120MP Canon FF SLR showed off again yesterday

[Sporgon]

When I consider the difficulty of getting maximum performance out of the 5DSR, I cannot see why I should want a 120MP FF sensor.

Because your phone will have 40MP with better IQ than your 30MP DSLR.

In what world ? Dilbertland ?

You've forgotten the influence of original capture size on IQ.

 

[Sporgon]

When I consider the difficulty of getting maximum performance out of the 5DSR, I cannot see why I should want a 120MP FF sensor.

Because you'll get an increase in the detail in your photos even if you don't "maximize" performance.

Perhaps for people who's pleasure is looking at images on their computer screen at 100%. And if that's the case, fair enough. Put for those of us that are producing printed images, even quite large ones, it's zilch. Alarmingly so in fact.

If you are chasing extreme detail you are better off with fewer pixels on a larger format IMO.

I disagree. If you are chasing extreme detail, you are better off with more pixels on a larger format.

I was meaning a larger format with less pixels than a smaller format, not specifically a lower resolution large format. For instance I would expect a 50 MP DMF to be better than 120 MP FF.

 

[rrcphoto]

When I consider the difficulty of getting maximum performance out of the 5DSR, I cannot see why I should want a 120MP FF sensor.

Because you'll get an increase in the detail in your photos even if you don't "maximize" performance.

Perhaps for people who's pleasure is looking at images on their computer screen at 100%. And if that's the case, fair enough. Put for those of us that are producing printed images, even quite large ones, it's zilch. Alarmingly so in fact.

If you are chasing extreme detail you are better off with fewer pixels on a larger format IMO.

I disagree. If you are chasing extreme detail, you are better off with more pixels on a larger format.

I was meaning a larger format with less pixels than a smaller format, not specifically a lower resolution large format. For instance I would expect a 50 MP DMF to be better than 120 MP FF.

don't know why you would.

assuming you are talking about the 44x33 MF size, the 120MP would run circles around it the amount you can oversample that 120MP over the 50?

 

[9VIII]

I want more pixel density.

Why ?

Focal length limited.
Primarily Macro, that's most of what I love to do with cameras, you just take tiny snips of however much detail you can get.
But it would be good for wildlife and landscapes too. Lenses these days are crazy sharp.

 

[3kramd5]

When I consider the difficulty of getting maximum performance out of the 5DSR, I cannot see why I should want a 120MP FF sensor.

Because you'll get an increase in the detail in your photos even if you don't "maximize" performance.

Perhaps for people who's pleasure is looking at images on their computer screen at 100%. And if that's the case, fair enough. Put for those of us that are producing printed images, even quite large ones, it's zilch. Alarmingly so in fact.

If you are chasing extreme detail you are better off with fewer pixels on a larger format IMO.

I disagree. If you are chasing extreme detail, you are better off with more pixels on a larger format.

I was meaning a larger format with less pixels than a smaller format, not specifically a lower resolution large format. For instance I would expect a 50 MP DMF to be better than 120 MP FF.

You'd expect better detail out of half the spatial resolution?

 

[jeffa4444]

Just arrived home after three days a Photokina. Took along the 5D MKIV (more on that when images are downloaded).
First stop was a tour of the C700 (only one camera at show, one was also at Cinec in Munich), the camera functionally is very close to the Sony F55 (which was also on the Sony stand with its new 4K raw recorder). The C700 expert was unable to say what the anamorphic output would be and was unaware of a future 4:3 sensor (however CVP say its in the works likely at NAB 2017).
Onto the expert bar and mount the EF 24-105mm f4L IS USM MKII and fire off a few shots, could not get to try the EF 16-35mm f2.8L IS USM MKIII, headed to the 8K camera and 8K monitors, yep they work but didn't really telling me anything. Stopped by the 120MP camera which I was told is housed in a 5DS body. 220MB files, hell the 88MB 5DS files are bad enough.
The M5 was mobbed, looks like it will be a hit for Canon and is well made, it reminded me of a G7 X MKII control wise with inter-changeable lenses but that's no bad thing.
Rest of the stand was unimpressive and not up to Canon usual standards at Photokina which seems to get smaller every two years.
Highlight of the show? Sigma remounted Art primes & zooms for cinema in both super 35 and vistavision/ff they were very well made cannot wait to MTF them and get them on a projector.

 

[Sporgon]

When I consider the difficulty of getting maximum performance out of the 5DSR, I cannot see why I should want a 120MP FF sensor.

