Will there be any announcements from Canon ahead of CP+ this month?

[Joules]

I do not find that to be true at all.
Try shooting in RAW and editing the photo in post.
The IBIS in the R5 even allows for longer handheld exposures.
"Night Shot" is meant for a tripod anyway.

The implementation on my Google Pixel 3a definitvely is not meant for a tripod. In a simple test I did once, it definitively produced a better straight out of camera result of a static scene than my 80D + Sigma 35 mm 1.4 were able to deliver hand held in that situation. With IS and IBIS thrown into the mix, it becomes a more fair comparison of course and as I said, that was not a proper test. Still, the results are impressive. No surprise really, as I use the same technique of stacking numerous shots to reduce noise in Astro all the time.

I'd also like to see the camera companies to offer some of these automized processing techniques, rather than relying on the user performing them manually in post. That's not the same as as strpping the control away from the user. Ideally a format to save this in would keep the underlying raw data in tact and allow tweaking of the processing steps. Of course it would require using proprietary software. But Canon has already taken some baby steps into that direction.

 

[Maps]

I can see someone jumping from a $1K phone to a $1K camera but not to a $4K camera.
M6 II, 90D, and RP are all around that price point.
A big problem is that a $1K smartphone has the same specs as a $4K camera.
Most people who want cameras have $500 smartphones and end up with M200, M50, or a Rebel.
The biggest problem is that most people are perfectly happy with smartphones just like most people were perfectly happy with point and shoots.

I was trying to think of some real world examples to illustrate my point and realized not only do I not know a single person who has gone from phone to FF, I don’t know anyone who has gone from phone to ILC, period. Can anyone share their experiences? I’m wondering about the last two years or so. A decade ago doesn’t count, that might has well have been a different planet.

Unfortunately, I know 3 people who’ve gone the other direction. I wouldn’t say they actively ditched their DSLRs for phones, it was more… as their DSLRs aged they decided not to upgrade them, instead opting to just use their phones. One of those three will still take the DSLR birding on occasion, but otherwise it’s been shelved.

 

[koenkooi]

I do not find that to be true at all.
Try shooting in RAW and editing the photo in post.
The IBIS in the R5 even allows for longer handheld exposures.
"Night Shot" is meant for a tripod anyway.
Most modern ILC will embarrass an iPhone at high ISO at long exposure.
Maybe an entry-level full-frame camera that does all of that automatically is a good idea to bring new people in but pros and enthusiasts mostly want control over that type of thing.
M200 is an entry-level APS-C camera with features like that but it also has entry-level specs.
Even a fully spec'd out camera with computational photography would not be as convenient as a smartphone camera as we would still have to transfer the photos.

I can get an handheld 3 second exposure sharp with my phone, not so with the R5. And the night shot mode on phones wasn't made for tripods, quite the opposite.

 

[amorse]

A lot of conversation on comparing the computational photography opportunities utilized in phones versus capability native in a full frame camera, specifically the night-shot systems. Personally, I couldn't see myself jumping on those tools if they were built into an ILC since I often end up doing them in post where I have more control, but I do see the value in offering those in an ILC. In my experience, I've known a number of people who buy their first ILC, don't immediately get the results they envisioned, then roll back to their phone to access those automated tools. I suspect making those tools available in a full frame camera could keep new entrants to the ILC world interested for longer, and maybe buying more lenses. That's complete speculation though.

Since a lot of those techniques require the camera taking many images in succession then using software to combine and reduce noise using an electronic shutter, I'd wonder if many ILCs would need faster readout speeds to make it workable in a similar way.

 

[Joules]

Since a lot of those techniques require the camera taking many images in succession then using software to combine and reduce noise using an electronic shutter, I'd wonder if many ILCs would need faster readout speeds to make it workable in a similar way.

Faster than 20 FPS? I don't think so. That's plenty do do handheld Stacking for noise reduction or super resolution. There's also a lot of cool stuff one could do with that in combination with the depth information coming from Canon's dual pixel design.

Official support for these types of applications in camera would improve the results even if you'd rather perform the processing yourself. The depth information is one example, exposing offsets between the frames through the camera's IMU data would be another.

 

[amorse]

Faster than 20 FPS? I don't think so. That's plenty do do handheld Stacking for noise reduction or super resolution. There's also a lot of cool stuff one could do with that in combination with the depth information coming from Canon's dual pixel design.

Well yes, the R5/R6 clearly have a fast enough readout speed to likely support it in electronic shutter, but I'm thinking of the RP as an example with a much slower max burst rate. What's its max burst rate - 4 FPS? My thinking has been that these automated tools may be more attractive to entry level users who are familiar with using the tools on a smart phone - maybe they either don't know how to perform those multi-image combinations in post, or don't care to go through the hassle, and just want the same ease they had when using a phone for photography. The utility of those tools may be greater in entry level bodies which may not have the burst rate, or computing power to support the function yet. The burst rates being put up by the R5 and R6 imply that it could become more accessible soon, but I couldn't see that getting into an RP as a firmware update.

