Delderby Loc: Derby UK

GAS496 Loc: Arizona

I have a horse. And he is a high horse at that, 17 hands tall. I suppose since I shoot film exclusively mostly large format 8x10 then I can ride as high as I want.

As photographers we are all concerned with getting the exposure as correct as possible in camera. No matter what platform we prefer to shoot from, high or low.

Cheers! Jerry

rmalarz Loc: Tempe, Arizona

I like incredible things.

Regardless of what camera settings are being used, blinkies apply to the in-camera jpg, with the camera processing the photo. Subsequent to the testing I do, I rarely chimp, as I have no need to do so.

If I were to choose using blinkies, they'd work the same for me as for anyone else. But, again, I don't chimp.

Rawdigger is one of the tools I use extensively during testing any new camera prior to taking photographs with it. And that's a fact, jack.
--Bob

GregWCIL Loc: Illinois

I think you are missing the advantage of ETTR. It is to capture the most photons so when you post process you get even deeper color depth and less noise. Its about maximizing the signal to noise ratio. One of the negative trade-offs is often a slower shutter speed or wider aperture which may limit what you can practically do.
Yes, I agree with you wanting to end up slightly darker after post-processing.

TriX Loc: Raleigh, NC

Not only that but losing a substantial amount of the dynamic range (and as you said), the signal to noise ratio of the camera. To my mind, there is only one possible of advantage to underexposure, and that is allowing a little extra headroom to prevent blowing highlights if you’re unsure of exactly how your cameras behaves, and even then, 1 stop should be enough. You can fix brightness and color saturation in PP or with in-camera settings if you shoot JPEG.

Delderby Loc: Derby UK

Now that is one very nice horse - and a good picture of him.

GAS496 Loc: Arizona

Thanks Delderby! iPhone photo credit goes to my wife. I will let her know.

selmslie Loc: Fernandina Beach, FL, USA

You said earlier, "RAW, at least on my camera has about 2-1/2 stops perhaps 3 (depending on ISO used) past where the blinkes start."

When I said incredible I meant that "2-1/2 perhaps 3" is not credible. It doesn't make any sense. I don't believe you.

I have found that the blinkies start at a 14-bit raw value of about 8000. One more stop and you are at 16000 and the raw file has no more headroom. The results for my A7 II, D610 and Df are in close agreement. For it to start two stops earlier would mean that they start at a value of 2000. That's about the upper limit of Zone V and only about 1/2 stop above middle gray.

You can easily find the raw value for middle gray by exposing a gray or white card at any ISO with the exposure set to meter off the center of the image. Raw digger will record a value between 1000 and 2000. I have found middle gray at around 1400 with all three of my cameras.

Since you don't use blinkies and you don't chimp, I doubt that you have ever conducted the tests using RawDigger that I have to determine the raw value levels that your camera records at the point that the blinkies just begin.

I'm not alone. Nearly everyone else who actually look at blinkies has stated that there is only between 2/3 to one stop of headroom in the raw file past the start of the blinkies.

You might want to take another look at this using a white balance other than UniWB.

rmalarz Loc: Tempe, Arizona

Scotty, I don't have to justify or explain anything to you, you apparently already know everything. At this point, I'll use a title to a Robert Mapplethorpe book, "Look at the Pictures".

For those that are interested in learning, it does take some diligence, I'm more than happy to help.

And just what is your justification for suggesting a different white balance? Perhaps a lack of understanding the extra data collected using what I use?
--Bob

selmslie Loc: Fernandina Beach, FL, USA

Here is an example from real life. I found the point just before the blinkies were barely visible in the clouds. Then I added about 2/3 stop by lowering the shutter speed. This brought back the highlight warnings because the green channel went above the 8000 level.

I did a little highlight recovery and some shadow lightening in Capture One just to show what can be accomplished.

Anyone can try this for themselves with a real scenario. It's a lot simpler than spot metering or doing a lot of testing under "laboratory" conditions.

It saves a lot of time and effort. Just try it.

JPEG SOOC

(Download)

A little highlight recovery and some shadow lightening in Capture One

(Download)

RawDigger histograms

(Download)

tomcat

Can you help some of us that have to shoot sports in low light, around ISO of 20,000 to obtain more ETTR? I'm shooting at 1/500s, which is as low I can go and still be sharp. Shooting at f/1.8, which is wide-open. I don't know how to get more ETTR without boosting up ISO and it's already high.

rmalarz Loc: Tempe, Arizona

First, let me explain that each camera is different, that is even the same model to the same model. They all vary a bit. So, without testing, it's going to be a wing a guess. You've pretty much maxed out your limits as far as ISO, Shutter speed, and f-stop are concerned. Additionally, I'm not aware of the noise created by your camera at that high an ISO, but ETTR might help. What I'd do is spot meter the lights illuminating the field, court, etc. Once you have the meter centered with that measurement, decrease your shutter speed 3 stops. If you have the f-stops to play with, decrease those. That will increase the brightness of the lights in the image. placing the lights in Zone VIII or IX. If the shutter speed is too slow, increase your ISO and meter the lights again.

This assumes you are using RAW and manual for settings. Also, I'm giving you advice based on what I'd do. Additionally, and coupled with ETTR/EBTR, one has to adjust their processing steps and adjustments. Just increasing exposure without taking into account the capabilities of your processing, is a job only 1/2 done. There has to be a matching of both the shooting and the processing for the ideal outcome.

One other note. By increasing the exposure, you're adding image data to the lower levels of exposure. This will help overpower the inherent noise within each sensel. Your best bet is to experiment and keep some notes as to what you did. Those will prove helpful.
--Bob

tomcat

Thanks for the suggestions Bob. I'm shooting raw and using a Nikon D3s. The gym is very poorly lit, right at the minimum level permitted for safe public occupancy. I'm using spot metering on the players.

selmslie Loc: Fernandina Beach, FL, USA

For properly lit night sports you should be seeing an EV of 9 - about 6 stops more light. See the table I posted earlier. That would permit ISO 800, 1/1250 @ f/1.8 or equivalent.

Amateur sports in a somewhat darker environment would probably not be more than a stop or two darker. You could bump the ISO to 1600 or 3200 and get a decent image.

ISO 20,000 1/500 @ f/1.8 comes to a net EV of about 3. If your night sports are really that dark it would be surprising if anyone can see the ball. ETTR is the least of your concerns at this point.

selmslie Loc: Fernandina Beach, FL, USA

There is no extra data collected in the raw file as a result of UniWB. Look at the result in RawDigger.

The white balance affects only the JPEG and the camera’s histogram, which you apparently ignore because you don’t chimp.