Saturday, 7 May 2011

White Balance Targets: WhiBal vs Digital Grey Kard

A colour correct image is important for the basis of any post processing: there are a number of white balance targets on the market (such as Xrite ColorChecker) but the 2 reasonably inexpensive options: the WhiBal vs Digital Grey Kard (aka DGK).

The Need

The idea of a white balance target is simple: place the WB target at subject location, illuminated by the key light source(s) and in post, use your RAW processor to adjust white balance based off the white balance reference point.

It is suggested that any neutral object (exposure grey card, white wall, white napkins, white paper etc) can be used as a white balance reference point. However this isn't strictly true, even though such neutral objects can be used in a pinch. The problem with the aforementioned objects is that that are not neutral in colour: for instance, white paper tends to exhibit a blue cast.

Therefore a dedicated WB target may fit your needs if you wish to be exact/precise in your colours.

The Targets

So, to the 2 products: the WhiBal states that all of their cards are measured for spectrally neutral (i.e true grey) and goes to great length to show it's measurements but the DGK only states that their cards are calibrated and tested.

Both products are available in numerous sizes from credit card/pocket sizes to larger (A4) sizes intended for studio.

When I was deciding on my WB target, google returned a few good pages related to WhiBal and Digital Grey Kard comparison, but the most appropriate was Mincel (aka Mustafa Incel)'s blog post.

After consideration, I went with the WhiBal.

However, I recently obtained a DGK and being the type of person I am, I wanted to see the compare the WhiBal and the DGK myself, under my own controlled conditions. From the exif from Mincel's post his comparisons were done with mixed lighting conditions: my comparisons were done under studio conditions - all artificial light from a speedlight in a 45" Westcott bounce umbrella with careful attention to ensure each WB target receiving the same amount of light, measured via a Minolta IVF

Availability in the UK is non-existent for the WhiBal although shipped directly from the US was no problem (also not subject to import tax due to the cost). The DGK is available via Amazon UK, although it is also shipped from the US.

The Images

The images are 14bit lossless compressed NEFs from a Nikon D300, processed in CNX2 as AdobeRGB - the colour correction was performed using the marquee tool selecting top-bottom section of the unobstructed WB ref card. The images were then converted to SRGB and resized for web.

Whilst I appreciate that it is impossible for anyone who hasn't observed my test-setup, to my eye, on my calibrated monitor, there is a subtle but definite red/pink cast to the DGK corrected images.

There are a number of issues that I can see on the soft toy cube from the DGK image, notably the white smiley face has a slight red cast and the gold background for gold X appears muddy (perhaps a green cast?) - conversely, the WhiBal's rendition of the test scene appears much more natural and as it appears in my eyes. Furthermore, examine the colour of the WhiBal card in the DGK corrected image in the image above and again the one at the top of this post - the WhiBal card seems to have a slight reddish/pink/magenta tone.

CNX2 can show red/blue values used in the colour correction stage to render a neutral scene.

Looking at the example above the DGK image (red/blue: 1.15/0.86) has more red being introduced by the WB correction stage when compared to the WhiBal image (red/blue: 1.11/0.87). This is also true of the image at the top of this post: DGK red/blue: 1.16/0.84, WhiBal red/blue: 1.12/0.86.

So, which one is better?

Certainly the 'less clinical' tone that the DGK corrections may appeal to some people but I found it a little off putting. Alternatively, the WhiBal produces what may appear to be cooler images but the manufacturer claims that the cards to be tested and assured to be neutral.

Personally, I prefer having the neutral image (no colour casts). If one then subsequently wishes to creatively warm/cool their images post colour correction then they can without worrying about the initial colour correction introducing a colour cast in the first instance.

Normally in such comparisons I would present my findings and then let others make their own conclusions. However, for this particular comparison I find it difficult to recommend one product due to what I see as inconsistencies in the very task that the product is meant to solve.

In this case, I am glad of my original choice for the WhiBal

No comments:

Post a Comment