Too many pixels for hand-held?

[Dikdik]

Advertisement (gone after registration)

I have read several times that high-resolution, or many pixels, makes it more difficult to do hand-held photos. Presuming that reading the sensor is the same speed for a 18mp as a 50mp, and final print size is the same, then what accounts for the difficulty?

 

Hello pico,

 

The problem depends on the number of pixels. The more pixels are on a chip the more accuracy is necessary to obtain really sharp results.

Please, let me explain a little bit more in detail.

 

There is a direct coherence between number of pixels and size of chip.

The most important influence on sharpness of photos is in this context the so called pixelpitch, the distance from pixel to pixel.

The newly delivered Canon cameras 5DS and 5DSR are working with a chip of 50 MP. According to specifications of the company size of this chip is 864 mm² (24mm x 36mm). There are in a total 8688 x 5792 pixels. That means a number of 50.320.896 pixels. All these pixels are compressed in a small chip of 24 x 36 mm. So, you have on this chip a distance from one pixel to the other from 4,1432 µm; that is round about 4 µm.

 

Once, the very familiar Canon 5D worked on a full format sensor with only 12.719.616 pixels but, on a chip of the same size. Therefore the pixelpitch was 8,24 µm. This is compared to the latest camera equipment from that company nearly the double space between one pixel unit to the other.

The sensor of the Hasselblad digital back CFV 50 C has a size of 32,9 x 43,8 mm, 6200 x 8272 pixels and therefore a pixelpitch of 5,3 µm. The Leica DMR is working with a pixel pitch of 6,8 µm, the Leica S with 6 µm.

 

There is a trend by the manufacturers to concentrate more and more pixels on the chips and therefore, to diminish the distance from one pixel to the other. But for the photographer that means a highly careful workflow while taking photos.

Times, when we've got sharp photos by handheld the cameras are gone. The modern chips will record even the smallest shutter. That can be movements by wind, air disturbances even by temperature, shuttering of ground by cars or simply by touching the equipment.

We – my wife and me - are working in nature and medical-scientific photography. But we are not professionals. It's just still a hobby. But some photos we sell to earn the costs. The primary rules in our workflow to obtain sharp photos is: use a tripod, use mirror lock-up and use an electrical shutter release.

The distance between the pixels is nowadays so small, that handheld photos will become indeed mostly unsharp looking at 100% at the screen.

 

There is in our field of photography – to my understanding – only a small advantage using these big chips: you can print the photos bigger and you get more detail - sometimes. The finest print the human eye can see is 300 dpi. Whatever might be over these 300 dpi we will not recognize a difference. It depends on the anatomical structure of our human eye. For instance, eagles and vultures can distinguish much more 'pixels' and therefore can see extremely more details. But what will they go sightseeing these big photoprints?

 

I hope, we could explain a little bit your questions. Sorry for my clumsy English.

Have a nice weekend and all times: 'good light'.

     

 

Dikdik

[Paul J]

I have shot 22, 30, 40, 60, 80 (I've tried 100MP too) digital medium format cameras for the last decade. I don't like tripods, I rarely use them. No real problems to talk about.

 

I have shot with a 5DS R for 6 months. No problems here either.

 

I have more issues with medium format hand held, because of the size, weight, bulk and mirror slap of these cameras which are very large. It does not stop me.

 

I remember reading arguments 10 years ago how Nikon's 12MP D2x made it so much harder to get sharper images because of it's high res. They were wrong then and they are wrong now.

 

There are those talking about it, and then there are those doing it, day in, day out without problems.

[MarkP]

Have you seen the Apple advertising campaign, large posters of people's iPhone pictures? Not limited by blur or close observation.

 

 

Likely carefully screened and more careful processed photos I'll bet.

...and we're seeing them from 50m away or more.

[pop]

Likely carefully screened and more careful processed photos I'll bet.

...and we're seeing them from 50m away or more.

Besides, we don't know what cameras  were used for making those pictures.

[edwardkaraa]

It seems that many people are happy with 24 mp, as it offers enough resolution with less motion blur issues. So let's imagine that Leica introduces the next M with 50 mp, and suddenly everyone who were taking perfectly sharp photos at 24 mp now notice blur at the pixel level. Well, just downsize the shaky shots to 24 mp and the problem is solved. No big deal

[Paul J]

Take a 24MP and a 50MP file that are equal - View the 50MP file at 100%, and in some circumstances (depending on your lens, your lighting, your shooting technique and your subsequent settings) you may notice the blur more because of the extra resolution and because the file is physically larger. But the amount of blur does not magically increase.

