A year and a half ago I found myself thinking “it’s time for me to take the plunge and buy a dslr.” My reasoning was that I take a thousand or so pictures each month and I’ve been attempting to shoot short films and music videos with my phone. I had the money: (I thought) it was time to commit.
I am going to share the process I went through along with numerous links and videos and write an absurdly long post on this subject. The reason being that I have not put the issue to bed. Four cameras, several lenses and numerous pieces of gear later and I am asking, daily: “should I sell X and start over or should I just make do with my iPhone and forget this nonsense.”
My hope is that by putting this out into the world I might just spare someone else the frustration of going down this path.
To begin – when I decided to buy a camera I was only considering DSLR (digital single lens reflex) and DSLM (digital single lens mirror less) cameras. Why? Because my initial searches all indicated that a person interested in DIY filmmaking should consider these. The selling points are: they are small, lightweight, shallow depth of field can easily be attained and you can use interchangeable lenses with these cameras. These are the features that stood out to me.
Price was also key. For $1,000 or less you could purchase a camera that would deliver quality video. So far, so good. Unfortunately this is where the Internet stops being helpful. You now enter a space where no two people can agree on which features are most important or, better stated, which are the ones a filmmaker truly needs.
I, absurdly, had gone down this path before (in 2007) and it was much simpler then. Twenty-four frames per second shot progressively (not interlaced) and with a LCD display were the key features. Now, whether the camera has focus peaking, variable frame rates it can shoot at ( for either smoother, more video-like footage or to be slowed down for the effect in post production) an EVF display, high dynamic range, or does it suffer from moire and aliasing are just a handful of the debated points with these cameras.
The most baffling new issue relates to crop factors due to the CMOS sensors the cameras use. Some are considered full frame (they have a physically larger sensor and use the entirety of it in video mode) while others use a portion of the (already physically smaller) sensor and effectively crop out the unused section. This is incredibly important when making a purchase because this crop factor effects the amount of light the sensor receives (which when applied decreases the size of the aperture – increases the f stop) and increases the focal length of the lens used. This is confusing. A crop factor effects both the field of view and the f stop. If your camera, like mine, has a 2x crop factor that means your 25mm lens set to f 2.0 is in fact behaving like a 50mm lens at f 4.0.
So what does all this mean? Essentially people really like full frame cameras. The can obtain a shallow depth of field very easily. They perform better in low light situations. They also do not change the focal length – if the lens being used is 50mm it stays a 50mm.
Okay so get a full frame camera. Only, a year and a half ago, there were two available adjacent to my budget (this is untrue, the Nikon D800 was also available but since I have found only one person online who seems to shoot video with it I ruled out this option quickly) – the 5d mark iii and the 6d: both Canon cameras. Only they each camera cost more than $2,000 dollars – without lenses. So if you are anything like me and are working with no budget and have no expected income from making this purchase: start over.
People all over the Internet sing the praises of micro four-thirds cameras. These cameras almost all have a crop factor of 2x meaning you multiply the aperture and the focal length of the lens by two. Suddenly your 50mm set to 1.7 is a 100mm at 3.4. Yet the cameras are mostly cheaper, much cheaper than their full frame counterparts. They are small and here is the main point of praise: they have a micro four thirds lens mount – which allows them (with an adapter) to use virtually any lens in existence. Everyone is very excited about this feature. The reason being, if you already own seven Nikon lenses you do not have to sell those lenses if you decide to start using a m4/3 camera. You now can purchase a lens adapter and use those very expensive lenses on your new, small camera.
So a year and a half ago all anyone was really talking about were the Panasonic LUMIX GH2 cameras. At that point the camera was old and outdated, Panasonic had released the next camera (GH3) in this line. What the GH2 had going for it was it was tested. Already used in many different settings by many different filmmakers this camera absolutely had a lot of things going for it. And it had many flaws. Weird work-arounds had to be employed for capturing sound and the headphone jack was an irregular size and people had to purchase adapters in order to plug in head phones. A third party hack was required to add features that allowed filmmakers to get cinematic images from the camera.
