Ad
Ad
Ad
Pages: [1] 2 3 »   Bottom of Page
Print
Author Topic: Megapixels and print size, a small experiment  (Read 11979 times)
ErikKaffehr
Sr. Member
****
Offline Offline

Posts: 7652


WWW
« on: March 12, 2013, 12:58:21 AM »
ReplyReply

Hi,

There was a recent discussion on print sizes, some posters argued that little difference can be seen in large prints from 12 MP and 36 MP cameras. I decided to make a small experiment. I took a reasonably well made picture from 24 MP Sony camera and made A4 print from 35% crops of

Original image
Image downsized using nearest neighbor to 12 MP and scaled back with bicubic
Image downsized using bicubic to 12 MP and scaled back with bicubic

Those prints would correspond to about 55x81 cm or 21"x31". What I have seen visually.

At distance, say 1.5 m, the images are quite similar. The image downsized to 12 MP with nearest neighbor has a bit more bite.

At medium distance the images are similar. The image downsampled with nearest neighbor is most crisp but gritty

At short distance the 24 MP image is sharpest, the nearest neighbor image jaggy and the 12 MPixel image still quite OK.

The enclosed screen shots shows a part of 300PPI scan of prints at actual pixels. This essentially shows that effects of the LR printing pipeline, dithering in printer driver, ink diffusion and adds some modifications in scanning. Processing was identical for all images.


Best regards
Erik
Logged

hjulenissen
Sr. Member
****
Offline Offline

Posts: 1683


« Reply #1 on: March 12, 2013, 02:47:32 AM »
ReplyReply

I don't think that nearest neighbor is an accurate emulation of a lower-resolution camera (if that was your intention)? More like an OLPF-less, perfect lens, non-Bayer, poor micro-lense camera? Bicubic is probably not an accurate emulation of low-res cameras either, but perhaps better (?), and certainly more relevant to how people scale images in their computer.

Adding another dimension to your test could be tweaking sharpening until the 24MP and the bicubic up/down-sampled one are perceived as having similar crispness/bite to the NN. Does that also bring unwanted artifacts to the same level?

-h
« Last Edit: March 12, 2013, 02:54:38 AM by hjulenissen » Logged
RobertJ
Sr. Member
****
Offline Offline

Posts: 597


« Reply #2 on: March 12, 2013, 03:47:35 AM »
ReplyReply

Try PhotoZoom Pro with S-Spline Max.  There's never any jagged edges, no matter how big you enlarge, because it kind of turns the image into a vector.  You'll know when you go too large, because it will start to look like a painting, but I like it so much more than Bicubic, it's insane.  I haven't used Bicubic in years.
Logged
ErikKaffehr
Sr. Member
****
Offline Offline

Posts: 7652


WWW
« Reply #3 on: March 12, 2013, 03:49:54 AM »
ReplyReply

Hi,

I wanted to emulate a non OLPF low res camera on the idea that it would cause fake detail that may give impression of better perceived sharpness.

I wanted to change as few parameters as possible.

Thanks for suggestions.

Best regards
Erik


I don't think that nearest neighbor is an accurate emulation of a lower-resolution camera (if that was your intention)? More like an OLPF-less, perfect lens, non-Bayer, poor micro-lense camera? Bicubic is probably not an accurate emulation of low-res cameras either, but perhaps better (?), and certainly more relevant to how people scale images in their computer.

Adding another dimension to your test could be tweaking sharpening until the 24MP and the bicubic up/down-sampled one are perceived as having similar crispness/bite to the NN. Does that also bring unwanted artifacts to the same level?

-h
Logged

BartvanderWolf
Sr. Member
****
Online Online

Posts: 3765


« Reply #4 on: March 12, 2013, 04:17:18 AM »
ReplyReply

Hi,

I wanted to emulate a non OLPF low res camera on the idea that it would cause fake detail that may give impression of better perceived sharpness.

Hi Erik,

For that purpose, you'd probably get closer to the effect by not simply downsampling with nearest neighbor, but by (first) applying a binning operation with Photoshop's Filter|Pixelate|Mosaic... filter. It is limited to fixed pixel size blocks for binning though, so you may not be able to achieve an exact image size. After this you can use nearest neighbor with much more predictable results.

Anyway, it will probably not change your overall conclusions about how the image quality compares.

Cheers,
Bart
Logged
PhotoEcosse
Sr. Member
****
Offline Offline

Posts: 636



« Reply #5 on: March 12, 2013, 05:03:07 AM »
ReplyReply

You have saddled me up on one of my favourite hobby-horses, Erik - i.e. that the main reason for opting for a good hi-res sensor is not to be able to produce massive prints (Which should, in any case, be viewed from a reasonable distance) but, rather, to obtain much more data for processing.

