Here is what Canon is announcing next, including the EOS R7, EOS R10 and RF-S lenses [CR3]

[adrian_bacon]

True, I enjoyed a lot the pancake on a rebel

Yep. Canon has a handful of unsung heroes in their EF lineup that I'd love to see in their RF lineup. Small, inexpensive (relatively speaking), killer optical performance for the price, and perfect for double duty if a FF lens. Even more awesome is the 40 STM on an old rebel film body like the 3000 or the 620/650. Makes for a really nice walk around film kit.

 

[jordanisaak]

They're both about 24MP, but other than that, no, not really. The 80D sensor is ever so slightly larger at 22.5mm across the long edge vs 22.3 for Canon's other APS-C sensors, and has a different color response (which you can see via the WB multipliers). Canon has more than one 24mp APS-C sensor design. They could really save themselves some money by getting it down to just one or two sensors instead of seeming to have a different sensor for each line.

Funny how Canon lists the same 3.72 µm pixel pitch and 6000x4000 resolution for all their 24mp sensors, including the 80D. That doesn't line up with the 80D having a slightly larger sensor unless the extra size is from extra pixels outside the imaging area.

 

[adrian_bacon]

Funny how Canon lists the same 3.72 µm pixel pitch and 6000x4000 resolution for all their 24mp sensors, including the 80D. That doesn't line up with the 80D having a slightly larger sensor unless the extra size is from extra pixels outside the imaging area.

Yeah, I don't know about that. All I know is the color balance multipliers are different and the technical specs that canon lists for the 80D lists a slightly larger sensor than what they usually do for APS-C. Of course, they couch it with an "approximately", so who knows what it actually is, all I know is that if you compare different canon APS-C sensors from different cameras (that have the same resolution) on DXO-mark, they don't all have the same white balance multipliers, which tells me that they aren't exactly the same sensor. You could also use dcraw to extract the full sensor array from the files and compare the dimensions if you wanted to. I wouldn't be surprised if they were different dimensions. You'll need to use Adobe DNG Converter to convert the CR3 files to DNG, then use dcraw to extract the raw sensor data from the DNG, and just use dcraw directly on the cr2 files.

 

[jordanisaak]

Yeah, I don't know about that. All I know is the color balance multipliers are different and the technical specs that canon lists for the 80D lists a slightly larger sensor than what they usually do for APS-C. Of course, they couch it with an "approximately", so who knows what it actually is, all I know is that if you compare different canon APS-C sensors from different cameras (that have the same resolution) on DXO-mark, they don't all have the same white balance multipliers, which tells me that they aren't exactly the same sensor. You could also use dcraw to extract the full sensor array from the files and compare the dimensions if you wanted to. I wouldn't be surprised if they were different dimensions. You'll need to use Adobe DNG Converter to convert the CR3 files to DNG, then use dcraw to extract the raw sensor data from the DNG, and just use dcraw directly on the cr2 files.

For the 80D, Canon USA and Canada list 22.5mm x 15.0mm, Canon Europe lists 22.3mm x 14.9mm, which lines up with their other APS-C sensors.

If there were different dye formulations in the Bayer filters that would explain the different white balance coefficients.

 

[neuroanatomist]

From the 80D brochure:
The EOS 80D camera features a newly developed 24.2 Megapixel (APS-C) CMOS sensor that not only captures high-resolution images, but also features refined individual pixels that enable high ISO speeds.

But a small tweak to the AA filter or CFA would constitute ‘new development’.

 

[jordanisaak]

From the 80D brochure:
The EOS 80D camera features a newly developed 24.2 Megapixel (APS-C) CMOS sensor that not only captures high-resolution images, but also features refined individual pixels that enable high ISO speeds.

But a small tweak to the AA filter or CFA would constitute ‘new development’.

The 80D predates the M50 and, as far as I know, all the other 24mp APS-C Canon cameras. So yes, its sensor was newly developed at the time.

 

[Chig]

It makes even less sense to think Canon would keep alive and maintain 2 APS-C systems. It doesn't make sense to mount a 24-70 lens but something like the R7 is perfect for the 100-500, giving plenty of reach.

For the entire lifespan of the M mount Canon had the EF aps-c system so carrying on with M and R aps-c is surely much the same .
It's the EF system that is doomed !
The M mount may well continue if Canon chooses to keep it going .
Personally it's not "my cup of tea" but plenty of others like it.

