ok Arthur, why do reds oversaturate?

[ricktas]

*now tries to find Ricktas to see where he's run off to to hide also*

*not hiding, just watching*

[wayn0i]

This red mushroom image is not far from clipping in the red channel from what I can't see?

- - - Updated - - -

Steve,

Completely off the subject, these mushroom images of yours are amazing, love that bioluminescent green image!

[Mark L]

Many words have been typed but I'm going to try this potentially simple thing,

Then I'd follow that up with an application of -2% contrast reduction which would miraculously recover the blown red channel.

[Kym]

The colour red is the fastest colour (according to physics), so when taking an image, the red hits the sensor first.

Que? Is not the speed of light constant.
The wavelengths are different.

Visible light is usually defined as having a wavelength in the range of 400 nanometres (nm), or 400×10⁻⁹ m, to 700 nanometres – between the infrared (with longer wavelengths) and the ultraviolet (with shorter wavelengths).^[2][3] Often, infrared and ultraviolet are also called light

Update: It depends on the medium http://www.quora.com/If-purple-light...r-inside-glass

[bitsnpieces]

*not hiding, just watching*

Ah, you're right... where did I get hiding from... ???

Que? Is not the speed of light constant.
The wavelengths are different.

Update: It depends on the medium http://www.quora.com/If-purple-light...r-inside-glass

Yes, red is part of the spectrum of light, and light is only one speed.

My remark was more on how physic classes teach if you split light up, red is the fastest colour of them all, so just playing on that.
Which is why playing on the lens such as red's going to refract in onto the lens faster than the other colours. :P

But yeah, I'd have no idea really

[ricktas]

Maybe it is lens design.

Take the ocean, it looks blue cause the sky is blue, why does the sky look blue, cause the water is blue. But go deeper than that. When light hits the ocean at the surface the water appears clear to humans if we are in it. As you dive deeper, light spectrums drop away and then the water around you does appear blue, after that as you go deeper still it appears as a blackness.

What causes this, the water does.

So all this talk about red is based on the assumption that our lenses are perfect, that they capture exactly what is in front of them and pass that perfect light onto the sensor. What if... the elements in our lenses are not all that perfect, and just perhaps the exact modern lens design that makes taking photos so much damn fun, also has a yet undiscovered issue related to the light spectrum that causes some frequencies to pass more readily than others? Thus those frequencies end up being over-saturated.

Just thinking out loud. Cause I agree, that reds do over-saturate more readily, but to find out why, we need to consider all possibilities.

[arthurking83]

I'm pretty sure this is why we see chromatic aberrations in some lenses.

As the light travels through the different mediums(of glasses) the wavelengths all begin to travel at different speeds due to this interference.
Because they travel at different speeds plus the combination of bending light through the various concave and convex glass surfaces .. the now separated wavelengths all land at different locations at the sensor .. and it looks like chromatic aberration to us.


And for Mark L: that -2 contrast reduction is only for ViewNX2 users(on NEF raw files).
I generally only tend to use this software, sometimes CaptureNX2 .. very very rarely Lightroom(4 point something tho!)

I doubt that the same step would work in Adobe(if you're using that).
I think from memory you also use DPP(can't remember?). I used to have it(for testing stuff) but haven't reinstalled it. Try some settings for yourself and let us know what happens.

[arthurking83]

sorry got disturbed by a knock at the door.

To add further to what I was saying to Mark.

My understanding of this thread is more along the lines of why do reds expose more relative to the other two channels.
Which is different to why are my images over exposing.

So my -2 contrast method is really relevant to images where the reds are seemingly ok looking in general, but appear to be overexposed relative to where the the blue and green channel.
This is different to an over exposed image!

My screen shot of the ViewNX2 rendering is more closely referencing the issue.
Look at the blue green cyan histogram. All the data is in the lower(below mid range) in the ViewNX2 rendering but exposed quite ok considering both the real life condition at the time, and the rendered look of the image.
By comparison, for the ACR rendering the histogram shows that the red should be a lot darker, which it isn't in the actual image. The red colour(which is a window blind .. now gone thank god!) appears about right.
But the histogram shows that it should be a lot darker .. maybe a dark red/burgundy colour rendered image.

The blue-gree-cyan histograms tho .. while they are obviously different between the two software, they a lot more closely related.

So the question is more about why is the red channel displaying a wider variation while the variation in the other channels aren't?

[ameerat42]

...I haven't dabbled in Sigma fovean sensors yet but do they have the same reds (or other colour channels for that matter) saturation issue?...

Swifty. Here are some (at last!) from today, in middish-day sunlight, as the afternoon sun proved bothersome with a breeze blowing. They're all in FULL sun, so... Only the 1st image shows a straight
conversion to jpeg, and a toned down one by comparison. The others are all straight conversions except for the odd 1/2 stop decrease in exposure, and a complementary 1/2 stop lightening of
shadows, as notated.
Am.

1. Poinsettia - Left: as taken, Right: 1/2 stop less exposure in conversion

SDIM5860a-b.jpg


2. ??Rhododendron (as taken) and Camellia (1/2 stop reduced)

SDIM5864-65.jpg


3. Euphorbia and a nice YELLOW (for comparison) succulent (1/2 stop down on exposure)

SDIM5866-67.jpg

[swifty]

Thanks Ameerat. I'll have to have a better look when I get home but just to clarify, which camera model were these shot on and with which converter?
And out of curiosity, are the Fovean colours true to what you see in real life as I'm unfamiliar with the flower varieties.

[ameerat42]

ΣSD1M and SPP5.4 (Sigma Photo Pro).
Your Q: In SPP, Yes. and pretty much (most time) in the browser.
I do have some problems on this laptop (Asus) if I don't pick the right color space.
For these, I seem to think they are a bit warm, but not much. Photoshop is where I get some trouble,
and the only way to "solve" it is to discard color profiles when opening files, even though they are the same.
Oh, well!!

- - - Updated - - -

PS: Looking at this post for about the 5th time, I will say that ALL but the Euphorbia look "normal".