Archive for May, 2010

Haven’t been shooting much lately but here’s some pics:

Created a camera obscura in my room.

Was bored so I went outside to take a self portrait. You can see how bored I was from my blank stare :p

Cat being lazy as usual.

Till next time…

Sharpness – it’s something every photographer cares deeply about whether they admit it or not. For some photographers sharpness is inseparable from a “good” picture. A soft photo is, in their minds, a “bad” photo. The problem is, most photographers don’t know what “sharpness” really is or what they really mean when using that word. In photography, sharpness is made up of two distinct concepts – resolution and acutance.

Keeping with tradition, I’ll keep the technical stuff to a minimum which means this will be a brief overview. In optics, resolution/”sharpness” is measured in line pairs per millimeter (lp/mm). What that means is the ability of the human eye or other “sensor” to perceive the number of lines in a space of one millimeter. The best color films can resolve up to ~100 lp/mm. The best lenses in the world can barely resolve up to 100 lp/mm (this might have changed since the last time I looked up the data). However, none of this really matters because the human eye can only resolve up to about 10 lp/mm. And that’s under OPTIMUM conditions and assuming you have perfect eye sight. Under anything other than the best conditions, it’s closer to about 5-7 lp/mm.

That doesn’t mean that having higher resolution isn’t better. To most people, having 5 lp/mm of resolution would be considered to be really sharp. That’s assuming other things (for this discussion it’s acutance) don’t cause a loss of “sharpness” that is visible to the human eye. In the beginning, I said that sharpness is made up of resolution and acutance. What do these terms mean?

Resolution is something most photographers understand. It’s basically the ability to resolve detail in a scene that we’re photographing. Acutance on the other hand can be thought of as the “contrast” between edges. Basically, the brightness of one edge to the brightness of another. Sharpening or applying a high pass filters in photo editing makes the brightness between edges greater. It is these two concepts that together make up what we call sharpness.

In Photoshop, The GIMP, or any other editing program, there is a filter called Unsharp Mask. This filter is used to sharpen a photo. You’re probably thinking why is it called Unsharp Mask. The reason is the process it uses to sharpen the photo by first blurring the image. Unsharp Mask increases acutance. Same goes with the High Pass filter method of sharpening.

No discussion of sharpening will be complete without discussing viewing distance and viewing medium. I’ll tackle the latter first. Viewing an image on a computer screen has different requirements than viewing a print. A classic example that I see all the time is looking a billboard. From a “normal” distance, the photo on the billboard looks sharp and you can see detail in the scene being displayed. Walk up to within a few inches and all you see is big ass dots of ink. This concept, magnification, is why medium format and large format seems sharper in general compared to 35mm format or smaller formats.

Let’s think about this. Viewing a photo at 100% on a computer screen may looks soft but when viewed at it’s output size (a 12MP camera has a 100% size of about 30×20 printed – I could be wrong about this and I’m too lazy to check) it will look perfectly sharp. People who look at their pictures and complain that it’s soft at 100% are basically walking up to a foot of a billboard and complaining that you can’t see the details. /* Before I continue, I actually finished writing the article but reformatted my laptop and forgot to back up this article first. I’m really pissed right now. */ An image that is soft at 100% on a computer screen can still be sharp depending on the output medium. For example, an image that is soft at 100% when viewed on the computer may be “sharp enough” on a small print (say 4×6) or displayed at low resolution on the web. There are photos that I have posted on the blog that were soft viewed at 100% but appear to be sharp on the web displayed at 800×533.

