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 :)

Exposure ASA and Latitude

B

Brian Broz

Well-known member
Joined
Dec 28, 2006
Messages
457
Reaction score
0
Points
0
Website
www.brianbroz.com
Had the pleasure of shooting some outdoor tests with RED last weekend.
Only had up to an ND .9 filter and obviously shooting outdoors (on a fairly bright day) 3 stops of light reduction wasn't enough at 320ASA. Found myself adjusting the shutter constantly if I wanted to shoot wide open for shallowest DOF.

My question is, if one lowers the ASA rating to 100 does that effect latitude in any way? I realize the camera shoots RAW but am trying to wrap my head around this. My previous experience with the F900 running negative gain -3db was that the signal may have been "cleaner" in terms of noise but at the expense of latitude.

Hoping for any experiences and thoughts on changing the ASA values from the baseline 320ASA for those waiting for hot mirror IR ND's to be readily available.

Cheers,

BB
www.brianbroz.com
 
ASA does not affect the raw image data, only the (monitoring) display in camera and the QT proxies. The raw data will contain metadata w. the asa exposure settings which can give you a jumping off point in post, but the raw data itself will be identical no matter what ASA you use, therefore it'll have no effect on the latitude of the camera.

cheers,

JT
 
Processing it at a lower iso will indeed clamp the Dynamic range a bit.
 
Thanks guys.
That's what I thought about adjusting ASA and clamping DR.
My main concern is obviously maintaining maximum dynamic range so I'll stick to 320ASA and occasionally step up to 500ASA if required for low light shooting.
So heavier ND it is then:-)
 
Brian Broz said:
Thanks guys.
That's what I thought about adjusting ASA and clamping DR.
My main concern is obviously maintaining maximum dynamic range so I'll stick to 320ASA and occasionally step up to 500ASA if required for low light shooting.
So heavier ND it is then:-)
Click to expand...

Isn't it the opposite? Contrary to instinct, wouldn't you LOWER the ASA for low light shooting, forcing yourself to shoot wide open? Since it doesn't change the sensor gain, you want to let in as much light as possible so if anything, you bring light levels DOWN in post.

Raising the ASA in camera will only make you close the iris and maybe even start using ND to compensate for the higher exposure. Then in post you will pull the image UP, raising the noise levels in the shadows.

So which is it?

Jim
 
Jim, I think we're on the same page.
I'm talking about low light situations where you are already shooting wide open at 320ASA and don't have enough light.
I know...either lower the shutter speed, raise the ASA or rent more lights!
 
Lowering your ISO to 100 in a bright, highlight heavy and high dynamic range scene will LOWER your dynamic range in your highlights because your light meter/monitor will tell you you need MORE light. They will therefore clip faster, so instead of 3.5 stops latitude in your highlights, you'll be clipping out whites sooner. Setting your light-meter to 500 ISO wil make you stop down/decrease your light/add ND etc which will mean MORE DR in your highlights but more noise in your shadows. You can then create a soft gamma curve in your highlights in post for a film-like roll off rather than a hard clip, and possibly crush your blacks a bit for less noise.

So, your options if you are wide open are more ND/faster shutter really.

Also, if you set your light meter at 500 in a bright scene, process it at 200 ISO for grading/post, then go back to 500 you will also have more control over your highlights.

All this will most likely change in build 16 though!!
 
Of course you will be clipping highlights sooner but it's easier to regain 1 2/3 stops out of the usable blown highlights than it is to deal with than 1 2/3 stops of noise.
 
supernovafilms said:
Of course you will be clipping highlights sooner but it's easier to regain 1 2/3 stops out of the usable blown highlights than it is to deal with than 1 2/3 stops of noise.
Click to expand...

Not sure I understand "usable blown highlight". If the highlight is blown, what do you do to recover 1 2/3 stops of information from it? In digital systems a blown highlight is just plain gone, is it not?
 
Everything "depends". I typically would rather protect important highlights and find the image in the shadows. Higher ISO will give you a more natural highlight rolloff. I typically shoot ISO 500 if I meter or in a contrasty situation. Then "open it up a bit" in REDCINE or RED Alert! Build 16 will help everyone a LOT in this area. I shoot 320 in a lower contrast scene with enough light.

If you "run out of light"...

1. Grab a faster lens.

2. Add light.

3. Increase ISO... up to about 800

4. Open the shutter

5. Increase ISO to 1000

6. Pray and then use some Noise Reduction in post.

Jim
 
Jannard said:
Everything "depends". I typically would rather protect important highlights and find the image in the shadows. Higher ISO will give you a more natural highlight rolloff. I typically shoot ISO 500 if I meter or in a contrasty situation. Then "open it up a bit" in REDCINE or RED Alert! Build 16 will help everyone a LOT in this area. I shoot 320 in a lower contrast scene with enough light.

If you "run out of light"...

1. Grab a faster lens.

2. Add light.

3. Increase ISO... up to about 800

4. Open the shutter

5. Increase ISO to 1000

6. Pray and then use some Noise Reduction in post.

Jim
Click to expand...

This is pretty much standard operating procedure among DP's. You open up the lens as much as you feel you can, you add whatever extra light you feel you can, you increase the shutter speed (if shooting digital), you start underexposing and pushing the film (or digital image), you hope you can use some noise reduction in post if forced to push the image very far, etc.

When shooting on the F900, my first trick when dealing with lower light levels was to boost the gain by +3db and open the shutter from 1/48 to 1/32. Separately, those tricks get you a 1/2 stop more exposure, so together you gain a stop without much image quality loss. After that, the question was whether to go to 1/24th shutter or +6db, and often that depended on the amount of movement in the shot.
 
Slightly OT...

I'm seeing it more and more in digital features (21, Get Smart) where the DP shoots at 1/24 shutter to get that extra light in...

I hate it. It screams video to me, yet it's becoming more and more common. I haven't tried 1/32, but I would imagine that it's not as ugly as 1/24.
 
There is a big difference in 1/24 and 1/32 or even 1/40. Unless it's action I find that shutter can be better than ASA(noise) or lens(super shallow DOF). Of course it depends on the situation.
 
kenn michael said:
Slightly OT...

I'm seeing it more and more in digital features (21, Get Smart) where the DP shoots at 1/24 shutter to get that extra light in...

I hate it. It screams video to me, yet it's becoming more and more common. I haven't tried 1/32, but I would imagine that it's not as ugly as 1/24.
Click to expand...

Kenn, 1/32 gave me superb results with the RED.
 
I've been glancing over several threads regarding ISO/DR and also processing some clips from the redrelay site...

Are most people exposing based on incident readings? Are they doing any spot metering based on 18% gray cards? I think a lot of confusion is coming via peoples metering methods, etc..

As for ISO 100... I plan to shoot a test using that ISO (next week), because I've looked at several .r3d files with clipped highlights shot at ISO 320, and guess what... I saw all kinds of non clipped detail when I set the ISO to 100, which leads me to believe that there is some kind of disconnect with how people are approaching digital exposure VS how people are used to doing it with film VS the whole 18% gray calibration thing... I've never believed the red to be ISO 320 actually, but ISO is forever subjective. Red said they would be re-calibrating the camera's ISO... So I take that to mean they are just going to more or less relabel some settings so that film shooters can immediately use it correctly with a "film/18% gray/incident metering, etc..." mindset. Not a bad thing.

I see a very complex miscommunication / misunderstanding going on between how Red built the system, how shooters interpreted it, popular metering methods that have been around for decades; and lots of people who don't necessary have a strong grasp on how metering really works. See what I mean? Complex...

As for how I go about doing things personally, my first step is to custom ISO rate EVERY camera I use based on clipping via spot meter reading, a proven method. It's pretty simple to do really, just shoot a white card, evenly lit, at several different f/stops. While you do so, you will need to take notes about what f/stop you use on each clip. Keep the lighting level consistent.

