Why not rotate the sensor 90 degrees instead of grip?

[Tailslol]

Would it be technically too hard to have a rotating sensor in a small body instead of a gripped body?

 

[HJVN]

Yes!

It is also much faster to just rotate the whole camera than it is to find a button, press it, wait for the sensor to rotate into vertical and lock into place, take a picture, find a button, press it, wait for the sensor to rotate back to horisontal and lock into place, take a picture, find...

 

[Stavrosf]

Would it be technically too hard to have a rotating sensor in a small body instead of a gripped body?

This would create another point of (mechanical) failure. And, I would prefer the R&D of the companies to go to more excited technologies than that. For me it is not too hard to change the orientation of my camera.

 

[P_e_t_e]

In case when the camera is mounted on a tripod and focused on the subject, sometimes it might be interesting to flip between H/V orientations to see which one works better. But the mechanical sensor rotation is highly unlikely due to reliability issues.

I think you're more likely to see a square sensor with H/V crop selectable by a button. Some medium format cameras may even already support this.

While this solution has the advantage of reducing the mechanical overhead to just a button, it wastes a lot of pixels on the sensor. Therefore I don't see this idea catching on.

 

[jjuncal]

Instead of rotating why not square then?

 

[dreyko]

Instead of rotating why not square then?

Mostly due to cost for little benefit, people are perfectly happy to turn the camera when a vertical shot is needed, plus all the petal lens hoods would no longer work.

I think its mostly a cost thing though, its a lot of wasted pixels most of the time, and it wasn't possible with dslr cameras due to depth requirements of the mirror.

 

[Kapiro]

Would it be technically too hard to have a rotating sensor in a small body instead of a gripped body?

As someone else mentioned, that could potentially introduce more points of failure. More than likely it would be make the camera more expensive.

You might want to look at this teardown of a medium format Fuji GFX 100, especially the disassembling of the sensor's IBIS unit. It is a pretty tight fit in a pretty large camera.

https://www.lensrentals.com/blog/2020/08/the-fujifilm-gfx-100-vs-salt-water-teardown/

 

[shuncheung]

Would it be technically too hard to have a rotating sensor in a small body instead of a gripped body?

As someone else mentioned, that could potentially introduce more points of failure. More than likely it would be make the camera more expensive.

You might want to look at this teardown of a medium format Fuji GFX 100, especially the disassembling of the sensor's IBIS unit. It is a pretty tight fit in a pretty large camera.

https://www.lensrentals.com/blog/2020/08/the-fujifilm-gfx-100-vs-salt-water-teardown/

No kidding. You need to leave a lot of space around the sensor so that it can rotate by 90 degrees, and a sensor has a lot of electronic connections; all of those connections must work the same way in either orientation.

And then the EVF will have to rotate also. Think about that space, electronic connections and then a separate motor to turn the EVF ....

Seems to be a really complex, error prone solution when a very simple one already exists.

 

[jthomas39]

Instead of rotating why not square then?

It's an obvious improvement. All the control buttons don't move. The hand positions and tripod config doesn't change either. The screen and EVF would need to be square and letterboxed which would affect the camera design.

Would the raw be a square, with a defaulted outline location embedded? A big file though. (Raws now are slightly larger than the jpg. In the edit, corrections need to be turned off to show the full size.)

 

[atdsatds]

The thing that I think would cause the most issue would be the precision that the sensor would need to be set in after rotation each time. That would probably be the expensive part. In machines to get precise placement after each move can be exponentially more expense as you need to get more precise.

It may be easier and less expensive to create a square sensor and just turn on and off pixels depending on the orientation wanted. And as stated, the EVF would need to do this also. But then people would ask why not just have all the pixels on and let the user decide how to crop or not crop.

Technically you could do that now. Keep your camera horizontal and crop it to vertical in post. How many want to waste those pixels?

 

[Joseph S Wisniewski]

Instead of rotating why not square then?

Remember, the lenses only have an image circle around 43mm, so you have two choices:

1. Make the sensor 36mm square, and force the orientation to be a shooting-time choice.
2. Drop the sensor down to 31mm square.

Either of these choices mean you'll have an image quality reduction, because you have to open up the lens hoods (and internal rectangular baffles on some lenses) to the larger format, letting in 50% more stray light to cause flare and ghosting.

It also means flashes will either have to autorotate or support a square format, resulting in a small, but annoying, loss of flash range.

Something else to keep in mind is that every other company to focus on square images is out of business. It's a cursed format.

--
Christine Fleischer 1947-2014.
My soulmate. There are no other words.
-----
Rahon Klavanian 1912-2008.
Armenian genocide survivor, amazing cook, scrabble master, and loving grandmother. You will be missed.
----
Ciao! Joseph
www.swissarmyfork.com

 

[Joseph S Wisniewski]

Would it be technically too hard to have a rotating sensor in a small body instead of a gripped body?