 

[Joules]

Well yes, the R5/R6 clearly have a fast enough readout speed to likely support it in electronic shutter, but I'm thinking of the RP as an example with a much slower max burst rate. What's its max burst rate - 4 FPS?

Well, I would hope with the level of technology they have in their latest sensor generation, such slow frames in lower end models will be a thing of the past. Quick reminder that the M6 II is a little cheaper than the RP and with 14 FPS mechanical at 32.5 MP delivers almost exactly half the throughput of the R5 and 14% more throughput than the R6. It also has a 30 FPS boot mode cropped to 18 MP that these two FF bodies lack. So Canon has set the precedent for putting serious speed into mid tier bodies. The RP is handycapped by its now two generations older sensor.

 

[amorse]

Well, I would hope with the level of technology they have in their latest sensor generation, such slow frames in lower end models will be a thing of the past. Quick reminder that the M6 II is a little cheaper than the RP and with 14 FPS mechanical at 32.5 MP delivers almost exactly half the throughput of the R5 and 14% more throughput than the R6. It also has a 30 FPS boot mode cropped to 18 MP that these two FF bodies lack. So Canon has set the precedent for putting serious speed into mid tier bodies. The RP is handycapped by its now two generations older sensor.

I agree with you - several cameras are fast enough now, and several are not - especially older ones. With speed moving into lower and lower end bodies, implementation of computational photography is becoming more possible, but I think that speed limitation may be (only) one of the reasons why we haven't seen ILCs incorporate some of the tech yet. Computing power may be an issue too - obviously several higher end ILCs from other manufacturers use multi-shot tools to increase resolution or colour depth, but some manufacturers still require you to do the combining outside the camera. Even the lowly RP has a multi-shot focus bracketing function which needs to be completed on a computer. I think it's safe to say that manufacturers are aware of the benefits of multi-shot computational photography, regardless of whether that's for low light improvements, resolution increases, depth of field improvements, or colour depth. I do think, however, that there are still some technical developments that must be brought to lower-end camera lines before the seamlessness expected by a cell phone photographer can be delivered on an entry-level body, but we're definitely getting closer especially with improvements in frame rates.

 

[festr]

where is RF 35 1.2?

 

[TravelerNick]

I was trying to think of some real world examples to illustrate my point and realized not only do I not know a single person who has gone from phone to FF, I don’t know anyone who has gone from phone to ILC, period. Can anyone share their experiences? I’m wondering about the last two years or so. A decade ago doesn’t count, that might has well have been a different planet.

For stills? Not sure. For video quite a few.

For stills I'd assume most of those sort of people are students in various types of photography/art programs. I think we're past the point of the average family being a standalone camera .

 

[TravelerNick]

Faster than 20 FPS? I don't think so. That's plenty do do handheld Stacking for noise reduction or super resolution. There's also a lot of cool stuff one could do with that in combination with the depth information coming from Canon's dual pixel design.

It's not just the time to take those frames. It's the time to process them.

But look at the Canon DGO sensor in the C70. My guess is they're going to expand on that. The C70 isn't full frame which makes me guess the costs are still too high for a similar sensor in a full frame hybrid camera.

 

[Pape]

It's not just the time to take those frames. It's the time to process them.

But look at the Canon DGO sensor in the C70. My guess is they're going to expand on that. The C70 isn't full frame which makes me guess the costs are still too high for a similar sensor in a full frame hybrid camera.

Maybe they just test new tech with c70 .They need info how flawlesly it work before they launch big production line with new kind sensor.
Price for mass produced sensor is still unknown.

 

[EOS 4 Life]

It's not just the time to take those frames. It's the time to process them.

But look at the Canon DGO sensor in the C70. My guess is they're going to expand on that. The C70 isn't full frame which makes me guess the costs are still too high for a similar sensor in a full frame hybrid camera.

According to Canon C500 does not have DGO due to power constraints.
ARRI ALEXA LF has a full-frame DGO but it eats through batteries like there is no tomorrow.

 

[Michael Clark]

R1 is a different price point so it should not impact R5 and R6 sales very much.

True, but the R5 being cheaper can certainly affect the potential sales of the R1.

 

[Michael Clark]

internet hype posters

It's easy to jump ship and switch when they don't actually own anything from either system...

 

[SteveC]

It's easy to jump ship and switch when they don't actually own anything from either system...

And will continue to not own anything from the new system afterwards.

I own everything Nikon ever made...however every single item of my Nikon gear is average in every way except for not existing. That minor attribute is not there. But I could switch to a bunch of S*ny gear with the same minor handicap at any time.

[Inspired by someone's example demonstrating why "existence" is not an attribute of an object.]