Welcome to our community

Be a part of something great, join today!

Log in
  • Hey all, just changed over the backend after 15 years I figured time to give it a bit of an update, its probably gonna be a bit weird for most of you and i am sure there is a few bugs to work out but it should kinda work the same as before... hopefully :)

Best Practices. Obtaining Proper Exposure. Share your methods please.

ISO800 doesn't result in more noise, Yousuf... it's metadata...

You are shooting 800 to protect the highlights. Think of the "native" sensitivity as 320... shooting 800 with a right bias on the histogram allows you to expose the sensor well and essentially makes you more careful of the highlights because your view of the image is showing more light sensitivity.

I really don't know how better to explain it - you just need to go shoot. Expose your shot nicely with 800 ISO then switch to RAW view to see exactly what the sensor is seeing... the image is pulled down a bit but still looks great.

800 is your view
320 is what the sensor is "seeing"

Hope that helps.
 
Troy Bakewell said:
As someone who is new to RED and about to be proud new owner of a Scarlett, I'm trying to wrap my head around the Raw/metadata concept. It seems to me to be quite different from what I'm used to. It's not a floating ISO exactly but from what I've read so far, choosing the ISO to rate your exposure at can be counter intuitive sometimes. To protect highlights rate and expose faster for example?


I don't have my Scarlett yet so I haven't had a chance to play around with it yet. For the benefit of us RED newbies, please share your method of obtaining proper exposure.
Click to expand...

Leave the iso alone, simply put it at 800 about. Then you got the bars left and right of your histogram. if you don't se the bars you got no noise and nothing burnt out. Then you got your whole image within the sweet spot of the sensor. Simply think of it as a 800 ASA film stock where you also have an electronic light meter connected... what you see picture vice on your screen is quite irrelevant, just as irrelevant as if / when you where shooting film and looked at the video tap. The image there is just one way of showing the image... since it's raw there are million more ways to have that image "developed". So while shooting you concentrate only on your exposure, frame rate and shutter everything else is quite irrelevant. Offcourse it's nice if it looks nice allready in camera.... But on the other hand I never seen a DP shooting celluloid that where bitching about the fact that he could not get the perfect look he wanted on the client monitor or in his view finder, on the contrary, most DP's are asking for a reflex/mirror view finder even now when cameras actually can show the captured image... this I find a bit odd since to that degree I think it's quite smart to actually look at the result.
 
Shoot properly exposed at 800ISO, lower to 320 in post and watch how much you can bring up the low end and the high end while still not clipping(using Redlogfilm). Now shoot the same shot at 320ISO, leave at 320ISO and push and pull the image and see what you get (not as flexible as 800ISO in regards to DR, but still good if exposed correctly). The main issue with shooting 320 is you clip highlights faster than when you expose at 800 which means you have to be much more precise with exposure and controlling contrast in your scene as your top end has fewer stops DR than at 800ISO (middle grey is closer to the top end at 320 then at 800). Vice versa the low end will be slightly cleaner at 320ISO than 800ISO because you have more stops of DR in the low end. But when you bring that 800 down to 320 the low end will dramatically improve and you have room to bring it up.

You will have better (more even above and below) DR capturing AT 800ISO rather than 320ISO. Ask Timur, he just found that out. RED, Brook, and Evin agree shooting 800ISO is optimal for the camera. I'd argue they have worked with and on some of the largest feature films/tv with RED camera's shooting at 800ISO (POTC4, GWTDT, Hobbit, Prometheus, The Great Gatsby, Southland, Leverage, and plenty of others). So their advice is from experience and testing, not just speculation...
 
Matt Ryan said:
Shoot properly exposed at 800ISO, lower to 320 in post and watch how much you can bring up the low end and the high end while still not clipping(using Redlogfilm). Now shoot the same shot at 320ISO, leave at 320ISO and push and pull the image and see what you get (not as flexible as 800ISO in regards to DR, but still good if exposed correctly). The main issue with shooting 320 is you clip highlights faster than when you expose at 800 which means you have to be much more precise with exposure and controlling contrast in your scene as your top end has fewer stops DR than at 800ISO (middle grey is closer to the top end at 320 then at 800). Vice versa the low end will be slightly cleaner at 320ISO than 800ISO because you have more stops of DR in the low end. But when you bring that 800 down to 320 the low end will dramatically improve and you have room to bring it up.

