Canon EOS R1 to have a version of a DGO sensor?

[Uneternal]

If Canon has figured out a way to make this work for a stills camera

There's nothing to be figured out, all dual pixel cameras already capture 2 different exposures. And with the right software you can even extract both images and combine them for higher dynamic range. I'm just not sure why Canon hasn't done that yet in camera. They've been sitting on the dual pixel technology since 2013, it's about time they make it dual ISO.

 

[Richard CN]

There's nothing to be figured out, all dual pixel cameras already capture 2 different exposures. And with the right software you can even extract both images and combine them for higher dynamic range. I'm just not sure why Canon hasn't done that yet in camera. They've been sitting on the dual pixel technology since 2013, it's about time they make it dual ISO.

because of parallax it's no where as simple as you think.

 

[Navism]

I guess R1 sensor would be some kind of C400 sensor. The question is Canon doesn't explain how Triple-Base ISO work.
I will choose R5 mkii but I love R1 power with big battery.
Long time ago, I like 5D3 and 1Dx mkii didn't release the same time

 

[frankchn]

I guess R1 sensor would be some kind of C400 sensor. The question is Canon doesn't explain how Triple-Base ISO work.

C400 doesn't have DGO, so if the R1 has a DGO sensor, then I don't think it would be that similar.

 

[Richard CN]

There has been a ton of talk about the EOS R1 in our inbox. Most media outlets are not NDA'd about the EOS R1, as its not in their hands. The EOS R5 Mark II has been shown in-person to retailers and a bunch of media outlets have the camera in hand. However, the EOS

See full article...

I am really unsure about how this would work for stills. I haven't seen any real deep dive (just dumbed down explanations) into how this works without subsequent images. Global shutter I guess would get around that, but without global shutter - I don't know. it may even be DGO in video, but normal in stills, or normal and expanded DR in stills (with the expanded DR having cavaets).

 

[DBounce]

I am really unsure about how this would work for stills. I haven't seen any real deep dive (just dumbed down explanations) into how this works without subsequent images. Global shutter I guess would get around that, but without global shutter - I don't know. it may even be DGO in video, but normal in stills, or normal and expanded DR in stills (with the expanded DR having cavaets).

So many possibilities. But in the end, I’ll bet our wild imaginations will conjure up a way more exciting camera than is actually released. I’m going to expect to be disappointed… and maybe it will turn out that I’m pleasantly surprised.

 

[PhotoBob]

Not likely in what\'s primarily a sports camera. This feature would be virtually of no importance in the real world of its intended use.

 

[Arturo+Lauren]

DGO in cinema cameras is enabled and disabled on some modes, eg C70 and C300III. DGO just up to 4k60 but not 120.
DGO sensors doesn't mean DGO is always enabled

 

[Arturo+Lauren]

The sensor on the C400 is likely based on R3 tech, the R1 could potentially be more advanced than that

 

[Sporgon]

For some people, DR doesn't matter. They will claim, that they can't see any difference and give some bs excuse why they can't.

You sound like someone who believes that even if a scene has an EV range within the limits of a ‘low DR’ sensor, another one that has a higher DR will still produce a better quality image of the same scene.

 

[Maximilian]

Sounds thrilling! DGO in Canon's stills cameras would really be very attractive, it would amend the possibilities of photography substantially.

Fully agree.

For some people, DR doesn't matter. They will claim, that they can't see any difference ...

Esp. when they use Sony sensors and talked 180° the opposite when Canon sensors were (a little bit) behind

On the one hand, more of any 'feature' is usually better...more MP, faster frame rate, more DR, etc. Typically, such things can be dialed back if the consequences are undesirable ...

That's it! Exactly!

 

[DSB]

I think we would've heard something about this magical sensor by now. Would huge DR it be great? Absolutely.

It would instantly make R1 the most capable compact-form video camera on the market. Photography wise it would also be a big improvement, but I have a feeling that target market for R1 photography wouldn't find it as significant.

To be perfectly honest I don't think there's any realistic chance of this happening.

 

[HarryFilm]

Yep! Where are the blind tests indicating that people can or cannot see the difference? Until I see those, I think the best conclusion is that buyers are trying to justify to themselves and to others that they weren't stupid.

Depends on your eyes! If you are male and have red+green or blue+green colourblindness then of course you can't see the difference and probably don't know it anyways since you can only see around three million separate colours/shades. Conversely, a few women are TRUE Tetrachromats having not only RGB but also an extra set of V (violet) cones in the eye for HUNDREDS OF MILLIONS of separate colours which can see far beyond what most current cameras can.

