A Canon Global Shutter Sensor – Would the next 3 series camera be the place to introduce one?

#21
PinholeR5 said:
So, is it useful to have a mecanical "protective shutter", cheaper to implement since it doesn't have any real performance requirement? Or is the benefit not large enough to justify having one?
Click to expand...
One big advantage of a protective shutter (or a mechanical one) is that without it the sensor is always exposed optically unless you put a lens cap on the front of the lens when you aren't shooting. That isn't normally a problem but is if the camera, slung on your shoulder, happens to point to the sun or to a bright reflection of the sun. This used to cause shutter holes in the days of cloth shutter rangefinders and now causes sensor burn-out in mirrorless cameras without a mechanical nor protective shutter.
 
  • Like
Reactions: 1 users
Upvote 0
#23
neuroanatomist said:
I think the answer is that it doesn't make much difference in the field. But consider that you might use the camera for a few hours but for most people the camera probably spends more time powered off than in use. So I think there is a real benefit to a protective shutter, but it's not during lens changes it's when the camera is sitting idle giving any dust in the box plenty of time to settle out, where some of it will land on the sensor.

I use my R3 in silent mode a lot, and when I don't remember to go back to normal shooting before putting the camera away, I have to clean the sensor more frequently.
Click to expand...

I avoid the silent shooting for that reason. I instead set the volume down to 0 on everything. It does mean I get the audible shutter close sound whenever I turn the camera off, but for my uses that is ok.
 
  • Like
Reactions: 1 user
Upvote 0
#24
David - Sydney said:
GS has half the pixel as the storage for the readout if I understand correctly. If the storage can be moved behind the sensor ie part of the stack then it could be the ‘simple’ solution… but I am sure that they have thought of that already!

SPAD sensor?
Click to expand...

SPAD really doesn't have anything to do with it and has it's own series of problems.

but yes, moving the storage to a secondary substrate is what most of them are doing .. Sony, for instance, was getting rid of storage all together and putting an ADC under each pixel.

but the cost comes from having 24 million, etc interconnection pads between the substrates and all of them working.

but it's tech that does already exist.
 
  • Like
Reactions: 1 user
Upvote 0
#25
David - Sydney said:
We now have 2 integrated grip bodies so the option to differentiate by sensor makes sense (a la 1D prior to 1Dx).
Click to expand...
I don't think that would be a R3 / R1. though if they were going to do that, they would have already with the R5 series ala 5Ds / 5DsR

no need for a integrated grip camera for that anymore - the need for a much larger camera to house the processors, etc disappeared with the 1Ds Mark III when the 5D could do it all.

The R1 Mark II will probably get a resolution bump, when Canon can actually deal with it.
 
Upvote 0
#26
With the R1 readout speed better than a mechanical shutter a global shutter seems a bit pointless for stills although it has an edge in video and flash sync speeds.
An intermediate resolution of about 33 MP might be a better compromise for an R3 ii body
 
  • Like
Reactions: 2 users
Upvote 0
#27
Richard CR said:
SPAD really doesn't have anything to do with it and has it's own series of problems.

but yes, moving the storage to a secondary substrate is what most of them are doing .. Sony, for instance, was getting rid of storage all together and putting an ADC under each pixel.

but the cost comes from having 24 million, etc interconnection pads between the substrates and all of them working.

but it's tech that does already exist.
Click to expand...
The cost effective solution would be to egrow the sensels in either an epitaxial layer or organic layer (as Panasonic and Fuji were developing) on top of the logic layer. That would make a fast FSI sensor and for global shutter, the secondary storage could be in the same location as current FSI sensors with the light gathering pixel on top. A lot has been done with layering technology with flash memory leading the way. Sensors are very different from flash memory, but layering is still a possibility. E.g. the Foveon sensor is a layered design and as I recall, that sensor was/is grown directly on top of a standard CMOS process, so actually an FSI design. The RF lens license to Sigma might have more layers (pun intended) than meet the eye.
 
Upvote 0
#28
Canon have created this camera linage confusion and only have themselves to blame and the only obvious thing they can do with the R3 linage is to totally re-invent it. Has Canon done this before? Yes...the 5D series was a camera linage that moved from consumer / semi pro to serious pro in a few generations.
I can't see Canon changing their 24mp & 45mp niches for a long time to come. Because Canon had the R5 as it's flagship camera for so long, many are pre-conditioned to think that Sports shooters want 45mp, they do not. The R1 and R3 are targetted primarily at sports shooters and that's why their sensors are the same resolution and why their frame rate is similar. 24mp is a perfect resolution to do minimal croppng, speed, quick workflow and file transfer size.

