Jump to content

Red Scarlet W with Dragon sensor - Exposure


Ruben Arce

Recommended Posts

Thanks in advance for sharing your experience with me. I'm going to be DP on a friend's shortfilm and they want to use the Red Scarlet W with a Dragon sensor. I'm excited about that because I have never used a RED camera, but I have some questions, specially since I won't have acces to the camera pretty much till the shooting day.

 

How do you expose for this camera? I know it sounds like a dumb question but I'm really asking, where do you put your 18% gray? and how's your experience with your technique? I like using my light meter to set exposure, but now with a variety of cameras available and manufacturers putting 18% gray wherever they want it's complicated.

 

Here is my point:

 

I have an Ursa Mini 4.6k and it's a great camera. I tested the camera using an 18% gray card and some other tests and I determined that in order to get proper exposure without correcting in post I have to overexpose the camera by almost two stops. I set ISO 800 on the camera and ISO 250 on my light meter, and I get what I consider proper exposure. There is a video where Shane Hurbult tests the camera and he gets pretty much the same results. I guess it was sort of embarrasing for Blackmagic to say their camera is rated at ISO 250 so they said it's a ISO 800 camera like REd and Alexas, but by doing that they are forcing people to under expose a camera that doesn't have tolerance to underexposure. It's a great camera, if you expose it properly and that is not at ISO 800. I know they are trying to protect the highlights, so their camera looks good, but I know how to take care of my dynamic range and I rather take it to the limits taking my own decisions.

 

Is there something similar with RED cameras? How do you expose for RED specially with a light meter? Do you over expose or underexpose for some reason? Do you get more noise if you underexpose it? How far would you take ISO and still get a decent image?

 

I'm not lazy, I did read the manual, but knowing that manufacturers play games with specs I rather have some info from people who have experience using the camera or I may say the sensor.

 

Thanks in advance for your help.

Link to comment
Share on other sites

  • Premium Member

Only thing I can think of is that you're not shooting in LOG or RAW mode with your URSA. If you aren't taking the footage, putting it into DaVinci and grading it, then you aren't using their workflow. The Red cameras work the same way, you have to work with the image in post to get anything out of it. Honestly, the end result depends on what LUT you apply. Sometimes I find with the Blackmagic stuff especially, I have to build my own LUT for a few things. The RED luts seem to be a lot better in DaVinci AND in RED cine X. Just remember, Red's shoot RAW only, so there is no ISO or color on the original files. The debayer is done in software, which means you're basically constructing the image out of metadata in post.

 

Personally, I've found the Blackmagic cameras to be overly sensitive, when you take into account the most important things like highlight clipping. For instance, it's impossible to shoot with the cameras without filtration at 200 ISO without clipping the imager in direct sunlight. I shoot 250D stock all the time and never use filtration and never have highlight clipping issues. The problem is the Blackmagic camera are rated at 800, so it just allows for adjustment of ISO which dramatically changes the dynamic range of the imager, over driving it in this case.

 

Red is no different then Alexa which is no different then blackmagic. They all work the same way, they all have the same issues, they all require some sort of cleanup in post to even get a decent image out of them. Even if you really work hard to make it perfect on set, the LUTS would need to be identical between the camera, monitor and in post, which for a young filmmaker borrowing a camera, is a tricky proposition. On shows with a real camera assistant and DIT, it's a lot easier because those guys know to keep that consistency.

 

Personally, I don't use a light meter with digital cinematography. I use zebra's to detect highlight issues. I use the histogram to monitor the overall image and on the red, false color mode rocks. With those tools, ya don't need a meter in my opinion, unless you're doing a very big project with a lot of lights and you're just checking for balance within a scene, I can see how a meter can come in handy.

Link to comment
Share on other sites

Thanks for your reply Tyler,

 

I agree in a few things. I'm shooting log with Ursa Mini and I understand the postproduction process, I still believe that the best and in many cases the only way you can create the right mood is in the camera. I mean lighting the scene in an interesting way and creating three dimensionality and the mood that you want for the picture, a good color palette, good blocking and good framing are things that you cannot get or fix in post. I know DaVinci is very powerful, but still has it's limitations and I don't think just shooting and fixing stuff in post is the way to go.

 

I have had very good results lighting by eye and meter, and treating digital cameras like film cameras. I'm not against the modern tools available for filmmakers these days. Basically what I do is... I find middle gray I test the limits of the camera on the highlights and the shadows with my meter (when I can test the camera before) and I create a mental map of the limits of the camera. That allows me to have information coming from my meter and my brain pretty close to the info I can get from false color, waveform scopes, histogram, etc. but without being an slave of the monitor. I may at the end check the monitor and the scopes, but I'm the one taking the decisions not the camera or the scopes, just like in the old days. You know because I know you are an expert on film as a medium, that people have been pushing an pulling film for a long time, and they test or used to test the limits and behavior of the film in that way and some times people just want to under or over expose a window or the sky and that is what they want to do even if the scopes say that's wrong.

