Six years of photos statistics

It has been 6 years since I started submitting images to various microstock agencies. Today I am going to look at it from purely hardware perspective. Thanks to ExifTool, free software program for reading image metadata, I can analyze information stored in my files.

exiftool -csv -r -sort -exif:ApertureValue -exif:ExposureCompensation -exif:ExposureMode -exif:ExposureProgram -exif:ExposureTime -exif:FocalLength -exif:ISO -exif:Make -exif:MaxApertureValue -exif:MeteringMode -exif:Model -exif:ShutterSpeedValue -exif:SubjectDistance -exif:WhiteBalance -xmp:ColorTemperature -xmp:Contrast -xmp:Exposure -xmp:FillLight -xmp:FlashCompensation -xmp:FlashFired -xmp:FlashMode -xmp:FlashRedEyeMode -xmp:HighlightRecovery -xmp:Lens -xmp:LensInfo -xmp:Tint -xmp:Vibrance  . > keywords.csv

I have been Canon user for 20 years and when I started working with microstock photos I was on my second digital camera Canon EOS 20D. Now I am shooting with Olympus OM-D E-M1.  I tried submitting images from point & shoot cameras, smartphones and even some old stuff scanned from slides (which probably are not included in this analysis cause I did not fill EXIF data).

I am not tracking success rates because different agencies got totally different criteria and same photo might be accepted in one and rejected in others. Surprisingly I tried to sneak 2 photos from Palm Pre+, which I think was 4 megapixel camera, long before agencies start accepting photos from mobile devices ;-)

Current minimum requirements on most agencies are 6 megapixels so most of modern smartphones and point & shooters would pass this but it is much harder with this type of devices to produce images less than ideal light conditions. Most important piece of hardware, optics, on these devices is fixed. What would you rather have best camera body with crappy lens or crappy body with best lens?

Since most of my experience was with Canon gear I can only tell that I went through 3 bodies during this time but in the same period I tried 12 different lens. My first professional lens was EF 17-40mm F/4L and on cropped sensor it was fantastic lens. When switched to full frame I noticed  that at 17mm people look comical so I switched to 24-105mm F/4L. I never really invested in faster lens because before microstock I was doing mostly landscape photography which employes tripod at most times.

Now I switched to different system and I am trying to figure out which lens I would need/like to have. Micro four thirds lens are usually half weight and half price comparing to Canon equivalents so I most likely end up with better ones. I read a lot of good reviews about Olympus 12-40mm F/2.8 PRO lens so it was my first choice when I was buying OM-D EM-1. Is it as good as Canon's EF 24-70mm f/2.8L time will show but for start it is half of the price :-)

Now let's bring subject of low light performance. I switched from full frame 5d mark II to half size sensor. I was wondering if this would have affect on noise of my photos and in result higher probability of rejections? Let's look at the data, do I need to have camera that produces low noise images at ISO 3200?

I shoot 80% of my photos below ISO 800. I need 1600 or higher for around 5% of my images. I can live with this :-)

Going back to lens, which one do I need? It seems that 24-105mm was very convenient for me. From the data I have I cannot really tell if I got any favorite focal length.

Besides the fact that I am often going to extremes on my zoom lens other data is scattered all over the place.  I was not using my tele zoom lens to often too.

When I looked at aperture data it seems like I am staying between 4 and 11 most of the time but I guess that two of my main lens were F/4 so I could not go wider.

Exposure time data shows that I could live with shutter speed shorter that 1/2000s and barely do long exposure over 1 minute :-)

Ula's cutlets

2 chicken breasts, cut into cubes

1 leek (white part) finely chopped

1 red pepper cut into cubes

6 mushrooms, chopped into cubes

2 tablespoons of corn starch

1 tablespoon breadcrumbs
3-4 eggs

Salt and pepper

Fry in oil.

Sweet potato crescent rolls

Ingredients

• 1 14- to 16-ounce red-skinned sweet potato (yam)

• 1 cup half and half
• 6 tablespoons packed golden brown sugar
• 1 envelope dry yeast

• 1/4 cup (1/2 stick) unsalted butter, melted, cooled
• 2 teaspoons salt
• 1 teaspoon ground cardamom
• 3 1/4 cups (about) all purpose flour
• 3/4 cup dried cherries or golden raisins (about 4 ounces)

• 1 large beaten egg, beaten to blend (glaze)

Preparation

Beat until potato is very smooth.

Heat half and half and 1 tablespoon brown sugar in heavy small saucepan over low heat until thermometer registers 105°F. to 115°F. Remove from heat. Sprinkle yeast over; stir to blend well. Let stand until yeast dissolves and is slightly foamy, about 12 minutes.

