[leekil]

Two questions:

What's the pinhole made of? I've seen some where the hole is a bit reflective inside and that will scatter light.

Do you have a really deep lens hood on it? The effective image circle on a pinhole is huge. With no hood more than 10X as much light will be bouncing around the mirrorbox and fogging the image as is part of the image.

My pinhole might be a little reflective; I seem to have misplaced it, so I'll have to find it to check.

I also did not use a hood. It's a pinhole "body cap" lens, and doesn't have any sort of hood attachment mechanism, but I could try something with cardboard & tape.

[swanlefitte]

the 4 lenes I checked adjusted brightness to be almost the same. There are differences. Comments anyone? Whites are almost identical. How?


sigma 70, pentax 77, 50-200dal, nikkor 105 2.5 ai

[Na Horuk]

Whites are almost identical. How?

Lenses are really complex, so they don't have just a "shift" in colour, but various differences. This is due to lens materials, lens coatings and optical design. This is why lenses render colours and contrasts differently. Pentax is known for good colours and contrasts. Sigma has traditionally had a different "look"
Some lenses might even be optimized for specific colours. For example portrait lenses must be great at rendering good skin tones. A product photo lens must be more clinical.
With digital PP this is becoming less important because people just slather on a bunch of Insta filters that change everything anyway

[stevebrot]

So how can I find a difference or is the human eye that good at tiny differences?

The latter. The human eye does not readily perceive "blue hour", "golden hour", or bluish open shade. The mind does a sort of auto white balance. Oh, and BTW, the in-camera auto white balance, while usually quite good, often is a little warm with tungsten/golden hour and a little cool in open shade. I consider that to be intentional so as to not cheat the photographer from the effect.


Steve

[stevebrot]

Whites are almost identical. How?

Are they identical when sampled for RGB values?

Steve

[leekil]

Whites are almost identical. How?

Did you have Auto White Balance set?

[angerdan]

Did you have Auto White Balance set?

flickr EXIF:

• White Balance - Auto

So i guess this applies for all images.

[Alex645]

the 4 lenes I checked adjusted brightness to be almost the same. There are differences. Comments anyone? Whites are almost identical. How?


sigma 70, pentax 77, 50-200dal, nikkor 105 2.5 ai

The Sigma 70 is the coolest and the Pentax 77 the warmest. The Nikkor 105 is the most neutral, but AWB is a factor.

Are you willing to do this again with manual white balance?

In sunlight, set it to daylight/sun, or if you have to use flash, set it the white balance to flash. Turn off any other lights in the room and don't let any natural window light into the room. Shoot with wall, ceiling, carpet, etc, that is as white, black, or neutral gray as possible including your own clothing. And if any of these lenses have filters, take them off. Also best to shoot in RAW as jpegs will be processed and the color balance possibly affected.

[stevebrot]

flickr EXIF:

• White Balance - Auto

So i guess this applies for all images.

...or at least for the Sigma 70mm file. The EXIF read by Flickr was originally from the Sigma 70mm file and did have auto white balance. I would have expected it to be warmer since the Sigma macros tend to veer that direction. Well, I guess that is a head's up when doing composites in Photoshop.

What's interesting, if AWB was on, is the variation in the color patches despite the white patches being essentially identical. There is more to color cast and color rendering than can be addressed by white balance.


Steve

[swanlefitte]

the auto white balance didn't matter they were RAW and I changed the white to 5600 and set the tint to 0 as stated above on all images because that is what yongnuo says the flash is. The only other thing I did for this composit was to set the RGB brightness so the values on the white square were close to 200 for red and thus GB are slightly more.
Unfortunately my phone crashed and I don't have time to get on a computer much for a few days so I can't check back here much.

---------- Post added 12-18-17 at 07:12 PM ----------

the auto white balance didn't matter they were RAW and I changed the white to 5600 and set the tint to 0 as stated above on all images because that is what yongnuo says the flash is. The only other thing I did for this composit was to set the RGB brightness so the values on the white square were close to 200 for red and thus GB are slightly more.
Unfortunately my phone crashed and I don't have time to get on a computer much for a few days so I can't check back here much.

[ScooterMaxi Jim]

You know what they say about RAW. It is raw in name only, and in-camera processing alters what's rendered from the sensor. Not that I have any scientific evidence, but the warmth in Pentax imaging varies based on particular bodies and - most certainly - the raw processing engine's profiling. It is pretty clear that Adobe (and therefore LR) profiles are put together with far less care than, say, Canon profiles (having shot both systems extensively in the days when LR was my main processor). The magenta shift problem is almost never corrected entirely when dealing with the Adobe profiles. In Capture One, Pentax and Canon profiling are pretty much equally meticulous and you get the magenta problem largely corrected.

Sorry to throw another challenging variable into the mix, but there's no doubt its another factor that makes finding "neutral" or "clinical" a nearly impossible task. You might be best off using JPEG and manual color temp - at least then you are getting what Pentax engineers intended (which, frankly, isn't neutral). You might be best off just accepting that the various lenses can be ranked from warm to cool, and you will have to accept what is closest to neutral based on your own preferences.

[swanlefitte]

Interesting point. I didn't use any lens profiles but it would be good to compare them with profiles enabled. I think my next thing to figure out is if the RGB % changes by color between lenses at a different rate from white.

[ScooterMaxi Jim]

The lens profiles - whether LR or Capture One - are remarkably unsophisticated - focusing entirely on distortion, light fall off and CA (sometimes) - and certainly have no color correction other than basic CA adjustments.

The point I was making relates to the coded changes made at the Pentax body sensor level which is somewhat ballpark with the Adobe profiling, and more precise with Capture One. Pentax introduces significant shifts in the sensor raw processing toward magenta bias (hence "warmth") that you don't find otherwise in the Nikon handling from the same Sony-sourced sensor. (Not that Nikon doesn't over-manipulate in other ways, such as unabashed underexposure followed by heavy smoothing compensations to claim DR that in reality isn't any different than the more honest approach taken by Pentax that properly labels DR highlight expansion rather than implementing it secretly...) Bottom line, RAW is not raw at all... and getting to a "neutral" treatment involves cracking something of a trade secret.

[RAART]

Interesting point. I didn't use any lens profiles but it would be good to compare them with profiles enabled. I think my next thing to figure out is if the RGB % changes by color between lenses at a different rate from white.

Not Lens Profiles are changing the colors in LR... you have to go to color profile and then choose "embeded" for K-3 and similar or for the K-1 yo will have multiple options there. I guess it will help also if you choose lens profile first to get rid of CA and some distortions.

[fehknt]

I have a commercially made pinhole (f/150) and it works fine on film, but not on digital. I was told the reason is that film emulsion is thin enough to tolerate light hitting it at various angles but that digital sensors need the light to hit it directly (90 degrees to the sensor). On film, I can get a relatively sharp, wide angle image with tons of vignetting. On an APS-C sensor, it's unacceptably soft.

Can you post examples of some of your 'fine' images made with a commercially made pinhole and any details?

pinhole examples - Google Photos

It's certainly not sharp but, diffraction-only photography is not going to be sharp. The above was with K3-ii and "rising pinhole standard" body-cap style pinhole. I've only used lenses on film so I don't know how they compare. But, to the point of the original thread, it's certainly sufficient for comparing color patches, as long as the lighting is from a controlled source that won't vary across the fairly long required exposure time.