 

What I'm saying is I have noticed that manufacturers are playing games with ISO rates in order to make their cameras more appealing or at the same level as other cameras. The old concept of pushing and pulling, the problem with this has always been grain, and in the case of some sensors fixed patterned noise. If I just assumed that the camera is delivering ISO 800 and I get under exposed footage, I'll have to take it to "proper exposure on post" and that will bring the noise floor up too.

Link to comment
Share on other sites

  • Premium Member

I agree in a few things. I'm shooting log with Ursa Mini and I understand the postproduction process, I still believe that the best and in many cases the only way you can create the right mood is in the camera. I mean lighting the scene in an interesting way and creating three dimensionality and the mood that you want for the picture, a good color palette, good blocking and good framing are things that you cannot get or fix in post. I know DaVinci is very powerful, but still has it's limitations and I don't think just shooting and fixing stuff in post is the way to go.

Ohh agreed wholeheartedly. What comes out of my camera, digital or film, doesn't need much work in post, but it ALWAYS needs something.

 

My point is that unlike film which theoretically can be one-light printed to another stock and in most cases still be acceptable, digital isn't like that. You can't just add a LUT and expect to be done. So a lot of finding what the camera is capable of doing, includes adjustments in post, especially when shooting RAW like the RED does.

 

I have had very good results lighting by eye and meter, and treating digital cameras like film cameras. I'm not against the modern tools available for filmmakers these days. Basically what I do is... I find middle gray I test the limits of the camera on the highlights and the shadows with my meter (when I can test the camera before) and I create a mental map of the limits of the camera. That allows me to have information coming from my meter and my brain pretty close to the info I can get from false color, waveform scopes, histogram, etc. but without being an slave of the monitor. I may at the end check the monitor and the scopes, but I'm the one taking the decisions not the camera or the scopes, just like in the old days. You know because I know you are an expert on film as a medium, that people have been pushing an pulling film for a long time, and they test or used to test the limits and behavior of the film in that way and some times people just want to under or over expose a window or the sky and that is what they want to do even if the scopes say that's wrong.

It's true, understanding the behavior is important. I was just saying the post production process is what generates the image with the RED cameras. Unlike the Blackmagic which you'd normally shoot Pro Res in LOG mode, the RED you'll be shooting raw which has no debayer on the file and no ISO baked in. So when you make adjustments, you are basically pushing and pulling just like in film. So the ISO setting on the camera really means nothing in the long run. You'll always be shooting around 800 ISO range and pushing or pulling in post to get your image. So it's really more subjective/creative than "scientific" because you can make that chart look like anything you want in post.

 

What I'm saying is I have noticed that manufacturers are playing games with ISO rates in order to make their cameras more appealing or at the same level as other cameras. The old concept of pushing and pulling, the problem with this has always been grain, and in the case of some sensors fixed patterned noise. If I just assumed that the camera is delivering ISO 800 and I get under exposed footage, I'll have to take it to "proper exposure on

post" and that will bring the noise floor up too.

 

Yea, I mean they all play games, but I haven't seen Blackmagic OR RED play any games that are unexpected.

 

I mean, I know what 50, 250, 500 ISO stocks feel like, so when I grab a digital camera and set it to those ISO numbers, I know what I need to do in order to get the shot. As I said earlier, I would actually say, especially the blackmagic cameras, they are "overly" sensitive compared to the selected ISO in my experience. I haven't done tests or anything like that, but just from gut intuition and experience tells me they're a bit more sensitive than the rating in the camera. Now I rarely shoot at base ISO because the noise floor is too high for my liking, so I generally try to shoot higher than that. However, I just shot something at 800 two weekends ago, in a very dark room and it came out great. It came out like 800 ISO film would come out to be honest and better than that, I had a lot of dynamic range to work with.

 

I'll have to take yee ol' meter out and shoot some stuff just with the meter on digital and see what it's like. Honestly, I'm so use to shooting by the seat of my pants with digital, I haven't done any real world tests.

Link to comment
Share on other sites

"Now I rarely shoot at base ISO because the noise floor is too high for my liking, so I generally try to shoot higher than that."

 

But by shooting at a higher ISO than base.. are you not making the noise worse.. your under exposing the LOG footage.. then having to "pull up" in post.. to improve the noise floor you want to lower the ISO and "over expose " your log to pull down in post.. ie base level 800.. rate camera 400.. 1 stop better noise levels.. you lose 1 stop off your highlights but thats usually not a problem with 12-14 stops DR..