You will have better (more even above and below) DR capturing AT 800ISO rather than 320ISO. Ask Timur, he just found that out. RED, Brook, and Evin agree shooting 800ISO is optimal for the camera. I'd argue they have worked with and on some of the largest feature films/tv with RED camera's shooting at 800ISO (POTC4, GWTDT, Hobbit, Prometheus, The Great Gatsby, Southland, Leverage, and plenty of others). So their advice is from experience and testing, not just speculation...
Click to expand...

Ok I see your point.
So are you monitoring your histograms and ETTR in RAW view or REDColor view?

and, What does RedLogFilm have to do with anything?
 
Yousuf Abbasi said:
What I don't understand about the "always shoot RED at ISO800 because it is native" is the following:

Let's say I shoot THE SAME scene using the two approaches below...

Case #1: Shoot at ISO320, and compensate exposure with a wider aperture (faster f-stop), while ensuring that RAW histograms are ETTR and highlights NOT clipped.

Case #2: Shoot at ISO800, compensate the exposure with slower f-stop, ensure that RAW histos are ETTR and highlights unclipped.

In Both cases, one could argue the image is "properly exposed" ... however, to me it seems that case #1 produces a better image because you have less noise due to the lower ISO!
(i.e. higher ISO brings more noise)

Therefore, if you know how to monitor your image and exposure properly, why wouldnt you prefer to shoot at ISO320 whenever possible ??
Click to expand...

I think you and others are missing something. In general you cannot ETTR, use your choice of ISO, and have the luma value of your midtones look the way you want all at the same time. You have to chose two of the three. If you ETTR, and chose your ISO based not on your scene luminance but on your preference for ISO 320, your images will most likely appear underexposed/overexposed all the time unless you have painstakingly lit and designed your scene so that your brightest subject in the scene happens to be exactly the same number of stops over as that particular ISO setting has in overexposure latitude. It's not always possible to light for an exact, limited ISO or for an exact amount of highlight latitude, but fortunately if you know what you are doing you dont have to.

Personally I think the flexibility of having many ISO settings with various over/under latitude ranges is great because it means you can always ETTR and have your images look the way you want straight out of the camera. I did an interior day shot recently where we had bright sunny out the windows and a limited lighting package and didnt want to use HDRx, so I shot at ISO1280 (almost did 1600 for an even higher key look) which meant I could ETTR and keep the window detail while also having things look bright inside the location. If I felt stuck with ISO320 or ISO800 I would have had to either blow out the windows or have the talent look underexposed. I'm very happy with the images, and ISO settings let you squeeze exactly the look want out of the camera by giving you control over where you put your luminance range. If you need to compress highlights and bring up mids, use a higher ISO, if you need to compress your shadows and bring down your mids use a lower ISO.
 
Hi Yusuf.

Your basic understanding of this is totally fine.

The advice of shooting 800 is a practical one. The camera calibrationpoint (not really "native point") is 320/daylight. 800 Is a good practical place to start. To set your ISO higher, forces your exposure down. Especially good for tungsten shots, to avoid clipping RED. It also defeats ETTR. Check this:

http://www.gunleik.com/tests/alexmx/RED1_MX_and_Arri_Alexa/ISO_&_Kelvin_-_an_introduction.html

As to noise and exposure and post, I guess one might probably touch into some relevant issues in this thread, too...
http://www.reduser.net/forum/showthread.php?78453-RAW-workflow-a-discussion

Here's a couple of things to think of when fiddling with in-camera metadata:
http://www.reduser.net/forum/showth...RedCINE-Curves&p=995758&highlight=#post995758

BUT, as long as you get your desired results, you are fine, whatever you do!

Cheers!

