47greyfox Loc: on the edge of the Colorado front range

Personally, George, most recommend keeping FW up to date regardless of what the manufacturer gives as reasons, as they often make subtle changes that aren't included in the notes. None are known for "full disclosure." Just a thought.... :-)

goldstar46 Loc: Tampa, Fl

===
Yes, I know. Unfortunately for me, I just got lazy. Your statement is, a general rule oh, the practice that I use. I'll get around to it for sure.

Cheers
George Veazey

PHRubin Loc: Nashville TN USA

Guyserman Loc: Benton, AR

The Ruby throated hummingbird beats it's wings about 50 times a second. I think the total distanced traveled by the tip of the wing in one beat would be less than 8 inches but suppose it was. That would be a total of 400 inches in one second. At a shutter speed of 1/4000 second the wing tip would travel 1/10th inch plus very slightly more due to the rolling shutter. Not enough to be perceptible in a picture I would think. Have you ever seen a diagram of how a hummingbird wing beats when it's hovering? It's a figure-eight pattern with constant twisting of the wing. This looks perfectly normal to me.

Ltgk20 Loc: Salisbury, NC

At shutter speeds above 250 or so, the mechanical shutter becomes a traveling slit (which is to say the entire sensor isn't exposed to the light at the same time). As shutter speeds go up, the distance between the opening shutter curtain and the closing one gets closer together, but they still move at the same speed. Likewise, the electronic shutter doesn't read the sensor at one time. It is read row by row (or possibly groups of rows), regardless of the shutter speed. On the R5, the electronic shutter traverse speed is ~1/60 sec. This is fast enough for most action, but for really fast motion the object can move as the shutter progresses down the frame). I haven't found any reliable source for the R6 electronic shutter traverse speed, but I've seen things that suggest it's slower than the R5 (maybe ~1/30 sec). For information, the a9 and a9ii were the first full frame fast e-shutter cameras and traverse at ~1/160 sec and the current speed champion is the a1 at ~1/240 sec).

All this to say, I can't say with certainty if what you're seeing is shutter distortion or not, but it could be.

SuperflyTNT Loc: Manassas VA

Or he could just go to Canon’s website and check on updates.

LEWHITE7747 Loc: 33773

What about the Canon R3? Do the Stacked sensors eliminate this?

amfoto1 Loc: San Jose, Calif. USA

If you Google "rolling shutter effect" for images you'll find many examples of it. Typically it causes moving things to lean or curve.... unnatural looking distortion. It's unique to electronic shutters, due to the way they "read" the captured data line by line from top to bottom. Camera makers have worked to minimize the rolling shutter effect by increasing the speed with which the data is gathered from the shutter. However, it can still occur with very rapidly moving objects (such as the wing of a hummingbird).

Another rolling shutter effect is called "banding". You particularly see it occur under fluorescent lighting, which cycles on and off many times per second. The shutter being read out line by line captures the light cycling by recording alternating light and dark lines across the image. I suspect it's something similar to this you are seeing in this hummingbird photo, even though you weren't using that type of lighting. Some cameras can be fine tuned to reduce or eliminate banding, but have to be adjusted for each and every situation. Obviously you won't be able to do that with a hummingbird in flight, the way you could with indoor lighting causing the problem.

As sensor resolution has steadily increased over the years, the problem of rolling shutter effect has increased along with it. This is simply because more pixels means more rows of pixels on the sensor which take longer to complete reading from the top to the bottom of the sensor.

I don't know exactly how, but apparently stacked sensors also reduce rolling shutter effect to some extent. I think it might be because stacked sensors effectively reduce resolution on one sensor, by splitting it across two sensors, one on top of the other. But don't quote me because this is just a guess.

But the only way to completely eliminate the rolling effect is to use a "global shutter". The type of sensor in this case has its data read out at the same instant across the entire sensor. As you might imagine, this requires some pretty powerful processing... a very "wide" channel for all that data to pass through coming from the sensor. As a result, camera with global shutters are pretty expensive. They also can have problems with heat. But I'm sure it's what many camera manufacturers are working on.

Ltgk20 Loc: Salisbury, NC

Stacked sensors can typically read out faster so I expect the R3 to be better in this respect than the R5 and R6.

In comment to a different post: As far as I know, the readout speed is designed into the sensor and electronics and not something that can be changed via firmware.

PHRubin Loc: Nashville TN USA

As reported on a PBS Nature program, hummingbird wings do a corkscrew motion for some maneuvers.

goldstar46 Loc: Tampa, Fl

===

Yes, and they can actually rotate up to 80 Strokes per second. Wow.

George Veazey