[Cannikin]

A month ago when I first got the K-3, before all this hysteria about "soft" images got into full swing, I went out doing random tests with my new camera. Normally I shoot RAW only, but that day I decided to test RAW+JPEG on two separate cards. Luckily I still had all the JPEGs lying around, so I dug them back up to see what all the commotion is about.

These are all 100% crops (~1 to 2% the area of the full images), straight out-of-camera JPEGs, default settings (didn't touch in camera JPEG settings at all). No post-processing done (except for cropping obviously). All shot with the SMC DA 15 Limited @ f/8.









Now, one thing I've learned over the years with my Pentax lenses is to be very perceptive about the difference between a soft lens/image and a slightly out-of-focus image. It very often looks suspiciously similar. Case in point, this image which was taken a few seconds before the above image. Same location, same camera, same lens, same aperture.



This last image is soft due to field curvature, but can easily be mistaken for a soft lens/sensor. Remember, this is a 100% crop. It is not immediately obvious at full size that this is actually OOF. I deleted the RAW immediately and retook the photo, because I chimped and recognized what slightly OOF looks like with this lens. Deleting the RAW doesn't delete the JPEG as well if it's on a separate card apparently, which is lucky for this demonstration.

I always spend hours getting to know all the rendering characteristics of my lenses to properly diagnose why a picture doesn't look right, and dialing in AF micro-adjustment, before I declare a lens or camera a dud. Here are some quirks I noticed:

- Don't count on DOF to get everything at infinity in focus, especially if there's field curvature involved. On modern AF lenses with short focus throws, very slight movements of the focus ring will throw infinity totally off. PDAF can be maddeningly inconsistent near infinity.
- Remember that with higher MP sensors, 100% view is at a higher magnification which redefines DOF (i.e. what appears in focus at 100% 16MP may not look in focus at 100% 24MP).
- Atmospheric haze can produce all sorts of image distortions and degradation for the same scene on different days.
- Foreground bokeh on several of my telephoto lenses can look remarkably similar to shutter-induced motion blur (double image blur).
- Shooting before the SR is primed and ready (the hand symbol in the VF) can cause its own motion blur.
- And something I've noticed with photos taken these past few very cold days (for Seattle): "shivering-induced motion blur"
- High ISO noise can often make it difficult to tell if the image is in-focus or has motion blur.

---

Last edited by Cannikin; 12-07-2013 at 08:17 PM.

[rawr]

I always spend hours getting to know all the rendering characteristics of my lenses to properly diagnose why a picture doesn't look right, and dialing in AF micro-adjustment, before I declare a lens or camera a dud.

You are doing it wrong, I'm afraid.

At a maximum, 5 minutes should elapse between you taking a photo, declaring it out of focus, then declaring it as a camera fault, and then starting a thread about how you have a defective camera (or lens) on this site or others (especially DPR). ...

[dansamy]

Still wrong. You must immediately whinge on about it on EVERY photography related forum on the Internet, tweet it with samples, Facebook it, blog about it and purchase a billboard next to Amazon headquarters, who sold you the "defective" item in a plain brown box but charged full retail.

[sholtzma]

I'm afraid I have to agree with rawr, and I have to disagree with him, whichever will create the most ruckus. All of your photos look soft, and I blame the camera. That convinces me not to buy a K-3--well, that and the stupid "Ricoh" insignia. ;-)

[sholtzma]

I'm going to have to disagree with dansamy, but for the life of me I don't know why....

[dansamy]

Because I make sense, therefore the only response is irrational opposition.

[Winder]

A month ago when I first got the K-3, before all this hysteria about "soft" images got into full swing, I went out doing random tests with my new camera. Normally I shoot RAW only, but that day I decided to test RAW+JPEG on two separate cards. Luckily I still had all the JPEGs lying around, so I dug them back up to see what all the commotion is about.

Now, one thing I've learned over the years with my Pentax lenses is to be very perceptive about the difference between a soft lens/image and a slightly out-of-focus image. It very often looks suspiciously similar. Case in point, this image which was taken a few seconds before the above image. Same location, same camera, same lens, same aperture.



This last image is soft due to field curvature, but can easily be mistaken for a soft lens/sensor. Remember, this is a 100% crop. It is not immediately obvious at full size that this is actually OOF. I deleted the RAW immediately and retook the photo, because I chimped and recognized what slightly OOF looks like with this lens. Deleting the RAW doesn't delete the JPEG as well if it's on a separate card apparently, which is lucky for this demonstration.

I always spend hours getting to know all the rendering characteristics of my lenses to properly diagnose why a picture doesn't look right, and dialing in AF micro-adjustment, before I declare a lens or camera a dud. Here are some quirks I noticed:

- Don't count on DOF to get everything at infinity in focus, especially if there's field curvature involved. On modern AF lenses with short focus throws, very slight movements of the focus ring will throw infinity totally off. PDAF can be maddeningly inconsistent near infinity.
- Remember that with higher MP sensors, 100% view is at a higher magnification which redefines DOF (i.e. what appears in focus at 100% 16MP may not look in focus at 100% 24MP).
- Atmospheric haze can produce all sorts of image distortions and degradation for the same scene on different days.
- Foreground bokeh on several of my telephoto lenses can look remarkably similar to shutter-induced motion blur (double image blur).
- Shooting before the SR is primed and ready (the hand symbol in the VF) can cause its own motion blur.
- And something I've noticed with photos taken these past few very cold days (for Seattle): "shivering-induced motion blur"
- High ISO noise can often make it difficult to tell if the image is in-focus or has motion blur.

I've probably shot 50 JPEG images over the last 5 years. Since getting the K-3 I have shot 500+ B&W JPEGs. It's a very impressive camera.

[Giklab]

I've probably shot 50 JPEG images over the last 5 years. Since getting the K-3 I have shot 500+ B&W JPEGs. It's a very impressive camera.

I posted an OOC shot, Natural mode w/ tweaks, to a "Recent Photo /p/ thread a while ago. Several people asked if I had found a DIY way to develop Kodachrome.