CMOS vs 3CCD -- Redefining Broadcast Quality
So which is better?
In regards with something being "better" especially when considering aesthetic value, it's really very subjective. Beauty is in the eye of the beholder, they say. What I mean by this is that there is no standard as to which is better and in the end it really boils down to personal preference. Although I have observed that technological advancement is favoring the CMOS sensor, perhaps because the market is very responsive with the DSLRs coming out that can capture video, as you read this article you will see why I believe that the CMOS sensor is winning the battle.
I've always been concerned with this issue and have always wanted to write about it. Just a few years ago, I'm sure all amateur filmmakers were all under the impression that 3CCD is "broadcast quality" meaning it is really something that can be used for TV but as technology advanced the concept of "broadcast" has also changed. I remember being so impressed with the 3CCD because it has three separate sensors for the colors red, green, and blue thus I've also always believed that the color produced by the 3CCD sensors are superior but nowadays as I observe many videos shot with the CMOS chip, the colors are so vivid and spectacular (of course bigger sensors tend to have better quality and at the same time broadens the possibilities in post-prod).
Also, we now have HD televisions and laptop monitors. Filmmakers are uploading their works on YouTube or Vimeo and are instantly published for the world to see so the concept of what makes something broadcast quality has evolved.
This isn't to say that there aren't any problems with the CMOS sensor especially with the low-end DSLRs coming out that are lacking many features and have many issues such as rolling shutter and video artifacts (these are discussed below). Also, take into consideration that I am not an engineer or an expert with the more technical aspects of this issue. I am merely an observant and passionate young man who enjoys writing about these kinds of things.
DSLR with video vs Camcorder
There are a couple of problems with the DSLRs that capture video. Of course this isn't specifically a CMOS sensor issue because as I have pointed out on the image on the right side, the RED One camera uses a CMOS sensor and filmmakers know how superior that camera is (and way more expensive of course). So I guess in the end it's also about how powerful your camera is and what kinds of features and settings it has.
One issue I have with DSLRs with video, especially the lower-end ones, would be the rolling shutter that causes different kinds of distortions or artifacts caused by using a CMOS chip. My eyes are really so sensitive to this and I've noticed that even some producers who use DSLRs aren't really bothered by it (I guess it's another personal preference issue). These visual effects include wobbles or jello effect (especially in fast camera movements), skewing and smearing of the image, and partial exposure (some parts of the video get different or rapidly changing exposure). This is probably the biggest turn off with the CMOS sensor and again I emphasize that this is more prevalent in lower-end models.
Another is human error. I am not saying that I know everything, I just want to point out that a great deal of research is needed about proper settings and handling. It is common that people who have DSLRs with video capture treat their video in the same manner as they would with photography. The thing is, the rolling shutter works differently from just capturing one shot and many times the visual issues arise because of this lack of knowledge with settings. For instance, it is easily assumed that faster shutter speed would mean higher speed recording but this isn't the case. Having 4000 shutter speed will not mean that you are capturing in high speed because if it's speed you want then you should increase frame rate (some cameras have a feature of 60p). From my research, optimized settings would be choosing a shutter speed that is double your frame rate.
Now with these complexities, more issues arise when it comes to post-prod, rendering, compression formats. Messing up frame rate settings in post-prod and rendering settings can mess up your video. Also, compression formats can have a significant effect on the quality of your video. Of course you want your file size to be small especially if you will be uploading it online but you wouldn't want to sacrifice the quality. Personally, I use H.264 video compression when I plan to upload the videos (maybe I'll write about this soon in a different article).
A video shot with the same camcorder as mine
Actually this was shot with an HV20, the first installment of my camcorder (that means mine is prettier and more advanced). Aside from the specs I've already mentioned in my Canon HV40 review, this is the video above is the first I watched that really convinced me to purchase the Canon HV40. Of course the greatest downside of this camera is that I am stuck with just one lens (there are some accessories though). With the DSLRs, you can easily change your lenses. Using prime lenses will also give you that beautiful depth of field that will really bring your audience much closer to the story you are telling. But of course the lesson here is that one must first gather information before just impulsively buying any gadget. The growth of technology is exponential and we must really learn more about it to avoid wasting your money on something. Ask yourself questions like for what purpose will you use it? What are the key features and settings that you need? How much are you willing to spend? And of course how much are you willing to learn?
A lot of it is really in the user and not the gadget. Talent is behind the camera, they say. It doesn't matter what gadget you have or how expensive it is because if you are not willing to learn and experiment, it will be useless.