If I print an A3+ print from the full frame of my D800 or D800E (36Mp) and compare it with a similar print from my D3s (12Mp), then there is no discernible difference in print quality. If I do an A3+ print from 10% of the full frame in each case, then I do start to see a difference. One of the ways in which I can use the 45Mb of lossless-compressed Raw file from the 36Mp sensor is to produce much tighter crops and still have printable resolution. But, of course, the other ways in which I can use all that data in the 45Mb Raw file are much more exciting and liberating than merely heavy cropping.
Logged

************************************
"Reality is an illusion caused by lack of alcohol."
Alternatively, "Life begins at the far end of your comfort zone."
BartvanderWolf
Sr. Member
****
Online Online

Posts: 3765


« Reply #6 on: March 12, 2013, 06:01:59 AM »
ReplyReply

You have saddled me up on one of my favourite hobby-horses, Erik - i.e. that the main reason for opting for a good hi-res sensor is not to be able to produce massive prints (Which should, in any case, be viewed from a reasonable distance) but, rather, to obtain much more data for processing.

Hi,

I don't view the matter as an either/or proposition. Both the upsampling potential is better, not only for the closer inspection of details but also for a higher fidelity in rendering surface structure, and we have more detail which allows more control over local contrast adjustment (and easier retouching of small details).

Quote
If I print an A3+ print from the full frame of my D800 or D800E (36Mp) and compare it with a similar print from my D3s (12Mp), then there is no discernible difference in print quality.

While it won't be a huge difference for some subject matter at that modest size, if you can't see any difference, then something is wrong. Maybe your subject matter doesn't require the realism that additional resolution potentially offers?

Cheers,
Bart
Logged
PhotoEcosse
Sr. Member
****
Offline Offline

Posts: 636



« Reply #7 on: March 12, 2013, 06:08:31 AM »
ReplyReply



While it won't be a huge difference for some subject matter at that modest size, if you can't see any difference, then something is wrong. Maybe your subject matter doesn't require the realism that additional resolution potentially offers?

Cheers,
Bart

Not really Bart.

My caveat, of course, should have been "with my printer" (an Epson R3000).

With a printer like that, printing at 300 dpi, you get as perfect an A3+ print from a 12Mp image as from a 36Mp image. What an amazing number of people fail to understand - including some experienced journalists on some of our most respected magazines - is that dpi and ppi bear no direct relationship to each other. You don't need a 300ppi digital file to get a 300dpi print.
Logged

************************************
"Reality is an illusion caused by lack of alcohol."
Alternatively, "Life begins at the far end of your comfort zone."
jrsforums
Sr. Member
****
Offline Offline

Posts: 744


« Reply #8 on: March 12, 2013, 06:25:32 AM »
ReplyReply

I do not have an R3000.

However, if it is like other Epson's, you will get best results if you feed it 360ppi, not 300ppi....and usually, best results if you properly interpolate to 720ppi.

Logged

John
BartvanderWolf
Sr. Member
****
Online Online

Posts: 3765


« Reply #9 on: March 12, 2013, 07:45:16 AM »
ReplyReply

However, if it is like other Epson's, you will get best results if you feed it 360ppi, not 300ppi....and usually, best results if you properly interpolate to 720ppi.

You're absolutely correct, and it has been discussed in several of the LuLa fora on a number of occasions.

Things start with good shooting technique (high enough shutterspeed, tripod, limited diffraction), then proper (deconvolution) Capture sharpening and image processing, then proper resampling to the printer's native output resolution (requires good resampling algorithms and proper printer driver settings), and finally sharpening the print file data (to compensate for the upsampling and pre-compenate for print medium losses). Larger format output obviously benefits more from such an approach.

Cheers,
Bart
Logged
ErikKaffehr
Sr. Member
****
Offline Offline

Posts: 7652


WWW
« Reply #10 on: March 12, 2013, 07:58:29 AM »
ReplyReply

Hi,

I was using 360 PPI, upscaling by LR 4.3 and LR 4.3 Output sharpening. Note that the images I have posted were scanned from prints at 300 PPI.


Best regards
Erik

You're absolutely correct, and it has been discussed in several of the LuLa fora on a number of occasions.

Things start with good shooting technique (high enough shutterspeed, tripod, limited diffraction), then proper (deconvolution) Capture sharpening and image processing, then proper resampling to the printer's native output resolution (requires good resampling algorithms and proper printer driver settings), and finally sharpening the print file data (to compensate for the upsampling and pre-compenate for print medium losses). Larger format output obviously benefits more from such an approach.

