Sullivan Bay from the top of Bartolome

Sullivan Bay from the top of Bartolome - Canon 1Ds3, 24-105 @ 47mm, f/8, 1/15s, ISO 100

On the fifth day of our photographic trip in the Galapagos we arose early in the morning so that we could climb to the top of Bartolome Island in time to photograph the surrounding area shortly after sunrise. I struggled with this scene for a while, trying all kinds of compositions. In the end, left it unprocessed for the last couple of years not really finding much inspiration to work on it until this morning. I remember that I used a soft graduated neutral density filter in the field to get the sky under control but I still had to do some local tonal working to get the overall image to look right.

Lonely House

Lonely House - M9, 50 Summilux, ~f/6.7, 1/500s, ISO 160

I drove past this building several times last year as I was going from Mono Lake to Bishop.  Finally, one morning I stopped and tried to make an image, though it was a challenge due to the presence of a fence.  The image looks like it could be an infrared capture, but the sky is so dark because of the pure blue skies you can get when you gain altitude.  

5D Mark III Shadow Pushing

Can you push the shadows on the 5D Mark III farther than the 5D Mark II?  Thats one of the burning questions a lot of upgraders have and its something I wanted to test.  Technically speaking how far you can push the shadows at base ISO is largely a function of the noise floor and thus is drive by how much read noise there is.  The latest generation of Sony EXMOR sensors have exceptionally low read noise at base ISO and many were hoping that the latest Canon sensors could deliver the same.

I captured the above scene with the intention of pushing the file in post so that the menu on the black board was readable.  The closeup above is a view of actual pixels. 

If I push the exposure by 3 stops, you get the image above, now lets take a look at how much noise there is in those shadows.  First lets take a look at images with no noise reduction applied at all.

5D2 with no noise reduction

5D3 with no noise reduction

Now lets see what happens if we apply some color noise reduction (about 35 on Adobe Camera Raw):

5D2 with color noise reduction

5D3 with color noise reduction

And finally lets add some luminance noise reduction as well since if were doing this to an actual image thats probably what we would do to try and get the cleanest image possible:

5D2 with color and luminance noise reduction

5D3 with color and luminance noise reduction

The way it looks to my eyes is that the 5D Mark III has a little less structure to the noise and noise amount appears to be slightly less (very slightly).  Though even with noise reduction, we are left with large color blotches which aren't as appealing.  I don't see this affecting my photography since it hasn't affected me with the 5D Mark II, however I know this is a big deal for some photographers out there.  In short if you were hoping for a dramatic improvement in this area with the 5D Mark III, you are going to be disappointed. 

More 5D Mark III Autofocus

Unfortunately I didn't get a chance to test out the 5D Mark III on birds this weekend (due to rain), so my 3 year old hopped up on sugar will have to do. 

Autofocus tests are very hard to do because the competence of the photographer plays a pretty big role in it.  So let me be up front.  I don't photograph sports and its been a while since I've done birds in flight too.  To help, I have overlayed the location of the active AF point so you can get an idea of what might have been a camera goof and what might have been operator error.




I set the ISO to be 800 to keep the shutter speed high. The first two are from the 70-200 f/2.8 IS II shot at 200mm wide open, shot handheld. These are 100% crops of a much larger frame. The AF point used as one of the off-center points (though still cross-type). I believe if you click on full screen, you will see the full resolution 1080p video.



Another series done with the 70-200.

I also had a chance to shoot with the new 400 f/2.8 IS II. The jump in focal length definitely starts to stress both the AF system of the camera and the strength of my arms due to the increase in weight and the fact I was handholding.




Again, you can reach your own conclusions.

Morning Call

Morning Call - 1D4, 70-200 f/2.8 IS II + 1.4x TC @ 200mm, f/6.3, 1/250s, ISO 1600, flash fired.

Two Gulls on the cliffs of La Jolla call out in the early morning before the sun has had a chance to rise.  Because of the way the cliffs are arrange, its possible to photograph these gulls against the beautiful pink skies during pre-dawn.

Fern Spring

Fern Spring - Leica S2, 35 Summarit-S, f/19, 6s, ISO 160

I had intended to go out to the Palo Alto Baylands this morning but had made my plans without first checking the weather.  This is not a problem 95% of the time since its usually sunny most of the time; not this weekend.  Since its raining outside, a water themed image seems appropriate. 