So many reasons:

1. Image quality. High end lenses are "cropped rectangular" by their petal lens hoods (which are built-in on some lenses like ultrawides or fisheyes) or internal baffles (like my 200mm f4 micro-nikkor). Opening up the hoods and baffles to accommodate two orientations would let in some 50% more stray light to cause flare, ghosts, etc.
2. Mechanical complexity.
3. Size. Not only the added space for the rotating mechanism, but the added height for a square viewfinder and rear screen that would accommodate an image 50% higher (or accepting that vertical images just would have to have smaller finders).
4. Redesign of the sensor for multidirectional scanning and cropping. Right now they scan top to bottom, which is great for landscape but horrible for the vertical orientation (this is also a problem when you rotate your camera).
5. Square flashes, which is basically a 33% power reduction since you have to cover a larger area.

That aside, the ergonomics are obviously superior. Rotating a camera is just a bad idea in general, and a lot of existing solutions (Hel-L brackets, flip flash brackets, etc. are an ergonomic disaster. #BanLBracketsNow)

----

Christine Fleischer 1947-2014.
My soulmate. There are no other words.
-----
Rahon Klavanian 1912-2008.
Armenian genocide survivor, amazing cook, scrabble master, and loving grandmother. You will be missed.
----
Ciao! Joseph
www.swissarmyfork.com

 

[jjuncal]

Would it be technically too hard to have a rotating sensor in a small body instead of a gripped body?

So many reasons:

1. Image quality. High end lenses are "cropped rectangular" by their petal lens hoods (which are built-in on some lenses like ultrawides or fisheyes) or internal baffles (like my 200mm f4 micro-nikkor). Opening up the hoods and baffles to accommodate two orientations would let in some 50% more stray light to cause flare, ghosts, etc.
2. Mechanical complexity.
3. Size. Not only the added space for the rotating mechanism, but the added height for a square viewfinder and rear screen that would accommodate an image 50% higher (or accepting that vertical images just would have to have smaller finders).
4. Redesign of the sensor for multidirectional scanning and cropping. Right now they scan top to bottom, which is great for landscape but horrible for the vertical orientation (this is also a problem when you rotate your camera).
5. Square flashes, which is basically a 33% power reduction since you have to cover a larger area.

That aside, the ergonomics are obviously superior. Rotating a camera is just a bad idea in general, and a lot of existing solutions (Hel-L brackets, flip flash brackets, etc. are an ergonomic disaster. #BanLBracketsNow)

----

Christine Fleischer 1947-2014.
My soulmate. There are no other words.
-----
Rahon Klavanian 1912-2008.
Armenian genocide survivor, amazing cook, scrabble master, and loving grandmother. You will be missed.
----
Ciao! Joseph
www.swissarmyfork.com

Indeed! You’re right didn’t thought even befor suggesting!

in this case a cross like sensor would work… turning off the part of the sensor that isn’t needed depending the orientation

 

[sybersitizen]

Would it be technically too hard to have a rotating sensor ...?

Yes, and also a correspondingly rotating viewfinder and rear screen. Fortunately, all three already rotate together with cameras as they are now.

 

[Mark B.]

Would it be technically too hard to have a rotating sensor in a small body instead of a gripped body?

I wouldn't want to worry about the sensor moving around inside the body and getting jammed up or the mechanism failing in some other way, or things going out of alignment. No thanks.

 

[Mark B.]

Instead of rotating why not square then?

This is brought up every once in a while in a dedicated thread, typically results in very long and oft-repeated arguments for & against this idea. I expect that if any of the ILC companies thought there was enough profit in it for digital, it would have been done by now.

 

[john isaacs]

Would it be technically too hard to have a rotating sensor in a small body instead of a gripped body?

It would be technically hard and costly. Any precision mechanism is expensive to produce and is a reliability risk.

But, there are multi-aspect sensors that allow selection of aspect ratios and could allow for selection of portrait versus landscape.

But, many lenses are made to support landscape only imaging (along with their lens hoods). They do not support multi-aspect sensors that switch from landscape to portrait.

 

[Graham Meale]

And another reason ... look at the back of the lens, which is often a rectangle to match the shape of the sensor.

 

[MediaArchivist]

1. Image quality. High end lenses are "cropped rectangular" by their petal lens hoods (which are built-in on some lenses like ultrawides or fisheyes) or internal baffles

This seems like a deal breaker to me. Almost all my lenses have a rectangular rear baffle/plate/opening, and many (but not all) have petal hoods. This also leans against the idea of a square sensor.

1. Redesign of the sensor for multidirectional scanning and cropping. Right now they scan top to bottom, which is great for landscape but horrible for the vertical orientation (this is also a problem when you rotate your camera).

I've not heard of the scanning orientation mattering that much. I'm curious— why?

 

[YWG]

Would it be technically too hard to have a rotating sensor in a small body instead of a gripped body?

I'm sure it could be done but it adds complexity and points of failure, cost.