Cheers,
Bart
Logged

mac_paolo
Sr. Member
****
Offline Offline

Posts: 423


« Reply #11 on: March 12, 2013, 07:59:06 AM »
ReplyReply

Not really Bart.

My caveat, of course, should have been "with my printer" (an Epson R3000).

With a printer like that, printing at 300 dpi, you get as perfect an A3+ print from a 12Mp image as from a 36Mp image. What an amazing number of people fail to understand - including some experienced journalists on some of our most respected magazines - is that dpi and ppi bear no direct relationship to each other. You don't need a 300ppi digital file to get a 300dpi print.
I have the very same printer. I usually print on Ilford Gold Fibre Silk.
I recently did a 30x45 cm print.
The shot have been made with a Nikon D300, 50mm f/1.8 D at around f/6.3, MLU and 3 seconds of wait between mirror rising and shot.
Focus adjusted in Live View.
Winter landscape with almost no haze at all, perfectly clear.

I -definitely- see margins of improvements here. Looking closely the detail is not that sharp at all.
Logged
BJL
Sr. Member
****
Offline Offline

Posts: 5163


« Reply #12 on: March 12, 2013, 08:21:42 AM »
ReplyReply

There was a recent discussion on print sizes, some posters argued that little difference can be seen in large prints from 12 MP and 36 MP cameras. I decided to make a small experiment.
...
Those prints would correspond to about 55x81 cm or 21"x31". What I have seen visually.

At distance, say 1.5 m ...

At medium distance ...

At short distance ...
Thanks for this experiment Erik (from one of those "arguing posters"!).

I have one question: can you specify more precisely what you mean by  "medium" and "short" distances?
I am interested in "pixels per viewing distance" as a measure of what our visual systems detects.

Hopefully one is close to the effective full image width of 81cm, since my observations in galleries suggests that this is a common range for the viewing of large prints. (Paintings by the way are typically viewed from further away, further than the "normal" distance of image diagonal length. But most paintings are very low res. by photographic standards!)
Logged
Ellis Vener
Sr. Member
****
Offline Offline

Posts: 1820



WWW
« Reply #13 on: March 12, 2013, 10:18:54 AM »
ReplyReply

A better way to do this is to start with a moderately high (20 to 26 MP) to higher resolution camera  and a high quality zoom lens with sufficent range to let you frame the subject at full frame resolution and then zoom out so that you keep the same framing of the subject and subject to camera distance when you crop the full resolution frame to lower pixel dimensions to emulate lower resolution cameras. The lighting needs to remain constant as well.

It is also important  that that the printing methodology, including  interpolation method and printing resolution remain constant from print to print as well.

These steps will eliminate as many variables as possible except for possible changes in resolution of real world detail caused by the lens being used at different focal lengths.

« Last Edit: March 12, 2013, 10:22:39 AM by Ellis Vener » Logged

Ellis Vener
http://www.ellisvener.com
Creating photographs for advertising, corporate and industrial clients since 1984.
ErikKaffehr
Sr. Member
****
Offline Offline

Posts: 7652


WWW
« Reply #14 on: March 12, 2013, 11:28:25 AM »
ReplyReply

Hi,

Medium distance may be something like 80 cm, about arm's length. Short distance perhaps 25-35 cm. I'm middle aged and have progressive glasses. Being myopic (?), I can see pretty decently at short distances without glasses.

Best regards
Erik



Thanks for this experiment Erik (from one of those "arguing posters"!).

I have one question: can you specify more precisely what you mean by  "medium" and "short" distances?
I am interested in "pixels per viewing distance" as a measure of what our visual systems detects.

Hopefully one is close to the effective full image width of 81cm, since my observations in galleries suggests that this is a common range for the viewing of large prints. (Paintings by the way are typically viewed from further away, further than the "normal" distance of image diagonal length. But most paintings are very low res. by photographic standards!)
Logged

BJL
Sr. Member
****
Offline Offline

Posts: 5163


« Reply #15 on: March 12, 2013, 11:44:25 AM »
ReplyReply

A better way to do this is to start with a moderately high (20 to 26 MP) to higher resolution camera  and a high quality zoom lens with sufficent range to let you frame the subject at full frame resolution and then zoom out so that you keep the same framing of the subject and subject to camera distance when you crop the full resolution frame to lower pixel dimensions to emulate lower resolution cameras.
Yes, I was thinking the same thing.

