Capture One CH

[DP]

Right, but the rolloff occurs *before* the single-channel-clipping point. My reading of what you wrote implied the difference is at the clipping point and beyond.

sure, roll does start gracefully to make a nice picture...

[arobinson7547]

Here is use case;

When creating a profile it’s essential to provide the profiling software an image of the color chart that has an absolutely linear response. “Linear Scientific” provides absolutely linear tonal values. This is what makes that Curve of value to me and why/how I use it.

[Doug Peterson]

I don't think there is a "beyond" beyond the clipping point...

When one of the three channels clips, Capture One uses an adaptive algorithm to maintain pleasant looking color and gradation, which is especially noticeable when using local burns or highlight recovery.

That adaptation is disabled when using linear scientific; if one channel clips and you go to recover it in some way no such adaptation is made and abrupt hue posterization and other artifacts can occur.

Needless to say linear scientific is meant for scientific/technical applications rather than aesthetic ones.

[ErikKaffehr]

Hi,

If the any of the channels clip, the image is overexposed. That would be a cardinal sin in reproduction, I guess. It could be argued that a raw converter should not invoke highlight recovery by default.

Keep in mind, we are shooting with cameras having a dynamic range of 13-14 EVs while the subject has a density range of 7-8 EVs in the best of cases.

But, really, raw processing matters very little. The digital original is the raw file. What is important is that the raw processing should not mislead the photographer to clip the exposure.

Best regards
Erik

When one of the three channels clips, Capture One uses an adaptive algorithm to maintain pleasant looking color and gradation, which is especially noticeable when using local burns or highlight recovery.

That adaptation is disabled when using linear scientific; if one channel clips and you go to recover it in some way no such adaptation is made and abrupt hue posterization and other artifacts can occur.

Needless to say linear scientific is meant for scientific/technical applications rather than aesthetic ones.

[DP]

When creating a profile it’s essential to provide the profiling software an image of the color chart that has an absolutely linear response.

you can however simply use raw data or SSF - which is the way to go (you can extract a 'transfer function' in a separate step)

[DP]

It could be argued that a raw converter should not invoke highlight recovery by default.

commercial software shall please the masses, by default ... PS: you can however create a curve for yourself with whatever flag inside and make it default in C1 for your own use.

[jeremyrh]

for me it was Lab readings and a Linear Scientific curve. v11 now shows LAB measurements and there is a workaround for getting a truly linear curve

Am I correct in thinking that you can't display Lab readings for an offline image?

[ben730]

Thanks all (especially Doug) for the informations. I have a job for about 1000 large repros
and wanted to know if it's worth the investment, but I think although one thousand is a lot, it's
not enough for this software. It's better to invest in an electronic shutter for my SK72 to safe time and power...

Regards,
Ben

[Paul Steunebrink]

A minor addition to the CH Edition story from Doug, is that I understand the the CH Edition is an extension of the DB version (not Pro) in that you can not process your DSLR images with CH.

Correct me if I am wrong.