For film I believe certain cameras like arri, red, select canon professional camcorders, the aja cion and certain DSLRs that can shoot in 4k are often used, but are those same cameras used for reality television. What type of cameras are usually used for reality television versus film? Are there certain specifications cameras used for that need to have, that ones used for film don't have? What specific makes and models of cameras are commonly used? Why is that? Can the cameras mentioned above be used for realty TV, but they need to be used differently for that type of content? For example film usually is shot in 24fps with a shutter speed of 50 and other variables, but are there different needs for reality TV? Is it good to shoot in 4k like it is in film, since 4k is supposed to closely resemble actual film --performing similarly to a film camera from what I understand? To shoot in something else? Is the color grading/correction process/need the same? Other processes?
Have you ever shot for realtiy TV before? If so, what did you use? Why? Know anyone who has shot for reality TV? If so, what did they use? Why? When I say "reality TV", i'm mainly describing shows dedicated to the following of subjects in primarily an observational manner --not a competition show for instance.
Here's is an example of the type of show i'm describing:
There are no rules, all sorts of cameras get used for reality television, some of the choices are due to budget (same goes for post options like how it gets color-corrected, what it gets mastered to and delivered as, etc.) But many choices are due to the nature of the shooting, i.e. how mobile does the camera have to be, how long does it have to record, can it be cabled or does it have to transmit a live picture by wi-fi or microwave to a director, how big and heavy can the lenses be if they have to be zooms, etc.
I haven't done reality television except back in 2001 just after the Sony F900 HDCAM came out, at a time when all reality shows were shot in standard definition, I was asked to visit a set with a Sony F900 and shoot side-by-side with the reality crew, who had some sort of XDCAM camera and a smaller DV-type camera for b-roll, run-and-gun stuff. Their operators tried out the F900 and all complained about the size and weight. Eventually the show switched to HD a year later but found something lighter than F900's.
Most reality shows shoot 60i (in the U.S.) for that "live" immediate look rather than 24P for that "cinema" look, but a couple shoot 24P. They actually look a bit odd sometimes -- I remember a live MTV concert that was broadcast in HD but shot in 24P and some viewers complained that it didn't look like a live show, it look like a taped show, i.e. it looked like it was shot in the past, not happening live. I've seen some cooking shows shot on 24P and they look more like a film documentary about a cooking show, which was fine, but it took a little time to get used to.
Funny thing is that if you go back far enough in time, "reality" was shot with film cameras, mostly 16mm. If you were going to recreate the look of news footage shot during the Vietnam War, it probably would be better to shoot in 24P and mimic 16mm grain, depth of field, and optics than to shoot in 60i. And of course, there are the NFL films shot in 16mm for decades, until very recently, and that look is very different than the 60i used for the sports broadcasts.
I've seen some reality shows that use a DSLR for some B-roll, like inserts of food, but a DSLR would not be good for a scene where the camera had to run for 20 minutes straight, so they use more of a traditional HD camcorder for that.
A lot of reality these days seems to go C-Series Canon, but the HPX series and the XDCam series. I often see them recording XD to disk when I walk around Noho past this little diner thing. Also I recall Dirty Jobs was on the sony's. The physical media helps in some instances.
Sure they use gopros and Dslrs as well-- some I am sure (pawn stars probably) shoot their intros on something larger format (red maybe) but the rest on some flavor of ENG/EFP camera mixed with prosumer camcorders.
Most of the event and reality-esque stuff I've done which isn't much at all was on the Sony EX series. Sometimes even on things like the Z1U.
The reality show we faked for a horror film-- it being self shot in style-- was a mixture of GH2, XHA1 and some other XH or XA canon camera (it had IR in it for night shooting), and a lot of go-pros (akin to what you'd see in Survivor Man which was part of the style aimed for).
Yep, I see a lot of C series canon and XDCAM/XAVC cameras mostly because the cameras are cheap, have XLR inputs, built-in filter wheel, zoom lenses (auto iris) and the workflow in post is pretty easy since the cameras shoot decent REC709 color space. These are the basics that are necessary for shooting reality.
Reality needs to post very fast, so they don't have the time to deal with coloring before watching. So they stick to standard broadcast formats in camera, 1080p, 29.97, REC709 etc. This gives the editor everything they need within each clip, so they can throw it together quickly. They need this speed because the story is made in post, so they need to edit and report back to set with the next plot point for the following day's shooting. They move very fast, almost at a newsroom speed, so they've gotta deal with standard media.
