A high megapixel camera is coming [CR2]

[keithcooper]

I don't see any limit to the desire of more pixels - as long as they actually provide more detail, there's obviously no point in having more pixels than your lens can resolve, for example.

Ah, outresolving lenses... comes up every time high MP is mentioned ;-)

I'm minded to suggest there is no 'obvious' about it - better sampling enables better processing if you need/want it. In this instance, a rising tide does indeed float all boats - I've tested all kinds of old/new lenses and from a detail point of view, more MP to sample the lens output helps (other annoyances/inconveniences notwithstanding)

Particularly illuminating was seeing what difference the 180MP multishot mode of the S1R made with some of my older EF glass.

Now, there probably is a point where more MP just fills your cards quicker, but not [quite] yet ;-)

 

[Fischer]

Why do people want more than 50 mpex on 35mm? Honest question. Thx.

Apart from the obivious sharpness advantage: Ability to crop more, better noise control above lowest iso-levels and flexibility when shooting a prime lens (think shooting almost anything with a 35mm lens - from landscape to near-macro-flower shot) are probably the three main factors. The second of course becoming more important as we see people move much more into 400+ iso shooting than previously.

 

[entoman]

None of this is true.

You may be able to see technical faults better because you are tempted to zoom in more but at the same output size - which is the only thing that matters - high MPIX cameras are never a disadvantage - only possibly at an advantage.

The whole point of high MP is that you *will* zoom in more (by cropping) or by enlarging, so you *will* see errors in focusing exagerrated, you *will* see subject movement exagerrated, and you *will* see camera-shake exagerrated.

What's the point in buying a high MP camera if you're only posting on instagram, and not utilising that extra resolution?

 

[unfocused]

A few thoughts:

This is not an R1. Canon is nothing if not predictable. The R1 will have either equal or slightly higher resolution than the Z9.

Because Canon is predictable and consistent, this will likely be another variation on the R5, just as the high resolution 5Ds and 5Dsr were based on the 5D.

Canon is not going to redesign their bodies to accommodate a different style flip screen. Not going to happen.

Beware of the buffer. The R5 can barely handle 45 mp. Generally okay for birds in flight, but it becomes a real problem with any kind of sports shooting. Even the R3 can run into buffer problems at the highest frame rate. I've gone exclusively to CRaw to reduce buffer issues.

The buffer is another reason this will not be an R1. Just too much data to push through for sports shooters.

Don't expect any major feature improvements (such as quad pixel autofocus). Canon will reserve new features for the R1 and R5II.

Where does Canon go with the R5II in two years or so? It seems like the resolutions are converging if this is 75mp and the R5II is...what...60mp?

If the R7 comes to fruition, expect it to be about 29 mp and share the same sensor, just as the 5Ds and the 7DII.

 

[Joules]

The whole point of high MP is that you *will* zoom in more (by cropping) or by enlarging, so you *will* see errors in focusing exagerrated, you *will* see subject movement exagerrated, and you *will* see camera-shake exagerrated.

What's the point in buying a high MP camera if you're only posting on instagram, and not utilising that extra resolution?

But in those instances where you want to crop or enlarge to show more detail, you are absolutely at an advantage compared to a lower resolution sensor. After all, if you do select the right settings to maximize detail, you will get more of it compared to a lower resolution sensor with the same settings.

In other words, a higher resolution sensor will at worst match the results from a lower resolution one, and at best exceed them by providing enhanced details. Which does have use cases. But there is no compromise on image quality with high resolution shooting. And when you take pictures of something that you don't intend to crop or enlarge massively otherwise, you also don't need to select the settings to produce detail past what is actually needed.

The only true downside to higher resolution are file size (which in turn negatively affects FPS, buffer depth, storage volume and post processing) and cost.

 

[Richard CR]

50% happy - had hoped for 100 MPIX. Will still get it! Since I'm also into buying a car and a - ridiculously - expensive bike - this year, I'll certainly be doing my part keeping the economy running against all Covid and Ukraine odds.

hahaha... me too! I thought it would be around 105, or something slightly over 100mp

 

[Joules]

If the R7 comes to fruition, expect it to be about 29 mp and share the same sensor, just as the 5Ds and the 7DII.

