IR Chrome Raws from Canon 6D Mark II

[Andrea B.]

Kolari had 3 Canon 6D Mark II (CR) files available for download made with their IR Chrome filter. So I got them and looked at them in Raw Digger. I made a very informal analysis of the raw colours and the final colours.

 

The raw colours surprised me. They are mostly cyans, lime greens, some yellows & a few slightly yellow-oranges. This makes sense for a cyan peaked filter I suppose.

 

The conversions show reds which are just a few degrees to the right of 0°, about half-way to red-orange (15°). I did not make the conversions. They were done for the download, so I don't know how the white balance was made. You can just click on the raw cyan and produce almost the same thing as these conversions.

 

As I mentioned, this was a very informal analysis. So I wouldn't make too much out of it. I was just having some fun musing about the IR Chrome and other such filters.

 

Color samples were made in Photoshop Elements with the dropper thingie. I used a 5x5 sampler. Once the position on the colour wheel was determined (in degrees), I then took saturation and brightness all the way to 100% so we could "see" the colour. Changing sat/brightness does not change position on colour wheel.

 

The Raw Digger photos all look rather dark because the shots were underexposed up to a stop. It's not easy to get good exposures with a color filter. My trick to avoid this with color filters is to shoot in black and white and to thus balance the overall brightness as best I can without hitting the right wall.

 

Photo #1 Raw Digger Conversion

When I look at the foliage I do NOT see lime green.

But when I sampled the foliage, that is what I got. Strange....

 

Photo #1 Blow-up and crop

As a sanity check, I blew up some foliage in Photo #1 and carefully resampled it.

Still got a lime green. OK, then.

 

Photo #1 Conversion from Kolari)

Not much variation in the reds for this one because of uniform foliage type.

 

Photo #2 Raw Digger Conversion

Here the raw colours had much more variation. The foliage was more yellow than lime green.

 

Photo #2 Conversion from Kolari)

All nice reds except for the grass which was headed towards orange.

 

Photo #3 Raw Digger Conversion

I was tired of using the text tool, so did not label these.

 

Photo #3 Conversion from Kolari)

Note the greens in the water. Again, the grass wants to become orange.

The white building isn't really white. It is a almost completely desaturated blue.

 

 

Here are some color patches 15° apart around red (0°).

To my eye there is "more difference" between red-orange (15°) and orange (30°) then there is between red (0°) and red-orange (15°). Most of the reds in the preceding photos are midway between red and red-orange.

[Andrea B.]

Added a blowup and resample for Photo #1.

[Andrea B.]

Here is the Raw Histogram from Raw Digger for Photo #1 (vines against building).

 

Linear Raw Histogram

Maybe the photo was underexposed by more than one stop. Could be two. (I don't know how hard you can push a Canon 6D/II.)

 

Logarithmic Raw Histogram

The log version better shows the noise on the left in the shadows and highlights on the right.

[Cadmium]

Now your getting somewhere. Not bad, Andrea, not bad at all.

[Andrea B.]

I should also try converting one of the CRs myself. I no longer have a Canon converter, but I'll try some others and see what happens. I'm curious to know how difficult or easy it is to bring out the red.

[Andrea B.]

OK, here's the world's simplest conversion in Photo Ninja. Just white-click some bright cyan area in each photo. Then add 1/2 to 1 stop of exposure on a simple unadjusted demosaic with no color profile. I also brightened the shadows a bit with the illumination slider (just personal preference). It doesn't get much easier than this.

 

The reds this time are a bit to the left of red (0°) range from around 350° to about 3°. So the reds are slightly more towards the cherry patch in the color patch photo in Post #1. A slight turn of the color wheel 5-10° would take the reds a bit more towards the true red to the red-orange as is the case in the supplied conversions from Kolari. That would take another 5 seconds using the Photo Ninja Color Enhancement Tool. (Slide the red patch hue shift slider a few degrees towards orange.)

 

The main point to be had is that it is quite simple to get the red foliage look from the Kolari IR Chrome filter if you are willing to make a white-click in some converter.

 

Can you get the red foliage look SOOC? I don't know cause I don't have the actual filter to experiment with.

I suppose if you set an in-camera white balance against grey stone or a white building, that would probably work. If I had the filter, my next step would be to try using a Spectralon white standard for setting WB either in-camera or in the converter. (I'm not saying that would be definitive. I'm just saying that's what I would try next.)