The problem we have as consumers of Canon's product line is tha we are confused by the need for 2 professional sports cameras in the linage. As far as Canon are concearned they already have created a 45mp professional camera...it's called the R5II.

The obvious way Canon can resolve this consumer dilema, is to move the R3's target audience into a new area and effectively make a R5 in a 1 series body. Or they continue with the current R3's target audience, splitting the pro sports shooters options between 2 similar cameras. Introduce new tech in each iteration and keep the two cameras competeing with each other on a technological level...eg...better AF, EVF and Shutter readout speed etc, flippng between the R1 and R3....Or they re-invent the R3 entirely as something new. In my mind this would be R1 for fast action / sports / keep it at 24mp and the R3 for landscape / wildlife and birding professionals who want the 45mp sensor in a gripped body that can handle a war zone type of build level.

In terms of global shutters...do we really need one when we have a 3ms read out speed already?

However, I'm sure when Canon finally get around to making a R3II....we will all say...ah....er....ohhh....hmmmm.....
 
  • Like
Reactions: 1 user
Upvote 0
#30
The digital photography experts in the forums here at caonrumors have long decried the reviews of DXO, and now you expect people to get on side because it shows Canon being better than Sony? Hmpf, you could almost be forgiven for thinking that people were biased...
 
  • Like
Reactions: 1 user
Upvote 0
#31
GMCPhotographics said:
Canon have created this camera linage confusion and only have themselves to blame and the only obvious thing they can do with the R3 linage is to totally re-invent it. Has Canon done this before? Yes...the 5D series was a camera linage that moved from consumer / semi pro to serious pro in a few generations.
I can't see Canon changing their 24mp & 45mp niches for a long time to come. Because Canon had the R5 as it's flagship camera for so long, many are pre-conditioned to think that Sports shooters want 45mp, they do not. The R1 and R3 are targetted primarily at sports shooters and that's why their sensors are the same resolution and why their frame rate is similar. 24mp is a perfect resolution to do minimal croppng, speed, quick workflow and file transfer size.

The problem we have as consumers of Canon's product line is tha we are confused by the need for 2 professional sports cameras in the linage. As far as Canon are concearned they already have created a 45mp professional camera...it's called the R5II.

The obvious way Canon can resolve this consumer dilema, is to move the R3's target audience into a new area and effectively make a R5 in a 1 series body. Or they continue with the current R3's target audience, splitting the pro sports shooters options between 2 similar cameras. Introduce new tech in each iteration and keep the two cameras competeing with each other on a technological level...eg...better AF, EVF and Shutter readout speed etc, flippng between the R1 and R3....Or they re-invent the R3 entirely as something new. In my mind this would be R1 for fast action / sports / keep it at 24mp and the R3 for landscape / wildlife and birding professionals who want the 45mp sensor in a gripped body that can handle a war zone type of build level.

In terms of global shutters...do we really need one when we have a 3ms read out speed already?

However, I'm sure when Canon finally get around to making a R3II....we will all say...ah....er....ohhh....hmmmm.....
Click to expand...
I'm sure most consumers are not confused, nor do they think there is any dilemma. They need a camera, they buy the camera that best suits there needs. If they wonder why 2 sports oriented cameras? The R3 is 3 years old and the R1 is new and essentially replaces the R3 for sports shooters. Confusion gone. If you want the cheaper one, buy the R3. If you want the newer, more advanced one, buy the R1.

In terms of global shutter, I agree that with stacked sensors now down to 3ms readout, they are not necessary, especially if they continue to have any drawbacks. That's from a technical or actual shooting perspective. From a marketing perspective, well, that unfortunately is another matter, especially if internet influencers continue to have such an influence on the camera buying public, which really is a shame.
 
  • Like
Reactions: 5 users
Upvote 0
#32
David_B said:
The digital photography experts in the forums here at caonrumors have long decried the reviews of DXO, and now you expect people to get on side because it shows Canon being better than Sony? Hmpf, you could almost be forgiven for thinking that people were biased...
Click to expand...
Speaking for myself, I find that DxOMark's measurements are generally good (though they do screw up occasionally, e.g. with the EF 70-200/2.8L IS II). But the Scores that they derive from those measurements are useless.

The easiest way to see that is that a Lens Score is affected by the dynamic range of the sensor behind it. Because DR is a critical property for the evaluation of lenses. :rolleyes:
 
  • Like
Reactions: 2 users
Upvote 0
#33
David_B said:
The digital photography experts in the forums here at caonrumors have long decried the reviews of DXO, and now you expect people to get on side because it shows Canon being better than Sony? Hmpf, you could almost be forgiven for thinking that people were biased...
Click to expand...
Well, if even an anti-Canon biased site declares a Canon to be better than a Sony, there must be some truth in it!
 