That being said what that camera could achieve is really very impressive: Shane Carruth filmed Upstream Color using a GH2 and if you saw the film I doubt anyone would ever question if it was filmed on a “real” camera. In 2012 Zacuto held “The Revenge of the Great Camera Shootout” and one of the competitors was the GH2. If you follow the link there are lengthy explanations and videos showing how exactly they conducted their tests but the quick explanation is this – they pitted all kinds of cameras against one another in a situation designed to level the playing field between them. They then showed the finished results to audiences and had them pick their favorites, without knowing which camera was used. This resulted in a $900 camera (not to mention an iPhone) holding their own against the best cameras available.
So in an attempt at brevity (ha!) I ruled out buying this camera because 1) it was old 2) it did not simply work out of the box without hacks, add ons and enhancements 3) people wanted far too much money for an old, used camera on Ebay. I opted for the next model, the GH3 for a number of reasons.
- It has variable frame rates. Shooting at 60p allows you to play the footage back in “slow-motion”
- The camera can shoot time lapse photography in camera.
- No hacks are required to capture cinematic images.
- It has a fully articulating LCD screen
There are other things about the camera that appealed to me but these were the main factors. To not have to buy an external monitor to see what you are shooting seems like a simple request but apparently the people who make cameras are not in agreement on this matter. Slow motion is a fun tool to be able to use in music videos and narrative films. Time lapse photography is interesting and used in all kinds of filmmaking today and yet many cameras require you purchasing additional items if you wish to create them. Not having to run a hack to use the camera appeals to me greatly. The fewer modifications I have to make the better.
I had the camera for about nine months, I got to know how to use it and produce results that I was pleased with. I learned a number of valuable things.
- When people talk about poor low light performance you should listen.
- Crop factors are tricky. Some times you cannot get far enough back to get the image you want.
- Shallow depth of field is not that hard to obtain on a m4/3 camera.
- Underexposing images is strange but effective.
When people talk about poor low light performance I tend to be a bit skeptical. Largely because film is notorious for performing poorly in low light and all of us have seen plenty of films that have great, low light scenes. That being said once you start trying to use a camera like the GH3 you start to understand that what constitutes low light is a very subjective concept. For example, my living room has four lights in it, two 75 watt bulbs and two forty watt bulbs. This isn’t scientific or professional, just an observation. The room is not cavernous but it is not a broom closet either. Trying to shoot video in this room at night with all of the lights on, the aperture set to 1.8 (so really 3.6) and the iso at 1600 I can get almost nothing usable. The image has a lot of noise, very little is visible and when I take the footage into my editing software I cannot do much with it.
Dave Dugdale made a wonderful statement about crop factors in that there are times you cannot get far enough away from your subject to get what you want in frame. This could be as simple as you want to get an interviewee’s entire body in the frame and there is a wall preventing you from doing so. I have had this on many occasions, simply trying to get more than a person’s face into the frame using a 50mm (effectively 100mm) lens. This is a problem.
There are many things people like about shallow depth of field, the blurred background that allows for a separation between the subject and the background is very pleasing. The strength of m4/3 cameras is their ability to have a deep depth of field – to be able to get everything into focus easily. While I have found that to be true, shooting landscapes is easy for me, I also have found that setting my aperture at 1.8 allows me to have a fairly shallow depth of field (certainly enough that I cannot get all of a face in focus).
Slightly under exposing your image so that you can lighten it in post is weird. It is. The logic behind it is simple, video cameras do better with the dark portions of the image than with the bright. If you slightly underexpose you will protect the highlights better (you will have more detail in the bright parts) and when you brighten the image you will find that detail was hiding in the shadows. It feels wrong to not try and expose the image perfectly but it is, absolutely, an effective way to capture excellent images with these cameras.