I've done a lot of reality shooting with my blackmagic cameras and honestly, they suck. It's the little things that are missing... the tilt viewfinder, the motorized zoom lens, the auto iris, built in filters, XLR inputs without shit sticking out the side of your camera, the long battery life, long card life, you don't think about these things until you shoot reality and then you're like, shit... it's the wrong camera.
For a long time in the UK the go to camera for cheap stuff like this was the Canon XF305 but it varies depending on what kind of reality show it is. A fixed rig show like Big Brother or something of that nature might use a wide variety of cameras including tiny lipstick cameras and the like to be unobtrusive. Cameras are hidden about the place without operators so much and signals are fed back to a control centre. Then it will be very different to that for scripted reality which might use Sony cameras and go a bit more like a drama. Made in Chelsea shot on the Sony F3 and some Sony PDW cameras for example. Unscripted is likely to be Canon cameras. Probably the new C series EOS cameras are becoming more popular or this but traditionally Canon XF305.
By coincidence I'm sitting at Plan Check (a burger pub) waiting for my food... and a Food Network crew is working here -- they have two Sony FS7's and a Canon DSLR plus a case of GoPros sitting near me.
Camera operator tells me that they shoot S-Log at 30P (29.97). Canon zooms on the Sony FS7. The Canon 7D DSLR is for time lapse and stills. The GoPros are for trick cooking shots like dunking the camera in water, putting it near a fryer, etc.
Over at Republic of Pie recently they had a little shoot happening it was more a online blog thing than reality but they were doing their interviews on 2x C series, a gopro with a gyrohead mounted on a boom pole (to get a jib like shot which I though was rather brilliant) and 2x 5Ds. I found it rather interesting to watch.
C300s were really popular for a while. There are a few companies I could mention that shoot exclusively Z7 to this day. GoPros and 5Ds are ubiquitous. GH4 is getting thrown in the mix a lot lately. The last one I worked on was on FS7s with cabrio lenses. I expect those will be the new workhorses. A lot of shows are going to Log, which is nice. It used to be all F800 (XDCAM) and panasonic 3100s.
But variety is insane. I have used a half dozen different camera systems in the same day many times. If I recall correctly, the only thing broadcasters mandate is a minimum bit rate that the camera uses (either 25 or 50). There is also a look that is baked into the pitch to network, so if you pitch a "cinematic reality" on a big sensor camera, you can't really go to a 1/3" camera, except for specialty shots.
I've never shot 4K for reality, even when we use 4K cameras, they are set to record 1080. No production company wants to deal with the exta data and processing, if they won't be delivering in 4K. Frame rate is either 24 or 30, and that is decided by the DP/EP/Network. It seems these days there is quite a bit of color work done. Gone are the days of making sure your white ballance and paint settings produced an instantly broadcastable image. Now it is more common to do preset WB so the cameras match, shoot in log, and let the post color handle the final image.
It isn't such a meat grinder that nobody cares about the look, but everyone who has something to do with the image and story is aware not to tip the hat and make it look "produced". There is a lot of work done to keep the verite feel. Each show finds their look along that continuum, but it all has to feel within the lineage of documentary, not narrative film.
Why..? .. if they are shooting Slog,without burning in a LUT .. presume they are shooting XAVC I.. 10 bit 422 HD this is a very good , small file size codec supported by all the usual suspect NLE,s..
Yes in Intra-Frame mode XAVC-I/XDCAM, the files do work in NLE's no problem and they are 10 bit 4:2:2. However, all other modes are 8 bit 4:2:0 and are not compatible with any NLE. Yes, some NLE's transcode on the fly, like Final Cut Pro X and Premiere. Today was my first day on Premiere and I gave it some DSLR and GoPro MPEG's, which it did a quick transcode on each file before allowing me to use them.
With MPEG's (even intra-frame) the compression is extremely high and it's efficient, but not very good. So even though you're capturing S-Log, you're still very limited in how much manipulation you can pull off. I work with FS7 footage all the time in DaVinci and have been unimpressed. If you underexpose slightly to help with the highlight clipping (which is STILL a problem), when you bring up the blacks, they loose all detail. It's the same with highlights, you will get MPEG noise all over the shot. Banding and aliasing can also get "muddy" with MPEG cameras, very annoying when you're sitting in DaVinci trying to correct a shot and you can't do anything without introducing lots of unwanted issues.
