Jump to content

Electronic Shutter mode


Recommended Posts

ES mode :  I was photographing Kingfishers this weekend and wanted the 20frames/sec ES mode gives as the birds dive really quickly so frame rate matters. Got some great images however the one attached is an example of a challenge. The birds wings are backwards, in the reflection they are forward's. So on seeing this a quick google highlights that Fuji do not recommend ES mode for fast moving objects so fair play its a known limitation. What I don't understand is why using the mechanical shutter would remove the scan delays. This image was taken at 1/2000 sec, images shot in portrait mode don't have the rolling shutter effect. Also images shot at 1/8000 are pretty good. Overall I am happy this is a fairly narrow use case and the images stand up well against the fellow shooters using Canon 1DX mark 2 which shoot at a lower FPS. Does anyone know if the scan times are different when using the mechanical shutter. I think the best option is to use Portrait mode, the dive in was fairly predictable where the bird emerged was slightly more variable and landscape orientation gives a little more flexibility. 20FPS by the way generally gives a shot with the bird emerging from the water, one just clear and sometimes a third before it disappears out of frame. the DSLR's at 12FPS only really got one, so the issue of the reflection isn't show stopper. It may be worth mentioning that I don't have photoshop and this is an export of a raw file default profile through Capture 1 so not really post processed. XT-3 50-140 F2.8 lens at about 120mm F5.6 and 1/2000 awesome ISO 10000

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

Link to post
Share on other sites

This was something that bothered me as well.

The sensor scan rate should be the same for both mechanical and electronic shutters - its a function of the sensor and not how its exposed.

The difference is that with ES the sensor is exposed to light for the entire scan whereas with MS the shutter blades control the length of exposure.

With MS the scan process can continue even though the sensor is no longer exposed to light

Both ES and MS must introduce distortion (due to the sensor being unevenly exposed) but ES distortion is greater than MS because the scan rate is slower than the focal plane shutter speed.

If anyone has a link to a knowledgeable article on this entire subject it would be appreciated.

 

Link to post
Share on other sites

What Greybeard said. The mechanical shutter moves very quickly -- just a few milliseconds elapse from top to bottom. You can still get some rolling shutter with mechanical shutter, but the subject has to be moving really quickly for that to be picked up.

The electronic shutter moves very slowly. On my aged X-T10, it takes maybe 70-75 milliseconds to scan the sensor. The more modern cameras will be somewhat faster, but not even close to the speed of a mechanical shutter.

With both the mechanical and electronic shutter, at high shutter speeds the entire sensor isn't fully exposed at one time (that's the source of the rolling shutter effect). The mechanical shutter essentially reveals a gap between the two shutter curtains that travels down across the sensor. Then the shutter is left closed, stopping the exposure, while the sensor is being read out.

The electronic shutter, however, has no way to stop the exposure without losing the data. Exposure of a row starts when its reset line is released, and continues until the reset line is reactivated (which zeroes out the exposure data). From the photographer's point of view, the exposure effectively ends when the row starts being read out, not when read-out is completed. But the camera can't snapshot the exposure data on the next row until this row is completely read out, so it can't even start the exposure on that row until this row is almost completely read out.

Various approaches have been proposed for allowing the data from more than a single row to be locked in at the end of the desired exposure. The Global Electronic Shutter is the ultimate, allowing the entire sensor's contents to be saved at one time.

Side note: the old CCD sensors generally had global electronic shutter. Rolling shutter is a phenomenon of the way current CMOS sensors are implemented.

Link to post
Share on other sites

I think I understand that if the camera is in mechanical shutter mode then the sensor is exposed as described by the slot between front and back curtains.  The sensor retains the image electronic charges until read.  That is scanned sequentially by the read process.

But at say 1/2000 of a second the sensor is exposed for 0.0005 S.  I don't understand the mechanism that does this in ES mode.  Whatever mechanism that does this it introduces the rolling shutter.  I understand what rolling shutter is, but given the fast pace I don't really get why something exposing the sensor in a sequential manner for 1/2000 of a second produces a different result to a mechanical shutter.  At these speeds a mechanical shutter is moving really fast so a very narrow slot of light is scanned across the sensor, very like a sequential read.

I had thought that the speed of the electronic shutter would be the read speed for the sequential scan of the sensor but that can't be.

The good news is that in portrait orientation the effect disappears and bird and reflection align.  So I have a viable workaround.

 

 

 

 

 

 

Link to post
Share on other sites

18 hours ago, Cai Le Dao said:

But at say 1/2000 of a second the sensor is exposed for 0.0005 S.  I don't understand the mechanism that does this in ES mode.

It is the electronic equivalent of the mechanical shutter.

In conventional CMOS sensors like Fuji uses, there is only "voltage storage" available for a single row at a time. Readout of a row cannot begin until the readout of the prior row is complete. For that reason, each row is processed one at a time.

Conceptually, the processor resets the row to clear the old exposure data and then removes the reset to begin the exposure (first curtain). After the desired exposure time, the processor performs the readout on the row. The first step in readout effectively stops the exposure for the row (second curtain) when it converts the quantum electron counts in the photodiodes into analog voltage. Then the readout process converts each pixel's voltage to digital and transfers it to the processor. Finally, the processor moves on to the next row.

It's probably difficult to tell in practice, but it's quite possible that the processor overlaps the exposure of row 2 with the tail end of the readout of row 1. This would involve two simultaneously-running processor threads: a "first curtain" that resets and then releases each row one at a time, and a "second curtain" that reads out each row one at a time. The second curtain thread would begin running after the first curtain thread, delayed by the desired exposure time (a half millisecond for 1/2000).

Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

  • Similar Content

  • Posts

    • Because the sensor assembly is moved electrmagnetically. When there is no power it is essentially free moving.
    • Ahoy ye hearties! Hoist ye yon Jolly Roger and Cascade away. NGC 1502 The Jolly Roger Cluster:

      Welcome, dear visitor! As registered member you'd see an image here…

      Simply register for free here – We are always happy to welcome new members!

      This is the equivalent of 43 minutes, 40 seconds of exposure. NGC 1502 is a neat little cluster located in the Camelopardalis Constellation. This region of space was thought to be fairly empty by early astronomers, but as you can see, there is a lot there. Kemble's Cascade (a.k.a. Kemble 1) is named for Father Lucian Kemble, a Canadian Franciscan friar who wrote about it to Walter Scott Houston, an author for the Sky And Telescope magazine. Houston named the asterism for Fr. Kemble and the name "stuck". NGC 1501 is the Oyster Nebula. A longer focal length telescope is needed to bring this one into good viewing range, but it is well worth the effort. NGC 1502: https://skyandtelescope.org/online-gallery/ngc-1502/ Camelopardalis Constellation: https://en.wikipedia.org/wiki/Camelopardalis Kemble's Cascade (and NGC 1501: The Oyster Nebula): https://www.constellation-guide.com/kembles-cascade/ Arrrrrr Matey.
    • Looking for input; there are some decent deals and might want to take advantage to expand my lenses for my 100s already own: 110/2 32-64 35-70 100-200 + TC   Shooting mostly family shots, bringing my kit to capture family outings indoors and out. Tracking the 63/43 effective FLs on the two, but has anybody used both? Would the 55 (covered by two zooms right now) be redundant? Would the 80 be too similar in character to my 110 for portraiture?
    • See what I mean? Two instantaneous ads. Worthless.   
×
×
  • Create New...