# OMG...tell me thiis guy is wrong about micro four thirds

Started 6 months ago | Discussion thread
 Like? 1
 Why Print/Display Viewing Size and Viewing Distance References Matter In reply to Moti, 6 months ago

Moti wrote: Looks very exciting but why would I want to remember any of that when I have a great app on my smartphone that makes all these calculations and much more in a second.

Detail Man wrote:

Does your "great application" on your smartphone tell you what the reference print/display-size, viewing-distance, and visual-acuity assumed are when it outputs numbers.

No it doesn't and I don't usually need it. Of course I can always use the rule of thumb where roughly, the viewing distance should be 1.5 to 2 times the diagonal length and then calculate the minimum PPI needed for printing as follows:

minimum ppi = 3438/Viewing Distance
where 3438, a constant for human vision, was derived as follows:
1/ppi = 2 x Viewing Distance x tan(0.000290888/2)
1/ppi = Viewing Distance x tan(0.000290888)
ppi = 3438/Viewing Distance
where 0.000290888 radians (1 arc minute) is known as the 'visual acuity angle' and represents how much resolution a human can see.

Well, in practice, I don't do that. I usually do not print larger than 40" , I just use an image resolution for printing between 150 and 300ppi depending on the complexity of the image and what I intend to use it for. That is my rule of thumb and it never failed me.

When I exhibit, unfortunately I have no control on the distance that people would look at my photos but until now, I haven't got any complain about them not looking good or not having enough pixels.

However, there is one case when I do calculate the minimum ppi and that is when I have very small files, usually due to excessive crop and I need to make sure that they'll print ok at rge size I need but then I also use a calculator

http://www.pointsinfocus.com/tools/minimum-resolution-calculator/

That formula (above) addresses what human vision can (possibly, reportedly in some cases) discern in units of pixels/inch (ppi). It is premised on an ability to see 60 cycles/degree (CPD), which is quite high, indeed. Most studies that I have seen come up with 30 CPD at the very most, and findings on the order of 10-20 CPD (depending on viewing conditions, and age) are not unusual.

Using that assumed 60 CPD figure, for an 8"x10" viewing size at 25cm, around 349 ppi is implied. When viewed (by your statement above) at a viewing-distance equal to 1.5x the diagonal dimension of an 8"x10" (19.21 inches), around 179 ppi is implied.

The above calculations are (most surely) erring on the conservative side (where it comes to the matter of necessary display/print resolution itself). Lots of resolution is great, but it does not mean much of anything if the diameter of the Circle of Confusion used by a DOF calculating application is based on assumptions which correspond to lower print/display resolutions.

.

The reason that I inquired as to what print/display-size, viewing-distance, and visual-acuity assumed are when your (or any other) DOF calculator application outputs numbers has to do with the fact that people use DOF to make what they believe to be reliable judgments relating to the Near and Far Focus Distances of the Depth of Field of Focus, and the Hyperfocal Distance (in object-space) - and print/display-size, viewing-distance, and visual-acuity directly impact the value of those numerical values.

A 15 Micron (0.015mm) COC diameter is commonly used for M43 image-sensors. More info here:

http://forums.dpreview.com/forums/post/40717560

The enlargement-factor between a 13mm height M43 image-sensor and an 8 inch size print is around 15.631x. Using a 15 Micron COC diameter, that transaltes to a (final-image) COC diameter of 0.235mm - around a 108 ppi print/display resolution (just a bit higher than many flat-screen LCD displays provide).

.

Notes: In the most ideal (the very best-case of fantasy worlds ) interpretation (for an 8 inch size print), by using the (height or width, not the diagonal dimension) of a 7.5 Micron 2x2 photo-site array (of 3.75 Micron photo-sites of typical 16 Mpixel M43 image-sensors), the (final-image) COC diameter would be 0.117mm - around a 216 ppi print/display resolution.

In order to ever be able to make any actual use of a (what is a theoretical maximum) 216 ppi print/display resolution, the (actual) Depth of Field would only be one-half of the depth that your (likely 15 Micron COC diameter referenced) DOF calculator application (just may) be reporting to you, and the associated Hyperfocal Distance would be double the distance that your application (just may) be reporting to you.

.

So, if your DOF calculator application uses the commonly used 15 Micron COC diameter in it's calculations, then the (object-space) Depth of Field that it tells you is not even (quite) up to the task for display on a (100 ppi resolution) flat-screen LCD display (and this is in the case of a print/display dimension of 8 inches).

Increasing the enlargement-factor up to the 40 inch size that you mention (below in your quoted text), the relevant print/display resolution decreases to a bit less than 22 ppi. In terms of the number of visually discernible line-pairs (consisting of a dark line and a light line), that would be approximately 1/3 of the number of discrete points, or around 7 line-pairs per inch.

.

Judgments of (object-space) Hyperfocal Distance and Depth of Field of Focus made from the output of DOF calculators (such as the DOF Master DOF Calculator, and perhaps your application used) that use the common 15 Micron COC diameter for M43 image-sensors are not even quite adequate for many flat-screen LCD displays (with around 100 ppi resolution for an 8 inch print/display dimension), and provide only a false and inaccurate sense of confidence where it comes to larger print/display dimensions, any/or any higher imagined print/display resolutions.

Thus, the COC diameters commonly used in DOF and Hyperfocal Distance calculators limit the meaningful print/display resolution to somewhat less than typical flat-screen display resolution (at around 100 ppi) for an 8 inch print/display dimension.

Only by lowering estimates of the visual acuity of the viewer (in CPD), and/or by increasing the estimated viewing-distance of the viewer can the numerical results (for DOF and Hyperfocal Distance) output from such DOF calculator application (using a 15 Micron COC diameter) be considered valid.

DM...

Edited 6 months ago by Detail Man
Complain
Post ()
Keyboard shortcuts: