Auto focus is confusing?

I while back, I looked through the archives at Cambrigde In Colour, and found this article about auto focus technologies. The author tried to describe how the autofocus of dSLR’s are working, when in fact he was describing the contrast based autofocus most commonly used in compact cameras. Later I found Tim Jacksons D70 site, where he also describes a focus system that I don’t really think exists in any camera on the market. After looking around on the net, it occurred to me that there are many attempts to explain autofocus out there that are just plain wrong. It seems like the phase dection autofocus system used by all SLR cameras, both digital and film, is a bit hard to grasp, so most authors explains the conseptually more easy contrast based autofocus instead. Or maybe they just fail to see that there is a difference.

Contrast based auto focus

To start with contrast based auto focus. This type of system is most commonly used in compact cameras, and on some dSLR’s when the Live View function is used. It uses the actual imaging sensor to determine focus, and here is how it works:

1. It shoots a series of images at different focus settings.
2. For each of these images, it will determine how  much contrast there is in the designated focus areas.
3. It selects the image with the most contrast, and reset the focus to the value it used to capture this image.

This process is relatively slow since the camera must take a large number of shots and process them before finding the focus. On many cameras you can observe this if you look at the lens while trying to focus. It will cycle focus from one endpoint to the other before it snaps back to somewhere in between which hopefully is the right focus setting.

A major drawback with this kind of focus is that the camera has no way of predicting if the focus is in front of, or back of the current setting. It must just try in one direction and hope that it is right, and if not, it must go the other way.

Phase detection auto focus

This type of auto focus works kind of like the good old split prism in the days of manual focus. The focus system will take a part of the light coming from one side of lens and project it on one sensor, and a part of the light coming from the opposite side of the lens and project it on another sensor. These sensors are dedicated AF sensors, and are usually formed as narrow strips. When the image is in focus, the two images are perfectly aligned, but they will get more and more misaligned as the system gets more and more out of focus. Just like the split prism, all the AF system has to do is to rotate the focus until the two images are aligned.

A major benefit with this type of AF is that it can predict which way the focus shall be turned. Not only that, but it can even predict how far it shall be turned. This is accomplished by detecting to which side the misalignment is, and how much the images are off.

Here is a short drill down how it works:

1. The AF sensor detects the direction and amount of misalignment.
2. Based on the information about the lens, if predicts how far the focus must be turned.
3. The focus is then adjusted accordingly (focus is not measured when this happens since the focus changes too fast).
4. When the focus motor stops, a new measurement is performed. If the focus is not perfect the drill (1 through 3) is performed over again, but now the adjustments are much smaller, and therefor more precise.
5. This goes on until the focus is good, or the system times out because it is unable to find focus.

The big misconception is that this system optimizes contrast, when in fact it correlates two one-dimensional images. When the contrast based system relies on a sharp edge to focus, this system only need some sort of recognizable pattern. In theory, it should be able to focus on a smooth gradient that goes along the focus sensor.

Here is an illustration by Canon (go to the bottom of the page).

And here is the Wikipedia article.

So, this is an overview of my understanding of these systems. If anyone more knowledgeable than me finds this to be wrong, please write me a comment.

rant about dynamic range

TOP just posted a long article on dynamic range, trying to explain what it is all about. It’s an ok article, explaining the facts fairly well, but there is one thing that still puzzles me, and that is the comparison between film and digital. I remember from before anyone knew about digital photography, in fact I can do better than remember, I looked up in a couple of the old books I have in my shelves. It seems like the consensus in that age was that color transparency film could capture about 5 stops of range, which is what Mike is saying in his article. Then my old books tells me that color negative film is usually capable of capturing 7 stops, while the best B&W negative films are capable of up to 9 stops. As far as I know, this was the truth back then.

Back to present day, and modern dSLR’s. Most of the testing I have seen shows an ability to capture around 10 to 11,5 stops for most current cameras. I haven’t done extensive testing on my own Canon 5D, but I did some quick and dirty tests a while back which confirmed that I could capture least get 10 stops. Looking at the DxOmark tests, I see that my 5D scored 11.1 at ISO100, and didn’t go below 10 until the ISO was crancket up to more than 800. Of course there is very few bits allocated to the shadow range, so the number of levels in these dark f-stops aren’t that many, but there should be somewhere around 5-8 levels even at the lowest f-stop (due to the 12 bit conversion).

