The GetDPI Photography Forum

Great to see you here. Join our insightful photographic forum today and start tapping into a huge wealth of photographic knowledge. Completing our simple registration process will allow you to gain access to exclusive content, add your own topics and posts, share your work and connect with other members through your own private inbox! And don’t forget to say hi!

Print Resolution & COC

Shashin

Well-known member
It might nice to point out, at least with my Epson printer, it can't print at 300dpi. It can print at 720dpi, 1440dpi, and 2880dpi.
 

tsjanik

Well-known member
It might nice to point out, at least with my Epson printer, it can't print at 300dpi. It can print at 720dpi, 1440dpi, and 2880dpi.
Hi Shashin:

Just so there is no confusion, I'm referring to ppi not dpi. I print at 360 or 720 ppi at the highest resolution of the printer which is 2880 dpi.
 

tashley

Subscriber Member
My Canon iPF6300 allows a max specification of 600dpi in the driver but the brochure spec is 2400 x 1200dpi. I assume the '600' figure relates to the notional input resolution and equates to the same number of ppi... which may be a daft assumption knowing me...
 

tsjanik

Well-known member
Tim:

A file of 600 pixels per inch printed at 2400x1200 drops per inch by the printer would result in (2400/600) x (1200/600) or 8 drops per pixel on the print; i.e., the printer could use 8 drops of ink for each pixel in whatever combination of inks gives the desired color.
I apologize if you already know this.

Tom
 

torger

Active member
The rules I go by is to try print at 200 ppi or more if I have the resolution to do it. This makes a nice looking print on any distance, without risk of even subtle jaggies effects (which I dislike very much).

Concerning DoF I try to make the best compromise at the shooting occasion, and do not think about print sizes at all. The traditional formulas does not help me much at all, I'm with Tim on this.

For the goal of optimizing DoF I think setting the CoC = Airy disk is a good target, ie variable with the aperture (unfortunately no apps support this(?), but I'm old-fashioned and use printed tables instead). The rationale is that for deep DoF photography aperture will always be so small that resolution is a little diffraction limited, and with smaller aperture you get even lower resolution through more diffraction and then CoC should be larger as it will take more to make a visible deviation from the plane of focus. In other words, everything within this DoF will be about equally sharp at the pixel level and will sharpen well with the same deconvolution radius.

In addition I know in the back of my head I know how much smaller apertures hurts (diffraction) so I can think about if it's worth to really stop down to f/22 or f/32 or if I should tilt or focus stack, or let something be a little bit out of focus.

Attached examples. The first row shows the difference between plane of focus and the edge of the DoF at 100% pixel peep of my 33 megapixel digital back shot at f/11, with a traditional CoC scaled to the sensor size (48x36mm) which is 42um. The second row shows with the CoC = airy disk diameter for f/11. Slightly fuzzier at the edge, but very little. I should have had a f/22 or f/32 shot too to show the effect when the plane of focus is softened by diffraction and thus we can have larger CoC and still have an edge as sharp as the focus plane.

Have DoF tables that actually relate to how sharp the file will be at 100% is a much better tool for me to use in order to make the best decisions out in the field, rather than the traditional related to viewing conditions (=viewer behaviour). (And which makes you think you need nothing but a 8 megapixel camera and you can shoot at f/32 all the time.)

Depending on the scene I can then adapt. Is it important that the foreground is really critically sharp? Perhaps just large low contrast structures there. Is it better to let it be a bit fuzzy and render the distant horizon at f/11 instead of f/22 that would make the whole scene equally sharp etc...
 
wow - there's a lot of discussion going on while i was offline ...

i did some test shots with my SK43mm today and have to have a close look at them now, but i'm with you tashley, that i just want to know what i can expect and then return to fun again ;-)

i'm printing on a canon ipf5000 at 300ppi and i can stick my nose as close as i want and there's lots of detail - given that the image is properly focused.
and to know which DOF i can get without loosing this viewing/peeping experience is the goal i aim for.

i'm also with you graham to aim at at least twice the pixel pitch to get optimum results - would mean in my case about 14microns.

but i have to do some test prints to find out if that holds true, or if i can stretch that limit a littlebit...

great forum, i love being part of it - thank you guys!
 

tashley

Subscriber Member
Torger, that's really interesting. I particularly agree about large structure detail being enough in certain parts of the print. If you have an iPhone, you might really find the apps "OptimumCS-Pro" and "TrueDoF-Pro Depth of Field Calculator"very useful: they do exactly what you want, really exactly - allowing you to make your own global parameters for either modified COC (taking diffraction into account in the calculation) or print size and resolution. They are the best things I have found and are remarkably accurate in my initial assessment, allowing me to place to DOF exactly where I want, rather than merely calculate it...
 

torger

Active member
Thanks for the tips, maybe I'll look into those apps.

I'm quite attached to my printed laminated tables though, and just started to use a Leitz Fokos vintage rangefinder, so it's an all analog process for me apart from the digital back, I kind of like that :).

I typically use the rangefinder to check distance to the closest object and to find a suitable object to focus at (if I need hyperfocal technique, which actually is quite rare for my shooting style), and then have a printed table with hyperfocal distances for different apertures, and hinge distance for different tilts and "dof height at infinity" for different tilts/apertures, and a coarse DoF table for far edge distances up to 100 meters, all this fits on a two-sided small card (double credit card size) for my 6 lenses.

It's coarser than laser distance meters and a DoF app, but as a view camera user focus I deal with limited focus precision anyway and I'd say for f/11 and smaller apertures this way to work is adequate (DoF large enough to mask precision error). It's not for everyone though. Getting a good distance reading from the range finder is as tricky as focusing precisely with the ground glass. Good eyes and some training is required. Sometimes a HPF ring and a laser distance meter would be nice to have, but for me it's not just worth the extra money which is a lot with 6 tiltable lenses (plus that my shooting style fits a view camera well).
 
torger, i sometimes wish i would have a groundglass that gives me a clear view to focus so i could trust my eyes rather than mathematics. it's a littlebit like flying in the mist only trusting your instruments. (i'm not a pilot though)
but sadly the rm3di is not meant to work this way.
so i have to get familiar with it's physics till i can trust my instruments.

my test shots yesterday showed me that a CoC of 20-25microns will be my safe zone.
i searched for a good place to check different distances and found a parking spot at a sports stadium next to my home.

lens: SK43mm5.6

the app "TrueDoF-Pro" was my buddy. i inserted different blur spot diameters (CoC incl. diffraction as tashley mentioned above): 14 microns, 20microns, 25microns, 28microns, 40microns.
i did a shot at each of these settings set the calculator for hyperfocal distance (f5.6).
i only adjusted the calculated focus distance at each shot and noted the focus distance, as well as the near and far limit of the DoF.

back at home i looked at the results and compared the actual sharpness with the calculation. which setting for BSD would match the real sharpness of the picture?

i ended up with 20microns for "perfect" up to 30microns for "good" sharpness in the predicted areas. (100% view on my monitor)

see also attached one final shot with maximum DoF at f11 2/3 (accepted 28microns and some little diffraction)
focus distance: 7m (100% crop)
near limit: 3,5m (100%crop)
far limit: inf (100%crop)
no sharpening as well as no lcc applied - right out of the box.

i believe i'd be able to print this one at 300ppi 400x600mm with satisfaction...
if i want to go for perfect sharpness 20microns would do the job.

but what if my iphone doesn't hold up battery long enough for a day of shooting? but that's another story :D

you are right torger, laminated lookup tables are never wasteful ;)
 
Last edited:
Top