Fit mixer with dough hook. Add remaining 5 tablespoons brown sugar, melted butter, salt and cardamom to sweet potato in bowl and beat until smooth and will blended. Add yeast mixture and mix until blended. At medium speed, beat in enough flour, 1 cup at a time, to form soft dough. Beat 3 minutes (dough will be soft). 

Turn out dough onto floured surface. Knead until smooth and elastic, adding more flour by tablespoons if too sticky, about 6 minutes. Knead in dried fruits or raisins. Butter large bowl. Form dough into ball and add to bowl, turning to coat. Cover bowl with plastic wrap, then towel. Let dough rise in warm draft-free area until doubled in volume, about 1 1/2 hours.

Butter large baking sheet. Turn out onto floured surface. Divide dough in half. Using rolling pin, roll 1 dough piece to 12-inch-diameter round (if dough retracts during rolling, stop occasionally and gently pull and stretch dough with fingertips, then continue rolling). Repeat with remaining dough piece. Cut each round into 8 equal triangles. Starting at wide end of triangle, roll up toward point of triangle, stretching dough slightly with fingertips as triangle is rolled up. 

Brush point lightly with glaze; press to seal. Place rolls on prepared sheet, shaping into crescents and spacing evenly. Cover rolls with towel. Let rise in warm draft-free area until light and puffy, about 45 minutes.

Preheat to 400°F. Brush rolls with glaze. Bake until rolls are golden and sound hollow when tapped on bottom, about 25 minutes.

Upgrade continued - got flash

In my Canon life I was using flash on camera quite often so when I started evaluating Olympus OM-D E-M1 not everything could be tested. Eventually I decided to get FL-600R flash to make sure I can do similar things with new camera. Of course I do not have same range of lens for new system so making switch decision will be based on partial data.

After unpacking Olympus flash I compared size of these two devices. Definitively Speedlite is stronger and bigger but if FL-600R takes same amount of batteries its recycle time must be pretty good. With less power I might not be able to flash across big rooms but this is not what I am doing very often anyway.

When I got chance to bounce light off the white ceiling it looks pretty good. Sometimes I cannot do this so I decided to get some sort of reflector. As you can see there are velcros on my Speedlite. I was suing LumiQuest ProMax system with it. It has a holder for multiple reflector and frankly it was to complicated so I was using it infrequently. Especially since I got off camera flash bracket which seems to work better. Anyway I wanted something that can be easier attached to flash. I looked at flash soft boxes and and accidentally I found something looking pretty interesting: Rogue FlashBender. It can be quickly attached without any velcros. Also you can bend it anyway you want, sometime it could be faster than tilting whole flash head.

Great little bonus, you can wrap it and make quick snoot :-)

Big switch

I am testing new Olympus mirrorless camera, OM-D E-M1. Most likely I will change system, after almost 20 years with Canon. It is not going to be breaking news like when famous photographer switched from Nikon to Canon :-)

So far after 3 days of evening testing at home I could say that transition would require some effort. First of all I do not have all comparable lens with new system. I do not have decent flash unit so I have to make a decision based on what I can see now. For a while I got access to Olympus PEN camera so I am not entirely lost in their menu system. The advantage of OM-D is that it has all the buttons like DSLR so I do not have to dig into menu every time I need to make small adjustment to settings. In fact it probably got more that my Canon.

Lets start comparing. One of main reason for switch was size and weight. At 895g Canon is twice heavier than Olympus (443g). If you add professional L lens to the package, like EF 24-105mm it is additional 670g. It is again almost double of ED 12-40mm PRO lens which is 381g. So we start with 1565g vs 824g. If I go for a hike in a mountains every gram matters :-) I think one of the reasons I was suffering on my Mt Whitney hike was the weight of my camera bag it was 5lbs in addition to my 30lbs backpack.

Next "big thing" is sensor size. It seems that recently people believe that everybody needs full frame sensor camera. Yes full frame sensors have better noise performance but if simple bigger was better we would be running around with Hasselblads and I do not see it would happen any time soon :-). I bet current technology allows to put medium format sensor into smaller body but you cannot bend laws of physics it would require bigger and heavier lens. If only sensor is compared, yes 5Dmk2 has better one. I am not sensor expert so I checked DxOMark Sensor Scores for both camera sensors:

DxOMark	Canon EOS 5Dmk2	Olympus OM-D E-M1
Overall Score	79	73
Portrait (Color Depth)	23.7 bits	23 bits
Landscape (Dynamic Range)	11.9 Evs	12.7 Evs
Sports (Low-Light ISO)	1815 ISO	757 ISO