It is also important  that that the printing methodology ... remain constant from print to print as well.
I admit that I would be happy to avoid the printing issues with a simpler approach: final viewing on-screen of crops to few enough pixels that screen resolution is not a limitation, and with the screen size and viewing distances specified. In fact, I will try this. I propose trying with viewing distances corresponding to something like 2000, 3000, 4000 and 5000 times the effective camera pixel pitch (the width of the screen area occupied by each camera pixel.)


P. S. Erik, thanks for your reply, which arrived as I was writing.
So your medium distance is almost exactly what I asked for, and what I call "close normal" because it seems common in viewing of large prints. I take it that with good downsampling (not NN), the 12MP is barely distinguishable from the 24MP at that close normal range.
Logged
Ellis Vener
Sr. Member
****
Offline Offline

Posts: 1820



WWW
« Reply #16 on: March 12, 2013, 12:40:43 PM »
ReplyReply

Okay to follow up on my earlier post, I've just shot a series wit ha Nikon D800 and AF-S 70-200mm f/4G ED VR. rather than worry about what focal length would match exact standard resolution ( 36, 24, 12 mp , etc.0 I simply shot at the marked focal lengths- 200, 135, 105, 85, and 70mm.  Exposure settings were Auto aperture at ISO 160, aperture set set to f/6.3.

I'll process in Lightroom and make crops to match the framing at 200mm. After identical processing I'll have a lab I trust make matching prints to show what each resolution looks like at 20 x 30 inches.
Logged

Ellis Vener
http://www.ellisvener.com
Creating photographs for advertising, corporate and industrial clients since 1984.
ErikKaffehr
Sr. Member
****
Offline Offline

Posts: 7652


WWW
« Reply #17 on: March 12, 2013, 02:52:48 PM »
ReplyReply

Hi,

The way I print, using the same image from Lightroom would be quite consistent I think.

Best regards
Erik


Yes, I was thinking the same thing.
I admit that I would be happy to avoid the printing issues with a simpler approach: final viewing on-screen of crops to few enough pixels that screen resolution is not a limitation, and with the screen size and viewing distances specified. In fact, I will try this. I propose trying with viewing distances corresponding to something like 2000, 3000, 4000 and 5000 times the effective camera pixel pitch (the width of the screen area occupied by each camera pixel.)


P. S. Erik, thanks for your reply, which arrived as I was writing.
So your medium distance is almost exactly what I asked for, and what I call "close normal" because it seems common in viewing of large prints. I take it that with good downsampling (not NN), the 12MP is barely distinguishable from the 24MP at that close normal range.
Logged

PhotoEcosse
Sr. Member
****
Offline Offline

Posts: 636



« Reply #18 on: March 12, 2013, 05:20:42 PM »
ReplyReply

, if it is like other Epsons, you will get best results if you feed it 360ppi, not 300ppi....and usually, best results if you properly interpolate to 720ppi.



Hadn't heard that before - and will certainly give it a try to see what difference (if any) it makes. (I assume you meant dpi, not ppi, as the ppi of a file "fed to the printer" makes no difference whatsoever. It is the actual dimensions of the file that might make a difference.).

What a lot of folk don't understand is that, even with a bog-standard 3-colours plus black inkjet, 300dpi really means 75dpi for each cartridge.

The big difference between inkjet printing and traditional litho printing is that, in the latter, the dot-screen pitch was important, to the extent that, with a magnifying glass, you could actually see the printed dots. With an inkjet, of course, the dots of each ink spread on the paper and run into each other to produce something approaching a continuous tone., which is why the sensor resolution (my example of my D800 and D3s) makes much less difference to print quality than many people seem to imagine.

.
« Last Edit: March 12, 2013, 05:25:52 PM by PhotoEcosse » Logged

************************************
"Reality is an illusion caused by lack of alcohol."
Alternatively, "Life begins at the far end of your comfort zone."
jrsforums
Sr. Member
****
Offline Offline

Posts: 744


« Reply #19 on: March 12, 2013, 05:36:19 PM »
ReplyReply

While many use them interchangeably, they are different.

PPI = pixels per inch. It is what you feed te printer with.  Each printer, based on driver settings will expect a specific PPI.  If it does not get that, it will interpolate whatever it is fed to get what it wants.  While you may not think this taters, many testers have shown that it does.  Software interpolation is much better Han whatever is done in the driver.

DPI = dots per inch. It is what the printer puts on the paper. This is usually a spec...not necessarily what actually happens....and I really do not much care about the details, just the results.
« Last Edit: March 12, 2013, 05:41:28 PM by jrsforums » Logged

John
Pages: [1] 2 3 »   Top of Page
Print
Jump to:  

Ad
Ad
Ad