4
\$\begingroup\$

I was doing a timelapse recording of sunrise with my Canon 1100D. I set manual mode of exposure (M on the disk), exposure time to 1/4000 s (smallest), aperture to f/3.5 (maximally open), ISO to 100 (minimum). These aperture&shutter settings are meant to maximally avoid starburst effect while also avoiding overexposure. Then I just took photos with maximum frequency the camera was able to when connected via USB. Namely, I just simulated very frequent clicks on "Capture an image" in Entangle using xdotool, which effectively resulted in about 5 seconds between frames.

After having done the photo set, I opened the photos in Geeqie and started going next,next and so on. Looking at this "slide show", I noticed that brightness of the images randomly varies between shots, despite the weather was completely cloudless and couldn't be the reason.

Following advice in comment to a related question, I checked bracketing mode — it appeared off. I then tried setting smallest aperture (f/22) and much longer exposure time (1/80 s) — and the result didn't have any variation of brightness. So I conclude that my camera isn't able to precisely move its shutter to achieve prescribed fast timing.

Thus my question: is it normal for a photo camera like Canon 1100D to have visible variation in exposure times when they are at/near their lowest settings? Or is it a sign of wearing?

Improve this question
\$\endgroup\$
13
  • 1
    \$\begingroup\$ Were you using Manual exposure mode? Is Safety shift enabled in your camera's menu? What metering mode is selected? Are you sure you were using Manual exposure mode and not manual focus or manual AF point selection or manual Av or Tv selection? \$\endgroup\$
    – Michael C
    May 17, 2017 at 19:46
  • 2
    \$\begingroup\$ Were there neon lights as part of the scene? Those cause variations at high shutter-speed because of flicker. \$\endgroup\$
    – Itai
    May 18, 2017 at 1:30
  • \$\begingroup\$ @MichaelClark I set "M" on the mode selection disk; otherwise I wouldn't be able to set both aperture and shutter speed. \$\endgroup\$
    – Ruslan
    May 18, 2017 at 3:50
  • \$\begingroup\$ @Itai no, there weren't. And anyway, any such lights would be too dim due to high shutter speed. \$\endgroup\$
    – Ruslan
    May 18, 2017 at 3:51
  • 1
    \$\begingroup\$ Were there any light sources in/illuminating the scene other than the sun? If so, what? \$\endgroup\$
    – Michael C
    May 18, 2017 at 5:35

3 Answers 3

Reset to default
2
\$\begingroup\$

I assume you are using : Manual exposure mode? If that's the case, your conclusion is right: it seems like a physical problem with the camera.

In general 1/4000 is a bit extreme of an entry level DSLR.

Improve this answer
\$\endgroup\$
3
  • 2
    \$\begingroup\$ 1/4000s is a very short shutter speed for any DSLR. Noticeable variations in actual shutter speed is to be expected. \$\endgroup\$
    – jarnbjo
    May 18, 2017 at 13:14
  • \$\begingroup\$ @jarnbjo why is it present as an option then? Is it supposed to be compensated by post-processing? \$\endgroup\$
    – Ruslan
    May 18, 2017 at 15:08
  • 3
    \$\begingroup\$ @Ruslan Because when setting the shutter speed to 1/4000s, the actual shutter speed is at least closer to 1/4000s than it would have been with a setting of 1/2000s. The practical use of such short shutter speeds is usually to freeze moving objects. Remember that a still image camera is not primarly made to create time-lapse videos and unless you directly compare two separate images taken with the same settings, you are unlikely to notice the variations in shutter speed at all. \$\endgroup\$
    – jarnbjo
    May 18, 2017 at 15:57
2
\$\begingroup\$

The time lapse flicker you have experienced is commonly an effect of aperture variation rather than shutter speed.

For each individual shot the camera moves the aperture blades into the right position, and then returns them to their default position after the shot is taken. This process is pretty accurate, but even tiny differences can appear when viewing the frames rapidly.

