Declaration of Principles
Film Festivals: Playing the odds
DIY Film Projects
Cinema vs. Home Theatre
For Those With Writer's Block
So You Wanna Go to Film School: 1
So You Wanna Go to Film School:2
This is the second of two posts considering the rewards and challenges of using DSLRs for cinema work. If you’ve not read the first post, start there.
At the end of the last post we had assembled a Canon 7D camera, a Canon 17-55 f/2.8 lens with Image Stabilizer, a Zoom H4N audio recorder, and PluralEyes software to help us sync the picture and sound in Final Cut Pro. The cost: $3230. I hesitate to call this a “bare bones” package since it doesn’t even include a tripod or microphones. It does, however, get you picture and sound.
But you get picture and sound with a pixelvision camera. My intention with these posts is to compare DSLRs to a more traditional prosumer camcorder. And we still have a ways to go before it’s a fair comparison. So let’s continue…
For starters, the Sony and Panasonic cameras have ND filters built into their cameras. And while there may be some optional kit with the Canon DSLR rigs, ND is not one of them. Not, at least, if you want that creamy shallow DoF cinema look, which is probably the reason you bought the Canon in the first place.
Some people, including Philip Bloom, swear by the FaderND, which cuts out between 2 and 8 stops of light. If you find it on Ebay you’ll pay around $125. Very cool!
Others, though, argue that the FaderND can make color correction a problem later on. Indeed, the quality of your lens is reduced if you put inexpensive glass in front of it. (Or, as Shane Hurlbut warns, “beware the reaper of cheap glass“!) So if you do want to be careful, you would need to budget between $275 and $450 for a set of high quality Tiffen “Water White” IR ND filters.
If you’ve got multiple lenses with different filter ring sizes you’ll need to purchase step-down rings. But for now, we’re assuming we only have one lens.
Let’s throw caution to the wind and go with the Fader ND. That puts us at $3355.
We also need to power the camera and record to something. So we need some CF cards and we need some batteries.
We would obviously need batteries if were were going with a more all-in-one solution (i.e., an actual video camera) like Sony or Panasonic. But in my experience the batteries supplied by these manufacturers last about 2x as long as those supplied by Canon, in part because the Canons weren’t really built for, you know, constant video footage. And, a manufacturer like Sony or Panasonic supplies an AC adapter so you can run your camera off wall power. Canon does no such thing. So to be fair, we’ll add the cost of two batteries ($156), even though you’ll actually need four or five to shoot a day’s worth of footage.
As far as CF cards are concerned, for a starter package, we’ll figure you need 32GB of CF memory. That’s about $77. Hurlbut makes a compelling argument that you should use lots of 8GB cards instead, but we’ll stick with one card, which gives you about the same recording time as the 16GB SxS card that comes supplied with the Sony Ex1R (roughly an 1 hour).
What are we up to now? $3588.
Finally, in my experience, I’ve found you need some sort of way to monitor your footage. The on camera LCD focusing system is not large enough to accurately focus on the fly. And it is often impossible to use in broad daylight.
The focusing issue is, for some, a real deal breaker, and for good reason: Everyone I know that has used this camera has shot footage that appeared to be in focus but, upon later inspection on an actual monitor, learned that the take was a bust. You have to be very careful about monitoring your footage, and you need to check every shot on a large monitor (Hurlbut recommends a 24″ LCD) before you move on to the next setup.
I’m not going to include the cost of the 24″ LCD. We’re going bare bones here. So we’re going to use a Zacuto Z-Finder ($395), which magnifies the camera’s LCD viewfinder.
Another option is to use an external monitor while shooting. The advantage is, obviously, a larger viewing area to judge focus. The disadvantage is that once you add an external monitor (with battery pack, HDMI cable, and hardware) you lose the small, stealthy DSLR form-factor. Good monitors are expensive, too, often averaging around $800-$1000. The cheapest possible monitor option, however, gives the Zacuto Z-finder a run for its money. That monitor is the Lilliput 669GL.
The Lilliput is only $220, but you’ll need a battery solution. I recommend the Ikan 107S or P ($68) depending on whether you already might have some Sony or Panasonic batteries. And you’ll need a special MiniHDMI-to-HDMI cable ($12). And you’ll need an arm (Ikan’s MA206 is the cheapest somewhat decent solution at $70) to mount the monitor to your camera.
The total cost of the Lilliput option as I’ve described it is around $350. If you need Sony or Panasonic batteries to power it (and a charger to charge the batteries) then your total will exceed that of the Z-finder. So let’s just add $395 for the Z-finder and be done with it.