Link to comment
Share on other sites

It will be better.. empirically (when you correct it in post) .. by what ever ISO lower you go .. half the iSO I stop.. half again 2 stops.. its almost standard practice for LOG.. esp if you don't agree with camera makers idea of native ISO..making the ISO higher will make it worse.. its the opposite effect ..you raising the noise levels ..when you correct in post.. if you are in non LOG/Custom mode your just adding gain =noise ..

 

The no free lunch is you lose the same off your high lights.. but 90% of the time its a better deal to do that and have less noise .. when you have 6 stops over grey..

Link to comment
Share on other sites

I don't know how you post a graph Robin, but please try again. I have seen that chart that tells you what you are loosing mainly in the highlights when you change the ISO to anything else than 800. Thanks for your input.

 

And thanks guys for all your responses, specially Tyler I appreciate the point of view of a person who has experience and who reacts to a situation in certain way based in all the information and experience that they have accumulated through years. It helps me to put my ideas in order and to see things from a different perspective.

Link to comment
Share on other sites

  • Premium Member

"Now I rarely shoot at base ISO because the noise floor is too high for my liking, so I generally try to shoot higher than that."

 

But by shooting at a higher ISO than base.. are you not making the noise worse.. your under exposing the LOG footage.. then having to "pull up" in post.. to improve the noise floor you want to lower the ISO and "over expose " your log to pull down in post.. ie base level 800.. rate camera 400.. 1 stop better noise levels.. you lose 1 stop off your highlights but thats usually not a problem with 12-14 stops DR..

 

Sorry, I misspoke. I mean "lower ISO".

 

Since 9 times out of 10 I'm shooting exterior in broad daylight, I generally shoot at 200 ISO which is the lowest my camera will go. Since I rarely shoot RAW with my cameras, I have to live with the baked-in ISO with LOG, so I'm kinda stuck. Honestly, it takes A LOT of filtration to use 800 ISO outdoors in broad daylight and the more ND you use, the muddier the image is with the Blackmagic cameras since it doesn't have an IR pollution filter built in.

 

I mean, my entire kit; two bodies, 4 lenses, wireless audio, loop, even the backpack, cost me $3k. So ya kinda get what you pay for.

Link to comment
Share on other sites

  • Premium Member

Yes I thought a typo or something.. just wanted to point it out for the OP.. if you use Sony cameras you have built in ND,s of pretty good quality :)..

 

Ohh no doubt I love ND's in the camera body. That's why I'm saving up for an URSA Pro! :D

Link to comment
Share on other sites

  • 2 weeks later...

Well after a really nice time shooting the project where I used the Red Scarlet W, here are my thoughts... The camera was really easy to use. It is not much more different than Sony, Canon or even DSLRs, the basic controls are there and they are easy to access and control. It's just another camera. Now in terms of exposure the camera was as I expected "underexposing". For some time companies have been placing 18% gray or medium gray at 38 or 42 IRE instead of 50. Why are they doing that? To protect the highlights and make their cameras look better I guess. Why do RED do that? I don't know, because most of their cameras are going to be used by professionals but Sony, Blackmagic Design and now I know that Red are placing medium gray below 50 IRE, which makes putting exposure with a light meter kind of tricky if you haven't tested the camera.

 

As I mentioned in a previous comment I own the Ursa Mini 4.6k and a Sony FS700, the two cameras are design to place medium gray at around 38-42 IRE when shooting log, but in the two cases I found it annoying that the images were underexposed, and when I was bringing the footage to proper exposure in post I was bringing noise up too. If you watch the video of Shane Hurlrburt testing the Ursa Mini 4.6 he gets to the same conclusion I got with the UM4.6... The camera needs light, so you have to overexpose it. I rate my Ursa Mini at 320 ISO instead of the 800 ISO that the manufacturer gives me. I mean the camera is set to shoot ISO800 but my meter is set to 320 ISO.

 

Ok, back to the Red Scarlet W. I tested the camera with an 18% grat card, my light meter and the histogram on the camera and it was pretty clear that the camera places medium gray at a number below 50 IRE, something that looks like 38 IRE. I asked my friend who owns the camera if he felt like the monitor was not that accurate and he told me that the images that he shot with the camera felt always underexposed. That confirmed that my fears, so I overexposed the camera by one stop and when we saw the footage it was beautiful. It was really easy to grade because the exposure was right on the money and trying different LUTs was fun because they looked great immediately. We only had the 5"monitor on top of the camera and I didn't rely on it to illuminate the scenes. I was using my light meter and my eye to set the light and I was looking at the monitor at the end just to frame the shot and roll, and because I was putting light by meter the shots came out very close to each other on terms of exposure that we just put them in premiere and started cutting.

 

I know my Sekonic 558 Cine doesn't know that the RED Scarlet W, or my Nikon or my Sony cameras exist, but it's a great instrument that allows me to have good continuity and repeat-ability from one shot to the next one without relying on the monitor. I hope this help people like me who have been looking for that kind of answers to get a better understanding of measuring light and exposing scenes.

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...