With all these giant-sensor cameras being announced - is it too much?

Benjamin Marks

Veteran
Local time
1:06 PM
Joined
Mar 27, 2005
Messages
3,343
. . . How many of you are using most of the data your current camera produces? Lordy, I'm not.

Tell you why I am asking. My data sweet-spot appears to be between 12-18 MP. My Nikon D3 and Leica M9 give me a level of resolution that approaches what I remember medium format did, back in the old days. But even then I wind up jettisoning much of the data before I hit "print" or "send."

I also have a relatively new-to-me Pentax K-1. The thing is brilliant. Great dynamic range, great resolution. But it produces 50 MB RAW files. Heck, I can't eat that much data on an empty stomach! I usually throw out all but about 2 MB of that data. . .You see where I am going with this?

It appears that I am discarding around 95% of what the camera is producing. . . . what am I going to do with a chip that produces even MORE 101010110001?

So: how many of you are actually using all (or most) of the bits and bytes your camera is producing? If not, are you jettisoning data? What percentage? Are we on the road to data bloat?

Inquiring minds. . . and all that.
 
Use that sensor/pixel shift tech in that K1 and then look at your file size haha. But man look at the result.

As for me I get off the bus at 16mp. Perfectly adequate for me. 18X20 is the biggest I ever print but I think something even bigger is possible because my pickles are awful clean. Plus my puter loves me for this I know.

Another note. When I look at snaps made with cameras like 5D1, 5D2, D700, D3, S5Pro, etc I like them more than anything made by most everything newer. All of them way less pickles than this new stuff.
 
Some cameras now offer “small RAW” in the menu options, so maybe the manufacturers realize your concerns.

I just watched a YouTube video showing a billboard print... 2 mp.

I keep the big files, but they definitely burn up hard disk space.
 
I usually throw out all but about 2 MB of that data. . .You see where I am going with this? It appears that I am discarding around 95% of what the camera is producing.
If your final image files are 2MB, then your cameras are already overkill and you have no need for larger sensors.
 
I use all the silver my film develops.

Okay seriously my D800E is a great camera and for commercial work the ability to crop quite a lot when needed is also nice, while keeping the resolution good. For me I hate going back to 12mp when shooting my old D700. 24mp minimum for digital FF IMO.

Oh and as for file space, 1GB 4x5 scans are worse.
 
LOL, but who cares? Can you sell more MP to the public? Can we even use them, if we buy them? What would we use them for that we don't already have with the current crop of FF sensors?

Legitimately curious here. . .
It depends on what you mean by current crop of full FF sensors. For Leica, it is 24MP, with a sensor size of 6000x4000. Most good printers have an optimum resolution of 300dpi. That means a print size of 20x13. You can print at a lower dpi but you sacrifice quality. If you don't print larger than 20x13, and don't routinely crop your images, you are all set.
 
More megapixels can be valuable for two situations in particular that I can think of readily - for non pro users.

The first is where you want to crop the final image for a closer shot say, while still maintaining a good pixel count for a large final image file. That strikes me as useful. In effect this is more or less what I am doing when I shoot with a cropped sensor camera.

The second is for some landscape photography. It bothers me when I make a landscape shot with the best camera and best lens at the best aperture setting - only to find when I zoom in there is little real detail due to the camera I used having a smaller pixel count sensor.

For pro users of course the ability to produce big prints may be essential to their clients so the more data the better for that kind of pro user.

Having said this, I recall that back in maybe the year 2000 or a little before, when digital cameras were just kicking off and 1 megapixel cameras were the "newest and the best" I read an article that said that depending upon the assumptions made, a 35mm negative had about 25-30 megapixels of data in them. It followed that once we had sensors of that size (and back then it seemed an almost impossible dream) we would have all that was needed from digital cameras in order to match 35mm film photography. Well, we are now there on the mega pixel front. In fact, not just there - we are well beyond.

I think that particular technology race is coming to its logical conclusion and has already switched to a race for better dynamic range etc - something which has already improved greatly but still has some way to go.
 
Having said this, I recall that back in maybe the year 2000 or a little before, when digital cameras were just kicking off and 1 megapixel cameras were the "newest and the best" I read an article that said that depending upon the assumptions made, a 35mm negative had about 25-30 megapixels of data in them. It followed that once we had sensors of that size (and back then it seemed an almost impossible dream) we would have all that was needed from digital cameras in order to match 35mm film photography. Well, we are now there on the mega pixel front. In fact, not just there - we are well beyond.
I don't think we should be satisfied with 35mm film equivalent quality from digital. Lots of us sought higher quality with MF and LF.
 
I think that particular technology race is coming to its logical conclusion and has already switched to a race for better dynamic range etc - something which has already improved greatly but still has some way to go.

I agree with all of that, but particularly this last bit. I am certainly "there" in terms of base-level sensor quality. I think I have been since 2008 or so. But dynamic range is really where I'd like to see improvement.
 
If you look at some of the cameras pros are using, the MP count isn't huge. For a long time, many ADs were comparing digital camera output to a scanned 8x10 chrome. With the demise of big film, companies like Phase One took up the market with big sensors. They just graduated to a larger format sensor with a big MP count.

I could argue that no one needs that much information for publication. I was competing with 8x10 when using Kodachrome for AD work. I'm sure that some scientific recordings make good use of a huge MP count. But, it's not of much use to the photo artist unless you print really big like Gursky.

