be acceptable. So the proportionate percentage going to the camera department can be reduced. One can also factor in renting cheaper and more efficient units like LEDs (but I don't know much about that so I'll stop here).

I was just grossly generalizing in that last post. I believe no rules can be set since there are too many variables involved. Plus, I can't wait to get my hands on an AF100!

[Brian Drysdale]

It's possible to shoot a feature on many current cameras, its more if they're the best choice for the job and if they're going to give problems that need to be "sorted out in post". Much depends on your budget and the proportion that's going on the camera, also you may find it's better to rent than to buy.

[Sareesh Sudhakaran]

Yes, I agree, except for DSLRs, every other camera system should be rented. There are some great deals out there.

[Shaun Roemich]

Originally Posted by Sareesh Sudhakaran

Yes, I agree, except for DSLRs, every other camera system should be rented.

I think a lot of the perceived animosity here is because of BLANKET statements like this one.

Rental versus purchase should be weighed against the ACTUAL financials.

An average rental is usually 3 - 5% of value per day on large capitalization items like cameras and usually higher on items like cards and such.

Most rental houses will give you a break on multi day rentals (1.5x day rate for a weekend, 3.5x day rate for a week).

BUT... if your shooting schedule extends past the "break even" sweet spot, you should consider buying and reselling afterward.

For example:

The Indie feature I lensed some years ago spanned a year of shooting. My Sony PD150 (again, this was YEARS ago...) kit ALL IN cost me $8000. Rentals for a weekend back then would have been around $250 a day, so even with weekend rates it would have cost $400 (if we had rented instead of using a camera I already owned...) each weekend we took the camera out. MINIMUM. 10 shooting weekends over the course of the year equals $4000. Would we have been further ahead to buy an $8000 kit and sell it one year later for half price? Probably... AND had use of the camera for other stuff in the interim, possibly to help further offset purchase price.

One simply CAN'T make a blanket statement around which is better, buying or renting, without first "running the numbers".

[Sareesh Sudhakaran]

Originally Posted by Shaun Roemich

I think a lot of the perceived animosity here is because of BLANKET statements like this one.

:) I agree with you - My views are opinionated. I'm hoping by reading different points of view a newcomer might benefit from our combined wisdom.

BUT... if your shooting schedule extends past the "break even" sweet spot, you should consider buying and reselling afterward.

For example:

The Indie feature I lensed some years ago spanned a year of shooting. My Sony PD150 (again, this was YEARS ago...) kit ALL IN cost me $8000. Rentals for a weekend back then would have been around $250 a day, so even with weekend rates it would have cost $400 (if we had rented instead of using a camera I already owned...) each weekend we took the camera out. MINIMUM. 10 shooting weekends over the course of the year equals $4000. Would we have been further ahead to buy an $8000 kit and sell it one year later for half price? Probably... AND had use of the camera for other stuff in the interim, possibly to help further offset purchase price.

I had a similar issue with my feature. I bought a JVC 111E camera with accessories, plus a sound 'system', and a laptop - the whole thing cost me around $10,000. The cost of renting a similar system (the only option was a Z1 - the XL-H1 and the HVX200 were unavailable in India) was around $6,000 (for the period I needed it). It seemed like common sense at the time to buy the camera, use it and then rent it out. I figured even if I didn't rent it out, I could sell it for around $6,000. After I completed my movie, I realized there was no demand for the camera in India (This is Sony land) either for rentals or for resale. Finally, I managed to sell it this year for around $2,500 (not including the laptop and the sound system). My bad, of course.

One simply CAN'T make a blanket statement around which is better, buying or renting, without first "running the numbers".

True. I did run the numbers, and for every other system excepting an HDSLR, I found it was cheaper to rent for a period of up to 30 days (the typical testing and shooting period of an indie feature).

For HDSLRs, I realized it was not worth it to rent simply because owning a DSLR has other benefits - like shooting stills - for storyboards, for artwork, for production stills, for casting, to shoot the making of, and so on. I have realized (being a director, not a DP) that having a DSLR just made me visualize better, and I'm improving every day with the 550D.

[Dylan Couper]

One thing the renting vs. owning arguement hasn't covered is the tax benefits of renting over purchasing. In many cases, the cost of renting can be effectively zero (depending on the laws in your country of course). This should be discussed with your accountant.