Open up all the files in Red Cine at ISO 100. Find the first clip that is as far to the right on the histogram as possible but NOT blowing out.

Go back to the test chart, take a spot reading. Adjust the frame rate and shutter speed/shutter angle to match the camera, then adjust the ISO until you hit the f/stop of the clip you identified as the highest exposure before blow out.

With this setup, all exposures will be determined base on spot metering the brightest object in the scene.

Try it! It works. Yes its different from old methodologies. You don't have to base it at the very edge of over exposure either, you can rig it to set highlights a little lower.

Last, but not least, keep in mind that while meters are precision tools, they will only be as good as the operator using it.

Just my 2 cents. Have fun!
 
Cardmaverick said:
I've been glancing over several threads regarding ISO/DR and also processing some clips from the redrelay site...

Are most people exposing based on incident readings? Are they doing any spot metering based on 18% gray cards? I think a lot of confusion is coming via peoples metering methods, etc..

As for ISO 100... I plan to shoot a test using that ISO (next week), because I've looked at several .r3d files with clipped highlights shot at ISO 320, and guess what... I saw all kinds of non clipped detail when I set the ISO to 100, which leads me to believe that there is some kind of disconnect with how people are approaching digital exposure VS how people are used to doing it with film VS the whole 18% gray calibration thing... I've never believed the red to be ISO 320 actually, but ISO is forever subjective. Red said they would be re-calibrating the camera's ISO... So I take that to mean they are just going to more or less relabel some settings so that film shooters can immediately use it correctly with a "film/18% gray/incident metering, etc..." mindset. Not a bad thing.

I see a very complex miscommunication / misunderstanding going on between how Red built the system, how shooters interpreted it, popular metering methods that have been around for decades; and lots of people who don't necessary have a strong grasp on how metering really works. See what I mean? Complex...

As for how I go about doing things personally, my first step is to custom ISO rate EVERY camera I use based on clipping via spot meter reading, a proven method. It's pretty simple to do really, just shoot a white card, evenly lit, at several different f/stops. While you do so, you will need to take notes about what f/stop you use on each clip. Keep the lighting level consistent.

Open up all the files in Red Cine at ISO 100. Find the first clip that is as far to the right on the histogram as possible but NOT blowing out.

Go back to the test chart, take a spot reading. Adjust the frame rate and shutter speed/shutter angle to match the camera, then adjust the ISO until you hit the f/stop of the clip you identified as the highest exposure before blow out.

With this setup, all exposures will be determined base on spot metering the brightest object in the scene.

Try it! It works. Yes its different from old methodologies. You don't have to base it at the very edge of over exposure either, you can rig it to set highlights a little lower.

Last, but not least, keep in mind that while meters are precision tools, they will only be as good as the operator using it.

Just my 2 cents. Have fun!
Click to expand...

Keep in mind that the least dynamic range will be found at ISO 100. If you use your technique at a higher ISO in contrasty situations, you will have better results. Build 16 helps a LOT in exposure setting.

Jim
 
Jannard said:
Keep in mind that the least dynamic range will be found at ISO 100. If you use your technique at a higher ISO in contrasty situations, you will have better results. Build 16 helps a LOT in exposure setting.

Jim
Click to expand...

But Jim, isn't this assuming I use incident readings based on 18% gray? From what I have read in the cameras manual, all ISO changes are done in post, and therefore just meta data. So when you shoot at ISO 500 and meter, your just underexposing for highlights, which is more or less what I do with my technique (but plan to test with ISO 100 Red Cine settings). It seems akin to exposure compensation settings in a light meter honestly.

My big question here is really this... Will I see improvements operating at ISO 100 and getting the highest highlight exposure I can before blow out? I guess I'll find out next week, albeit on build 15 though...
 
You must log in or register to reply here.
Back
Top