  • Like
  • Haha
Reactions: 2 users
Upvote 0
#34
Quack said:
In terms of global shutter, I agree that with stacked sensors now down to 3ms readout, they are not necessary, especially if they continue to have any drawbacks. That's from a technical or actual shooting perspective. From a marketing perspective, well, that unfortunately is another matter, especially if internet influencers continue to have such an influence on the camera buying public, which really is a shame.
Click to expand...
I recognize that this is a edge case, but global shutters still have an edge where there is a lot of PWM-dimmed LED lighting. I shoot a lot of theater and dance, and you can still see artifacts except leaf shutter and global shutter cameras.

1732207553352.png

1732207565900.png
 
  • Like
Reactions: 2 users
Upvote 0
#35
Quack said:
In terms of global shutter, I agree that with stacked sensors now down to 3ms readout, they are not necessary, especially if they continue to have any drawbacks. That's from a technical or actual shooting perspective. From a marketing perspective, well, that unfortunately is another matter, especially if internet influencers continue to have such an influence on the camera buying public, which really is a shame.
Click to expand...
There is also a difference in flash sync speed with eshutter with GS.
A9iii is 1/80000s. Leaf shutter also effectively don't have a constrained shutter speed.
The R1 is faster than any other canon body at 1/320 (some say 1/400 but maybe that is for crop mode).
Definitely an edge use case though eg RF50/1.2 wide open with fill flash for backlit bright sunlight or fast moving subject at night.
 
Upvote 0
#37
About 5-6 years ago Canon patented a sensor with two charge buckets per sensor pixel that could switch between them globally. This let you do the following:

1) global shutter, with one stop less DR. Start off exposing on bucket 1 (B1). Once the shutter is open, switch to B2. Collect charge there. When exposure is done, switch back to B1 and read out B2.

2) one-shot HDR with nearly double the DR. Every millisecond, spend 1 microsecond on B1 and 999 microseconds on B2. Keep alternating. Exposure would then be 10 stops darker than usual for the image captured in B1. Afterwards camera could download as two separate images or combine into 1 with 23 stops DR. The patent didn't describe the switching speed or ratio, this is just an example

3) electronic ND filter with one stop less DR. Every millisecond, spend 1 microsecond on B1 and 999 microseconds on B2. Keep alternating. When you're done just throw away B2 and your final image is in B1. You could use a tripod to take a multi-second exposure in daylight with a big aperture, for instance. The patent didn't guarantee the switching would be fast enough, but I've done some calculations suggesting you could get, say, solid lines from moving highlights in many or most situations even if not all.

A sensor like this would add a lot more freedom to shooting, I think, much as Canon's AF-tilt lenses will.
 
  • Like
Reactions: 1 user
Upvote 0
#40
PinholeR5 said:
Even with the R7 I have been 100% on ES.
Click to expand...
Same here, despite the R7's sometimes very distorted outputs caused by i'ts slow read-out speed. I mostly shoot 30 fps (wildlife), and there are nearly always keepers in these massive bursts. I only switch to MS when I know I will shoot e.g. smaller birds in flight with fast wing flaps, and the quite loud machine gun like noise of the R7 will not so much disturb. So, if one can use even an R7 frequently with ES for action in the field, this shows that a camera with faster sensor read-out and no global shutter could be in real life a better choice than one with GS and visible losses in IQ. I personally would always prefer better IQ.
 
  • Like
Reactions: 1 users
Upvote 0
You must log in or register to reply here.

Similar threads

Canon Rumors
Canon has a global shutter camera in the pipeline
Replies
76
Views
4K
DBounce
DBounce
Canon Rumors
Canon releases a full frame 19MP Global shutter sensor to Market
Replies
20
Views
3K
Objective!
O
Canon Rumors
Sony Electronics Releases the Alpha 9 III; the World’s First Full-Frame Camera with a Global Shutter Systemi
Replies
75
Views
11K
David - Sydney
D
Canon Rumors
Canon not going global shutter with next round of EOS R camera bodies
Replies
205
Views
31K
EOS 4 Life
E
Canon Rumors
Canon Celebrates 21st Consecutive Year of No. 1 Share of Global Interchangeable-Lens Digital Camera Market
Replies
4
Views
1K
John Wilde
J
Top Bottom
Back
Home
Forums
Media
Menu