Apart from similar pixel densities, those sensors have nothing in common. The 7D II is a dual pixel sensor, the 5Ds one is conventional.

 

[Richard CR]

The whole point of high MP is that you *will* zoom in more (by cropping) or by enlarging, so you *will* see errors in focusing exagerrated, you *will* see subject movement exagerrated, and you *will* see camera-shake exagerrated.

What's the point in buying a high MP camera if you're only posting on instagram, and not utilising that extra resolution?

yes, but there is a lot of fallacies with that. an APS-C camera with Canon's 32.5MP is just as difficult as a 75MP+ full frame camera with the same lens. However people get horrified thinking about the full frame one moreso than the APS-C one.

 

[Richard CR]

I don’t know how quad pixel AF will work but the 75MP sensor rumour probably have more to do with the technology used in this system rather than the output resolution of the camera.

I just don’t think Canon sees high resolution as the future or that 120MP sensor from way back would have made an appearance by now. Speed, AF and video are what future Canon camera buyers want. When Canon jumps to 12K or 16K video resolutions is when we will see a massive bumps in stills camera resolution IMO.

That 120MP sensor DSLR though probably took a hit because Canon was internally switching to working on the RF mount, and then a 120MP sensor without DPAF would have been DOA.

Canon's been working on Quad for a long time now. I've lost count to how many of those patent applications I've reported on. There's been tons more I've seen, but for the life of me i couldn't describe the patent and unlike some sites (cough.. CW...) I try to always explain the patent application.

AF speed will be impacted the most. Image processing - no.

BUT - now that Canon can do stacked sensors, the world of opportunity is open, for Canon do to much faster processing of AF data actually on the sensor itself.

Canon is also doing tricks - for instance looking at ROI (Region of Interest) versus reading the entire sensor. That can help too.

but really a 75mp camera isn't going to be a speed demon and I really doubt it will be quad AF, as no matter how good Canon does the fab, it's less efficient than dual pixel AF in terms of DR, and this camera is meant for landscape aka high DR.

 

[Fischer]

The whole point of high MP is that you *will* zoom in more (by cropping) or by enlarging, so you *will* see errors in focusing exagerrated, you *will* see subject movement exagerrated, and you *will* see camera-shake exagerrated.

What's the point in buying a high MP camera if you're only posting on instagram, and not utilising that extra resolution?

I can see that you misunderstand the relationship between sensor resolution and camera output.

This discussion took off big time when the 5DS/R was released when several well reputed reviewers created confusion with claims such as these - until the dust settled and all serious reviewers contradicted their spurious claims. However, the myth seems to live on. I also blame Canon marketing for creating some of the confusion when they released the 5DS/R.

The only thing that counts is your output. The actual image you are viewing on screen, print etc. Either you are comparing two identical images or you are not. What end viewing result does one camera deliver compared to another? Nothing else matters. Ever.

If you compare the same image taken from a 1 - 20 -50 - 5000 MPIX picture there will be zero (0.0) more shake, zero (0.0) more subject movement and zero (0.0) more focus miss between them. Its simply not physically possible as all these three are optical properties and have nothing to do with sensor size. The larger pix picture will very often be sharper - and shooting at high MPIX is thus almost always an advantage - it can only never be a disadvantage in relation to these three issues. Not possible.

As for your second point about Instagram I am not sure why you are posting on a Canon forum, if this is the media you are interested in. An iphone is likely a better option for you. YMMV.

(I will not expand or continue this discussion on MPIX shake etc. as I cannot be bothered anymore, but refer to the explanations on normalized viewing sizes that you can find on the web and elsewhere if you want to learn more about how this works out in real life.)

 

[entoman]

I can see that you misunderstand the relationship between sensor resolution and camera output.

This discussion took off big time when the 5DS/R was released when several well reputed reviewers created confusion with claims such as these - until the dust settled and all serious reviewers contradicted their spurious claims. However, the myth seems to live on. I also blame Canon marketing for creating some of the confusion when they released the 5DS/R.

The only thing that counts is your output. The actual image you are viewing on screen, print etc. Either you are comparing two identical images or you are not. What end viewing result does one camera deliver compared to another? Nothing else matters. Ever.