 

Note: I could have sharpened a bit more?

 

White-click made on bright area of the left column.

 

 

White-click made on building in background.

 

 

White-click made on small grey tombstone just to the right of middle of photo.

 

---

Thank you, Kolari Vision, for offering these raw IR Chrome files to play with !!!

---

[Andrea B.]

attempted sharper version.

I've fallen apart on sharpening over the last year. Can't ever seem to get it right any more.

What has happened?

 

 

[Cadmium]

Looks good Andrea. Even better, I think. Your going to make Birna want her filter back...

[dabateman]

Still might be camera specific. Wait for the sun to come out and have Cadmium take some images with a Nikon. That seems to be the hardest to white balance camera brand.

Shouldn't matter from a raw file, but it might. Cadmium ealier tried from raw and had a harder time than the in camera WB setting. So maybe some play there.

 

[Cadmium]

Oh please, David, get your own Nikon.

Andy Broomé has the filter, he gets orange, play with his RAW.

David, Correction: I have not tried white balancing this filter ever at all yet, you must have that confused with other filters. I have never yet even had this filter on a lens.

By the way, the really snazzy slim filter ring design isn't worth the trouble. Filter and retention ring (from the back) pops out when holding it in my hand,

quite unsafe, I am going to have to remount it in a normal 67mm ring to make sure the glass is secure and doesn't pop out on me and get broken.

Nice looking ring, very slim, but very insecure.

[dabateman]

Oh please, David, get your own Nikon.

Andy Broomé has the filter, he gets orange, play with his RAW.

David, Correction: I have not tried white balancing this filter ever at all yet, you must have that confused with other filters. I have never yet even had this filter on a lens.

By the way, the really snazzy slim filter ring design isn't worth the trouble. Filter and retention ring (from the back) pops out when holding it in my hand,

quite unsafe, I am going to have to remount it in a normal 67mm ring to make sure the glass is secure and doesn't pop out on me and get broken.

Nice looking ring, very slim, but very insecure.

 

Its tricky WB on my Df. But its not converted to full spectrum, so can't play too much in that area with it. The Kelvin range on Nikons is limited compared to Olympus, where I find it much easier to WB or even dial in a Kelvin setting.

 

The cool thing about the Nikon is you can set a custom WB from a previous image on the card. So it is fun that way. But of the cameras I have used, Canon, Panasonic, Olympus, Sony, Sigma, Kodak SLR/n, Phase one, Nikon was the hardest to white balance.

I haven't used a Samsung, Pentax or Fuji yet.

[Andrea B.]

Yes, Nikon is well-known for not being able to make in-camera white balance under some dark filters. Sometimes I want to write fierce letters to Nikon about this. (Like, they'd ever get read! La!) But the Nikon dynamic range and ISO capability has always been leaps & bounds better than most, so some of us continue to use Nikons for UV/IR and simply make our WB in the conversion step. Where, I should add, you are going to be anyway even if you make an in-camera WB. So it is hardly any effort to correct Nikon WB in the converter.

 

Please do note that for those cameras which can make an in-camera WB under UV, IR or UV/IR filters, the resultant WB is not always "accurate" in the sense that you get the same result in-camera as you would from photographing a white standard and setting the white balance in the converter by white-clicking the white standard.

 

The take home message here is that UV/IR white balance is always a tricky business. Those darned dark filters and false colours !

 

Sometimes I think that there is too much emphasis on false colour white balance to the detriment of artistic exploration. UV/IR white balance is useful for technical reasons (standardization, in particular, or for such filters as the IR Chrome to achieve the desired result), but is not mandatory. You, the photographer, get to decide where you want to go with UV/IR WB. (woo! acronyms abound......)

 

---

 

P.S. David, Pentax will make in-camera WB under most of our filters. I have an entire series posted here somewhere.

 

P.S. Steve, you should mention the thin filter ring to Kolari so that they can improve their product. I can see that a thin ring for a 52mm might not work once you get up in to the 60s and 70s.

 

---

 

Members, I would like to hear from a few of you as to whether this "raw color analysis" is useful to you or not. I thought it might help figure out the practicalities of such color filters as the IR Chrome.