By the way, I’ve not been tallying shipping costs on these items, but lots of places like Amazon, B+H, and Adorama offer free shipping on certain items, so you might get lucky.
I think this does it for a bare bones kit. Remember, my estimates do not include the things you’ll need to actually shoot for an entire day — things like extra batteries, multiple CF cards, a camera bag or case(s), a shoulder mount, or a tripod. Nor does it include things like quick release plates for your tripod and shoulder mount. Nor does it include any sort of rod system or a follow focus, which you may want since the whole purpose behind using these cameras is to have that all-important shallow depth of field.
The final total? $3983.
That’s twelve dollars cheaper than Panasonic’s HPX-170.
So the question is, which do you want?
Traditional low-level professional camcorder:
- not a stills camera
- less cinematic depth of field
- fixed lens
- somewhat video-ish handling of light
+ actual HD resolution
+ accurate focusing
+ less pronounced jell-o problems
+ single-system sound with XLR inputs
+/- all in one build (pros: it’s meant to be used this way; cons: looks like a video camera)
+ solid HD codec
+ stability/durability as a camera intended for video
+ great stills camera
+ cinematic shallow depth of field
+ option of interchangeable lenses
+ beautiful handling of light
- difficult-to-edit codec with “reversal film” (i.e., limited) flexibility
- less-than-HD resolution
- issues obtaining accurate focus
- aliasing and moiré problems
- jell-o problems
- double system sound with separate sound recorder
+/- modular build (pros: pick what you need; cons: you’re only a strong as your weakest link)
- some overheating problems*
* Did I mention that there have been some issues with overheating since video on these DSLR’s is so demanding? If the camera overheats, it may not work for a while. One solution is to have another camera body on hand (+ $1700).
Look, I’m not advocating one camera over another. And I am not trying to diss on the DSLR revolution. I’m just trying to cut through the hype to talk realistically about the choices that exist for a low-budget filmmaker.
Cameras — like life, art, and love — are full of compromises. The question is, what are the compromises you can deal with, and what are the deal breakers?
Just how badly do you want that bokeh?
Are you willing to sacrifice reliability?
Are you willing to risk losing half a day’s worth of work?
Are you willing to endure slow-downs because you have to re-shoot footage?
If so, how much?
I don’t have the answers. At the beginning of this series I said I was ambivalent. And I meant it. I haven’t made up my mind about these cameras. I doubt I will. It will be a case-by-case, project-by-project thing.
I think there will be some times where these cameras are appropriate for me to use. They’re great for clandestine filming. I like them for filming in/with/around cars. I like the way they handle close-ups. If I was single-handedly making a shot-for-shot remake of The Passion of Joan of Arc, this would be my camera. (Hmm…)
But if I had to choose only one camera to own, a DSLR would probably not be it. I won’t even consider it for documentary, or documentary style, filming because of the shallow focus and overheating issues, never mind the moire and jello.
Do I think DSLRs are game-changing technology? Only sorta. These cameras have been handicapped by the corporation that produces them. Whether intentionally or not, it doesn’t matter. Either way, it’s the same old corporate routine. Call it the corporate camera cha-cha: One step forward, one step back. What has people intrigued about DSLRs is that the steps forward and back are not the ones we’re used to.
With time maybe I’ll come around to love these cameras whole-heartedly, but even if that happens I will not argue that DSLRs have “democratized” filmmaking in any meaningful way:
First, as I think I’ve demonstrated, at their current price point these cameras aren’t that much cheaper than other things on the market. When we talk about “democratized technology” we must be talking, on one level, about cost. And on this score, they do not pass the test. (The T2i makes a somewhat better case, but it’s also the most handicapped of the bunch.)
Secondly, DSLRs — as they are currently designed — actually require more know-how to use effectively than other cameras that can be used for filmmaking. In this sense, DSLRs are actually less “democratic” than other existing movie-making technologies.
Finally, even if — especially if — I allow that these DSLRs are getting more people to make movies, let me address a bigger point:
“Democratized” technology serves little purpose if it isn’t being used in the service of stories that otherwise couldn’t be told. Otherwise, what’s the point of democratizing it?
Put another way, if you have the means to make a movie, and you only use that technology (not to mention your time and talent) to make another frigging zombie movie, well, pardon me for not caring. If the storytelling is out of focus, who cares how beautiful the bokeh is?