 

Moreover - Resize the 50MP file to 24MP sizes and it will still look better than the 24MP file. This is the affects of supersampling.

 

Resize the 24MP file, larger, to 50MP, and the 50MP will look far better compared to the 24MP one.

 

It's a win/win scenario.

 

Most photographic problems are solved by throwing more light at it - OR - more sensor sensitivity, and this is one of them. High megapixel count no longer comes at the cost of high ISO. What you gain from double the resolution 24 => 50MP far outweighs what you loose with an extra stop of ISO - this is generally all that is needed in terms of shutter speed increase to balance the difference between what is seen at 24MP compared with what is seen at 50MP.

[zlatkob]

Advertisement (gone after registration)

With so very many photographers in world and with cameras having ever more megapixels, presumably someone has already proven that more pixels make it more difficult to do hand-held photos. There are countless tests and comparisons showing just about everything else.  But I have yet to see visual proof of this alleged phenomenon.  That seems rather odd to me, considering how common it is for people to claim that it's true.  Has anyone seen the visual proof?  A link somewhere?  I'm taking about proof in a normal photographic print, not something only visible when viewing pixels at 100%.  If it only exists a the pixel level, then we can't see it without extreme magnification.

 

Print size has always affected hand-holdability.   That hasn't changed.  A 50-inch print will reveal much more blur than a 5-inch print.  If you want to see maximum detail in a 50-inch print, you need better technique than for a 5-inch print.  This much is obvious:  if you enlarge a picture more, any blur in the picture is easier to see.

 

Did a similar difficulty happen in the pre-digital era?  Was it more difficult to do handheld photos with Panatomix-X or Plus-X than with Tri-X?  I don't recall that being the case.

[pgk]

Motion blur is motion blur - end of story. FWIW I use Nikonos SB105 flash (strobe) units underwater. Despite their age they are good, robust and reliable units. They do though have a quirk. Their output is almost instantaneously to full power but tails of ever so slightly over time (at full power they run at about 1/1000s with the 'tail' dying at about the same - read on). If the camera is moving this results in very small specular highlights imaging as though they were a comet - with a slight 'tail'. More MPixels means that the 'tail' is easier to see when magnifies, that's all and revisiting film shows the same effect. So to get back to the question, if your technique or camera or equipment don't freeze motion then you will get motion blur regardless of sensor or pixel pitch or MPixels. The degree to which is apparent will depend on various parameters too. I have motion blur on low and high MPixels and equally I have images which I could not want to be 'sharper'.

[Exodies]

Besides, we don't know what cameras  were used for making those pictures.

 

 

They were iPhone 6 and 6+. The ones I saw were at bus stops, about 1.5*1m.

[pop]

They were iPhone 6 and 6+. The ones I saw were at bus stops, about 1.5*1m.

How do you know? It's not exactly news that advertisement is not always in all things honest.

[Jeff S]

Ming Thein's view...  https://blog.mingthein.com/tag/shot-discipline/

 

And some other thoughts from Ctein on megapixels (with multiple linked articles)...  http://theonlinephotographer.typepad.com/the_online_photographer/2012/02/d800-megapixels.html

 

Jeff

[zlatkob]

Ming Thein's view...  https://blog.mingthein.com/tag/shot-discipline/

 

And some other thoughts from Ctein on megapixels (with multiple linked articles)...  http://theonlinephotographer.typepad.com/the_online_photographer/2012/02/d800-megapixels.html

 

Jeff

 

Ming Thein writes:

 

"Cameras with higher resolutions, for a given sensor size, are harder to hand-hold and get a good result than those with a lower resolution. And I define a good result as one which the image is critically sharp at 100% actual-pixels view."

 

But that's not the definition of a "good result" in photography generally.  That's his own definition of a good result, and it relates exclusively to 100% pixel viewing.  This definition stacks the answer by enlarging the image from the camera with higher resolution more than the image from the camera with less resolution.  In other words, to see a difference one needs to be looking at actual countable pixels.  When viewing at that level, one can't even see the photograph.  

 

Ming's conclusion is:

 

"The upshot of all of this in practical shooting is that you’ll need to use a higher shutter speed than you expect, which lowers your shooting envelope by a corresponding number of stops as noise increases as you ramp up the sensitivity to keep shutter speeds high."  

 

But he fails to qualify that conclusion.  That conclusion only follows if you're also going to be enlarging one image more than the other.  Enlargement increases blur; resolution doesn't.