So my question is: When did film become capable of capturing more than 10-11 stops. After all, there hasn’t really been any improvements after films like T-Max, and I think that was the one with around 9 stops about 20 years ago. Did they lie to us back then?? When reading Mike’s article, not only is film better, it is much better. From that I would say that it must be at least 1 stop better, maybe 2. Can anyone back with hard facts and testing that film is capable of substantially more than 1o stops?? I’m just asking. Not that it matters all that much since I will not start using film anyway, but it’s still interesting.

local corrections, Lightroom or Photoshop

It’s been a while since LR 2 came about, and I’m increasingly fond of the new gradient and brush tool for local correction, but it’s definitely not for everything. Like with everything that is new, I have used this tool maybe a little too much lately, making it a bit difficult for myself. The thing is, even if it is local adjustments, they are not as precise as in Photoshop. First of all, unless you have one hell of a computer, the time you wait for the spinning beach ball (hour glass in Windows) will limit how much tweaking you do to a mask. My computer, an 8 core mac pro, really kick ass, so this isn’t that much of an issue for me, but I have tried to use it on a 2.16GHz iMac and it was not that fun. The second thing that limits the precision is that you have to paint the mask by hand. Ok, you can get some help, but the vast number of methods for creating masks in Photoshop is unbeatable. As a side note, the lack of precision when creating masks was my main issue against LightZone. This was even more annoying than LightRoom since the masks were made by creating vector drawings. Anyway, it seems like it is very important to not get stuck in LightRoom  when Photoshop can do a much better job, even now when LightRoom has become this versatile.

lots of subtle changes

This was so obvious when it suddenly dawned on me. The reason why I mostly end up with very few layers in my Photoshop editing is that I tend to try and make all the curve adjustment in one layer. One layer for dodging and burning, one layer for correcting colors and so on. So figuratively speaking, I wanted to paint the entire scene without lifting the brush from the canvas.

When I analyzed Diane Varners example a bit closer, I see her doing many small adjustments, and when I looked back at some of my more successful images, that is also what I have done. While doing this, you should pay attention to whether you start to create layers that reverse the effect of a previous layer instead of removing the previous. When I have done this, I have made my worst failures.

So, adding many small and subtle adjustments seems like an easier approach than doing it all in one stroke. And why not, this is what you do if you draw or paint.

post processing

I’ve visited Diane Varners picture blog, Daily Walks from time to time, and just now Paul Lester had a link to her in one of his posts. Since it was a couple of months since I dropped by last, I clicked on the link, and I was rewarded by one of her beautiful images. 

She definitely has a distinct look to her photographs that is created through post processing, and she has made a couple of short tutorials that describes what she does. You’ll find them here. Well worth reading even if you don’t like the result all that much. I like much of what she does, but not all. When I look at her example, and compare it with my own effort, I rarely use that many layers. Usually it is only my failed attempts that has that many layers. I seldom see the use of that many correction. 

That might be because I don’t have the need for it in my photos, or it may be that I don’t have the ability to see what I should do. Probably a combination, but I wonder what the mix is.

primes and walking-zoom

Just read an article by Mike J about his affection for 40mm lenses.

Nothing wrong with that article, it just got me thinking. I don’t get that affection some people have for that one prime lens. Well, that’s not entirely true, I do get that affection, what I don’t get is how people who don’t really share that affection, but wish they were, are arguing for it. I’m not saying Mike J is one of them, again, his article just spurred my thought. It usually takes the form of; “My hero uses that lens (or that film format, or whatever…), so that must be good. I don’t really know why he/she think that’s good, so I come up with something myself. “

Back to the heading of this post, and my second thought of the day on this topic.

I have had my fair share of primes, and I still own a 50mm and a 100mm macro that I only use for special purposes. The 50mm when I want something small, and the 100mm for, well you know, macro. The thing is, I’m not that fond of the primes. It is always a limiting factor when it comes to the crop I want. A prime advocate would then just say, you just step back (or forward), and then you get both a zoom and exercise. Can’t argue against the exercise part, but the zoom part I strongly disagree with. I love my zoom. With them I can first pick my viewing point, and thus select the right perspective. Then I zoom in and out until I get the right crop. If I had to move back or fore, the perspective would change, and I would have made a different image. With a prime, I would have to select. Either settle with the crop I happen to get, selecting the perspective, or select the crop and settle with whatever perspective I happen to get. 

This is my take at walking vs. zooming.

• Your feet:  A perspective changing device. Excellent for changing the relationship between objects in front of you.
• Zoom. A cropping device.

Two very different, but complementing tools.

workflow, what to do with finished files

I’ve been pondering for a while what to do with finished files. That is pictures that I have processed in photoshop and that I am now happy with. So far I have saved them with all the layers intact. I do non destructive processing on layers, which is very valuable when the work is in progress, but it takes up a lot of space, and makes the psd files harder to handle. Not that space in itself is that important, hard disk is very cheap these days, but the handling when printing is a drag. But there is an even more important reason. Some times when I look at a picture that was previously marked as finshed, I find something that need to be corrected. Then I start to remove layers, change layers and add new ones. This sound good, but in reality it is not. All the history that lies in these layers prevent me from looking with fresh eyes. I remember the decisions I made the last time. It also prevents me from applying new techniques and more refined craftmanship that I have acquired since the last time. It may even be that I should make some adjustments in the RAW converter, which even may be a newer one. It’s really better to start with a fresh copy.

I have a color scheme in Lightroom that I use to keep track of work in progress. A raw file marked with red is found worthy of post processing. Once I creeate a psd file of it, it gets labeled yellow to indicate work in progress, and once I declare it finished it becomes green. Last night I went through the green ones and flattened all the layers within them.