A Shake-up Coming to the APS-C DSLR Lineup? [CR2]

[Talys]

A perfect video-purposed camera will have lower resolution/pixel density, because that's how you get less noise and better low light performance. It should have an electronic viewfinder, because you can't see the display screen in the sun (and monitors with shades are often not appropriate). Factors like drive speed (fps) and autofocus modes don't matter much. Instead, in the consumer-ish market, you're looking for features like dual pixel autofocus, subject tracking, and facial recognition.

A perfect wildlife camera for enthusiast types will have higher resolution, because we're forever too short on reach and are forced to crop. We want optical viewfinders, because they don't have refresh issues, autofocus modes and precision are very important, more fps is always welcome, and things like subject tracking don't matter at all. Plus, we're looking at how big the buffer is and how quick it can empty out, something that video people don't care about, because what they want is the ability to constantly write data at a speed fast enough for whatever resolution they're recording.

As usual, that's well-reasoned. But I don't quite agree.

First, I don't think we are talking perfect anything. Perfect is the realm of Canon Cinema and the 1D line, not the APS-C line. So, in my mind it's all about the compromises.

Since video cameras generally shoot at lower shutter speeds than stills cameras (double the frame rate) and since random noise can be less apparent on video (it moves around from frame to frame) I think a case can be made that a video camera can sustain higher pixel density than a stills camera.

Note that I'm only saying "a case can be made" and I'm sure some video folks will disagree. But, I'm just saying that the 1/60 of a second for video allows for a lot more light to hit those pixels than the 1/800 and above needed for birds and moving wildlife. Plus, with video, you are much less likely to be using a 400mm plus lens, which drives up the shutter speed needed for stills photography.

You may be correct about an EVF, but any video DSLR is a compromise and this is a rumored DSLR, not a rumored mirrorless with an EVF.

On the other hand, a wildlife camera needs to have the highest possible ISO performance because most birds and mammals are active during the lowest-lit parts of the day. Yes, reach is important, but if the image is noisy that will only get worse as you crop. And, as I mentioned above, those high shutter speeds are needed with those long lenses.

I'm not sure what you mean by "things like subject tracking don't matter at all." Subject tracking matters a whole lot if you want to shoot a bird in flight.

So, while I respect your opinion, I'd have to say that from my perspective, I tend to reach the opposite conclusion.

Thanks I'm quite fine with different perspectives -- I have had my mind changed about many things over the years The things that matter and the way I shoot now, whether it is wildlife or portraiture is dramatically different than a decade ago, and a lot of that is simply people convincing me that there's a better way.

So first of all, I'm not a video guy at all. For me, aside from the occasional family thing that I could just as easily take from a smartphone, I don't do video. I totally agree that a flagship APSC DSLR is not an ideal format for a video-centric rig, which I didn't articulate well -- but I what I was driving at is that basically, nobody should be spending $2,000 (or whatever) for an APSC DSLR and hoping for an awesome 4k video rig.

I mean, maybe the 7D3 will do 4k, but I just can't imagine that 4k video is a reason to buy the 7D3, 90D or model in between.

On the birds in flight, I have a miserable track record with AI tracking (using sensitivity/acceleration/point switching on 5D/7D bodies), but it's not really the fault of the tracking technology. Basically, what happens is, if the bird is small enough in the field of view to be effectively tracked and cropped, I will be probably be unhappy with the final result, because the bird will be too small, even if it's perfectly in focus.

If the bird is big enough for me to be happy with the result, I will be tracking the bird anyways. At the end of the day, it's just easier to use spot AF or center + expanding points and make sure I'm pointing at the bird in flight

I am happy to be convinced otherwise; I just haven't been able to get good enough results out of tracking to use it, and it was a large contributing factor to my deciding to buy a 6DII despite not having any of the fancy AF modes.

I totally agree with you with wanting high ISO performance so that we can raise shutter speeds. However, I have been convinced by people in this forum that getting much better than 80D or Sony A6300 (given their pixel densities) is an unrealistic expectation given today's technology. So with APSC, what I'm primarily looking for is a bright daylight camera with lots of reach and shooting at lower ISOs.

At over ISO 800, I am not really happy with my 80D's results for most bird shots, and prefer cropping from a full frame image on my 6DII. The additional noise, especially in the eyes, just kills the shot for me. Not because they're terrible photos -- just, it'll be another at-best mediocre shot, and I generally just cull those because I already have terrabytes full of them.

There are exceptions, of course -- especially if it's something I've never photographed before... I'm just talking about a general rule of thumb for me.

 

[nchoh]

Maybe

90D 24mpx sensor

7D Mark III 24mpx sensor

7Dr (same as 7D Mark III but 30 mpx sensor and no low pass filter)

What fps's would you put on those?
A 7Dr sounds juicy but please not at to much expense of fps, 8 would be minimum for me.