Thursday, June 24th, 2010 at 8:00 am | by Paul Harrill
| Filed under Basics, DIY, DIY Filmmaking, DSLR, Production, Technology, Tools/Equipment |
| Follow responses
Both comments and pings are closed.
First, many of us are not chasing bokeh (although a nice bonus). It’s latitude, color and low-light that you need to spend tens of thousands or more to get in a video camera. That point simply cannot be overstated.
Here’s my bare bones kit.
7D with kit lens (only used for occasional telephoto or macro)
Tamron 17-55 f2.8 (much less the Canon and good lens for stills and video)
Canon 50mm 1.4 (awesome lens)
2 extra batteries, 3 X 32 GB Kingston Cards
Cost – $2700
I would like to add the Small HD 5.6 720p onboard (you can see aliasing/moire clearly on it) which brings it to $3800 when I can afford one. I already owned Pluraleyes (a near must have for any editor with any camera), tripod, lights and $250 24″ cheap LCD monitor with HDMI in.
The 7D is my sole camera for still and video including docs – I was nervous shooting a doc for it but I’m currently a DP for a doc and found it’s a great doc cam much like shooting with 16mm or 35mm. Sure, a couple of minutes of 30 hours has some aliasing. But at least 15 hours would have be crappy looking footage due to lighting issues (we have to shoot without lights) shooting on any other video camera but look great on the 7D.
About story – I’ve had the opposite experience. First, I love my 7D (I’ve never loved a video camera). This is creatively stimulating. I love shooting and lighting by eye – the great sensors in the cameras just handles light, color, actors, especially the low light/bad light/no light of indie filmmaking. And audiences love films made with them. This had been well demonstrated at our last two 48 Hour Film Festivals here in Greensboro.
It’s revolutionized the way I shoot – and they way people respond to what I shoot. When you watch your film project off blu-ray or other quality HD source in a real movie theater, then you really experience how it feels/seems/looks like watching 35mm movies. That’s the reason so many of us got inspired to this – to make “motion pictures” shown in movie theaters. That’s democratized technology for filmmakers. That’s why it’s been embraced like no other camera technology before it. And it’s only going to get better and better.
Sure – there are issues to solve, but it’s the closest non-film camera to the 35mm motion picture camera that you can buy as a normal person at your local Best Buy.
Great post! But isnt the real problem that manufactures dont listen to their customers? The demand for bokeh has been since the start of 35 mm adapters. So im not chocked that people are jumping of bridges to get their hands on the pandoras box of bokeh. The demand has been out there of years. If the customers / us would have complained from the beginning we might already had a sony ex1r with a fullframe chip with alpha lenses, or a xh a1-s with canon lenses. So it was just a fluke that they succeded with a extra feature for marketing and sales boost, instead of the old mega pixel war. The little extra feature turned out to be a gold mine. So why not produce what people want? Nah, we are like flies( flocks to a pile of shit)that why this will keep reapting itself. If it smells it must be good ,right? or is it just a good old pile of shit in the end anyway when the smoke has cleared?
@ Mark: I think manufacturers do listen to customers, but usually in fairly paternalistic — and always in profit-driven, ways. I thought this was an interesting read: http://www.eoshd.com/content/262-Do-companies-differentiate-ranges-differentiation-by-firmware-switches
@ Steve: No need to defend your choices to me. As I said in the comments of the last post, if it works for you, that’s great. I feel the way you do half the time, otherwise I wouldn’t be blogging about my ambivalence. I definitely have a love-hate relationship with these Canons. Also, you bring up a point that I left out of my discussion: I do appreciate their mass availability. If a battery dies on you, you can probably get one at a Best Buy (or similar kinda store) so if you’re not in a major production center (as neither of us are) your chances of having a shoot stop dead are far, far less than, say, shooting with a Red or a film camera or whatever.
@Paul : Thanks for the link! Great read, i love the whole dslr revolution, beacuse now every mancufature must rethink there old strategy. The semi-pro market video customer won´t pay 5000-9000 dollars for a camera with a little little censor with sucky low light capabilitys, and telephoto- bokeh. If you have tried a 5d or 7d you know what im talking about. Yes lights will to the trickl, but since production is getting smaller , and you have to to more and more yourself . My back is happy the less lightning equipment i bring with me.
It´s time for manufactures to do a camera that worth 5000-9000 dollars!! They make money and i get what i want, that what these is all about at least for me. Until then i will keep on editing materials from these little cameras and that make me pay my rent. But is very funny to see more and more of “anti-aliasing” from fine texture in commercials. Customers seem to be happy, but for me these seem to me a slap in the face ,spending all these years trying to tweek every single frame to perfection, ? Passion is the world, but now i have to accept a new artifact “anti- aliasing” as a new standard. Life goes on and so will i. Im a fly to…..