If you compare the same image taken from a 1 - 20 -50 - 5000 MPIX picture there will be zero (0.0) more shake, zero (0.0) more subject movement and zero (0.0) more focus miss between them. Its simply not physically possible as all these three are optical properties and have nothing to do with sensor size. The larger pix picture will very often be sharper - and shooting at high MPIX is thus almost always an advantage - it can only never be a disadvantage in relation to these three issues. Not possible.

As for your second point about Instagram I am not sure why you are posting on a Canon forum, if this is the media you are interested in. An iphone is likely a better option for you. YMMV.

(I will not expand or continue this discussion on MPIX shake etc. as I cannot be bothered anymore, but refer to the explanations on normalized viewing sizes that you can find on the web and elsewhere if you want to learn more about how this works out in real life.)

You seem to misunderstand my attitude to high MP. For the record, I shoot on a 45MP R5 and I consider that to be the "sweet spot", the best compromise for my own genres and methods of shooting.

I'm most definitely NOT interested myself in instagram, Flickr, fbook or any other social media - until recently a lot of my work was published in books and/or magazines (for which 20MP incidentally is more than adequate), but for the last couple of years I've concentrated on images for my own personal use. Many such as BIF shots need to be heavily cropped as I prefer to leave space around the subject in the EVF to make them easier to follow in flight, which is why I use a 45MP camera instead of a 24MP model.

The explanations about exaggerated camera shake, subject movement and focus errors are referred to currently in various reviews and fora, and seem quite logical to me. If indeed they are myths that need to be exploded I'm happy to learn, but in addition to these and the file-size issues that you acknowledge, are you also telling me that the higher pixel density images resulting from more MP do not adversely affect noise and diffraction, or is that another "myth"?

 

[Czardoom]

Everything is a compromise and we all have different needs and priorities. Some prioritise high ISO, others prioritise resolution or DR, others prioritise fps. Unfortunately there is no "one size fits all". You say 45-50MP is your current minimum requirement, but a couple of days ago you were extolling the virtues of the 20MP OMI, and saying how good the 20MP Nikon D500 still is, which seems a little contradictory.
...

I don't want to put words in Alan's mouth, but I think it is pretty obvious that he was referring to a FF camera when he gave 45-50 MP as his minimum. Less than that and you are getting less resolution than a 20 MP crop camera. So, not contradictory at all.

 

[entoman]

yes, but there is a lot of fallacies with that. an APS-C camera with Canon's 32.5MP is just as difficult as a 75MP+ full frame camera with the same lens. However people get horrified thinking about the full frame one moreso than the APS-C one.

Perhaps the reason why people get horrified more about high MP FF is because those that buy them are likely to be more advanced users than those buying a 90D, and hence fussier?

Or perhaps it's just an irrational fear of high numbers, based on the widespread statements about perceived negative effects on noise, diffraction, camera shake, subject movement and focus accuracy.

There are so many conflicting claims and theories expounded on the internet that it's very difficult to distinguish fact from myth

 

[AlanF]

Everything is a compromise and we all have different needs and priorities. Some prioritise high ISO, others prioritise resolution or DR, others prioritise fps. Unfortunately there is no "one size fits all". You say 45-50MP is your current minimum requirement, but a couple of days ago you were extolling the virtues of the 20MP OMI, and saying how good the 20MP Nikon D500 still is, which seems a little contradictory.

There is nothing contradictory about it whatsoever - it's 100% completely consistent. The D500 is a 20 Mpx APS-C not a 20 Mpx FF. It has the same resolution as the 45 Mpx FF Nikon D850. It's pixel density that determines resolution, not total Mpx. The Canon 7DII is another one of my favourite cameras and its 20 Mpx sensor pixel density is exactly the same as the 5DSR, which I also used a lot.

 

[AlanF]

I don't want to put words in Alan's mouth, but I think it is pretty obvious that he was referring to a FF camera when he gave 45-50 MP as his minimum. Less than that and you are getting less resolution than a 20 MP crop camera. So, not contradictory at all.

Thanks. I didn't see this until after I posted my reply identical reply.

 

[entoman]

There is nothing contradictory about it whatsoever - it's 100% completely consistent. The D500 is a 20 Mpx APS-C not a 20 Mpx FF. It has the same resolution as the 45 Mpx FF Nikon D850. It's pixel density that determines resolution, not total Mpx. The Canon 7DII is another one of my favourite cameras and its 20 Mpx sensor is exactly the same as the 5DSR, which I also used a lot.