In terms of more modern digital image recording needs, WE MUST be able to record at 16-bits per RGB colour channel along with a fourth Z or Depth channel for RGB+D imaging which uses millimetre wave, infrared, RADAR or SONAR scanning to provide a distance from sensor to subject value which can be in millimetres, centimetres, metres or km (or Imperial measure 16ths of an inch, inches, feet, yards and miles!) so that newer virtual reality and augmented reality applications can be facilitated. Apple's upcoming DSLR and Cinema cameras are coming with those RGB+D sensor types to be possibly announced sometime in 2025 for use in their new AR/VR applications and for 3D printing and 3D scanner use!

With 16-bits per channel recording, you also have the BEST colour rendition for initial high quality and you can then SUB-SAMPLE the colour down to lower bit-depths such as 14, 12, 10 or 8 bits per colour channel to satisfy the requirements for any given output display, print medium and/or communications bandwidth limitation.

For speed of processing, it is far easier AND CHEAPER to simply use a more modern CPU (i.e. ARM/Qualcomm 855 series or higher) which is ALREADY OPTIMIZED for 64-bit processing including for high dynamic range colour depths (i.e. 16-bits per channel RGB+D) and then use the on-board DSP processing to convert that HDR stills or video imagery on-the-fly to the required file format and bit-depth for sending over a wireless connection or via wired network/hard drive connector.

I am pretty sure Canon will UP their level of colour processing and FINALLY REALIZE that 16-bits per channel is the BEST WAY to record initially and let better CPU/SOC processors do all the fancy legwork in making the imagery FIT the available bandwidth and/or storage space!

If they DO NOT do this for the R1or the R5 mk2 ......... then .......... CANON IS DOOOOOOOOOOOOOOOOOMED !!!!!!!!

V

 

[yankiefrankie]

Guys, we're less than a week away now, and there's little to no information on this thing and lots of "it may be this, but might not be" going on. Anyone on the internet can take some shots with asterisks. What's going on? I am getting messages in my inbox from random people that I may or may not trust, as well. You guys have been very timid and overly cautious lately with little to offer. Also, you said weeks ago you'd be publishing an article about what's going on at Canon and continue to just ignore questions about that.

I am waiting for that article as well. I thought it was "finished by needed to pass a few desks" (not exact quote).

 

[wtlloyd]

All I care about in this next generation of cameras is that the AF be improved to the levels Sony has achieved. Everything else is nice, but not critical.

 

[John Wilde]

OMG DGO.

​

 

[TedYork]

On the one hand, more of any 'feature' is usually better...more MP, faster frame rate, more DR, etc. Typically, such things can be dialed back if the consequences are undesirable (files too large, too many images in a burst, etc.). But more isn't always necessary, and some people know that.

During the DRone Wars of the early 2010's, I didn't argue against the importance of more DR, but rather against the significance of an additional 1-2 stops. Sure, there are some shots where that can make all the difference. But there are many situations where the scene DR is not so high that 10 stops can't capture it, and many others where the scene DR is too far great for a single image even at 16 stops.

If we're talking about going from 12 stops of real, usable DR ('photographic DR' as opposed to 'engineering DR') up to 16 stops, that's a meaningful difference that will have a positive impact for many people. But I'm not holding my breath in the hopes that we get a 4-stop increase in either the R5II or the R1.

 

[uPhotic]

I am really unsure about how this would work for stills. I haven't seen any real deep dive (just dumbed down explanations) into how this works without subsequent images. Global shutter I guess would get around that, but without global shutter - I don't know. it may even be DGO in video, but normal in stills, or normal and expanded DR in stills (with the expanded DR having cavaets).

I would assume it would use parallel pipelines. Takes one image from the sensor into two amplifiers which are offset by whatever amount they see best. Basically one sensor exposure/readout that is put through two gain paths which the processor then combines to take the best of both based on peaking/crush.

This will basically take two images with one signal/exposure going through the photosites then combine images.

 

[koenkooi]

I would assume it would use parallel pipelines. Takes one image from the sensor into two amplifiers which are offset by whatever amount they see best. Basically one sensor exposure/readout that is put through two gain paths which the processor then combines to take the best of both based on peaking/crush.

This will basically take two images with one signal/exposure going through the photosites then combine images.

ARRI has a patent for that, not sure how generic it is. Canons DGO in video mode seems to be different enough to avoid lawsuits.

 

[DhlcadR6]

All I care about in this next generation of cameras is that the AF be improved to the levels Sony has achieved. Everything else is nice, but not critical.

If Canon is going to Sony's level, it's evolving backwards. Have fun with eye-lash AF and AI chip marketing gimmick.