What are the shortcomings of the T2i you were talking about?
@Sam: The limitations are not awful with the T2i. But there are a few things. Quickly, they are:
- build quality (7D is far more rugged)
- ISO increments (7D has 1/3 stop increments)
- manual WB (7D has it, T2i doesn’t)
- recording to rugged Compact Flash cards (7D) vs SDHC (T2i)
Philip Bloom discusses the T2i here in the context of the 7D:
Hope this helps!
This is really helpful. I love the potential DSLR cameras have. I think that the next model to come out will become a standard in the industry.
Thanks for “doing the math,” about what I would have expected cost wise. Unfortunately it doesn’t help me any to make a decision, which I’ve been equivocating over for five to six months now.
““Democratized” technology serves little purpose if it isn’t being used in the service of stories that otherwise couldn’t be told. Otherwise, what’s the point of democratizing it?”
This says it all. I really don’t give a ff about another zombie movie shot on a 7d. I guess this is the history of cinema in a nutshell: art vs. commerce. Will people bother to watch anything but that which shocks them? It’s the rare beast that manages to combine both.
@Kent: Thanks for readings — I’m glad the post spoke to you.
RE: your “equivocating” — I sympathize. After wrestling with this stuff, my first advice would be: Don’t buy equipment until you have a specific project that you need it for. I will say that I’m doing some more testing with yet another camera that I overlooked in this discussion and I might end up doing an unexpected “Part 3″ to this series. No promises, but it could be interesting….
Excellent advice. Even with the three projects that I had in the past couple of months, I still didn’t bother to purchase: gear came from Lensrentals.com. But they have whet my appetite enough that a move could happen soon. The work I want to attempt in the future is less “project” oriented and more impressionistic, which would be helped considerably by owning rather than renting.
Your article implies that, depending on the project, you are not sure whether to use a pro-summer camcorder or a DSLR. What about the Panasonic AG HPX500. Would you choose a DSLR’s over the AG HPX 500; with or without a 35mm adaptor kit).
@Charles: Good question. I haven’t worked with the HPX500 so it wouldn’t be fair to comment on it. I can say that, as far as professional (i.e., higher than “pro-sumer” level) Panasonics are concerned, “Quick Feet, Soft Hands” was shot with the Varicam. We used it with the P+S Technik lens adapter and 35mm cine lenses. As far as image quality is concerned, I was very pleased. But I wouldn’t use that setup on a documentary or small-crew fiction film. Again, there’s not a one-size-fits-all answer.
Paul – testing the hacked GH1, perhaps?
@ Stephen: I’ll only say I just sold my Fisher Price PXL 2000 on Ebay, so it’s not that….
Are you saying the Panasonic AG HPX500 is out dated?
I took a look at the SvV website and I’m sure they aren’t taking that position :-)
–talk about a time to upgrade!
@Charles: I wasn’t saying the HPX500 was outdated at all. Just curious — what about my comment made you think I felt that way? I was mainly commenting on the Varicam and, in fact, talking about the good experience I had with it. To elaborate, the reason I wouldn’t use the Varicam + lens adapter setup for a small crew or run-and-gun documentary project is not that it’s outdated but that using such a setup is heavy, requires a fair amount of setup, and it demands a focus puller.
Actually, that’s what I thought Paul. SvV’s comment threw me.
This is a really well thought out and well done article.
I have personally never understood why people refer to this new wave of filmmaking as democratized. I mean, haven’t independent filmmakers always made movies without asking permission?
But even comparing these DSLR limitations to the work I had to go through with my Arri BL 16mm – I can honestly say there is no comparison. Even on the high end…
This is an exciting time to be a movie maker!
The author of this article has a creativity deficit that will cost him a fortune. DSLRs are not for the closed-minded cinematographers of last decade. Only a complete disregard for limits combined with unbridled creativity will deliver a serious return on investment in this new era of DSLR cinematography.
With $2,200 and a little creativity, I’m essentially running an entire studio.
Below is my equipment checklist for getting the most out of the DSLR revolution:
CAMERA & LENSES
-Canon EOS 550D DSLR (1080p @24fps)
-Canon 18-55mm f3.5-5.6 IS Lens (decent for wide-angle)
-Canon 55-250mm f4.0-5.6 IS Lens (decent for telephoto)
-Canon 50mm f1.8 Lens (amazing low light performance and bokeh for $100)
Category Total: $1,112.95
-Custom 7ft – 22ft telescoping jib (most powerful piece of equipment in my arsenal)
-Custom camera stabilizer (works incredibly well)
-Sturdy tripod and head combo (easily holds 40lbs)
-Tripod dolly (for stable tracking and other moving shots)
-32GB Class 6 Delkin Devices SDHC (very fast card. Holds 90 min. @ 1080p)
-2 additional LP-E8 batteries (generic versions are very cost effective)
-Azden boom mic (fantastic audio quality)
-1.5ft XLR to 3.5mm cable (connects Azden boom mic to DSLR)
-20ft XLR to 3.5mm cable (connects Azden boom mic to DSLR)
-25ft RCA cable (for external monitor)
-RCA coupler (connects RCA extension cable to Canon RCA output cable)
-Nikon Lens Pen (removes debris from lenses)
-$35 floor lamp with 3 tilting lights (for natural lighting. Includes three 60W bulbs)
-Yongnuo intervalometer (for remote shutter release and time-lapse photography)
-Vehicle suction mount (mounts cameras up to 6Kg to the exterior of a vehicle)
Category Total: $1,086.12
Grand Total (Category 1 + Category 2): $2,199.07*
*includes all S&H and tax expenses
Basil, no one is more aware than I of my creative “deficits”, but believing you need an expensive camera to make a movie is not one of them. If you think I’m somehow arguing that one must spend ~ $4000 on a camera to make a movie, you need to re-read this post and, for that matter, many of the other posts I’ve made on this blog over the last 4.5 years. I don’t care if you shoot on a PXL-2000, a VHS camera, or Digi8. Content and craft are what matters.
My goal was to reduce the deficit of realism in the conversation about DLSRs. If you want to participate in this conversation, then take a lesson from the posters above about how to respectfully disagree.
Thank you for your reply. Your article is of great significance in this time of debate over how to move forward with video technology. You made several important points which I should have acknowledged earlier. Realistic is a relative term. What’s realistic for you may not be realistic for me. For the average consumer or “Spielberg wannabe”, your article is very realistic indeed. Most cinematographers will not benefit significantly from DSLRs in terms of both cost and performance.
But what exactly do you mean when you say DSLR video is “less-than-HD resolution” while “raditional low-level professional camcorder” is “actual HD resolution”? I happen to be specialized in video codecs and I can tell you with certainty that it’s quite the contrary.
The HPX-170, for example, produces a mock 1080p produced through a complex digital image construction process called “pixel-shifting”. The native resolution on the HPX-170 is much lower than 1080p. For this reason, no professional would call the HPX-170 video “native 1080p”. Canon 7D, 550D, 5D Mk II, 1D MkIV on the other hand all capture native 1080p.
At the end of the day, which camera you decide to use is not entirely important. Ask not which equipment is in the hand, but rather which hand is on the equipment? “Who”, not “what”, determines the ROI on everything in this world.
Thanks again for your article. It must have been well written to provoke my response.
Video from a Canon DSLR is less than HD resolution, at least according to Barry Green:
Which brings us back to the HDSLRs – the GH1, the 7D, and the 5D Mark II all look like they’re rendering incredibly sharp, highly detailed images, but they’re not. In reality, according to the resolution charts, they’re rendering images that fall somewhere around a standard-definition camera, and maybe a 720p camera. Any additional “sharpness” you see in the image is fake – it’s aliasing, it’s smoke and mirrors, it’s image contamination.
As the author of some supplemental books regarding Panasonic cameras, Barry has some conflict of interest, but the charts are the charts. (It’s worth noting, he’s also commenting on the Panasonic GH1, not just the Canon line in the quote above.) The full article is here.
As far as saying that you can get true HD resolution from a video camera, I probably should have noted that I was thinking more of something like the Sony EX-1, which I mentioned in the first article. (Adam Wilt demonstrates that the EX-1 has “honest to goodness 1920×1080 resolution” here.) Admittedly, an new EX-1 costs more than an HPX-170, so it’s probably incorrect to list that as a pro/con. (I’m well aware of the HPX-170′s pixel shifting.)
Perhaps I should remove the “real HD” line and instead substitute the presence (HD video cameras, yes) or not (Canon DSLRs, no) of an OLPF?
On a different note, if I were to do this comparison over again, I would probably recommend getting a GH1 ($1100) over a Canon camera (7D or otherwise). If you get a good one (it’s been plagued with manufacturing problems) and hack its firmware, it can produce impressive for very little money.