Indeed, so let's concentrate on comparing low and high pixel-density for a given sensor size, i.e. FF. I'm not debating for the sake of pedantry, just trying to separate fact from popular myth. We are all here (I hope) to amicably exchange opinions and understand the truth about perceived facts.

Which if any of the following, in your view, are adversely affected when comparing hypothetical same-generation 20MP and 80MP FF sensors of similar technology, and when examining the final image occupying the entirety of e.g. an 8K monitor ? -

minor focus errors
subject movement
camera-shake
high ISO noise
diffraction-related softness
perceived edge sharpness
perceived fine detail

 

[twoheadedboy]

A few thoughts:

This is not an R1. Canon is nothing if not predictable. The R1 will have either equal or slightly higher resolution than the Z9.

Because Canon is predictable and consistent, this will likely be another variation on the R5, just as the high resolution 5Ds and 5Dsr were based on the 5D.

Canon is not going to redesign their bodies to accommodate a different style flip screen. Not going to happen.

Beware of the buffer. The R5 can barely handle 45 mp. Generally okay for birds in flight, but it becomes a real problem with any kind of sports shooting. Even the R3 can run into buffer problems at the highest frame rate. I've gone exclusively to CRaw to reduce buffer issues.

The buffer is another reason this will not be an R1. Just too much data to push through for sports shooters.

Don't expect any major feature improvements (such as quad pixel autofocus). Canon will reserve new features for the R1 and R5II.

Where does Canon go with the R5II in two years or so? It seems like the resolutions are converging if this is 75mp and the R5II is...what...60mp?

If the R7 comes to fruition, expect it to be about 29 mp and share the same sensor, just as the 5Ds and the 7DII.

If they want to continue the R5 as a stills-focused hybrid body which shoots 8k raw, it would behoove them to continue with 45ish MP in order to make it as easy as possible from a processing sense to do that. You could imagine an R5 MKII having the 45 MP version of Canon's new stacked sensor in the R3, quad pixel (assuming an R1 comes before an R5 II), next gen IBIS, next gen DIGIC for better performance, faster bus for card performance/buffer, better heat performance (though still not limitless as with the R5C), etc. It wouldn't have to be as much of a leap over the R5 MKI as the R5 was over the 5D MKIV to be worthwhile.

 

[iheartcanon]

I would not want to give up the flip out screen. However if they could figure a way to make it be able to either tilt OR flip out leaving you the choice, now that would be very cool.

Eith way, I'll be waiting for the R1 which will go nicely with my R5

 

[Fischer]

are you also telling me that the higher pixel density images resulting from more MP do not adversely affect noise and diffraction, or is that another "myth"?

From the release of the 5DS/R and the later 5DIV we know that having more pixels can compensate for increased noise even compared to newer sensors when iso goes to around 400 and higher and invariance starts setting in. I expect this to continue to be the case.

My hope is thus, that a high MPIX R will be at least as good as the R5 at base level iso - because the 5DS/R was better noise wise than the 5DIII even at base iso. I see no obvious reason to expect less this time around.

Of course there is no way telling if history will repeat.

However, in practical terms I do not see that there is much straight forward to be said about the noise advantage of the higher and lower MPIX sensors that we as consumers actually get to buy. Its much more about comparing the characteristics of the relatively few sensors manufacturers are selling us at the time they are released to the market. And here the picture is far more blurred than simple theory suggests as demonstrated by comparing the 5DIII, 5DS/R and 5DIV. Still maybe a new 75 MPIX would prove worse noise wise than the existing R5 - its just not an obvious assumption.

Diffraction can be calculated and affects outputs at various f-stops as you can easily see by taking a picture at f/22 or higher.

 

[Fischer]

Don't expect any major feature improvements (such as quad pixel autofocus). Canon will reserve new features for the R1 and R5II.

5DS/R came out with a lot of great new features compared to the 5DIII that later were incorporated in the following 5DIV. Why should this not repeat, if you think Canon is predictable, so that the high MPIX gets several new features compared to the R5 that will later go into a R5II?