Because you'll get an increase in the detail in your photos even if you don't "maximize" performance.

Perhaps for people who's pleasure is looking at images on their computer screen at 100%. And if that's the case, fair enough. Put for those of us that are producing printed images, even quite large ones, it's zilch. Alarmingly so in fact.

If you are chasing extreme detail you are better off with fewer pixels on a larger format IMO.

I disagree. If you are chasing extreme detail, you are better off with more pixels on a larger format.

I was meaning a larger format with less pixels than a smaller format, not specifically a lower resolution large format. For instance I would expect a 50 MP DMF to be better than 120 MP FF.

You'd expect better detail out of half the spatial resolution?

In terms of clarity in print, yes. In terms of spatial resolution ? Well my understanding of spatial resolution is that you have to compare the same size. Clearly DMF and FF are not.

Obviously I don't have a 120 mp FF camera to compare, but I do still have the 12.7 mp 5D and can compare this to a 24 mp crop sensor. Working on linear resolution the 24 mp M3 crop sensor has more resolution, producing an image that is about 37 larger along the long side, and a theoretical increase in resolution of about 15%.

However, that's the theory. The 5D at capture has twice the area of the M3, it's a greater magnification of the subject and that in itself creates more definition. To show this I'm attaching some screen grabs of what I see in practice. One is shot on the 5D, the other the M3. The former uses the 40/2.8 @ f/5.6, the latter the 28/2.8 @f/4, so both lenses about the same in resolution potential in the centre. Studio tripod.

Files from both cameras have been increased to an output size of 36" long side, so thats 8640 pixels on the long side at 240 dpi. So the M3 has gone up from 6,000, the 5D a massive doubling of the long side from 4368.

Now there is a discrepancy in the size of the subject due to the lack of a total match in FOV. But even so it is quite clear to me that the 12.6 mp 5D is better, despite the fact that it is "much lower resolution". Of course in actual fact it isn't much lower, but still, increasing the size of the file cannot create detail resolved that was not recorded in the first place.

You could argue that the 40 is better than the 28, but that is the reality of larger sensor vs smaller: the larger will use a longer focal length for the same FOV whether that be FF to crop or DMF to FF etc.

So the moral of the story is that pixel density resolution is one thing, sensor size is another. You can't take one without calculating the other, which is why "spatial" resolution applies to the same size capture.

Or at least that's what I think

 

[bjd]

This could be what, the 5ds II, or maybe a 1Ds IV? 220mb files- ouch

But I already have non-cropped RAW files between 33 and 40MB, and they equate to TIFFs
out of LR uncompressed (?) with around 170MB. Resaved from PS with LZW gave me 200MB.
Resaved from PS with ZIP compression gave me 150MB.

Obviously the 120MP would be even worse.

Cheers Brian

 

[brianftpc]

Im curious how this camera would be for astrophotography. Im starting to use my 5DsR for that now and playing with stacking tons of images. I definitely like the images more than with my 1Dx mk2 for astrophotography however the 5DsR's autofocus cant cut it on moving objects so its just a photoshoot/play around with camera.

 

[dak723]

In my limited experience, adding MPs has little benefit compared to going with a larger sensor. I know there are folks who have samples where the smaller pixels of their crop cameras resolve a similar amount of detail as the same amount of pixels on a FF sensor - but that is not my experience. Yes, with each increase in MPs you will resolve more detail than the lesser MP camera, but I believe without using a tripod, camera shake and the limits of your lens, will make the increase in MPs much smaller than you think.

Before buying my 6D a few years ago, I rented a 60D as well and then took similar shots with those 2 cameras, as well as my 300D (6 MP) camera. The 6D and the 60D have similar total pixels (20 MP to 18 MP) so one might suppose that the detail and sharpness would be similar - and certainly both would be much better than the 300D. Well, that's not how it turned out. The 6D image is much sharper and there is very little difference between the 18 and 6 MP cameras -even though I upsampled the 300d image to match the size of the 60d image.

Granted, I am sure there are folks who will and can do much more scientific tests. But the conclusion I come up with is that the resolving power of the smaller pixels of the crop camera are greatly reduced to to camera shake and potential limitations of the lens. (I was using an older 28-75mm non-L lens). In other words, even though it was 18MP, the actual result was similar to using, let's say, a 10 MP crop camera. So, hand held, due to the smaller pixels, the 50MP Canons may actually produce images equal to using a tripod on a 24-30MP camera. A 120 MP camera may not produce images any sharper than a 40-50 MP camera. These numbers are obviously total guesses, but hopefully the idea comes across.