Red code (JPEG2000) and CinemaDNG (Tiff's) are far better formats, both are 12 bit, both offer full color space per frame and unlike the MPEG cameras, there is no processing, what you get off the sensor is what's in the file. If you don't like shooting in a RAW format, shoot in Pro Res 444, which is also 12 bit and with the Blackmagic cameras, offer RAW color space within a Pro Res quicktime file at ANY resolution. I have WORKED Pro Res files and never once seen the codec break down. It's always the imager, processor or something else introducing noise into the image. With my cameras, the imager isn't quite good enough for 4:2:2, you do get banding on red/blue highlights. However, switch to RAW mode, it's all gone. The bigger cameras like the Alexa, CION, URSA and URSA mini, don't have those problems because the imager has so many more pixels to work with.
Besides those issues, most MPEG cameras (not XAVC-I) don't write time of day timecode (starting at zero every single clip), they don't have built-in accurate/good naming conventions and worse of all, files can be very slow to access and work with. I mean, I have one heck of a powerhouse computer and dealing with MPEG files compared to Pro Res, is night and day. Pro Res just works and MPEG's work about 1/10th of the time... on those rare occasions when someone shoots in intra-frame.
Canon zooms looked small to me, made for still cameras.
ah Ok thanks.. thought they might be using the 17-120 CN7.. great for run and gun.. but heavy !! .. a lot of people have been buying easy rig,s as now doc,s are also being shot s35.. but ENG style S35 zooms are not light !.. and for Movi etc.. (looks great in the promo video,s .. but not something you want to hold out in front of you for more than 30 seconds ..)
Don't know how they deal with non par focal zooms in that style of shooting.. unless they have 5 camera,s blasting away anyway..
Tyler... with due respect I think you have something wrong with your workflow for XDCAM and XAVC..(/slog.).(obviously you are set up for ProRes). A lot of high end drama/ documentary is shot with this codec..(XAVC 10 bit)..with and without Slog.. the XDCAM 422 HD 50 mbps.. is probably the most used codec on earth.. yes not great for any strong grading.. thats why you have XAVC/Slog .. but all these problems you have with it.. and the rest of the world using it for millions of hours of programs..
The Fs7 sensor,is the same as the F5.. and actually has more pixels than Alexa..
Maybe you are getting mixed up.. the FS7 is a full 4K S35 sensor camera ..
I said; "Yes in Intra-Frame mode XAVC-I/XDCAM, the files do work in NLE's no problem and they are 10 bit 4:2:2."
but all these problems you have with it.. and the rest of the world using it for millions of hours of programs..
The only reason XDCAM was used is due to big contracts with broadcasters and Sony. Whilst Panasonic was still messing around with metal particle tapes, Sony was building random access optical disk cameras. They integrated very well with NLE's and weren't that expensive. Panasonic then switched from tape to solid state media, which was at the time very expensive. Companies still invested in XDCAM because they preferred that workflow, a physical asset that holds data forever, makes more sense to them. As XDCAM switched from optical to solid state, the workflow didn't change. Those companies could use the same computers and software, so this made the switch even easier. Panasonic didn't stand a chance and since nobody else's ENG cameras were any good, companies stuck with Sony.
Nobody uses XDCAM because it looks good, they use it because they had no choice. 50Mbps XDCAM HD is the lowest quality accepted by broadcasters. It's the cheapest solution companies can use for ENG shoots. The last two full-time jobs I had was working with XDCAM HD media at company's producing promo's/reality programing. So no, it's not "my" workflow, it's the industries workflow. In contrast XAVC-I looks 10X better, but there are very few cameras that shoot with it.
The Fs7 sensor,is the same as the F5.. and actually has more pixels than Alexa..
Maybe you are getting mixed up.. the FS7 is a full 4K S35 sensor camera ..
Nope, I despise the FS7. I've used the F5 and F55, not just shooting, but also coloring. With the raw recorder, the thing looks absolutely amazing. With XAVC-I, S-log, it's not quite as amazing, but for sure workable. The FS7 in contrast is substantially crisper and not as cinematic looking. The natural tendency for the camera is a colder look. It can take A LOT of work to get the FS7 colored warm and even when your finished, I've found it to be very challenging because not every shot has the same problems, so you can't build a LUT to combat them because it's all over the map. Coloring with S-Log MPEG's, also has the breakup issues I talked about earlier, any part of the image that was under or over exposed and needs to be corrected, generally becomes mush.
