Focus bug in the user "Adjustable" Self-Timer in the E-M1 Mark II Version 3.0.
Re: The user adjustable Self-Timer with the variable time delay, number of frames taken, interval, and Auto-Focus "ON" or "OFF" settings.
If one has some Auto-Focus Fine Tuning data set in the body, this particular self-timer will ignore the user's auto-focus fine-tuning data if the AF is "ON". If you turn it to OFF it will accept the user's data for the auto-focus correction when shot. Example: If self-timer's AF is ON, no matter if you have a +20 or -20 in the AF Fine Tuning, all images will be exactly alike with no shift in focus (i.e. It ignores any user applied AF tuning info.).
The other self timer's do accept the user's AF tuning data and work as they should. The 12 second self-timer works well.
Discovered this bug when trying to adjust a lens. Although I could see the focus shift in the viewfinder from -20 to +20 on a slanted ruler by pressing down slightly on the shutter button to focus, the final shots all came out on the exact same number (Foreground setting only.) regardless if I had set -20 or +20.
Drove me nuts until I saw the C-AF being Zero ("0") in ALL shots no matter +20 or -20 set with that particular self-timer in FocusTune software.
I finally took some small change, purchased LA version 4 and gave it a spin yesterday.
I took Phocal's workflow and I believe I followed it by the book.
Camera: Oly E-M1X
Lens: Oly 300 f/4.
Oddly enough when I import the values the column for AF Mode shows S-AF. This explains why I had no way to de-focus properly between shots.
I ended up to create 11 lens data sets (from -5 to +5), pick the set and switch into "normal" shooting mode. Single shot, C-AF, Aperture priority at f 4.0, ISO 400.
My lens ended up to be sharpest at -1, so I do not expect a massive image quality difference but maybe it adds that tiny little bit of critical sharpness I'd like to see.
Time will tell.
During the "Stay Home" virus thing, I thought about trying to refine the 25 AF points on the 300mm f/4 on the E-M1X while waiting for the birds to appear. Came up with the slant target below. I printed it on a sheet of 8.5x11 inch photo paper.
I put it on the ground (BOTTOM is closest to tripod for the 45 degree slant to the targets.) and used the camera's single C-AF auto-focus point. I moved it around until the outer and inner AF points align on the center and outer four corners of the black targets on the paper. The inners are an approximation of the inner circle of AF Tuning points as there are 25 of them in the AF Tuning menu.
Fire a shot of the center one and examine the white line on the left and right of the middle target. Best to use the camera's magnification to examine the result, and try a few shots too. The sharpest line should be centered on the AF target. If not, apply a + or - number to the Lens Focus setting until the sharpest white line is moved to the center of the target. Same goes for the rest of the AF Tuning setting working around all the 25 focus points on the screen.
Shoot the target wide open (f/4) and about -0.3 to -0.7 stop. If the sharpest white line is higher than the target's middle where the green square got a lock-on, then dial in a negative AF tunning number to move it down. Opposite applies if the sharpest white line is too low which means you add a few point to the AF tuning number to raise it. Hit the playback arrow to show the JPEG and dial in maybe a x2 to x5 magnification to examine the result. It actually goes pretty fast overall once you get the hang of it. I had issues with the Micheal Tapes Lens Align as the ruler is too far to the right and putting the focus square on the Tape's target you cannot see the ruler on the right as it is out of the frame. With the Tapes setup,you need t move the caerma around a lot to center it, and use the software too if you have it.
With the target below, the ruler (white lines) is on both targeted spots so the camera stays put and you just move the AF point around to the target area. Hitting the blue arrow button (Image review) and spinning the dial to zoom in to 2x or 5x magnification you can see and examine the white ruler grid lines to either side. Focus Peaking is an option too, but it seems to work best in the middle of the frame and less to the top and bottom. You can see the white lines quickly snap into place when touching the shutter button and near the AF targets if it is tuned up right.
Don't know if this will work and align with other lenses, but one can try. I used Windows 10 Paint to draw the thing and kept the white horizontal lines about one grid square apart (Highly magnified in Paint) from the blacks so to get a sharp edge and so it will blur out if AF is off. If you make one yourself, best to save as a BMP file as the JPEG will gray some edges but it still can be used.
Subscribe to see EXIF info for this image (if available)
Another day:
Below is the image I used for the ~40mm lenses. It is skewed more so the target squares will appear square when shot at a 45 degree angle. It is printed on a 17x22 inch photo paper since it needs more paper area for the wider angle of those lenses.
