29
\$\begingroup\$

I just found a picture again I took years ago, and I'm wondering what's happened ever since then - I hope this is the right place to ask!

I sat on my swivel chair and spun a few times, while taking this picture of my CRT display at a random moment.

Distorted image of my CRT display

The digital camera I used was (according to EXIF, I don't remember anymore...) a Nikon E4600.

While I've heard of distortions due to a rolling shutter, I would expect it to affect the whole picture, not only the desktop screen.

What's going on there?

\$\endgroup\$
4
  • 8
    \$\begingroup\$ stuXnet getting some unexpected results when spinning around - oh the irony =) Great first question, welcome to photo.SE! \$\endgroup\$
    – null
    Feb 9, 2016 at 21:28
  • 6
    \$\begingroup\$ Incidentally, it managed to capture exactly one full screenful of scans, no more no less. So your shutter speed was probably 1/60 and monitor refresh rate 60Hz, though not guaranteed those are both fairly common values. \$\endgroup\$
    – Octopus
    Feb 10, 2016 at 0:07
  • 1
    \$\begingroup\$ "Johnson, what are you doing!?" "Umm, graphics research, sir." \$\endgroup\$
    – corsiKa
    Feb 11, 2016 at 0:10
  • \$\begingroup\$ It looks like the wizard on the screen is causing the image to fly off of it! Cool effect - whether intentional or not. Even better if you blurred the sharp edges from the "electron beam". \$\endgroup\$
    – Joe
    Feb 16, 2016 at 23:03

2 Answers 2

46
\$\begingroup\$

The reflection on the screen tells me you used a flash. The flash only lasts a very short time (at most 1/200 of a second), while the shutter was probably open for a much longer time (maybe 1/30 of a second). Since the frame of your CRT as well as the wall behind it do not emit light on their own, their appearance on the photo is mainly due to them reflecting light from the flash. The CRT however emits light on its own, thereby contributing to the image over the whole time of the exposure.

EDIT: Looking more closely, you can see that the electron beam was in the bottom quarter of the screen at the start of the exposure (and therefore at the time of the flash). It then traveled to the bottom and wrapped around to the top, getting more and more displaced to the left.

\$\endgroup\$
4
  • 4
    \$\begingroup\$ Thanks, that helped! For the sake of completeness: the shutter was open for 1/60 of a second, says EXIF :) \$\endgroup\$
    – stuXnet
    Feb 9, 2016 at 21:22
  • 5
    \$\begingroup\$ Ah yes, that makes sense. Thinking about it just now, if the shutter was slower, you would have gotten two images of the screen content on top of each other and if it was faster, you would have gotten an incomplete image of your desktop. \$\endgroup\$
    – Jules
    Feb 9, 2016 at 21:26
  • \$\begingroup\$ I believe that the flash is irrelevant. \$\endgroup\$ Feb 11, 2016 at 5:44
  • 3
    \$\begingroup\$ @200_success The flash is relevant because it explains why the rest of the image is not distorted. (Ie. the light in the rest of the image comes mainly from the duration of the flash, not the whole period the shutter was open) \$\endgroup\$
    – Taemyr
    Feb 11, 2016 at 9:48
5
\$\begingroup\$

The effect is created due to you rotating and the lines being drawn from top to bottom and your shutter was open long enough to capture all lines being emitted.

How did the image get skewed that way? You're rotating clockwise and when the flash went, a line at the top of the lower quarter of the screen was being emitted. Then you captured the next line when you were rotated just a fraction to the right. So that line appeared a bit more to the left and so one. Your shutter was open long enough to capture the emitting of all lines.

You can see that the lower line and the upper line are aligned horizontally. That makes sense, because the top line is being emitted right after the bottom line was emitted.

Update: finally done polishing my answer. Now when I read back the accepted answer, I guess mine tells the same story...

\$\endgroup\$
1
  • 1
    \$\begingroup\$ The same effect is noticeable when looking at many scrolling LED message boards, assuming one's eyes track the text, though many such message boards scan bottom-to-top rather than top-to-bottom so that someone whose eyes are moving right to left will observe the text as being slanted slightly to the right rather than slightly to the left. \$\endgroup\$
    – supercat
    Feb 10, 2016 at 16:24

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.