G9: Blurring the background

.
A blurred background is useful for isolating and emphasizing the main subject. Extremely blurred backgrounds add a particularly nice effect to portraits. Unfortunately, extreme background blur is not easily obtained using short focal length lenses such as are used with the Canon G series and most other small sensor digital cameras.

The concept for producing a blurred background is simple: use a combination of focal length, aperture and distance such that the background is outside the depth of field. In practice, this means selecting a telephoto lens with a large aperture, placing the camera near the subject and placing the subject away from the background. With the G9, this means zooming the lens to near maximum, forcing the largest possible aperture and placing the subject far from the background.

A previous posting showed how the depth of field varied with the focal length of the G9 at “portrait” distances. Another post showed the maximum aperture for those focal lengths. These two posts illustrate the relationships between zoom (focal length), shooting distance and aperture for the G9.

For this picture of Hannah, I used the G9 at maximum zoom, f4.8 and was about 12 feet from her. The shrubbery was another 12 feet away; the fence (barely discernable) was about 30 feet behind Hannah. These are fairly generous distances; in fact, more separation than is often available. Even so, the background behind Hannah is not particularly blurred (click for a larger image).

Depth-of-field calculations are based on small but noticeable blurs at certain enlargements and viewing distances. For more information, visit and study at DOFMaster. In contrast, the background blur that many photographers want for portraiture is extreme. My own personal favorite is of Wallace; however, it was not made with the G9.

Small sensor cameras like the G9 are great when the goal is to have nearly everything in focus but such cameras are just not the right tool for producing extreme background blur for ‘portrait’ photography.
.

G9: Portrait Distance

.
It seems to me that actual height for a tightly composed head shot is about two feet. How far away from the subject should the camera be to capture this two foot high subject? I suppose the distance can be calculated but frankly it is just easier to tape a yardstick to the wall and take a few shots.

For my G9, these are the camera-to-subject distances for various focal lengths.





The focal lengths in the chart are the same as the focal lengths from the previous posting which gave the maximum aperture for various focal lengths.

The depth of field can be calculated from distance, focal length and aperture. Taking some liberties in rounding off, depth of field for a portrait setting with the G9 is about one foot – give or take a few inches.

This little experiment would seem to indicate that a blurred background would be easily obtained with the Canon G9; however, the opposite is true. In fact, blurred backgrounds are difficult to obtain and extreme blur for portraits and larger scenes is nearly impossible to achieve with the G9. The apparent discrepancy is because the depth of field calculation is intended to predict the onset of blur and not the extreme.

Next: an example of typical G9 background blur for portrait setting
.

G9: Focus Vertically or Horizontally?

.
Although I’ve made several postings about focusing the G9 (sort of cross referenced here), a few days ago I began to think that my G9 was focusing much better when held horizontally (landscape) than when held vertically (portrait). In fact, sometimes it would not focus when turned vertically. Was my G9 broken?

On further review, I’ve come to realize that the small Flexizone frame works best on vertical lines when held horizontally and on horizontal lines when held vertically. Said another way, when shooting in landscape orientation (the “normal” shot), look for a vertical line on which to focus. When shooting a portrait, look for a horizontal line and focus on it.

My problem – when I feared my G9 was broken – was that I was shooting in portrait orientation and the subject lines were vertical. The G9 just would not focus on those vertical lines with the small Flexizone frame. But turning the camera to landscape orientation, it focused immediately.

In much the same way, when using the small Flexizone frame, the G9 struggles to focus on simple horizontal lines when the camera is in landscape orientation. Turned vertically, focusing is routine on those same horizontal lines.

The AiAF matrix of focus points does not have this problem – probably because one of the focus points usually finds an appropriate contrast pattern.

I still prefer the small Flexizone frame but now have a better understanding of those missed focuses. Strange, but true.
.

