[ozzi-paul]

I am currently doing a photography course & I have to make a zone ruler & measure the dynamic range & colour info of each shot. I have done this with the camera setup in the shade of the house in the morning (10am), grey card was on a white poster sized piece of paper & stuck to a very light coloured brick wall. Custom WB was set & aperture was set to f8 & ISO 100. These are the settings we have to use for the test. Also have to do it at ISO 400 & 800 but can reduce the aperture. Anyway, I am getting 9 stops in the range including pure black (RGB 0,0,0) & pure white (RGB 255,255,255) so only really 7 stops with detail. Does this sound about right for my K20D or should it be more. The stop before pure black only has value's of 5,5,5 so not much detail in that one. Images were shot in RAW (dng) & converted to jpeg with ACR & default settings which were very close to a jpeg image I shot (for zone 'V' at correct exposure) just to check the difference.
I have sent the results in for the tutor to check but thought I would ask here to see what other members thought & if anyone else has done this with the K20D.

Thankyou.

[ozzi-paul]

Forgot to say, each shot was taken by only changing the shutter speed by 1 stop at a time. Camera was on a tripod & 2 sec self timer with mirror lock up function used. Lens was a Sigma 70-200 f2.8 set to around 120mm & grey card filled the entire viewfinder.
Thanks again.

[dosdan]

Here's a DR test I performed on a K20D (from www.whirlpool.net.au photographic forum):

Here's a simple DR test you can perform yourself. Walk outside into the sunshine. Noon would be best, but I've just done this now, late in the afternoon. Hold a piece of very bright copy page – I used Reflex Ultra because it has high reflectance. I used a bundle of 6 as a single sheet in the sunlight will be slightly translucent. Put you camera in Av mode, CW or Spot AE mode (CW is better because it's a little broader). I used f/8 & ISO100. Now hold the copy paper at arms length and frame it so the paper is fulling most/all of the frame (45mm FL on an APS-C camera) – you don't need to be in-focus – we're interested in shutter speed.

At this time of the day, when the sun it not behind a cloud, the shutter speed is 1/2500s. Now swing the paper out of the way and frame the sun-lit grass. The shutter speed dropped to 1/100s. The ratio of the change in shutter speed is 25:1 = log 25 / log 2 = 4.6 stops.

Now look for a spot in the shade. I used a spot under a small tree/bush whose foliage stops only 25cm off the ground. I switched to Spot AE and "investigated" the contents of this shadowed area. I got down to 0.25s in some spots. The DR here if I was to include both a near 100% reflective object (Reflex Ultra) & the shadowed area is 2500:4 = 750:1 = log 750 / Log 2 = 9.6 stops.

At noon, it would be greater than 9.6 stops. Now I was seeing details in the shadow so you want more than 9.6 stops – otherwise something 9.6 stops down would be completely dark. Also some room to boost exposure is nice. So 11 stops DR would cover it, but if you start to pull up the shadows to fit them within the limited DR of a printout, shadow noise would become visible, unless you had more reserve DR.

Lens flare may reduce real-world DR. Read the thread off: Re: Realistic case: Nikon D3 - D1 / D700 Forum: Digital Photography Review

BTW, you're probably already aware that 8-bit JPEG can hold more than 8 stops of DR. From my post in the www.whirlpool.net.au photographic forum:

WerTicus writes...
8 Stops i believe is the entire range for a JPG.

No, the encoding is non-linear – a Tone Response Curve (TRC) with gamma compression is applied. According to GordonBGood, the max. possible DR for 8-bit JPEG with a sRGB TRC is 11.7 stops.
See:
Re: JPEG's hold 8 stops of DR, not correct...: Open Talk Forum: Digital Photography Review

Re: How Does High DR Translate to Printing?: Pentax SLR Talk Forum: Digital Photography Review

Tone reproduction - Wikipedia, the free encyclopedia

Gamma correction - Wikipedia, the free encyclopedia
"In any case, binary data in still image files (as JPEG) are explicitly encoded (that is, they carry gamma-encoded values, not linear intensities)"
...
"almost all standard RGB color spaces and file formats use a non-linear encoding (a gamma compression) of the intended intensities of the primary colors of the photographic reproduction; in addition, the intended reproduction is almost always nonlinearly related to the measured scene intensities, via a tone reproduction nonlinearity."

JPEG - Wikipedia, the free encyclopedia
"Many JPEG files embed an ICC color profile (color space). Commonly used color profiles include sRGB and Adobe RGB. Because these color spaces use a non-linear transformation, the dynamic range of an 8-bit JPEG file is about 11 stops"


Dan.

