[sinus007]

Hi,
I just got a new lens, DFA 24-70. Ran a quick test which prompted the following question.
Why do the images at the shorter focal length seem to be less in focus?
I shot the chart from about 900mm at FL - 24mm, 35mm, 50mm and 70mm. See attached images (crop about 200%).
Thank you very much.

Attached Images

View Picture EXIF

File Date

 

View Picture EXIF

File Date

 

View Picture EXIF

File Date

 

View Picture EXIF

File Date

 

[btnapa]

Why do the images at the shorter focal length seem to be less in focus?

Which body did you use? I can't tell if it is a focus issue or lens softness.

[sinus007]

Hi,
I'm using K-1

[c.a.m]

Why do the images at the shorter focal length seem to be less in focus?

I assume that the images are of the centre of the target, you used a tripod, and autofocused.

It's possible that the Autofocus Fine Adjustment for your lens requires different settings at different focal lengths for optimal AF performance. In other words, at a single common AFFA setting, the lens might be front-focusing at certain focal lengths and back-focusing at others.

For example, see this thread: K-1 soft images: AF Fine Adjustment or Body issue? - PentaxForums.com

It's not obvious at which part of the target the camera focused on in the various shots.

It might be useful to repeat the test while keeping the camera-target distance at a fixed product of the focal length, say, 25 times the focal length. So, 60 cm distance at FL = 24mm; 87.5 cm at 35mm; 1.25 m at 50 mm, etc. In that manner, the target will appear the same size in each shot and eliminates the potential blur caused by zooming in on the image by different amounts.

- Craig

[sinus007]

I assume that the images are of the centre of the target, you used a tripod, and autofocused.

It's possible that the Autofocus Fine Adjustment for your lens requires different settings at different focal lengths for optimal AF performance. In other words, at a single common AFFA setting, the lens might be front-focusing at certain focal lengths and back-focusing at others.

For example, see this thread: K-1 soft images: AF Fine Adjustment or Body issue? - PentaxForums.com

It's not obvious at which part of the target the camera focused on in the various shots.

It might be useful to repeat the test while keeping the camera-target distance at a fixed product of the focal length, say, 25 times the focal length. So, 60 cm distance at FL = 24mm; 87.5 cm at 35mm; 1.25 m at 50 mm, etc. In that manner, the target will appear the same size in each shot and eliminates the potential blur caused by zooming in on the image by different amounts.

- Craig

c.a.m.,
Yes, I was using tripod, autofocus, MUP, remote trigger; tried to eliminate all external factors. Camera was focused on the center of the chart for all shots.
As for Focus Fine adjustment, I believe it's for a lens, not for different focal lengths?
I'll follow your suggestion about fixed product of FL. Will let you know.
Thanks.

[c.a.m]

As for Focus Fine adjustment, I believe it's for a lens, not for different focal lengths?

Sorry, I might not have been clear.

Members here (myself included) have found that some zoom lenses have needed different optimal AFFA settings at different focal lengths. In certain cases, the AF calibration has been significantly different across a FL range. For example, on my K-3 II with my DA 20-40mm Limited, I found that the optimal AFFA calibration at 20mm was significantly different from that at 30mm and 40mm; the only setting that gave the sharpest autofocus at all focal lengths was AFFA = +2 when focused from infinity. On my K-3 Mark III, a single AFFA setting works well across the range.

Unfortunately, Pentax cameras provide for only one AFFA setting for a zoom lens, so a compromise setting might be necessary. Other brands (e.g., Canon, I believe) allow two settings.

- Craig

[AstroDave]

It looks like you are using a (version of the) ISO 12233 chart.

If your chart has all the details it should, I would suggest shooting it such that is just fills the frame of your camera (at the proper aspect ratio), at whatever focal length you are using. That's how the chart is meant to be used.

In that case, your K-1 should more than resolve all the lines at the "20"s on the chart if the lens is up to snuff.

I made my own version of the resolution chart, by blowing it up by about a factor of 3 or 4, so my printer, at 1200 dpi, would adequately resolve it. The pdf given here (ISO 12233 Test Chart) is a good place to get a pdf which can be scaled up to whatever size you need.

Here's my alternate version of the chart, and a zoom into the upper left corner, taken with my K-1 and 100mm Pentax macro. All the corners are nice and sharp with the macro. My chart is scaled such that it should fill the width of a normal ratio (1:1.5) frame and the resolution numbers will be correct. The letter C is there because I have three of these, which I used for my students to make test shots with their cameras. The crosshairs in the middle allow for good focusing, and the color chart - why not!

I'd be glad to make available high-resolution files of the resolution "wedges" or whatever they are called.

Attached Images

View Picture EXIF

File Date

   

[sinus007]

Hi,
It looks like it was backfocussing. I set the Fine Adjustment for the lens to +3 and images from 24mm and 70mm are much closer in-focus now.
Thanks all for help.

[clickclick]

I never use autofocus for shooting a chart like that. I would use live view and press the "OK" button to zoom the image to the maximum enlargement for best focusing results. And as mentioned, that chart should be used filling the full frame. The link @astrodave posted is where I downloaded my copy from too.

[zapp]

The target is flat, not well suited to judge fine adjustment. Keeping camera distant constant will change scale between images, making it even harder to judge anything.