G9: Aperture variations with zoom

.
The aperture in a camera’s lens is simply the opening through which light passes to strike the sensor or film. Instead of quantifying aperture size directly, for example, speaking of an 8mm opening, we usually speak of the “f-number” or “f-stop”. The f-stop is the relative aperture in comparison to the focal length of the lens; that is, the f-stop is the focal length of the lens divided by the diameter of the aperture. For example, if the focal length is 100mm and the aperture is 25mm diameter then the f-stop is 100 / 25 = 4. Since most people use aperture and f-stop interchangeably, we’d probably say that the aperture was “f4” in that example. For more details, check Wikipedia.

What about f-stops for zoom lenses? If the physical opening remains constant but the focal length changes, then the f-stop changes with the zoom setting. This is the situation with the Canon G9 and most digital cameras having built-in zoom lenses. By tinkering with my G9, I found the zoom points for which the maximum aperture (f-stop) changes and constructed the table below.




The table shows, for example, that the maximum aperture for the wide angle (7.4mm focal length) is f2.8 but near mid-zoom, say 25mm focal length, the maximum aperture is f4. At maximum zoom (44.4mm focal length) the maximum aperture is f4.8.

Notice that, contrary to the simple definition, the maximum f-stop for the G9 is not exactly proportional to the focal length. This apparent contradiction is an indication that the available physical aperture is also changing. If this were not the case, then the f-stop at 44.4mm focal length would be f16.8 instead of f4.8 !

Try this little experiment. Put the G9 (probably any camera) at maximum wide angle, Aperture priority (Av mode) at f8. Oh – you might want to turn off the focus assist light! Point the camera towards yourself and look deep into the lens. Push the shutter button halfway (you don’t have to actually take a picture) and watch carefully for a little bit of movement deep in the assembly. You’ll see a small circular opening become even smaller. Now change the f-stop to f2.8 and try again. You’ll see that the small opening doesn’t change. Now zoom to about mid-telephoto and repeat the tests. Notice that the initial size of the circular opening (the physical aperture) is somewhat larger at mid-telephoto. Zoom all the way in and repeat. Although it becomes more difficult to see deep into the lens, the initial opening becomes even larger at maximum telephoto settings and it changes from a large opening to a smaller one at f8.

Finally, remember that, counter-intuitively, a larger f-number (f-stop) actually means a physically smaller opening (aperture) that lets in less light.

.

Snow Day

.



Snow does not often come to south Louisiana but today we had about three inches of snow at my house. Our roads and systems aren’t built for snow and the disruption was considerable. Our local news featured “Snow Day” and even the national news mentioned snow in Louisiana.

Before the snow melted (yes, it’s all gone now), I grabbed a few shots to record the occasion. I especially liked the shot above. It was taken with my G9 in raw mode at ISO 400, 1/80 sec, f4.8. The sky was still overcast and the colors a bit drab so I punched up the saturation during processing.

.

Flash Links

.
On this blog, some of the more frequently visited postings are about using flash. In particular, one of the most viewed posts is about flash compatibility with the G9. Here are some links to posts on flash at Light Description:

General info on flash for the G9
LCD display in flash mode

Canon external flash
Canon multiple flash

Flash compatibility
Flash sync speed

Wireless flash triggers
Self portrait with external flash and wireless trigger

Attachments and more attachments

There are other posts on flash as well. The easiest way to access specific postings is to use the list of labels on the left hand side of the blog.
.

More Flash Gadgets

.

.
Flash photography can be difficult and photographers seem to be always seeking magical gadgets to solve flash problems. I’m at least as guilty as anyone (see previous posting on Flash with Gadgets) and recently acquired a few new gadgets. Above L-R front are Lumiquest Mini-Softbox, Lumiquest Snoot, a Sto-Fen diffuser on top of a red gel and a Lumiquest Softbox III. Across the back row are a Lumiquest Pocket Bouncer fixed to a Nikon SB-28 with Cactus wireless trigger, the Cactus transmitter and a Honl Grid. Surely one of these gadgets will help me!