XF IQ4 150MP Camera System (full frame medium format sensor)
The XF IQ4 150MP Camera System provides unparalleled resolution and detail, allowing you to produce RAW images that are larger than life, for limitless editing possibilities. The XF IQ4 150MP Camera System is much more than the world’s first 151-megapixel camera – it’s a forward-thinking companion for producing the most creative and ambitious photographic visions.
https://www.phaseone.com/en/Camera-Systems/XF-Camera-System/IQ4/XF-IQ4-150MP-Camera-System.aspx
 
We just keep on buying them so they keep on building them bigger and bigger.

As for dynamic range, digital passed film on that front quite some time ago but that old myth keeps getting kicked around anyway.

I hate to break the news to everyone; there are a lot of reasons to keep using film but using it because it produces better technical files than digital is not one of them.

Of course, as always, technical perfection is not the measure of a great photograph and never has been.
 
Years ago I experimented with printing, because I had a new printer and it was a big leap forward after the old one.

One of the things I did was some large prints (a little over 18" x 12") from crops so that they were 100 dpi. That had the advantage of being from the centre of the originals and so the best bit of the lens was in use and I reckon that the lens is an important part of this. Anyway, at that size no one held them in their hands and looked at them like we do 4"x6" and so on. So the prints hung on the wall and looked good.

That was just using about 2 megapixels...

So I reckon I could get a 5ft by 3ft 4ins poster from 24 mp and more as you'd have to stand so far back from it that you'd not see the individual pixels...

Now the question is; how many of us look at our photo's on the screen when the computer cuts them back to about 1 mp to fit the screen? Even smaller if they are using a smart phone and never back them up out of the 'phone...

And the second question is do people think or care about it?

Regards, David
 
Years ago I experimented with printing, because I had a new printer and it was a big leap forward after the old one.

One of the things I did was some large prints (a little over 18" x 12") from crops so that they were 100 dpi. That had the advantage of being from the centre of the originals and so the best bit of the lens was in use and I reckon that the lens is an important part of this. Anyway, at that size no one held them in their hands and looked at them like we do 4"x6" and so on. So the prints hung on the wall and looked good.

That was just using about 2 megapixels...

So I reckon I could get a 5ft by 3ft 4ins poster from 24 mp and more as you'd have to stand so far back from it that you'd not see the individual pixels...

Now the question is; how many of us look at our photo's on the screen when the computer cuts them back to about 1 mp to fit the screen? Even smaller if they are using a smart phone and never back them up out of the 'phone...

And the second question is do people think or care about it?

Regards, David
The viewing distance is important and does make a difference, someone like Burtinsky intends for his large prints to be somewhat studied closer up to reveal the information held with in. A lot of other work is made not to be studied in this manner but viewed as a whole and the further we move away the lower the DPI that is required. The massive billboards are very low resolution per inch as the viewing conditions don't require it. Personally I have done an A0 from a Ricoh GX100 but that was over laid images (ones in my avatar) and the 60x40 inch from a 16mp m43 camera no one complained about quality. I have been told by a retoucher that high res files do make their work a lot easier though.
 
. . . How many of you are using most of the data your current camera produces? Lordy, I'm not.
Me too.
If not, are you jettisoning data? ...

No. I keep them as they are. But I don´t do that RAW-stuff. My JPGs are from my 5 to
18(?) MP cameras.

I generally do not care for the amount of pixels a camera produces.

My pictures are taken for viewing as a whole only. So it doesn´t matter how many details there
*could* be. If I need big prints and there are too few data I can scale them up to any size the
printer is happy with :)
 
Since the camera in many phones is as good as the m8 was there is little room for the camera companies to not keep upping the sensors. Why lug around a camera if the phone makes equal or better pictures?

...As for dynamic range, digital passed film on that front quite some time ago but that old myth keeps getting kicked around anyway.

I hate to break the news to everyone; there are a lot of reasons to keep using film but using it because it produces better technical files than digital is not one of them...

Bravo on both counts. I am getting ready to sell off most or all of the 810 equipment since seeing what the X1D can really do.
 
24 X 36 mm Is No Longer Full Frame

24 X 36 mm Is No Longer Full Frame

It is amusing that some people who insisted 24 X 36 mm sensors are required pay no heed to the newer cameras with more surface area.

Granted, the increase in area is only 1 2/3 – 1,441 mm sq. vs 864 mm sq. However the total maximum signal level is higher. So there is a significant increase on potential signal-to-noise ratio.

The surface area difference is smaller than APS-C vs 24 X 36 mm cameras. But every single argument people made to justify 24 X 36 mm sensor areas is equally valid for the 43.8 x 32.9mm sensors.
 
... But dynamic range is really where I'd like to see improvement.

Get a 24 X 36 mm camera with dual conversion gain(1). Better yet, use a 43.8 x 32.9 mm (or larger surface area) camera.

Here's some data as measured and computed by Bill Claff (scroll down for the table.

The columns are

dynamic range| low-light ISO| low light EV


FujiFilm GFX 50S 11.90 7853 11.30

Hasselblad H6D-50c 12.01 6400 11.00

Hasselblad X1D-50c 11.98 9058 11.50

Leica S (Typ 007) 11.36 6400 11.00

Nikon D850* 11.63 4115 10.36

Pentax K-1 II* 11.60 7669 11.26

Phase One IQ3 13.06 12800 12.00

The * indicates dual-conversion gain sensors. The dynamic range data is about a constant 2 stops lower than engineering dynamic range. I only included cameras with DR > 11.

1. Photo-diode circuits can be optimized for dynamic range or sensitivity, but not both. The circuits' conversion gain level determines which property is favored. Dual conversion-gain cameras use two different gain settings. One is for bright light (low camera ISO settings) and the other is for low light (high camera ISO settings).
 
Back
Top Bottom