G
 
Noah Yuan-Vogel said:
I think you and others are missing something. In general you cannot ETTR, use your choice of ISO, and have the luma value of your midtones look the way you want all at the same time. You have to chose two of the three. If you ETTR, and chose your ISO based not on your scene luminance but on your preference for ISO 320, your images will most likely appear underexposed/overexposed all the time unless you have painstakingly lit and designed your scene so that your brightest subject in the scene happens to be exactly the same number of stops over as that particular ISO setting has in overexposure latitude. It's not always possible to light for an exact, limited ISO or for an exact amount of highlight latitude, but fortunately if you know what you are doing you dont have to.

Personally I think the flexibility of having many ISO settings with various over/under latitude ranges is great because it means you can always ETTR and have your images look the way you want straight out of the camera. I did an interior day shot recently where we had bright sunny out the windows and a limited lighting package and didnt want to use HDRx, so I shot at ISO1280 (almost did 1600 for an even higher key look) which meant I could ETTR and keep the window detail while also having things look bright inside the location. If I felt stuck with ISO320 or ISO800 I would have had to either blow out the windows or have the talent look underexposed. I'm very happy with the images, and ISO settings let you squeeze exactly the look want out of the camera by giving you control over where you put your luminance range. If you need to compress highlights and bring up mids, use a higher ISO, if you need to compress your shadows and bring down your mids use a lower ISO.
Click to expand...

Would love to see the shot, if you have a still.

Guys, I get what you are saying about not shooting at 320 ISO. It makes sense. I guess its just the idea of "shooting with film ISO" that messed me up. RED Digital is totally different, and I see now that in order to take advantage of the full DR, we should overexpose (without clipping) and bring ISO down in post, so that the whites are toned down (properly exposed) and the shadows have detail, and you are not adding noise by increasing gain/ISO in post either. Makes total sense. I actually did some tests today, and I think my image results look much cleaner and better exposed. You guys are pros and that's why I love this forum !!!!

But someone still explain to me: when shooting and making sure you ETTR, are you monitoring the RAW HISTOGRAMS or the REDCOLOR Histos? It makes a difference, because RAW histo seems a couple stops to the left.
 
Yousuf Abbasi said:
Would love to see the shot, if you have a still.

Guys, I get what you are saying about not shooting at 320 ISO. It makes sense. I guess its just the idea of "shooting with film ISO" that messed me up. RED Digital is totally different, and I see now that in order to take advantage of the full DR, we should overexpose (without clipping) and bring ISO down in post, so that the whites are toned down (properly exposed) and the shadows have detail, and you are not adding noise by increasing gain/ISO in post either. Makes total sense. I actually did some tests today, and I think my image results look much cleaner and better exposed. You guys are pros and that's why I love this forum !!!!

But someone still explain to me: when shooting and making sure you ETTR, are you monitoring the RAW HISTOGRAMS or the REDCOLOR Histos? It makes a difference, because RAW histo seems a couple stops to the left.
Click to expand...

I view in Redcolor. Reason being that shows you the correct representation for 800ISO. RAW view will show you what it will look like when you take the shot into RCX and bring it down to 320ISO. So in short I expose in camera with Redcolor as the view using the in camera meters/goal posts for 800ISO. I often switch over to RAW when I want to see exactly what in the image is clipped and that is the only time I use it. Then flip back over to Redcolor.
 
Always view redcolor preview/histograms. Like Matt said, that's what's showing your 800ISO preview and will make you protect highlights and retain dynamic range. The stoplights and goal posts are always showing RAW information... you will notice when switching between redcolor and RAW previews, they will not change.

I know it's a lot to wrap your head around - I'm only just starting to get accustomed to this, but the more you shoot the more it'll make sense. Keep testing, and enjoy!! You have an incredibly powerful tool at your disposal... and you're just touching the tip of the iceberg when it comes to knowing how to use it.
 
Yousuf Abbasi said:
Ok I see your point.
So are you monitoring your histograms and ETTR in RAW view or REDColor view?

and, What does RedLogFilm have to do with anything?
Click to expand...

Sorry for the delayed response to this, I must have missed it.