I decided on a simple setup to get a feeling for the relative effects of these gadgets: a model in front of a plain backdrop (white wall) in a large studio (garage) using each gadget individually. I’m not saying that I used each gadget optimally or even as recommended; I just wanted to practice and see what would happen with each one.

Instead of my previous models, I selected Hannah this time even though she is known to be independent and undisciplined. (Hannah has been put in “time out” by my granddaughter on several occasions.) Fortunately, except for needing help from duct tape (which I understand is not exactly unfamiliar to models), Hannah did well. She was patient and quiet – unlike Elmo who tends to speak out.

Hannah was placed six feet in front of a white wall and the camera another six feet in front of Hannah. The flash stand was placed six feet from Hannah and roughly two feet to the right of the camera. The flash (Vivitar 285HV with ¼ CTO gel) was about a foot above Hannah’s eyes and, initially, pointed right at Hannah. The camera, my G9 fitted with Cactus transmitter and tripod mounted, was roughly the same height as Hannah’s eyes – perhaps half a foot higher. The studio lighting was dimmed (closed the garage door) so that the effects of flash and gadgets would not be masked by the ambient light.

I dialed in the estimated (taking a guess) exposure in manual mode: 1/250 sec, f4.5, ISO 80, Vivitar at ¼ power, focused using the small Flexizone frame and switched to manual focus so that all photos would be the same. Setting Custom Mode 1 assured that those settings would be easily recovered if necessary. Time for the test shots.

Of course, that first shot was way over exposed. I fiddled with the aperture and tried a few more shots before settling on f6.3. I then reset Custom Mode 1 and, apparently, “adjusted” the manual focus setting such that nearly every shot taken is slightly out of focus. Experience is remembering that you’ve made the same mistake previously!


The images above offer a relative comparison of direct flash and the various gadgets. The idea was to place the gadgets on the Vivitar and take shots without any adjustment. But of course, exposure had to be adjusted in each case. The Vivitar had to be cranked to full power and the aperture opened to f4.5 for the softboxes. The snoot and grid shots were taken at ¼ power and f5.6.

The two softboxes seemed to change the color a bit but the effect was pleasing – in my view anyway. All the same, next time I won’t add a ¼ CTO gel for the “warming” effect. The snoot is a bit hotter than I thought it would be. The grid produces a harsher shadow than I expected. I liked the large Softbox III but the Mini-Softbox produces nearly the same effect and fits into my kit (well, just barely now – I need a larger bag!).




Everyone knows – or should know – about using bounce flash. The above three shots were taken with variations of bounce flash. The first shot was made by simply pointing the Vivitar 285 nearly straight up at the 12 foot ceiling height; however, the flash power had to be increased from ¼ to full power, the ISO changed to 200 instead of 80 and the aperture opened to f5.6 instead of f6.3. With the change to bounce flash, the harsh shadow of direct flash has completely disappeared; however, Hannah’s face needs a bit more light. For the middle picture, I held a white box behind the flash to reflect light forward; notice that a small shadow also appears. The Pocket Bouncer causes an even more distinct shadow and requires about the same amount of flash power and aperture.

Although my G9 was set for RAW + JPEG, I couldn’t resist tweaking the selected shots from RAW. All were processed in Adobe Camera RAW 5.2 using “Flash” as the white balance and auto exposure adjustments. Yes, some of my exposures were off a bit.

My conclusions? Well, I’ll tinker a bit more and report. The soft boxes, snoot and grid aren’t really meant for main, direct light. I’d already concluded to only use the Pocket Bouncer outside or with very high ceilings. I just needed to play -- ah, practice .

(The individual portrait of Hanna was made using a red gel inside a Stofen diffuser fitted to a flash pointed at the wall. The main flash was bounced at ½ power and the large softbox was placed on a flash at 1/16 power about two feet from Hanna. She seems to like it -- hasn’t complained anyway.)
.