However, those issues are nothing compared to my real big beef of the FS7; over exposing. Let's face it, the FS7 is an ENG camera with some bits taken from cinema cameras. When lit properly, the FS7 looks pretty good, but most of the look comes from the decent/large sensor. However, there is a problem with the image processing side of the camera. The added sharpness which is missing from the F5/55, makes the FS7's over-exposure areas look very harsh and too defined. You see them everywhere from nighttime shooting with car headlights, to daytime shooting with the sun in the background or reflections of sun on objects. Anytime there is over-exposure, there is harshness in the shot. I've been told the problem was fixed with a firmware update, but I just colored some stuff for a friend of mine with a brand new FS7 and all the updates, the problem was still there. If I had done the work at home, I would have posted some samples, but unfortunately it was done at his shop.
Needless to say, I'm a minority. Everyone seems to be completely enamored with the FS7. I just dislike Sony's look and am not happy with their design decisions. They already have a new version coming out, which goes to show you how many problems they had with the first gen. Hopefully the 2nd gen will fix many of those issues, but I would never own one. I don't care how good XAVC-I is, it's a substandard format compared to 12 bit RAW (Jpeg2000/Tiff).
XAVC is also Mpeg .. and all the PDW700,800, s PMW500.. main stay work horse broadcast camera,s of the last 10 years are all 8 bit Long GOP..
That's correct and those cameras need transcoding to work in most editing systems. My comments earlier were about XAVC-I which is "Intra-Frame" and doesn't need transcoding to work. I did mistype when I said HDXAM HD was native, it's not. Only the older standard def XDCAM/IMX format was native. I was mixing up DVCPRO HD with XDCAM HD… my bad.
Fs7 has more pixels than Alexa.. 422 does not lead to banding..
With 444 the Cr,Cb channels are using full resolution. With 422, the Cr,Cb channel use half the horizontal resolution.
If you're using a 1920x1080 imager @ 422, your active pixels on the Cr.Cb channels is 960x1080.
This creates banding between the Y channel and the Cr, Cb channels.
This is why you don't see as much banding with higher resolution imagers in 422 mode.
bad lighting/over exposure are usually user error and not a camera fault
Really… so you'd expose to the car's headlights at night if they were simply driving by your shot? How about expose to the sun in the background when you're trying to get actors faces? Over-exposing parts of the frame is standard practice, with film we didn't think much about it because it deals with over-exposure in a nice soft way. With digital however, we have introduced problems which aren't easily solved. Cameras like the Alexa have a very soft clipping, in fact they're the best at clipping according to a recent test I saw. Guess which camera was worse? It's made by Sony…
The Fs5.. its not replacement for the Fs7.. you haven't read the info..
I've never heard of an FS5. The FS7 is a cheaper, lower price, more ENG version of the F5.
Im not an Fs7 owner.. but a lot of your info is duff..
Naaa, it's just personal opinion based on working with these formats for years. The problem is most cinematographers aren't post production experts and most post production guys aren't cinematographers. The good news is that I'm both and have been both for two decades. I've also served for 10 years as a broadcast engineer, building facilities and restoring old equipment like 2" quad machines. I always joke with friends, the only format I've never touched is VistaVision. But I'm a nerd true to the bone. All that said, I've been around the "technology" side since I was a child and sure digital cinema is a whole new can of worms, but I bought in, experimented and have come up with my own conclusions. It appears what cinematographers are looking for and what a technical person like myself is looking for, are two completely different things.
In my eyes, most cameras are trying to please the cinematographer whilst leaving the technical side of things to be desired. This is how they sell cameras, make the people who buy them happy. Since most people who own high-end cameras, don't edit or color their material, this philosophy has worked well. This is why I've invested in Blackmagic Designs. They've turned that philosophy around, making cameras that may not be the best for the cinematographer, but are technically brilliant for the money. It's taken them four long years of market discovery to come up with the URSA mini and like the D20/21, it took Arri many years to come up with a decent digital cinema camera and they really have broken the mold with the brand new Alexa 65 and SXT 4k. Best imagers, best color science, best capture format, best post production work flow. The Arri currently makes the best digital cinema camera on the market, so the question is, how close with the new 4.6k URSA mini be? It will be 1/10th the price, higher resolution and the same post production workflow. So if it takes Blackmagic Designs 2 years to make the URSA Mini viable, they're still way ahead of the curve. Whilst Sony dicks around with MPEG's, the rest of the world will be working in full color space 12 and 14 bit formats.