All this is to say that after a good amount of use I found myself getting images from the camera I was happy with. Then, out of nowhere, Blackmagic Design cut the price of their Pocket Cinema Camera by $500 for one month. When I had been considering buying my GH3 this camera had cropped up several times. It sells for $1,000 normally and it is, in many respects, an excellent camera.
It does not, however, take still images. So when I was deciding on what to buy I chose not to purchase this camera based largely on it lacking this feature. At $500 (and already owning a still camera) it was too temping and I bought one.
This camera also has a m4/3 mount which meant my lenses and my lens adapters would work on both cameras. I forgot to mention my lenses, didn’t I? An account on twitter, Vintage Lenses proved to be very helpful. The user posts Ebay auctions most weeks of vintage lenses that work well for video. I followed up on one of these posts, spent $17 and bought two Minolta lenses that with a $30 adapter allowed me to start using my camera. Later I found a used lens made by Panasonic that would allow me to use the automatic functions of my camera (even though everyone will tell you never to use the autofocus or auto white balance features there are times when it is handy to do so).
But back to the pocket camera. Once it came it was time to try and learn how to shoot. It had a number of new, and according to the Internet, must have features and I was eager to see what they could do for me.
- It can record in either ProRes format or RAW.
- It captures footage in a Log profile.
- The color space (depth?) is 10 bit 4:2:2.
So what do these things mean? ProRes is the codec, the container, that the camera puts your footage into. Apparently ProRes is one of the better codecs that a camera can use. Since I have owned the camera Blackmagic has released several firmware updates that have expanded the varieties of the codec. This is really something because these allow you to have greater flexibility manipulating the footage in post production. RAW is another step further, where the footage is captured with the understanding that you have to manipulate it greatly to have a usable image. As such I have only attempted to do so once and found that it is not a format for me.
As for the color space, I do not pretend to know/understand all of this so I am going to directly quote someone here:
First some math: 8 bit color yields 16,777,216 possible colors. 10 bit color yields over 1 Billion possible colors. And for exponential reference, 12 bit is over 68 Billion possible colors.
All three of these items essentially add up to one thing – you can do a lot with the way your footage looks in post. Whether you are attempting to brighten or darken the image or create specific looks, the BMPCC gives you greater flexibility to do so compared to a camera like the GH3.
The downsides are plenty though. The GH3 has batteries that never seem to run out. I believe you can shoot continuously for four hours. At best you can get 40 minutes with the BMPCC. The screen does not move on the BMPCC which means if you are attempting to any kind of handheld, documentary-style shooting without an external monitor you are not going to be able to see what you are capturing a fair bit of the time.
Which is fine, that is not what this camera is for. This camera is for controlled shoots. It has settings for adjusting the aperture and focusing that are done by pushing a button. Nothing is automatic, nothing is set up for run and gun style shooting. It is a very different tool. Once I learned and accept this most things became simple.
So I said four cameras and I have written an epic saga and covered two. Fascinating! Let me just recap quickly.
I found myself with two m4/3 cameras which were made for very different purposes. I suddenly found myself with a new and unexpected problem, which camera should I use? I am in the backyard, trying to get some footage of my daughter running around but I brought the BMPCC and now I cannot take pictures. Or I brought the GH3 and I find myself in a low light situation – and although the BMPCC is not very good in low light I can get at better image because I can manipulate the footage more in post.
Suddenly I found myself either trying to bring everything I owned with me or I was constantly wishing I just had one camera that did everything well.
About six months after I purchased the GH3 Panasonic released the GH4. I learned something valuable: there is no good time to buy a camera. They come out with new ones at breakneck speed and there is always something bigger, better and faster than what you just bought. I let it go. For one thing, the GH4 cost more and now that it was out the sell price of my GH3 on Ebay was nowhere near what I spent purchasing it.
Flash-forward to a month ago, tax returns arrive, I am continually annoyed that I cannot shoot unless there is an abundance of light and I still have this “which camera should I use” problem. B&H runs a promotion on the GH4, my will is weak and I decide to give this camera that everyone has been praising a try.
It comes and I start testing. What had I heard that made this a “must have” camera?