 

[Blue Zurich]

True, I enjoyed a lot the pancake on a rebel

One of my favorite lenses on the 5D3

 

[adrian_bacon]

From the 80D brochure:
The EOS 80D camera features a newly developed 24.2 Megapixel (APS-C) CMOS sensor that not only captures high-resolution images, but also features refined individual pixels that enable high ISO speeds.

But a small tweak to the AA filter or CFA would constitute ‘new development’.

If memory serves, the 80D was the first APS-C sensor where the ADC was on chip, giving it much better DR over the 70D and 7DII.

All that being said, if you look at the color response to it compared to other 24MP APS-C sensors from Canon, the 80D WB multipliers have RGB 1.75, 1, 1.75, the M6 has 1.45, 1, 1.35, the M100 has 1.49, 1, 1.37, the 2000D has 2.08, 1, 1.75, etc. Despite that they're all APS-C and roughly 24MP, the different measured white balance multipliers seems to indicate that they're not all exactly the same sensor.

 

[neuroanatomist]

If memory serves, the 80D was the first APS-C sensor where the ADC was on chip, giving it much better DR over the 70D and 7DII.

All that being said, if you look at the color response to it compared to other 24MP APS-C sensors from Canon, the 80D WB multipliers have RGB 1.75, 1, 1.75, the M6 has 1.45, 1, 1.35, the M100 has 1.49, 1, 1.37, the 2000D has 2.08, 1, 1.75, etc. Despite that they're all APS-C and roughly 24MP, the different measured white balance multipliers seems to indicate that they're not all exactly the same sensor.

There were several ‘newly-developed’ 18 MP APS-C sensors before the 24 MP sensors came along.

 

[Chig]

This R10 is clearly not a 90d replacement, it's more a 1300d replacement, so the numbering system is changing, the 90d must therefore be an R8, then the 850d's replacement will most likely be the R9. This R10 is entry level. I'm wondering if the R7 is even worth the wait, especially if they use an old sensor from the 90d. I'm more interested in performance than cost as we are ripped off by canon anyways, the R5 is more in pounds than dollars, even though it's still 1.22 to the pound. How can they justify £4,300 against $3,800?

20% VAT added makes it $4,560 which equates to about £3,720 so yes Canon UK is gouging you.
I assume you're talking about the price of the R5

Weirdly in NZ if you take our GST sales tax off (which you can avoid by taking a trip overseas and buying it here Duty Free at a local shop which sends it to the airport for you to collect) the the price is only NZD 5,734.78 which converts to USD 3,604 which is a bit cheaper than the USA price and Canon NZ has a full 5 year warranty too

 

[Chig]

No in body stabilization for the R10!!!??

There shouldn't be any mirrorless cameras without stabilization. Especially on these small cameras especially prone to shake.
.
Damn it Canon is going back to its old shenanigans. The r7 will likely be a larger body following the seven series. That means the r10 is going to be the only crop option for carrying around everywhere.

Even worse, trash lenses that will come with it.

Doesn't sound good to me at all.

Why are you saying "Trash" lenses ?
All lenses whether cheap or eye-wateringly expensive from Canon, Nikon, Sony, etc have been of at least decent optical quality in the last 10 years at least.
The EF-s and M lenses are all excellent optically if a bit plasticy in construction.

IBIS isn't a crucial feature and for most uses isn't really needed

 

[Chig]

I feel like with the various software techniques IBIS is a bit dated anyway. Physically jiggling the sensor about was great when sensors were pushing limits, but right now we could make a pixel dense sensor that's oversized and achieve the same results without the complexity of moving parts.

Yep, I'd rather not have IBIS in an R7 personally , I'd rather have a more solid sensor mount with better heat dissipation and I'd much prefer a BSI stacked sensor too !

 

[adrian_bacon]

There were several ‘newly-developed’ 18 MP APS-C sensors before the 24 MP sensors came along.

I think they did the same thing with those 18MP sensors too where each line had a sensor that varied slightly from the others, even though they were all 18MP. Moving forward, hopefully, they can get some cost savings by just having the same sensors and differentiating on features, body size, buttons, etc. Fabbing sensors isn't exactly cheap, and they can't be doing themselves any favors to have so many different sensor variations, even if it is as relatively simple as just changing the pigments/dyes used in the CFA filters. From what I understand, that's an integral part of the chip making process, so it's not like they're just taking a sensor chip and slapping a different CFA filter on the top when they're putting the camera together on the production line.

