Zum Hauptinhalt gehen

⚠️ Please note that this topic or post has been archived. The information contained here may no longer be accurate or up-to-date. ⚠️

White Balance - weird values?

Permanently deleted user
Hi all,

I'd like to open a discussion about the numeric values shown on the White Balance Tag, especially for different preset WB settings such as "daylight", which I consider somewhat weird or at least far off values that I know.

For example: isn't "daylight" considered something around 5500-5600K?

C1 4.01 suggests for daylight 6854K. And so on.

I don't say that WB is off. But the numerical values seem so strange. Any ideas what might be the reason. I really would appreciate some explanation from Phase One staff too.

Cheers,
Michael Fritzen
0

Kommentare

9 Kommentare

Datum Stimmen
  • Jonathan Gilbert
    Hi Michael,

    The Capture One color temperature scale is not really a a meter of color temperature, but rather a color temperature adjustment scale that references color temperature principles. Because each camera capture light differently the same color temp value in CO4 will look different in the file. For instance if the actual measured color temp of a scene is 5500K on camera might reproduce correctly at 6300 in CO4 and another at 4800. The preset values you see are the balances from the camera converted to CO4 equivalent numbers.
    0
  • Henry1
    [quote="JonGilbert" wrote:
    Hi Michael,

    The Capture One color temperature scale is not really a a meter of color temperature, but rather a color temperature adjustment scale that references color temperature principles. Because each camera capture light differently the same color temp value in CO4 will look different in the file. For instance if the actual measured color temp of a scene is 5500K on camera might reproduce correctly at 6300 in CO4 and another at 4800. The preset values you see are the balances from the camera converted to CO4 equivalent numbers.


    Feels like Alice In Wonderland
    0
  • Jonathan Gilbert
    The basic point is that the color temp scale in CO4 is not a colormeter, but a reference for adjustment. judge color by the outcome and not by the numbers.
    0
  • Henry1
    But Kelvin (K) is a known standard scale and hence saying something is 5000K means there is a value people can "expect". Problem is this not only happens in Phaseone C1but also in PS and lightroom.
    0
  • Jonathan Gilbert
    Understandable. Problem is that the K scale for color, calibrated to blackbody radiators is standard as it relates to the calibration of film color filtration. CCD and CMOS are colorless of course and color is assigned by applying profiles to filtered values. It would be great if Digital cameras used built in color temperature meters to analyze proper color balance, however this is not the case. Instead color is often determined by neutralizing the highest values (this is an overly simple explanation for sure). In this way each camera operates different producing different values which are later assigned. If we were working with only one camera and one CCD it would likely not be to difficult to produce a relatively accurate color temperature scale, but of course this is not the case.

    I think your main point is that given this problem, why use the scale at all?? Well essentially we use it because it is how most photographers are accustomed to thinking about color. It provides a nice frame of reference. I remember some of our competitors used RGB adjustments at one point, and after many requests from photographers and digital techs moved to a Kelvin scale in more recent software releases.
    0
  • Permanently deleted user
    Hi Jon,

    thanks for the clarification. I've been aware that despite there may exist a standardized Kelvin light temperature metering scale the perception of colour is essentially subjective. For an example, who would think about adjusting a scene lit by candle light for correct colour temperature and thus killing the atmosphere.

    But in a certain way I believed both camera and conversion software were trying to keep close to the Kelvin scale in order to have a minimum of a reference. Actually I'm processing, or better re-processing, some KM 7D shots taken in autumn. First processing I did with RSE and I was never pleased very much by the colour rendition, especially of the yellow / gold of the leaves. Photos were taken on sunny days and camera's WB was consistently around 5500-5600K. From my point of view what else could it be on a sunny day? Now I re-processed a larger number in C1 V4.01 and I started to rise in 50K steps the WB setting. The result was that now the yellows are right the way I have leaves colour in mind from the moment I took the photograph. What would confirm your position that it is better to adjust in the conversion software WB to taste than relying too much on values.

    Quite interesting point. The only downside that one looses basicly any reference.

    Cheers,
    Michael Fritzen
    0
  • Henry1
    [quote="JonGilbert" wrote:
    Understandable. Problem is that the K scale for color, calibrated to blackbody radiators is standard as it relates to the calibration of film color filtration. CCD and CMOS are colorless of course and color is assigned by applying profiles to filtered values. It would be great if Digital cameras used built in color temperature meters to analyze proper color balance, however this is not the case. Instead color is often determined by neutralizing the highest values (this is an overly simple explanation for sure). In this way each camera operates different producing different values which are later assigned. If we were working with only one camera and one CCD it would likely not be to difficult to produce a relatively accurate color temperature scale, but of course this is not the case.

    I think your main point is that given this problem, why use the scale at all?? Well essentially we use it because it is how most photographers are accustomed to thinking about color. It provides a nice frame of reference. I remember some of our competitors used RGB adjustments at one point, and after many requests from photographers and digital techs moved to a Kelvin scale in more recent software releases.


    Over the past few years when I started shooting digital, I have found my Minolta Color meter to be quite useless. I can read color temperature of a light source and key the value into my cameras and it would be way off. In the end, a simple custom WB or shooting a reference card works better
    0
  • Permanently deleted user
    Hi,

    returning to this question after some months and now at release Capture One V4.6 the colour balance or better saying the numeric values related by C1 with specific lighting conditions is still giving me some headache.

    Actually there's a thread on depreview discussing possible WB problems of the A700 when using flash. Some are reporting the WB being off shifted to the "warm" side. This remembered me that when processing recently A700 cRAW files (FW V4) in C1 V4.6 where I had used external flash outside to compensate for insufficent lighting that all files reported a WB at around 10.000K and a tint of 2 to 4. Besides the question of the (far) too warm WB I was again wondering why Phase One is still maintaining the numeric value scheme so far off of what seem to be "common" numeric values.

    I understand based on Jon Gilbert's reply that WB should be adjusted to taste. But what I don't understand yet why Phase One continues using such off numeric values which turns it hard (or impossible) to compare and discuss settings and possible steps for improvements with useres of other Raw-conversion software. My take is if the numbers themselves don't count really for a correct WB why Phase One does not simply "transform" those numbers into values closer to what long-term-use would consider adequate, for example around 5.500 for daylight and 6500-6600 for flash. If this means "just" introducing a new count 10.000 - 3.500 = 6.500 why can't this be done? This would be much more user friendly - apart from any disputes about colour theories. What is the main point why Phase One has adopted a different scheme of numeric values?

    I hope this will be addressed some day by Phase One.

    Cheers,
    Michael Fritzen
    0
  • Permanently deleted user
    As a user, I understand exactly what Jon posted. Color temp is essentially imaginary except for heated blackbody physics, and any visual scale is highly subjective. While it is true that the old Minolta color meters don't work with digital...some newer models do work to an extent. My sekonic works with my canon in most situations, but reports a lower 4K value for my strobes...manufacturer says something different....not to worry. This is not a standard value like length or weight that can be easily quantified...so don't put any faith that there will be a table to use AFA color temp......and until then, there's no abiding reason that the lighting, camera, meter, or raw converter are right or wrong....it's just a direction to move in, changing to warm/cool might be better for a label.
    0

Post ist für Kommentare geschlossen.