- It shoots 4k (resolution) video.
- It shoots 96 frames per second video that it conforms in camera to slow motion.
- Improved codecs.
- Focus peaking.
- Better quality video if exported out via HDMI into an external recorder.
So numerous people, Philip Bloom being the first I heard it from, stated how wonderful shooting in 4k can be. One of the biggest selling points comes from shooting the footage at 4k then downscaling it to 1080p. What this allows you to do is move the footage an reframe without any loss of quality. If you shoot an interview you can fake having two cameras with this method of shooting by having a wide shot and a close up. If you shoot hand held and use some form of stabilization in your post production workflow the cropping that comes from this technique won’t result in a loss of quality. There are numerous benefits to shooting in 4k and outputting to 1080p, especially if you shoot by yourself, which I do.
What I found was my computer was very unhappy with this workflow. My iMac is from 2011 and although it had decent specs at the time it is not equipped with a graphics card that is meant for 4k footage. Suddenly the computer was performing tasks very slowly. It also was crashing often.
People praise the GH4 for conforming the 96 frame per second footage in camera. What this means is when you play the footage back it has already been converted to 24 fps and is playing at 1/4 speed. What no one mentions is that this method makes it impossible to record audio in this mode. Why is this a problem? If you ever watch any snowboarding or surfing videos (or music videos for that matter) one of my favorite editing tricks they employ is having the footage play at a normal speed and then at a key moment going into slow motion. If you only are shooting with one camera the method of doing this is shooting at a high frame rate (96 fps say) but having it play back at a normal speed until that critical moment. Much like downsampling the 4k footage this is very easy to do in post production. Unlike downsampling this does not tax your computer terribly.
So what I found was while this is a fun feature to use, suddenly my footage was devoid of sound (plus I have to stop shooting and change the settings to use the feature). If I were shooting footage where I did not care about sound this would be fine but for most situations I would want to retain the sound.
The GH4 has improved codecs and new picture profiles. All of this means that you can capture video with more information that should be easier to manipulate in post. I have found this to be true and the footage certainly can be made more cinematic than the GH3 but it is still nowhere near as flexible as the BMPCC. Which is disappointing because the camera costs $500 more than the BMPCC.
The same is true for focus peaking. This is a feature I was very excited for and once it came to the BMPCC I found I was shooting with greater confidence. The screens on these cameras are small and not always easy to see under the best conditions. The focus peaking on the GH4 seems to work differently than the BMPCC and I have found it is not as easy to use. Considering that these cameras are known for their ease of achieving a deep depth of field with the GH4 I have been surprised at how hard it is to know that my image is in focus. Part of the reason for this is that the camera is no better than the GH3 in low light, which means you are shooting at apertures like 3.5 when you do not wish to be.
This problem feeds directly into external monitors, something I have no experience with – I do not have the budget for them. Aside from the cost of these monitors you have to figure out where you are going to put them, which often require buying a cage for your camera to attach the monitor to. At the most recent N.A.B. two companies announced iPhone-sized external monitors but even the cheaper of the two, the Blackmagic Video Assist, will sell for $500. Even with something this small you leave the realm of easy, handheld shooting and need either a rig or a tripod, which can be problematic.
All of this led to me to decide that the GH4 is not worth the cost of upgrading. Yes it is a better camera than the GH3. B&H offered to buy the GH3 for $375 and is selling the GH4 for $1,500. For me, with my limited budget this just did not make a lot of sense. If I did not already own the BMPCC? Maybe. I found that what I can do without the GH4 is almost exactly the same as with it (splitting the work between the GH3 and BMPCC) and I don’t need to spend any money to get a similar result. So I am back to the two cameras one person dilemma.
On to camera number four. The Sony A7S. If you somehow have not heard of this camera I am amazed (if you are still reading this I am equally amazed and I love you). More disruptive than the GH4 and constantly compared to it, this offering from Sony sells for $2,600. Why am I considering it:
- This is a full frame camera. No crop factor is applied to anything.