I made this image of Fern Spring in Yosemite on a trip there last year.  This image exaggerates the size of this spring, its really quite small.

Canon 5D Mark III High ISO

There's a lot of confusion out there on just how much Canon has improved the high ISO performance on the 5D Mark III.  Many of the sample JPEGs shows a dramatic improvement (up to 2 stops claimed by Canon) however the question is how much of an improvement is there with RAW images.

I decided to capture some quick test frames from both my 5D2 and 5D3 at various high ISOs to answer this question.  This was done with a tripod, converted in Adobe Camera Raw.  Click on the images to view the full size versions.

5D2 @ ISO 3200

5D3 @ ISO 3200

5D2 @ ISO 6400

5D3 @ ISO 6400

5D2 @ ISO 12800

5D3 @ ISO 12800

5D2 @ ISO 25600

5D3 @ ISO 25600

Canon 5D Mark III Autofocus First Observations

My 5D Mark III is here and I've had a very brief chance to compare the autofocus performance with the 5D Mark II. I'll be doing some flight photography this weekend, but in the mean time I tested out my other use case for pro AF, photographing with large aperture primes in low light. I put a 85L on both bodies and went to photograph my 3 1/2 year old daughter in the living room. She was sitting about as still as a 3 1/2 year old tends to sit and I tried both the center and outer AF points (using AI Servo). I took a series of images at full speed with both cameras. I then processed the images, cropped them down to the point of focus and made a movie out of the frames in Photoshop CS6. Here are the results.
That is the 5D Mark II using the center AF point which is much more accurate than its outer AF points.


This is the 5D Mark III using the center AF point.


Now this is the 5D Mark II using the left-most AF point.


Finally this is the 5D Mark III using the left-most AF point.

This isn't the most scientific of tests so you are free to draw your own conclusions. For me, one of the things that drove me bonkers with the 5D Mark II was flaky AF with the outer points when working with fast primes. I am thrilled with the performance I'm seeing so far with the 5D Mark III in this regard. Whether this will translate to better tracking performance for avians and wildlife remains to be seen. The exposure I believe was around f/1.2, ~1/80s, and ISO 3200. Thus the light level was around 2 EV give or take. Its not exactly candlelight, but not bright either.

Common Loon and Lightroom 4 Goodness

Common Loon - 1D4, 800, f/8, 1/1000s, ISO 400

I made this image at Lac le Jeune near Kamloops, British Columbia in the summer of 2010.  Its been on my list to process for a while.  Loons are very difficult subjects with which to nail exposure.  If you try and expose for the head, the whites on the body will be blown and if you try to keep the whites well exposed, the head will be very dark and you'll have no feather detail. 

When I captured this image, I biased my exposure to get slightly more detail in the head and allow some of the highlights to clip.  After a bunch of work in Lightroom 3, this was the result I was able to come up with:

This was after doing some extensive local adjustments to try and recover as much detail as possible in the highlights.  I left it with the intention of completing the tonal work in Photoshop. 

Today, I opened up the image in Lightroom 4, converted it to Process Version 2012 and using the new targeted shadows and highlights adjustment tools I was able to get the tones where I wanted with the appropriate detail right in Lightroom.  The really neat thing about the new controls is that they keep haloing artifacts to a minimum.  Another major advantage of having local shadows and highlights is that I don't need to be as precise in my brushes.  With my LR3 local adjustment, I have to very careful with the brush (I have to use an exposure or brightness adjustment) because if I go into a neighboring region, that region will get affected.  With LR4, since I am only targeting shadows or highlights with a single local adjustment, if I go into a neighboring region and the tones there are very different, it isn't affected as much.  I should note that I also did some additional work on the background which is why its different from the LR3 version.

Pismo Pier

Pismo Pier - M9, 35 Summilux FLE, ~f/4, 8s, ISO 160

Hyenas in the Rain

Hyenas in the Rain - 1Ds3, 500, f/4, 1/200s, ISO 800

Almost exactly a year ago I had arrived in Arusha eager to start my photographic safari.  My buddy Andy Biggs is over there now running a couple of back-to-back safaris and he posted on Facebook today about the storms and rains they had seen.  I only saw one afternoon of downpour during my trip and this image was taken during a 'light drizzle' break.




I took the video above shortly before we stopped and I photographed the Hyenas. Its quite an experience driving through the Serengeti during a rain storm.