Awhile ago, I did an engagement shoot for my friend’s brother. Some of the pictures were soft when viewed on the computer. I gave these “soft” pictures to them because when printed at 5×7 (which is the size they wanted) looked sharp enough when viewed from a normal distance. If I printed those any bigger, then you’ll see that they were soft. As a side note, the best color photo paper can only resolve up to ~80-85 lp/mm. It’s important for photographers to remember and know what the final output will be. If I was printing a billboard, anything less than tack sharp would be thrown out. Of course, if the final output will be the web I wouldn’t care too much if it looked soft viewed at 100% on the computer. /* I actually had something much better written but I can’t remember what I wrote and I’m pissed so you get to read this crap instead of the beautiful poetry I wrote earlier */

So that’s a brief explanation of sharpness as it pertains to photography. Very brief but I hope it helps. If you’re interested in the topic, there are plenty of educational photography related sites that explain the concepts in great detail. Or if you’re a CPS member, the Canon EF Lens Works III book is a great resource as well. I’m not sure what the reaction to this post will be like. If the feedback is positive, I’ll consider writing similar articles on other topics. So let me know in the comments what you think!

Till next time…
Fun Fact #1: The difference between the “look” of Japanese glass and German glass is due to the way each manufacturer handles micro-contrast and resolution in their lens design. Leica traditionally favor resolution while Canon favors micro-contrast. In optics, you have to sacrifice more of one to get more of the other.

Fun Fact #2: The only “real” German manufacturer of photographic lenses is Leica. Carl Zeiss AG has moved quite a bit of their manufacturing to Japan. So technically speaking, if you’re buying Zeiss lenses you’re buying Japanese glass :p

Spent yesterday at the Huntington Library with my grandparents. Here’s some pictures:

The Huntington Art Gallery

I don't know why they put a mirror under this table but they did.

Inside the Huntington Art Gallery.

Ducklings following their mom.

US flag as seen from the Virginia Steele Scott Galleries of American Art.

Rain forest exhibit.

Rainbow patterns in a little tunnel.

Rainbows on my sister's hands.

Till next time…

Another interesting article from OSNews.com today with regards to h.264: http://www.osnews.com/story/23236/Why_Our_Civilization_s_Video_Art_and_Culture_is_Threatened_by_the_MPEG-LA

As with previous posts, the article written by the talented and knowledgeable writers at OSNews.com does a MUCH better job at explaining things than I can.

As explained before, you’re only allowed to use the h.264 codec for “personal and non-commercial” uses without paying royalties to MPEG-LA and/or other patent holders. Unfortunately, most HD cameras and camcorders shoot in h.264 format. So just by using these cameras, you’re not allowed to use the footage for anything other than personal and “non-commercial” use. The author shows proof of this from this Canon 5D Mark II manual which I will show from my own manual here:

MPEG-4 licensing info from my 5D Mark II instruction manual. H.264 is in Part 10 of the MPEG-4 specification (http://en.wikipedia.org/wiki/H.264/MPEG-4_AVC). Sorry for the shitty quality, this was taken on my phone.

A link to the PDF version of the 5D Mark II’s instruction manual is provided in the article.

The reason the word non-commercial is in quotes is the definition that MPEG-LA gives for what it considers to be non-commercial. According to their definition, non-commercial means that the video can’t be displayed on, with, by, anything that bring “income” to the owners, makers, and distributors of the video. What this means is that if you’re running a site supported by ads, you’re not allowed to use the video that is shot in, encoded, and/or distributed in h.264 without paying royalties to MPEG-LA. Keep in mind that Youtube uses the open source x264 encoder for encoding h.264 content and has ads on the site. The former point will be discussed later. According to the definition of non-commercial use, Youtube needs to pay royalties. MPEG-LA has so far been using what the article calls common sense and have extended the royalty free use of h.264 for online broadcasting.

Now assuming you’re going to use h.264 to deliver content to your friends and family for purely personal use (e.g. a video of Fido running into a wall). And you post the video to Youtube or Vimeo (which also uses the x264 encoder), you and Youtube/Vimeo needs to pay royalties. Why?! The reason is the x264 encoder devs didn’t pay to license the technology. For that reason, it is not considered an “authorized” encoder. I don’t feel comfortable to go any further. As stated in previous posts, I haven’t done enough research, fact checking, etc. to back up anything further. So read the article. I don’t want to be spreading the wrong info.

EDIT: Most of the articles I link to with regards to this issue is from OSNews.com. Here is an article about their “bias”: http://www.osnews.com/story/23235/A_Statement_Regarding_OSNews_Bias

Till next time…