 

[scyrene]

Wait, is the contention that different resolution sensors of the same phsyical size don't... resolve much differently?

I have a 5Ds and a 5D3... I'm happy to take side by side shots if it'll help anyone. I just tried, but tbh I'm not sure how best to display them so as to show whatever people want to see, or what subjects are best (text?).

Fwiw here's two shots. The only thing that leaps out at me (apart from the difference in resolution in the originals - these are resized) is the 5Ds has considerably better contrast. But is that just how Lightroom processes the raws by default? PS the framing is ever so slightly off, I must have moved the lens when I mounted the second camera.

(Camera attached to tripod-mounted lens; Live View manual focus - focus unchanged between shots; 70-200L IS II; 102mm, f/8, ISO 400, 2s timer delay).

 

[tron]

When I consider the difficulty of getting maximum performance out of the 5DSR, I cannot see why I should want a 120MP FF sensor.

Maybe but wouldn't you be tempted to use it with your 600mm to shoot birds? (at least standing ones)

 

[Maiaibing]

When I consider the difficulty of getting maximum performance out of the 5DSR, I cannot see why I should want a 120MP FF sensor.

Because you'll get an increase in the detail in your photos even if you don't "maximize" performance.

Perhaps for people who's pleasure is looking at images on their computer screen at 100%. And if that's the case, fair enough. Put for those of us that are producing printed images, even quite large ones, it's zilch. Alarmingly so in fact.

Try selling to magazines one day...

 

[Maiaibing]

In my limited experience, adding MPs has little benefit compared to going with a larger sensor...
(...)
...These numbers are obviously total guesses

Camera shake may sometimes have an influence. But mostly will have none at all because you can just bump the speed.

And if - if - there's camera shake, you will get as much or little camera induced blur no matter what MPIX you are shooting at; be it 6, 36 or 66 MPIX. So unless your claim is that camera shake makes hand held photography impossible (which is not the case) - there is no draw back due to camera shake with a high MPIX camera.

However, if you shoot at decent speeds, hold your camera still or have a support you will benefit greatly from the extra MPIX and get far sharper pictures with 50 MPIX camera than anything from a 25 MPIX camera.

 

[9VIII]

...but I believe without using a tripod, camera shake and the limits of your lens, will make the increase in MPs much smaller than you think.

Sorry, but you're doing it wrong.
There's nothing stopping anyone from accounting for camera shake with a faster shutter speed set according to pixel density.
If you're shooting correctly camera shake will have exactly the same effect on pixel level sharpness.

If you're light limited, then ISO can be a problem, but that's the sort of decision everyone has to make in limited shooting conditions.

 

[rrcphoto]

Im curious how this camera would be for astrophotography. Im starting to use my 5DsR for that now and playing with stacking tons of images. I definitely like the images more than with my 1Dx mk2 for astrophotography however the 5DsR's autofocus cant cut it on moving objects so its just a photoshoot/play around with camera.

umm I think you have a problem with your camera or your expectations perhaps if you think the 5DsR AF can't cut it on moving objects.

 

[rrcphoto]

In my limited experience, adding MPs has little benefit compared to going with a larger sensor. I know there are folks who have samples where the smaller pixels of their crop cameras resolve a similar amount of detail as the same amount of pixels on a FF sensor - but that is not my experience.

no offense. but you're wrong. if this was the case no one would be using 24MP cropped cameras that are more pixel dense than the current full frame cameras.

IMO .. crop users are just a little better and more understanding of learning how to use their tools apparently and NOT getting bent out of shape over pixel amount.

Pixel density affects:
- diffraction (resolved mostly with today's modern technologies)
- motion blurr
- camera shake.

what more Mp's allows you to do is still maintain the above, once you adjust for the different in pixel density.

You can still hand hold the 5DSr or even a 120MP camera as long as you realize what the 1/ focal ratio is.. as the pixel density goes up, so will the ratio. Croppers know this.. why don't full framers?

then you have what those pixels will do for you in the case of a 120Mp camera. look up oversampling.

you will get cleaner, sharper, more clarity, less digital artifacts at a given print size between a 30MP camera and a 120MP camera which is twice the resolution.

those 120Mp pixels also allow you to do more computational post processing and still have more pixels in the end than the 30Mp camera.