What does it mean? E-M1 only looses in Sports category. Let me put it into perspective. Why people need high ISO? Because when they shoot action hand held and they need higher shutter speed. I am not doing that. Shooting my daughter at playground does not count :-) Besides even with my current 5Dmk2 I would need faster lens anyway. If I disregard Sports category sensors are pretty much comparable.
Let me go a little bit further. Is 5Dmk2 with F4 lens at 800 ISO comparable to E-M1 with F2.8 lens plus IBIS? Because of camera weight and in body stabilization I most likely can go ISO 400 or less when shooting handheld and still got similar results. You cannot just compare bare sensors, lens and other features might totally change the equation. This is main reason I got E-M1 first. I will check if it can fit the type of photography I am doing and if answer is yes I can sell all my Canon gear. If it fails the test I can still return it.

MOTOACTV - my first sports watch

I was thinking for while about buying GPS-enabled cycling computer and Garmin Enge 500 was my favorite. I wanted to try my routes and connect other sensors via ANT+. Obviously Garmin device provides all these features. I can connect cadence and heart rate sensors. I can even upload it to my Bodymedia Activity Manager via Garmin Connect. Then few weeks ago Motorola revealed MOTOACTV device, Android based sports watch with music player. Moreover it has ANT+ and Bluetooth LE so potentially I can use more accessories than with Edge 500. It also has WiFi and FM Radio. Apparently you can also root the device and install some other Android apps. I guess these two devices are hard to compare. Edge is truly cycling computer while Motoactv is a hybrid. I decided to get Motoactv cause it can be used for other activities beside cycling and in theory I can expand it by installing additional apps in future.

I got it yesterday and it was pretty easy to setup. If you want to use WiFi you have to install desktop software in order to add networks (you cannot type passwords on this device). I tried desktop software on Windows and OS X and functionality is similar. I charged it for couple hours as directed before I tried it outside. I must say that GPS lock is instant in comparison to my Palm Pre+ which quite often cannot sync at all. I guess it is what you get when you got GPS integrated with a phone receiver. What I liked Motoactv is auto start/stop workout feature. It automatically sense that device is not moving and suspends logging data until you start moving again. I was not always sure about my average speed provided by SmartRunner app on my phone cause it kept logging even if I stopped. I hope it will be possible to install Android version of this app on Motoactv some day to compare. After you are done with you workout you may upload data to tracking website. You might do it wirelessly with WiFi or connect to the desktop with USB. Wireless option is great when you do not have your computer with you but still want to upload and see results. I would probably need ANT+ USB stick if I chose Garmin. I am still waiting for delivery of heart rate monitor to see how ANT+ integration works.

Finally I got Android Market on Kindle Fire

Finally, after couple unsuccessful attempts of installing Android Market without rooting my device, I decided to root it. Thanks to The Unofficial Kindle Fire Blog article I was convinced that I can unroot it later.

[caption id="attachment_914" align="aligncenter" width="300" caption="Android SDK Manager"][/caption]

I already got few first steps covered cause I got Java Development Kit, Microsoft .NET Framework and Android SDK installed. I had to download SuperOneClick. Since I haven't done any development on Android besides trying to setup some samples to run on emulator my KIndle Fire was not detected. Amazon Appstore Developer FAQ got precise description (PDF) how to make it happen. The missing step was to update USB driver in Windows Device Manager with the one that Google provides with Android SDK. Unfortunately reboot is required on Windows. After that I could follow steps, run SuperOneClick and reboot Kindle Fire.

[caption id="attachment_915" align="aligncenter" width="175" caption="Android Market"][/caption]

Installing Android Market was actually a little bit complicated. Again I got few steps covered by my previous attempts. I already tried to sideload some apps and this option is enabled on my device. In addition it is possible to install GoogleServicesFramework.apk without rooting device so I did it when I was loading some other Google apps. I only needed actual Market app (Vending.apk)and something given in this recipe: MarketOpener.apk. Up to this point it was going according to the plan. Then next step filed. I do not know what is wrong with current version of File Expert available on Amazon Appstore but it does not properly copy files to /system/app directory. Be aware that version 4.0.3 is not working with his recipe. I was trying to copy or cut Vending.apk from my Download directory and paste it to /system/app but it did not happen. I found Root Exploder on this blog. It was able to complete what File Expert could not do. I was able to change permissions of Vending.apk and install MarketOpener.apk. After rebooting I connected my device to my desktop machine and run SuperOneClick to unroot it. Now Androit Makert is functional. Of course it can only allow to install applications that Google marks as compatible with Kindle Fire (sometimes Amazon marks app compatible while Google claims is not).

[ad#adsense-1]