[Gabe Strong]

Originally Posted by Sareesh Sudhakaran

For HDSLRs, I realized it was not worth it to rent simply because owning a DSLR has other benefits - like shooting stills - for storyboards, for artwork, for production stills, for casting, to shoot the making of, and so on. I have realized (being a director, not a DP) that having a DSLR just made me visualize better, and I'm improving every day with the 550D.

Ah ha! I am guessing that you are a 'HDSLR guy'. I personally can't stand the things for
shooting video. I'd much rather have an FS 100 or AF 100 (and I too admit I am biased,
I own an FS 100). However, that being said....

would it not be true to say that owning an FS 100 or AF 100 or really
any of these cameras has other benefits as well?. I have actually used my
FS 100 to print off stills....smaller ones true...5x7's....and I will
say, that they are as close to HDSLR 5x7 stills as HDSLR video is to
FS 100 video :-)

But to each his own :-) I personally think the biggest
factor in the HDSLR craze is price. There are a LOT more people willing
to spend 1-2k on a camera than 5-6k. There is always a law of 'diminishing
returns' on these things. So a 550d gets you 85% of the way to a FS 100
(and is much cheaper). You chose a 550d because of that. To me, the extra
15% is easily worth the extra money (but I run a BUSINESS doing this, so
it's not a hobby and I don't make films 'on weekends for fun'. Not saying
you do, or disparaging those who do, but this is my ONLY source of income
so I tend to spend a little more on it than many do probably....but not as
much as some who have 'higher profile' clients. Kind of the typical small
town guy who makes corporate films and TV commercials, and weddings and so on.)

However, the FS-100 might get you 85% of the way to an F3. Why didn't I just
buy that then since I value quality? Simple, budget! The clients I currently
have, don't have that kind of budget! To me, there is a lesson here.
When looking at a camera that you are actually going to buy, the
very FIRST thing you want to look at, is not technical specs.....but
your BUDGET! Once you have that nailed down, you can figure out what
some of the best cameras in your budget range are.....and then compare
them and see what works best for you. It's pointless looking at
65K cameras if you only have 10k to spend. It's silly to look
at 6k cameras if you only have 2k to spend. Now I know that for
many, renting is an option. However, for those of us who live outside
the urban centers, and have no rental houses, having a camera on hand
can mean the difference between getting the job, and getting passed over.
It would take at least two or three days for any rental gear to get to
me in Alaska, so I don't really ever look at renting, unless it is
a very high end job with at least a weeks notice.

[Sareesh Sudhakaran]

Originally Posted by Gabe Strong

Ah ha! I am guessing that you are a 'HDSLR guy'. I personally can't stand the things for
shooting video. I'd much rather have an FS 100 or AF 100 (and I too admit I am biased,
I own an FS 100).

Don't get me wrong, Gabe - I'd rather shooting anything but HDSLRs too...but not on a low budget indie feature.

would it not be true to say that owning an FS 100 or AF 100 or really
any of these cameras has other benefits as well?. I have actually used my
FS 100 to print off stills....smaller ones true...5x7's....and I will
say, that they are as close to HDSLR 5x7 stills as HDSLR video is to
FS 100 video :-)

And a 5D Mark II can't make a good still image?

You chose a 550d because of that.

If you mean my own personal choice, then no, I bought the 550D for photography exclusively.
If you mean from the chart, then, yes, it was chosen comparing its advantages to its costs, and then coming to the conclusion that it was indeed 'good enough' for low budget indie features.

To me, the extra
15% is easily worth the extra money (but I run a BUSINESS doing this

For stuff other than low budget indies, like in your line of work, it would be pointless to shoot with DSLRs when you have the choice of an AF100 or FS100. In 2007, when I was hunting for a prosumer camera for my first feature, the only option was HDV, and I chose one which gave me the best signal, lens and form factor within my budget (excepting the XL-H1, which was out of my price range). I don't regret it.

the very FIRST thing you want to look at, is not technical specs.....but
your BUDGET! Once you have that nailed down, you can figure out what
some of the best cameras in your budget range are.....and then compare
them and see what works best for you.

Experience teaches one well, doesn't it? When I started out, I didn't have many people to ask, here in India. I have learnt that, for features at least, it's better to budget every non-technical thing first and then spend the remainder on whatever fits the budget. But a newcomer doesn't 'get' these things. I certainly didn't.

It would take at least two or three days for any rental gear to get to
me in Alaska, so I don't really ever look at renting, unless it is
a very high end job with at least a weeks notice.

You are right. The guy who bought my JVC is a wedding filmmaker. I never made a cent out of my camera. But he is earning and raising a family from it. That was a strong lesson for me.