- It can shoot at unheard of ISO levels with usable results.
- It can shoot at 120 fps (but only at 720p resolution).
- It has an APS-C crop mode.
The camera has other exciting features like focus peaking and an amazing viewfinder but I want to try and focus on these points. Full frame is neat, it really is. I had not had the opportunity to shoot with it prior to this camera and it is quite nice. People refer to the full frame look and after a limited amount of time with the camera I cannot say that I see this look, for sure. There is something different, especially with photographs but what it is I am not exactly sure.
Point number four ties into this one- the APS-C crop factor. This may seem counter-intuitive, why would you want a crop factor? This camera offers two shooting modes with no penalties. What this means is you can have all the benefits of full frame shooting or with the push of a button you can have a 1.5 crop factor that allows you to use the same lens but with a different field of view. Why is this good? Today I was testing the camera in my backyard. My daughter was quite far away and even though I was shooting with a 50mm lens I wanted to get closer to her. I was on a tripod and I did not want to move it, so with the push of a button I now have a 75mm lens. I found this to be quite handy.
If you know about this camera you know about its ability to show in very low light. This is absolutely a see-it-to-believe-it kind of thing and there are many, many videos online that do just that.
For myself I have already noticed how handy this feature is, in mundane settings. In my living room at night I cannot really shoot with my m4/3 cameras and get decent results. With the A7S I can not only get great results but I don’t have to go anywhere near the noisy edge of its ISO capabilities. What does this mean? Well for me it means quite a lot. I own only a couple of lights and they are not very good. I bought them for doing a basic interview set up. Which means 90% of the time I am shooting with available light. With this camera I see myself being able to get usable, decent results in all kinds of conditions. And I will be able to control what is in focus with greater ease because I will not be tied to any particular f stop. Whereas before I found myself not even taking out my camera, despite the fact that I was standing under a street light, or looking at a city street, simply because the camera could not find enough light to produce a usable image.
Unlike the GH4 the A7S does not conform its high frame rate footage in camera. So while I dislike that the resolution drops to 720p I do like that it captures audio. Since nearly everything I make is posted to the web this is not much of an issue for me. I can either make the entire project 720p or I can upscale the slow motion footage to 1080p and use it in my timeline. The footage from the GH4 was also lesser quality at these higher frame rates (in its own way) so while I wish all of the options were great I will take what I can get. With the A7S I can still have the sound and choose when I want to make things very slow.
So the conclusion: I really do not know. Isn’t that awful? Today is really the first day that I have gotten to play around with the A7S and once I started comparing the footage it captures against the BMPCC I was disappointed. Mind you the best mode for capturing footage is S-LOG which has a minimum ISO of 3,200 – which means a whole lot of neutral density filters (which I do not have) or you shoot in another mode. I have some filters but nothing strong enough to make that high of an ISO usable. I am also guessing that without filters that deal with infrared you get IR pollution and things look terrible.
That being said this camera has many more controls and options over the image than I am used to with the GH3 and the BMPCC. If I do decide to keep it I will most likely purchase an in-depth guide like this one to make sure I am getting the most out of the camera. A video shared by B&H of Philip Bloom explaining his settings for the camera has served as a useful starting point for me and given me some confidence that I am not doing everything completely wrong.
I am learning as I go and sadly this is a very expensive hobby. If you are smart you are already working with your camera and adding bits and pieces as you go. I am not so smart. My hope with this post is to share what I have learned and try and offer a way to bypass a number of the mistakes I have made.
I am in the position now of keeping the A7S mostly for the low light capabilities but still relying on the BMPCC for the more controlled shots or returning the A7S and getting my GH3 back. If the GH3 can be intercut with the BMPCC I know the A7S can, the question is it worth all of the extra money. If I were to spend half of what I spent on the A7S I could buy a number of accessories that would make my shooting life easier (slider, skate dolly, external monitor). Mind you if none of those are lights I am going to find myself unable to shoot in all sorts of situations again.