The notion of an additional model number suggests to me moving the top of the line 7D up a level and I find that exciting

This sounds the most logical. A higher pixel 7D would help drive sales of the newer EF-S lenses.

 

[RGF]

I wonder if Canon will put their best AF, metering in the top of line APS-C camera. Nikon has done this and the D500 is much superior to the 7D M2 in terms AF, DR, ...

 

[ritholtz]

Hoping it is be 90D in SL2 size. I want to go back to 80D. It is just too big.

 

[zim]

What fps's would you put on those?
A 7Dr sounds juicy but please not at to much expense of fps, 8 would be minimum for me.

Obviously the lower resolution version would have a faster frame rate than the higher resolution one.

I'd think more likely to be 8-9fps for the 24mpx version and 6-7 fps for the higher resolution version

Sadly that sounds about right, 1 or 2 less than I'd want. As always tho imaginary specs for imaginary cameras always leave room for hope

 

[zim]

Maybe

90D 24mpx sensor

7D Mark III 24mpx sensor

7Dr (same as 7D Mark III but 30 mpx sensor and no low pass filter)

What fps's would you put on those?
A 7Dr sounds juicy but please not at to much expense of fps, 8 would be minimum for me.

The notion of an additional model number suggests to me moving the top of the line 7D up a level and I find that exciting

This sounds the most logical. A higher pixel 7D would help drive sales of the newer EF-S lenses.

Actually I think a high pixel 7d would be more inclined to drive L glass. Lower Res video centric being more STM glass

 

[AvTvM]

an SL-2 sized DSLR with 7D III functionality would be of interest. very compact but full performance. all muscle, no fat. best sensor available, best DIGIC, best AF available, fully functional EOS UI and an LP-E6 battery for sufficient juice (yes, it will fit).

basically a mini john cooper works. just like the car industry took nearly a century to find out that many people prefer small cars but still want full performance ... camera industry still has not understood this. those "innovative" canikons still want you to buy humgonguos bricks if you want full functionality, performance and control. this fundamentalky wrong approach is the main reason why smartphones have been eating most of camera makers cake. people got sick and tired of their sorry, underpowered "compact camera" offerings and sorry, underpowered "rebels" and bulky, heavy and absurdly overpriced mirrorslapper bricks.

 

[Hector1970]

I wouldn’t see the he point at all of a higher density pixel 7D3.
The 7D series is all about frame rate.
You also want the fastest shutter speed possible so you’d want much better ISO performance than the 7DII
The second best upgrade it can have is better tracking.
It’s hit and miss on BIF - birds are small and tricky you need the best focusing you can get.
More MPs is just a waste unless you are photographing static objects. If you are an XXD or XXXD camera would be better value for money.

 

[mb66energy]

Funny how so many folks complain about choices of bodies which they will never use. The more the merrier...that is unless you're anti choice.

It depends on the availability of "clear choices" - 20D and 5D was a clear situation: APS-C or FF and a substantial price gap. Today you have the choice between flavours of basically the same thing like SL2 - 77D - M50 - 80D all sporting very similar or same sensor but slightly different feature sets and prices between 500 ... 850 EUR. O.k., the M50 as mirrorless has a very special property at least for me: I can adapt other optics without "mirror box hassle".

The problem (at least for me again) is that I have slow progress of decision processes if the options are too close together. At the moment I can make faster decisions in the lower price segment where the losses in case of a wrong decision do not hurt

 

[jolyonralph]

I wouldn’t see the he point at all of a higher density pixel 7D3.
The 7D series is all about frame rate.

This is exactly why they would differentiate into two streams, a lower-pixel faster frame rate model, and a higher-pixel lower frame rate model because everyone has different needs and you clearly would prefer the former while others would prefer the latter.

 

[fingerstein]

7d mark II has better performance for video than 6d or 5d mark III - higher framerates, DPAF, headphone jack...
I expect for 7D mark III to be somewhere below 1dx mark II but a bit better than 5D4. Of course, in low light it will suck.

 

[BillB]

I wouldn’t see the he point at all of a higher density pixel 7D3.
The 7D series is all about frame rate.

This is exactly why they would differentiate into two streams, a lower-pixel faster frame rate model, and a higher-pixel lower frame rate model because everyone has different needs and you clearly would prefer the former while others would prefer the latter.

Faster fps and more mp may be better, but I'm wondering how much practical difference there is in 20-30% bumps in fps or mp.

 

[Loswr]

Faster fps and more mp may be better, but I'm wondering how much practical difference there is in 20-30% bumps in fps or mp.

If your employer offered you a 20-30% raise, would you turn it down because you question the practicality of the difference?