 

[Sporgon]

In my limited experience, adding MPs has little benefit compared to going with a larger sensor. I know there are folks who have samples where the smaller pixels of their crop cameras resolve a similar amount of detail as the same amount of pixels on a FF sensor - but that is not my experience.

no offense. but you're wrong. if this was the case no one would be using 24MP cropped cameras that are more pixel dense than the current full frame cameras.

IMO .. crop users are just a little better and more understanding of learning how to use their tools apparently and NOT getting bent out of shape over pixel amount.

Pixel density affects:
- diffraction (resolved mostly with today's modern technologies)
- motion blurr
- camera shake.

what more Mp's allows you to do is still maintain the above, once you adjust for the different in pixel density.

You can still hand hold the 5DSr or even a 120MP camera as long as you realize what the 1/ focal ratio is.. as the pixel density goes up, so will the ratio. Croppers know this.. why don't full framers?

then you have what those pixels will do for you in the case of a 120Mp camera. look up oversampling.

you will get cleaner, sharper, more clarity, less digital artifacts at a given print size between a 30MP camera and a 120MP camera which is twice the resolution.

those 120Mp pixels also allow you to do more computational post processing and still have more pixels in the end than the 30Mp camera.

But the individual pixels are not as well defined, and so whilst this is fine for making images up to the native resolution at a given dpi, once you have to go larger than native size the interpolation can work much better with better definition pixels, one of the reasons why on my example the "low" resolution 5D sized up to 36" @ 240 dpi looks to have better definition than the 24 mp crop sensor. If you look closely at the two images at full size you can actually see that the M3 has slightly more resolution in the letters.

I would make a guess that with current lenses etc there would be little difference between a 50 mp FF camera and a 120 mp one - apart from the larger output size of the latter.

 

[dak723]

In my limited experience, adding MPs has little benefit compared to going with a larger sensor...
(...)
...These numbers are obviously total guesses

Camera shake may sometimes have an influence. But mostly will have none at all because you can just bump the speed.

And if - if - there's camera shake, you will get as much or little camera induced blur no matter what MPIX you are shooting at; be it 6, 36 or 66 MPIX. So unless your claim is that camera shake makes hand held photography impossible (which is not the case) - there is no draw back due to camera shake with a high MPIX camera.

No, I am not claiming that camera shake makes hand held photography impossible. Nor am I saying there is any drawback from camera shake with a high MP camera. You will gain more detail and resolution with an increase in MP. What I am saying is that the amount of gain may not be as much as you hope (or think) if you are not using a tripod. Again, I was not doing scientific testing - I was just comparing the FF camera with 2 crop cameras. Based on what I shot, the 18MP crop camera resolved more detail than the 6 MP camera. A 36 MP camera will also resolve more than the 18 MP. But because of camera shake and possible lens limitations, the higher MP camera may have similar detail and resolution as a lower resolution camera UNLESS you use a tripod.

I had a long discussion with an expert in optics who told me that smaller pixels enable one to get finer detail IF the lens has the resolution and the camera does not move as much. Otherwise, the higher MPS will have only a little benefit. My limited experiments agree with that assessment.

 

[dak723]

In my limited experience, adding MPs has little benefit compared to going with a larger sensor. I know there are folks who have samples where the smaller pixels of their crop cameras resolve a similar amount of detail as the same amount of pixels on a FF sensor - but that is not my experience.

no offense. but you're wrong. if this was the case no one would be using 24MP cropped cameras that are more pixel dense than the current full frame cameras.

IMO .. crop users are just a little better and more understanding of learning how to use their tools apparently and NOT getting bent out of shape over pixel amount.

I take no offense, but my EXPERIENCE is that the larger pixels in the FF camera gave me substantially better image quality than a more pixel dense crop sensor of approximately the same MPs when hand held and using that particular lens. So, no I was not wrong. That was my experience and I reported it accurately.

As I mentioned, the lens is an older one, and that certainly may have effected my results. It was not a scientific test, as I mentioned. If you have examples that show the opposite to be true, I would be glad to see them. Your insinuation that I do not know how to use a crop camera I will ignore. At the time I took the photos, I had ONLY owned crop cameras and know quite well how to use them. The next time I try out a camera may well be when I can get my hands on the new M5. I will definitely be taking comparison shots with the M5 and my 6D. I am definitely hoping that the images will be of equal IQ. since I would much rather walk around with a smaller camera.