[Gooshin]

if i set my camera's (k20D) white balance manually to a specific Kevlin figure

and do the same for on my friends Canon 7D,

should our white balances be 100% identical, in theory?

I think so, in theory.

After all, you don't get brand specific greycards.

[Ash]

Yes, in theory.
How often does theory equal practice?
I wouldn't be surprised if even the WB is slightly different between them.
But the colours will certainly be - depending on the lenses used...

[Blue]

To get a true idea of the difference, it would probably be good to compare 5 bodies of each camera.

[Eruditass]

Interesting points brought up, lets see someone test it! Of course there would be no way to tell which one is more accurate without certain equipment..

[Blue]

An integrating sphere and a white "standard" would be one way but would only be feasible on a printed copy which would open up a different can of .

[BB_Zone28]

Regrading WB and Color Temp(CT)...

Many cameras and published literature refer to Daylight CT at 5500 K.
I've notice the Pentax K20D uses Daylight CT about 5200 K.

Where or how does Pentax derive their value for CT that differs from so many other sources????????
Does that value include some offset corrections for the specific Bayer filter used on it's sensors or the specific color processing algorithms? If so, why not apply corrections internally and used the same numbers as most other literature uses.

[roentarre]

Canon images are so dulled coloured and heavily smudged by AI filters.

You would get mixed results. Canon typically has bad performance concerning yellow to cyan colour while Pentax does the red and blue badly.

[pingflood]

I think Canon does ok color wise... here's an escaping GBH who didn't want to participate in the shoot.



And 100% crop to show the smudging and other nasties.

[maxwell1295]

WB can differ between 2 identical bodies so I would expect it to differ even more for different brands and models.

[Blue]

WB can differ between 2 identical bodies so I would expect it to differ even more for different brands and models.

That's why I suggested 5 sample bodies of each. All 5 of both would need to either come from the same production batch or 5 different batches.

[KungPOW]

I understand the topic is WB, but the variation of colour between camera brands was brought up, so my two bits.

Yes colour profiles between Canon, Nikon, Pentax etc all differ. But it is so easy to make a colour profile in Lightroom, that the differences become irelevant. I would assume other PP programs allow this as well. So Nikon's images not as "punchy" as Pentax? big deal. Tweak the profile, apply to all images, and you are done.

[Blue]

I understand the topic is WB, but the variation of colour between camera brands was brought up, so my two bits.

Yes colour profiles between Canon, Nikon, Pentax etc all differ. But it is so easy to make a colour profile in Lightroom, that the differences become irelevant. I would assume other PP programs allow this as well. So Nikon's images not as "punchy" as Pentax? big deal. Tweak the profile, apply to all images, and you are done.

I see what you are saying . . . to a point. It still wouldn't hurt to compare images straight out of a camera. In the film era, no one would be saying, "yeah the film sucks but let the processor fix it . . . "

When I'm doing an image for scientific purposes, I won't it as close to what I see as possible whether it is for a presentation or journal.

[KungPOW]

In the film era people celibrated the various characterists of different film. kodachrome certainly does not have the first prize for acurate colour.

As for getting archival quality colur acuracy now, how do you do it? All cameras have variation in their colour profiles. How do you get the best possible colour reproduction? (i'm not asking this retorically, I really want to know).

[Blue]

In the film era people celibrated the various characterists of different film. kodachrome certainly does not have the first prize for acurate colour.

As for getting archival quality colur acuracy now, how do you do it? All cameras have variation in their colour profiles. How do you get the best possible colour reproduction? (i'm not asking this retorically, I really want to know).

There are actually color standards out there that the sensor images can be compared to. This has to be done with instrumentation such as integrating light spheres etc. to calculate the hue.

We now that individual units (i.e. camera bodies) will have some variation. That is why we would use at least 5 bodes. This variation is what allows us to statistically compare them.