 

[crashpc]

Bhaha, yes, 30% is easily noticable, 50% is very significant. It would be so awesome to have 32Mpx APS-C cameras around now...
I tried two teleconverters at once on my 55-250mm IS STM, and they still resolved more than the lens alone.
That is, 50Mpx sensor for that lens is not an issue, and would help to resolve more, crop more, or resample more...
After 150Mpx, moire issues would be healed almost completely. That´s what I´m after.

 

[BillB]

Faster fps and more mp may be better, but I'm wondering how much practical difference there is in 20-30% bumps in fps or mp.

If your employer offered you a 20-30% raise, would you turn it down because you question the practicality of the difference?

And why would a 30% increase in fps be comparable to a 30% increase in income? A 30% increase in fps gives you 9 images per second rather than 7. Better, sure? But how much better?

 

[Orangutan]

Faster fps and more mp may be better, but I'm wondering how much practical difference there is in 20-30% bumps in fps or mp.

If your employer offered you a 20-30% raise, would you turn it down because you question the practicality of the difference?

And why would a 30% increase in fps be comparable to a 30% increase in income? A 30% increase in fps gives you 9 images per second rather than 7. Better, sure? But how much better?

I think you've just made his point: more is always better unless there's a specific cost to the increased FPS. For BIF, sports, etc. it can improve the chance of getting the shot you wanted (e.g. wing position). Is it a lot of difference? No, but what's the drawback? A much higher frame rate can affect the choice of sensor tech used; but, other than that, more is better.

 

[BillB]

Bhaha, yes, 30% is easily noticable, 50% is very significant. It would be so awesome to have 32Mpx APS-C cameras around now...
I tried two teleconverters at once on my 55-250mm IS STM, and they still resolved more than the lens alone.
That is, 50Mpx sensor for that lens is not an issue, and would help to resolve more, crop more, or resample more...
After 150Mpx, moire issues would be healed almost completely. That´s what I´m after.

If you look at the effect on print quality, the questions become how many more pixels per inch you are getting out of a 30 per cent bump in the pixels on the sensor, and how much difference that will make to print quality. 300 ppi seems to be the magic number for print quality and opinions vary on how much quality you are buying when you go above that. The number of pixels per inch on the print varies with the square root of changes in the number of pixels on the sensor, so a 30% increase of pixels on the sensor isn't going to give you a 30% increase in pixels per inch on the print.

More pixels on the sensor are better, but how much better? Any time camera spec numbers get thrown around there is the danger of falling prey to DPR Magic Number Syndrome, in which all numerical differences are believed to be significant.

 

[BillB]

Faster fps and more mp may be better, but I'm wondering how much practical difference there is in 20-30% bumps in fps or mp.

If your employer offered you a 20-30% raise, would you turn it down because you question the practicality of the difference?

And why would a 30% increase in fps be comparable to a 30% increase in income? A 30% increase in fps gives you 9 images per second rather than 7. Better, sure? But how much better?

I think you've just made his point: more is always better unless there's a specific cost to the increased FPS. For BIF, sports, etc. it can improve the chance of getting the shot you wanted (e.g. wing position). Is it a lot of difference? No, but what's the drawback? A much higher frame rate can affect the choice of sensor tech used; but, other than that, more is better.

True enough, but is the improvement worth the cost of buying a new camera, or paying more for the higher specs?

 

[Orangutan]

Faster fps and more mp may be better, but I'm wondering how much practical difference there is in 20-30% bumps in fps or mp.

If your employer offered you a 20-30% raise, would you turn it down because you question the practicality of the difference?

And why would a 30% increase in fps be comparable to a 30% increase in income? A 30% increase in fps gives you 9 images per second rather than 7. Better, sure? But how much better?

I think you've just made his point: more is always better unless there's a specific cost to the increased FPS. For BIF, sports, etc. it can improve the chance of getting the shot you wanted (e.g. wing position). Is it a lot of difference? No, but what's the drawback? A much higher frame rate can affect the choice of sensor tech used; but, other than that, more is better.

True enough, but is the improvement worth the cost of buying a new camera, or paying more for the higher specs?

Each person has different values and disposable income.

 

[Loswr]

Faster fps and more mp may be better, but I'm wondering how much practical difference there is in 20-30% bumps in fps or mp.

If your employer offered you a 20-30% raise, would you turn it down because you question the practicality of the difference?

And why would a 30% increase in fps be comparable to a 30% increase in income? A 30% increase in fps gives you 9 images per second rather than 7. Better, sure? But how much better?

30% = 30%. What's confusing about that?

30% more MP means you can crop deeper with the same output resolution. Maybe that means a 400mm lens will do, instead of a 500mm lens.

30% higher fps means a higher chance of capturing the moment of peak action. Sports and birds in flight are obvious applications, but higher fps is useful in many situations.

"Pucker Up"

EOS 1D X, EF 70-200mm f/2.8L IS II USM @ 70mm, 1/200 s, f/2.8, ISO 100