[adelante]

I'm doing the same course (I think) and have just completed a zone ruler for my K20d. What I found interesting is the zone V on my ruler only has 35% white, Zone 6 is 50% so from what I can make out I'm a whole F stop under exposed. This really makes sense as I'm forever increasing the exposure in Photoshop. Making a zone ruler is a great exercise, my camera is going in to get re-calibrated as a consequence.

[jamesm007]

I'm doing the same course (I think) and have just completed a zone ruler for my K20d. What I found interesting is the zone V on my ruler only has 35% white, Zone 6 is 50% so from what I can make out I'm a whole F stop under exposed. This really makes sense as I'm forever increasing the exposure in Photoshop. Making a zone ruler is a great exercise, my camera is going in to get re-calibrated as a consequence.

If you have a warranty go ahead and have it checked. But if not save your money. All owners of Pentax dSLRs before the K5 have under-exposure. Read the reviews for proof. Pentax does this on purpose to save the high-lights at all cost. Typicall I will dial in 1/2EV and even then its not fully using all the room to the right. Also if you shoot raw you will see some pics (JPEGS) are darker than RAW files. There is a lot of info to be played with in RAW files versus JPEGs.

[adelante]

You're right James, there's a lot of comment on the under exposure. I called the agent here in Australia and was told that no adjustment was possible. I'm not concerned because I understand what's going on now and just add another Fstop. The camera is great I love what it can do.

[Lowell Goudge]

I would like to revisit the results the OP posted, and his "9 stops from complete black to complete white."

I did this exercise with my origonal *istD and just with JPEGs (no RAW processing as I am lazy)

What I measured was 10.5 stops total with all settings neutral, However, and this is the important part, you need to look at this not just from the total range of black to white, but how each stop moves in the range.

the middle roughly 5-6 stops (it kind of depends on where you start and finish) are approximately linear, and by this I mean a relitive change of 40 greyscale per stop. when you get below 30 or above 225 greyscale, the next stop has only a greyscale value of 15, the next 7 the next 3,

what you see, as a result is that each stop of detail is compressed into a much smaller range of values.

how did the OPs test work out. For me, you want the important detail only in the middle 6-7 stops, and beyond that, you simply don't worry.

[adelante]

Hi Lowell,

That's an interesting result got with your iStD, my K20d is a bit different, I basically have 9 fstops dynamic range, from pure black (0) to mid grey is 6 fstops then to white is another 3 fstops. mid grey to white is fairly linear with a little compression on the last stop. Mid grey to black is quite compressed. It's a great exercise to do as it gives you a much better understanding of your histogram. My results follow. (i really should check it with another 18% grey card to validate it.)

Zone
AVG
Zone 0
0
Zone I
1
Zone II
18
Zone III
30
Zone IV
50
Zone V
87
Zone VI
127
Zone VII
180
Zone VIII
227
Zone IX
255

[Lowell Goudge]

The grey card should not matter, and I have checked and had similar results with the K10 and K7 The K7 has a very similar sensor to the K20 I used a block wall and measured the medial of the histogram for the value. I started at 1/60 and set aperture to get mid histogram on the camera and then went through shutter speeds in 1/3 ev increments to generate the whole range

[dosdan]

mages were shot in RAW (dng) & converted to jpeg with ACR & default settings

The free Histogrammer program (GUILLERMO LUIJK**-**b i t s & f o t o g r a f í a) may help. It does not decode raw files, but you can do a flat dcraw convert first to feed it a TIF or JPEG. Due to the gamma 2.2 encoding in the sRGB tone curve in the TIF, there can be more than 8 stops of DR.


Here's an 8-bit TIF I had lying around from a K20D raw conversion in SilkyPix Pro. Probably for this, a 16-bit TIF would be better.





I then used the Zones button to export R, G, B & L zonal maps. Here is the Luminance 11-zone map:



Dan.

[TVRDriver]

I guess you're doing the same course as me (Photography Inststitute) ??

The following is my tabulated data which gave dynamic range of 10 Stops for my K20D at ISO 100 in answer to thee Grey Scale Assignment

Zone R G B
Zone -1 0 0 0
Zone 0 0 0 0
Zone 1 0 0 0
Zone 2 9 12 10
Zone 3 24 25 26
Zone 4 49 50 50
Zone 5 82 83 83
Zone 6 123 124 125
Zone 7 164 172 177
Zone 8 204 216 224
Zone 9 250 255 255
Zone 10 255 255 255
Zone 11 255 255 255