Redlogfilm has no pre set curve like redgamma2, 3, etc. It is a flat image with the full range of latitude leaving it up to you to bring the darks down manually and bringing the mids and highs to proper levels. Basically, you create your own custom curve for the shot. This gives you the most control over your images if you know what your doing and why you make the choices you do. In a nutshell you can get a wider dynamic range by using REDlogfilm and processing your images on a shot by shot basis. RED has created redgamma 2, and redgamma3 to get you 90% of the way to make it quicker, simpler, and less of a hassle to get good images. It's their own s curve applied to your shots which crushes the blacks and rolls off the highlights. So it's a pre built "curve/look" that you can then make minimal adjustments to get a good image. But, it is a basic curve which isn't ideal for every shot that everyone shoots. So in short Redlogfilm gives you the user/colorist complete control, and Redgamma2/3 gives you a pre built curve.

This is a very basic way of explaining it and I'm sure a real colorist could explain it better but this is an easy way of understanding it without getting overly technical.

Hope this helps.
 
Mathieu Ghekiere said:
Thanks for explaining, Matt. I'm also a new Scarlet user, and I was pretty sure that was what RedLog was, but it's nice to see it explained in such basic terms.
Click to expand...

No problem Mathieu.
 
Matt Ryan said:
Redlogfilm has no pre set curve like redgamma2, 3, etc. It is a flat image with the full range of latitude leaving it up to you to bring the darks down manually and bringing the mids and highs to proper levels. Basically, you create your own custom curve for the shot.
Click to expand...

Redlogfilm is a pre-set curve by definition, no? It's just closer to a log curve, which can be beneficial for retaining shadow bit depth when transcoding to lower-bit-depth formats and for those wanting to bring their footage into log space for certain transformations. But yes basically Redcolor gamma curves tend to give you images that are proper video gamma images that could potentially be viewed uncorrected by an audience, whereas Redlogfilm is not designed to look good or natural for viewing but to be transformed into properly corrected images and provide as much luminance data as well distributed as possible for that transformation.
 
Log is a gamma curve, it's not linear. I believe Redgamma 2 or 3 are meant to look decent within Rec.709 gamma while still containing most of the dynamic range of the original, whereas RedLog is an even flatter gamma designed to mimic a scan of color negative film -- on a Rec.709 monitor it will have lifted blacks, pastel colors, and dull, dim whites. It's not designed for viewing but for color-correcting in a D.I. system designed around 10-bit Cineon Log film scans.
 
Can someone explain me why changing ISO settings to lower or higher values moves the mid grey point down or up? It shouldn't change what physically hits the sensor as I understand, so is it some post algorithm behaviour or am I thinking completely wrong here? Also, why is ISO 800 said to preserve most dynamic range? I thought that a well exposed histogram should give the best dynamic range, no matter what ISO?
After reading the whole (great) thread those are the things I still struggle to understand.
Thank you!
 
When you're setting ISO with RED cameras, you're choosing where in the camera's dynamic range you're placing middle grey. And yes, a well exposed histogram (well, depending on camera settings and viewing mode) will give you the best dynamic range. ISO doesn't affect total latitude, it affects how much latitude you have above mid grey and how much you have below it.
 
Jaakko Rinne said:
When you're setting ISO with RED cameras, you're choosing where in the camera's dynamic range you're placing middle grey.
Click to expand...

Thank you Jaakko!
My brain jumps in circles when it comes to logic but that sentence makes the whole nature of exposing that sensor easy to understand for me.
 
If I use a light meter and set the ISO to 320 and the Scarlet to 800, then what do I set the light meter to if Scarlet is set to 320?
Or should the light meter always stay on 320 as it's only metadata? Any suggestions for operators using light meters?
 
Joel Mielle said:
If I use a light meter and set the ISO to 320 and the Scarlet to 800, then what do I set the light meter to if Scarlet is set to 320?
Or should the light meter always stay on 320 as it's only metadata? Any suggestions for operators using light meters?
Click to expand...

Incident meters dont really work. Well.... you can use one to rough in lights, but dont trust it 100% in digital. The ISO correlations never really seem to match up. I like to use a spot meter to measure the distance of sections of the image from middle grey.
 
You must log in or register to reply here.
Back
Top