Fwiw, the black rows are about 1/4" wide and the targets about 1" square. I tried using finer lines but they get more confusing as to what is sharp than when being spaced further apart. Also try and print it in a BMP or PSD mode with no gray area that might make seeing the sharp edge of the lines difficult.
I also found setting the Sharpness to +2, Contrast +2, and Monochrome to work the best for viewing the result.
Once you work through the sequence and AF tuning, I'd recommend shooting the entire sequence with the new settings going across row 1 (5 shots) and then dropping down to row 2 and continue to the end. Once done, load the RAW files into FastRawViewer software and set it to Green or Red sharpness mode and see if the green or red sharpness lines pass through the middle of all the targets. Being shot in sequence, it's easy to just right arrow through them and see the results of each targeted AF point. This works well to back up your AF tuning results and correct any if needed.
Subscribe to see EXIF info for this image (if available)
With a 150mm to a 420mm lens, place it about 4 feet from the front of your lens at a angle of ~45 degrees. Fill in the appropriate grid squares with black ink that correspond to a row of your five AF tuning squares. Shoot them with the normal single (larger sized) focusing square in each row across and examine them for an equal front-to-back green lines in FastRawViewer. If low, add to the AF Tuning number value. If they are higher, subtract from the AF tuning value and see if they become equally spaced on top and below the focused on square. Then move on to another row and test.
Having done the slanted targets thing above, I got to thinking about the supposed cross-target AF system Olympus uses for focusing. Not knowing where the AF cross is actually focusing at with the slant-type of targets, I went another direction later.
I decided to try and come up with a flat or perpendicular target (i.e Parallel to sensor plane.) and see what difference it makes. Results look pretty good doing it that way. I did a test with the E-M1X and the 420mm setup (i.e. 300mm +1.4TC) along with the camera's "Digital Teleconverter" turned 0n which made it around 840mm and was shocked how tack sharp it was after doing the AF calibration setup with the 420mm configuration. This one: https://www.mu-43.com/threads/share-birds.4838/post-1376102
The AF target I used is below. I made a copy of the 10x10 squares image off http://print-graph-paper.com/details/10-squares-per-inch and then marked off the AF targets for the Olympus using Windows Paint and numbering them for Row and Column.
An 8.5x11 inch print will work for the 300-420mm range. A larger one of 13x19 worked for the 75-150mm lenses. A 17x25 inch worked for the 40mm range, although I also pressed it to do a 12mm zoom too but a larger one of maybe 36x48 inch would be better for the wides and provide a better distance between the camera and target.
I did a ring-around shooting it with the AF set to -2 to start, then a round shot at zero, and then a last round set to +2 AF tuning number. I shot a dark frame between the three so I could tell when I switch the AF tuning number. I shot in rows as the target are labeled so I can find the image in FastRawViewer zoomed in to 500% and in the "E" (Or red sharpest contrast.) mode. It takes me about 15 minutes to shoot the three AF points in camera before examining them in FRV on the computer for best sharpness. I shot them in "Silent" (heart) shutter mode with a 2-second delay, aperture wide open, via a wireless Vello release made for Canon which fits the E-M1 II and E-M1X both.
Look for the zoomed in one showing the most red at 500% and on the same target you shot too. If one is red equally at -2 and zero, then call if -1 or do the test again. If none are red, then AF may be off further than you think or the lens is bad.
Aside, I found my 45mm f/1.2 Pro to take a lot more minus in the edges/corners than the center than the smaller aperture telephotos. More bizarre is you may have one AF square showing maximum red in FRV and really sharp, but scrolling around the frame may show less areas red and then some areas maybe more red. Just weird characteristics of the lenses and why you need to do all of them and refer to the one you shot at and why just doing the middle one may not be enough.
Subscribe to see EXIF info for this image (if available)
I used MTFMapper last week. I had the A2 target printed on a poster thru Officemax for $13. It was pretty easy following the instructions in this thread and the software wiki.
Links on this page may be to our affiliates. Sales through affiliate links may benefit this site.Mu-43 is a fan site and not associated with Olympus, Panasonic, or other manufacturers mentioned on this site.Forum post reactions by Twemoji: https://github.com/twitter/twemojiForum GIFs powered by GIPHY: https://giphy.com/