 

[mariosk1gr]

Why 90d?

80D have 24MP, 90D 32MP.

I don't see the point to go back in sensor tech with a new camera first and I imagine that tech would be more similar with latest dslr 90D Canon introduced back then and propably better in many sections.

 

[SnowMiku]

I don't see the point to go back in sensor tech with a new camera first and I imagine that tech would be more similar with latest dslr 90D Canon introduced back then and propably better in many sections.

That's a good point about going backwards with the MP with the R10 compared to the 90D. The 90D did get a few 7D like features with the Joystick, I think they done that because they already decided they were not going to release a 7D mk III. I think the R10 may go back to around 80D level again to differentiate itself enough from the R7.

 

[jordanisaak]

If memory serves, the 80D was the first APS-C sensor where the ADC was on chip, giving it much better DR over the 70D and 7DII.

All that being said, if you look at the color response to it compared to other 24MP APS-C sensors from Canon, the 80D WB multipliers have RGB 1.75, 1, 1.75, the M6 has 1.45, 1, 1.35, the M100 has 1.49, 1, 1.37, the 2000D has 2.08, 1, 1.75, etc. Despite that they're all APS-C and roughly 24MP, the different measured white balance multipliers seems to indicate that they're not all exactly the same sensor.

Hang on. Those white balance multipliers are massively different. Are you sure you took those multipliers from files that were taken in the same lighting condition with the same white balance setting on each camera? Even completely different sensors would not have white balance multipliers vary so widely in the same lighting conditions. But different lighting would explain the variance.

Take two shots with your 80D. One with tungsten light balance setting and one with shade white balance. Then compare the white balance multipliers with each other.

 

[ashmadux]

Why is it? I’ve never had a Camera with IBIS before and my pictures are fine. It’d be a nice addition but is in no way a crucial feature.

"No way crucial.." to you.

No need to debate IS. The tech available and mature. Its should be in all mirrorless bodies. I'm sure I'm more demanding about my files considering I have successfully used M bodies for professional gigs because I know what I'm doing and have a decade of experience using them locally and across the world.

Shutter shock/slap is a real thing that really impacts pictures, especially on small, light bodies like the later M series and denser, small pixel pitch ASPC sensors. It's been the same problem on the M3, M50, and the M62. I negated the issue on my original M1 with a korean made Phottix large metal grip. No such item is available for the m50 at all. I have the m62 in a smallrig cage for the same effect.

Lack of IS is a real problem. And unfortunately canon's segmentation shenanigans has decided that this is a pro-body feature only.

If there's anyone out there that knows the challenges of small mirrorless without IS, trust that it's me. The last few M bodies I have had have been the most unreliable (and broken) cameras ever for me. I have used them on zero shoots because I just can't trust them.**


** I was able to successfully use the m62 with a EF 24-105 attached for a assignment. Worked great for a few hours, however it's never been able to repeat that success. AF fails + shutter slap all over the place. Trust me, it stinks

 

[ashmadux]

Hang on. Those white balance multipliers are massively different. Are you sure you took those multipliers from files that were taken in the same lighting condition with the same white balance setting on each camera? Even completely different sensors would not have white balance multipliers vary so widely in the same lighting conditions. But different lighting would explain the variance.

Take two shots with your 80D. One with tungsten light balance setting and one with shade white balance. Then compare the white balance multipliers with each other.

The difference in color from the old T2i is closer to the M1 (original), and MUCH better than the flat colors of the m50 and m62. I own all of them all of them, and I'm extremely bothered by the desaturated color profiles of the m62 especially.

 

[ashmadux]

But, your current M bodies don't have IBIS either, so why would that stop you replacing them with a successor without it? If they are intending to keep the form factor (and $s) comparable to the M series, then IBIS is unlikely. Some of the competitors do (of course) have IBIS, but it's always a compromise on image quality / $s / features, no matter what body we're talking about.

Simple. Those are very old cameras, and I'm looking to upgrade - in the same small aspc form factor.

New bodies should have it as a standard feature. The sensor quality is debatable, however the lack of IBIS is killer when shutter slap is also a factor.