 

It's not the image from the higher res camera that needs a higher shutter speed.  Rather, it's the image that gets enlarged more that needs a higher shutter speed.  That would happen if you switched the cameras.  If you enlarged the 12mp image to 100cm and you enlarged the 36mp image to 10cm, the lower resolution 12mp camera would require the higher shutter speed.

 

The bottom line is that Ming is talking about 100% pixel viewing, not about prints.  I think some people are misreading this to conclude that prints of the same size will somehow show more blur from a higher res camera than from a lower res camera.  They won't.

[FlashGordonPhotography]

I also think the whole pixel count thing is somewhat misleading. Personally i think it should read, "the larger you view your images, the more important camera technique becomes. Making an image larger also makes any technical flaws larger until they become easily visible. Because higher megapixel cameras often result in viewing images at larger sizes an increase in MP count can also increase the ease at which you can see when less than perfect camera technique is employed."

 

Apart from detail, I don't perceive much blur (camera movement) differences between a 12, 24 or 42MP sensor at postcard sizes (all 35mm sensors). But i sure do at poster sizes. Sometimes at large sizes I certainly see lower resolution files having more edge sharpness than high res files but I can usually equalise them with pixel bining. With my own photography I can also tell when I've not used a tripod vs when I did, even at speeds approaching 1/500 of a second. I also know what my personal hand hold threshold is with each camera and lens i own. None of them are 1/focal length. Mostly I need around 3-4 x focal length to get images that are sharp enough for myself.

 

Gordon

[helged]

It is hard to understand that binning of a high-res image to match that of a lower res image, with appropriate post-sharpening of the binned image, should not give comparable sharpness to the low res image.

 

Viewing high-res and low(er)-res images at 100% on screen - or on prints resolving the underlying pixel resolution - can of course show differences; but how often do we really present (final) images at 100%? Certainly very seldom in my case. But other's practices may, of course, differ.

[flyalf]

It is not a myth. See  my slightly primitive sketch.

 

On a large pixel the motion blur is inside the sensel, thus invisible.

With small pixels the same amount of blur will affect three sensels, making it visible and even artificially increasing it.

 

So the problem is not the MP count, but the pixel size. A medium format 50 MP sensor will have less problems with motion blur than a 35  50 MP sensor.

 

pixel.jpg

 

Obviously, the larger the blur gets (thus outside the large sensel), the less the impact of the spillover -relatively- will be.

Imo it will be the same blur given the same magnification on the final rendering.

 

You can arange to divide the large sensor into several small inside it. The same amount of photons will hit these combined sensors as the large one (excluding "gaps" between them) at any movement.

 

The blur inside a large sensor being smaller than the universe will not be invisble because some photons that would have hit just inside the sensors border will hit the neighbouring sensor due to movement.

 

So imo the blur will be exact the same up to 100% of the large sensor.

[gsartori]

It is not a myth. See  my slightly primitive sketch.

 

On a large pixel the motion blur is inside the sensel, thus invisible.

I do agree, when the Nikon D800 came out, Nikon had many issues with customers claiming that it was a bad camera, not sharp at all, they did a study and came out with a paper saying that if you apply the classic 1/focal length rule and you see the photo at 100% you may have blur due to micro motion

[pgk]

"The upshot of all of this in practical shooting is that you’ll need to use a higher shutter speed than you expect, which lowers your shooting envelope by a corresponding number of stops as noise increases as you ramp up the sensitivity to keep shutter speeds high."

Which is of course 100% wrong as any target shooter will tell you (otherwise nobody could ever shoot straight). The point being that you simply need to stabilise the camera better and ensure that any movement is along the taking lens access (similar to ensuring motion is in the direction of a rifle barrel). In effect higher MPixels simply mean that you have to up your game in terms of stabilisation, much as a target shooter needs to increase control with a smaller target or longer distance. 

[IWC Doppel]

I have to say the logic for smaller pixels offering more motion blur doesn't cut it in afraid

 

The speed will be identical the resolution will be greater you'll simply end up with improved resolution of the blur. So it will be true that very small motion blur that might not have been resolved with a coarser grain of resolution could now

 

However on the same size image the only differences will be the subtle interaction between lens characteristics and image processing

[IWC Doppel]

PS I think Jaapv is definitely into monkey business

[Advertisement]

Hallo,

Look here: Leica M

[jaapv]

PS I think Jaapv is definitely into monkey business

Selfie:   

 

edit: found better file

 

 

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!