Mysterious Vortex caught on film in Sedona, AZ !!

[nicol_verheem]

Or at least that's the closest I can come to explaining this bizarre spiky histogram. This is straight out of my Canon 9950, which BTW is regularly IT8 color calibrated. This is Fuji Provia but Velvia 100 of the same scene and same exposure gives similar results.

No adjustments / corrections made yet. None. Zip. It looks the same if I scan 48b HDR as well. The image ends up looking completely psychedelic in PS CS2. The Canon s/w scan true to the original color, with a more natural looking histogram.

Any ideas ?

P.S. I would have love to post this to the Silverfast forums; but they have been down for several days when I try. Is it the same for other Silverfast users ?

[George Kara]

Nicol

Im sorry but don't see the histogram problems you are speaking of. There are some spikes in the lower portion of the value scale. I use this software myself and am familiar with it. A histogram is never right or wrong - it just is. The exposure seems fine. What do you want? btw I live in Scottsdale and often go to Sedona.

I dont know about your prints and associated problems but your negative seems OK.

George

[Daniel_Buck]

The spikes are just the change in tones, that low on the histogram it's probably the shadows on the rocks, changing with the different colored rocks?

[nicol_verheem]

Thanks George and Daniel for trying, but I' afraid you missed my point. Guess my corny title didn't help, huh ?

Statistically, looking at a naturally, evenly lit scene, with such a wide filed of view, that histogram is impossible. Basically that histogram says there are 100 pixels (or whatever unit) with brightness 0, a whole lot less, lets say 5 pixels, with brightness 1-9, 90 pixels with brightness 10, 6 with 11-19, 110 with brightness 20, 7 with 21-29, 120 with brightness 30, 8 with 31-39, 100 with 40, and so one. Basically, lots of suspiciously evenly spaced "brightness peaks".

That just doesn't make any sense to me, looking at the scene. It rather looks like a rounding / scaled integer / oscillation problem to me.

I attached the green channel histograms from the Canon 9950 s/w scan of the same slide, again no alterations, as well as the Silverfast scan, to show the difference.

Although, like George alluded to, a typical histogram "just is", in this case I had a really hard time to get the file to resemble the color on the light box, whether I scanned the Provia or Velvia slide, the colors of the rocks on the left in Photoshop were a ghastly pink. This is why I paid attention to the histogram in the first place, and compared it to the Canon s/w.

What am I missing ?

P.S. I upgraded to 6.5.5 this morning - just released last night, no change....

[Daniel_Buck]

Is this the first time you have seen this with your scanning setup? You are right it does look a bit odd, the regularity of the spikes. To me, that regularity might suggest an oddity in the scanning process. Do you know anyone near you who has a scanner you could ask them to scan it for you? Might be worth scanning on a different scanner if you think something is wrong.

[Dave Wooten]

it is not the vortex :)

[nicol_verheem]

Hi Daniel

This is the first time I noticed it. I do not need someone else's scanner - I just use someone else's software (Canon's) and the results look normal. So it's not h/w related.

Dave, I know it's not the vortex - there would have been hundreds of people in funny 'garb in my picture if it was ;-)