This doesn't happen at the aperture extremes in your case as it's much easier for the camera to position the blades accurately against their stop points.

You can avoid this effect by choosing a lens with a manual aperture ring that can leave the blades in place between shots (or provide a mechanical 'stop point' for the camera to set the blades against.

Improve this answer
\$\endgroup\$
4
  • 1
    \$\begingroup\$ Aperture variation is usually a LOT less variable with cameras that use electronic aperture control than with cameras that use a mechanical aperture linkage between the camera and lens. \$\endgroup\$
    – Michael C
    May 18, 2017 at 22:40
  • 1
    \$\begingroup\$ Since the OP is using manual focus and M exposure mode, it would be very easy to test if that were the cause. Use the 'Set' button configured as the DoF button to lock the lens at the set aperture by pressing the lens release button and rotating the lens slightly while the aperture is stopped down to the desired value. Of course since the camera will no longer detect the lens, M exposure mode must be used to take the picture. \$\endgroup\$
    – Michael C
    May 18, 2017 at 22:44
  • 1
    \$\begingroup\$ As a Nikon shooter with only mechanically-linked aperture lenses, I can't emphasize enough how right @MichaelClark 's comment is. Nikon's aperture linked to the mirror motion drives me absolutely nuts. It's not a problem wide open, but mid-aperture range and smaller... it makes producing timelapses more tedious than they already are. \$\endgroup\$
    – scottbb ♦
    May 18, 2017 at 22:45
  • \$\begingroup\$ At f/3.5 as indicated in the OP, the aperture of the lens in question is wide open, so how can aperture variability be causing anything? With an EF lens, when the aperture is set at the maximum value it does not move at all for each frame. \$\endgroup\$
    – Michael C
    May 18, 2017 at 23:02
-2
\$\begingroup\$

If you are using any metering mode other than Evaluative, or if it is set to auto and the camera selects any of the other modes, then, the following will happen in case of respective metering modes.

Spot metering: When the sun is not at the AF point, the scene framed, will be bright. The brightness will depend upon the distance of the sun from your AF point.

When the sun is at the AF point, then the frame will be dark. The darkness depends upon its distance from the AF point.

Center Weighted Average: This gives more weightage to the area around your AF point, and the behaviour will be similar to the above description, with more importance being given to the exposure value near the AF point. Thus, for sun at AF point, the frame will be dark, but not as much as it would be in Spot mode.

Thus, for photographing landsacpe, sunset or sunrise, or moon, use evaluative metering, or matrix metering on Nikons.

Best yet, shoot RAW, and do all this stuff manually in PS, LR, or ACR.

Improve this answer
\$\endgroup\$
11
  • \$\begingroup\$ I didn't use autofocus, it was switched to Manual by the mechanical switch on the lens. \$\endgroup\$
    – Ruslan
    May 18, 2017 at 12:52
  • \$\begingroup\$ I meant the focus point, if you have used a single point, or focus area, if you have used multi point focus. In any case, it is the point at which the camera lens achieved proper focus in your frame. \$\endgroup\$
    – ATG
    May 18, 2017 at 13:43
  • \$\begingroup\$ Why does metering mode matter if I set shutter speed, aperture and sensitivity manually? \$\endgroup\$
    – Ruslan
    May 18, 2017 at 14:30
  • 2
    \$\begingroup\$ @ATG Metering has nothing to do with in-camera processing if the ISO, Tv, and Av are all set manually. \$\endgroup\$
    – Michael C
    May 18, 2017 at 22:34
  • 1
    \$\begingroup\$ Since the light meter in the 1100D is a monochromatic one, it also doesn't affect WB. The scene contents may affect WB when the image is analyzed in processing IF 'Auto WB' is selected. But if 'Daylight' has been manually preselected as the WB, then nothing the metering does will have any effect on JPEG processing, either in-camera or later. \$\endgroup\$
    – Michael C
    May 18, 2017 at 22:37

Your Answer

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

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