The GetDPI Photography Forum

Great to see you here. Join our insightful photographic forum today and start tapping into a huge wealth of photographic knowledge. Completing our simple registration process will allow you to gain access to exclusive content, add your own topics and posts, share your work and connect with other members through your own private inbox! And don’t forget to say hi!

Questtions about Linear Response

Dan Santoso

New member
Hi Guys,

I just got my 1st DB with 645DB. I often shoot on location and my first impression about the LCD is :deadhorse:

Enough about the LCD.

I have some questions about the linear response.
My ISO 50 looks accurate with the light meter showing white at 243 (RGB). The image looks too contrasty under film standard but when I put under linear response, it is 1 stop under. I need to bump +1 stop to make the white patch 243 again. Shadow details is also better.

I tried to shoot over +1 stop so when I use linear, the image wont look underexpose but the white is blowout too much. I can get away at +0.5 stop.

What setting should I use?

Thanks,
Dan
 

Jack

Sr. Administrator
Staff member
Use the LCD histogram when you shoot, or spot meter for your white point. "Film Standard" is a contrast response curve, so it will brighten light tones and darken dark dark tones; linear is flat, allowing you to create your own desired response curve.
 

Dustbak

Member
I think you answered your own question.

When you are satisfied with the tone curve provided by the canned profile use the appropriate metering.

When you plan to do your own curves use the +0.5.

I have a similar issue with the Hasselblad tone curves. Lately I find the default ones downright ugly and prefer using 'reproduction mode' and tweak my own curves. Another thing is that I prefer to know what exactly is happening and the default curve I cannot seem to find what exactly has been done to it. I cannot entirely reproduce it via my own curves and I hate it when things happen that I cannot understand or reproduce :)

Anyway, there is no real right or wrong (unless you screw up your metering totally). Use what you feel shows best what you intended.
 

thomas

New member
re C1:
actually the linear curve doesn't lower the absolut light by 1 stop. It's the distribution of light that is different (think about it as a, well, as a curve within the levels captured). however in the highlights the standard film curve is quite steep. the linear curve might help if you want a lot of differentiation in the highlights as it lowers the contrast in the highlights significantly (compared to the "standard film curve").
as the histogram on a (Phase One-) back shows the linear curve (with the "no color correction" icc profile) and due to the design of the film curves you should be careful with the "expose "to the right" method.
"slightly expose to the right" is the way to go... IMO.
YMMV...
 

Dan Santoso

New member
Yah thanks. 0.5 + stop should be a great place to start.

btw I look over my 1st photoshoot today and I found out a lot of out of focus images. I was using fast AF with multi focus. I changed my setting to accurate and middle focus just now.

Any advices for accurate auto focus? I never have out of focus image with my sony a900. Phase one still has a long way to go.

Thanks
 

jerryreed

New member
I reproduce fine art. So, while my experience may be incompletely useful, it does speak to creating a predictable exposure/color workflow. I am using the SINAR 54H. After getting everything set up and the WHITE reference shot completed, I increase/decrease the power of the lighting in 1/10th stop increments until the white square on a GreTag Macbeth color checker reads 235 (measured using the eyedropper). I use the LINEAR curve processing because otherwise the software re-interprets what I want.

Then in Photoshop, I can place color checker eyedroppers on the image containing the GreTag MacBeth grey squares (the limit is four eye droppers) using the color balance adjustment layer, I can then ZERO out at each of the four selected luminance, the RGB values, making the values of R=G=B. One must ignore the values of the left-most square because it is blue deficient and making the values equal will create a warm color cast.

But, when interpreting the value for the brightest GTMcB square on can change the measure criterion from RGB to LAB or Grayscale either setting will allow you to ignore the hue and attend to making the luminance value equal the 235 that you exposed for using the EXPOSURE adjustment layer.

So, using this method one can go from known color values to reproducible equal values for RG&B at every luminance level. It has only taken me 8 years to work this all out - perhaps you will catch on quicker.

Jerry Reed
 

PeterA

Well-known member
Each manufacturer prepares their 'own' preferred 'look(s) or pre-canned styles etc..where efforts were made to deliver a good starting point for various lighting conditions and uses and manufacturer 'look'

out of the camerra for example daylight balanced Hasselblad files are warmer than their Phase One counterparts and Leaf files are generally brighter and punchier - whilst sinar files are beautifully and wonderfully flat and boring - very very linear and the least interventionist approach

doesn't matter which manufacturer you are working with -given the bit number and DR you are working with - you can (with a little practice) get whatever look you want regarding colour

from your post - it seems that you are asking about DR considerations for highlights - read what Jerry said carefully - because he has addressed the DR issue very well with regards to a specific workflow

btw Jerry some amazing examples of old fashioned tonal work in B&W on your site.
 

jerryreed

New member
Peter:

btw Jerry some amazing examples of old fashioned tonal work in B&W on your site.[/QUOTE]

Thank you. Here is some new work.
 
Top