Last week I received from B&H a Zhongyi Mitakon Super Macro 1x-5x (generation II) and put it through my normal lens set of tests. This is the results.
Short answer:
It is good considering the price, with one notable exception: It appears to have internal overall flair (lower contrast) then I expected. Image quality is OK, but not up to comparable Nikon microscope objective. Big advantage is the variable power (1x to 5x) and the long working distance (10" @ 1x, 4"@5x). No noticeable Chromatic Aberrations.
Long Answer:
This is typical of the charts I do for every lens. I like to be able to look up things like working distance, effective F-Stops, step size, etc. It’s not perfect, but it works for me. Included is a brief description of the process to make a lens chart.
Make an image of machinist ruler graduated in 0.5 mm. Into Photoshop, measure pixels between extreme marks, calculate Pixels per MM, MM across Sensor, Multiplier.
Calculate DOF, 70% DOF (will be step size), Effective F-Stop
In the chart:
Repro Ratio is the markings on lens, usually the ":" and a guess as to the half value
Ruler is a visual observation of the ruler, not calculated
F-Stop is the lens F-Stop as marked
Small Ruler and Big Ruler is used as setup (part of the Macro Stand)
Lens Clearance is approximate distance from front of lens to object (Working Distance)
The images in page 1 of the chart are single images, using continuous lighting (2 each LEDs, 6500K, 3W, 350lum). Don’t care about image quality, as only a single distance is measured in Photoshop.
Multiple images are used for the Diffraction Test. Setup is like you were doing stacking, but not stacked, just a single image is visually chosen as the diffraction start point.
Mitakon Chart Page 1
Calculations for Diffraction:
Images (20 or 30) of flat laser print for each F-Stop. Use step size = 20um for 1x to 2.5x, then step size = 10um for 3x to 5x. Taken with one to four Flash (Yongnuo YN560 IV Speedlights), various power out depending on F-Stop.
Choose the best image at 100% under Lightroom for a given F-Stop. (The first bunch of images will be out of focus, then the best focus image, then diffraction on the final images).
Compare the selected image for each F-Stop of a given multiplier to determine (read guess) where diffraction starts.
It appears that the usable (non-diffraction) is up to f/8 for up to 2.5x, and f/5.6 for the 3x to 5x range.
The good F-Stops are in BOLD typeface.
Mitakon Chart Page 2
Mitakon Chart Page 3
Mitakon Chart Page 4
Since I also have a Nikon Plan 4x on a Nikon 70-300mm, I thought I would run some comparisons to the Mitakon.
I did a 20 image pass at 4x and 5x on both lens setups of the flat laser printout. Run each group thru Zerene Stacker (PMax only). All my macro images are taken as JPEG (space becomes an issue), and presented to ZS as JPEG, are straight out of camera (Nikon D300s), “Standard”, no other processing. Any sharpening is done on the final image ZS output (no shaprening on these, tho). These have been sized for the web upload.
Because these are stacked in ZS, I don’t expect to see any CA here. (I really did not see any in the original images before stacking.)
The lower contrast on the Mitakon lens is shown in the Histogram screen shots. These also indicate that there might be some CA in the images. Note the blue and yellow on the Mitakon Histograms, compare to the Nikon Histograms.
Nikon 4X Diffraction:
Nikon 4X Histogram:
Mitakon 4X Diffraction:
Mitakon 4X Histogram:
Note the blacks are not there compared to the Nikon Histogram.
Nikon 5X Diffraction:
Nikon 5X Histogram:
Mitakon 5X Diffraction:
Mitakon 5X Histogram:
Note the blacks are not there compared to the Nikon Histogram.
Some real world images:
Here is a 2.5mm by 2.5mm Toshiba chip that I had set up on a 10x run last week. These are 50 images per stack using 20um steps. SOC and no sharpening. This is a 900 MHz cell phone chip, stuck in some black modeling clay.
Nikon 4X Chip:
Mitakon 4X Chip:
Nikon 5X Chip:
Mitakon 5X Chip:
Summary
Looking in the lens, there does appear to be some dust inside the lens. I cleaned both outer surfaces, and appears to resolve the issue. Trying to clean the inner (camera side) element is awkward at best, as it is inset several inches into the lens. I had to use a APS-C Sensor cleaning swab. I do not see dust showing up on the images, but I am sure it is there. Something does show up on the Diffraction tests at the higher F-Stops, and it looks like dust on the sensor, which it might be. There was white particles in the grooves of the outer surface of the lens, not sure where this came from. A toothbrush dislodged this nicely, then blown away. Zoom lenses have a tendency to suck dirt into the inside as they are zoomed, so hope the seals are good.
This should be a good lens for doing insects in the field, with the variable power and the long working distance. For the more critical images, like the chips above, I would stay with my Nikon setup. The contrast issue can be mostly handled by increasing contrast in post processing. Adding sharpening would help the image quality. Of course, RAW images instead of JPEG would be a big advantage.
If anyone wants to pixel peep, I can send some of these images into the cloud. But not all 1700!
Comments, of course, are welcome.
George
Mitakon Super Macro 1x-5x (generation II)
Moderators: rjlittlefield, ChrisR, Chris S., Pau
-
- Posts: 36
- Joined: Mon Nov 24, 2014 9:50 am
- Location: Colorado
-
- Posts: 36
- Joined: Mon Nov 24, 2014 9:50 am
- Location: Colorado
For anyone that wants to pixel peep, here is a link to OneDrive that has a sampling of 13 pictures.
https://1drv.ms/f/s!AvUX6lOYSxvraU-FAhs_h5dlX-E
Appears that OneDrive lets you zoom in to about 140%
Let me know if this works OK, or if you have issues using this method.
Thanks,
G
https://1drv.ms/f/s!AvUX6lOYSxvraU-FAhs_h5dlX-E
Appears that OneDrive lets you zoom in to about 140%
Let me know if this works OK, or if you have issues using this method.
Thanks,
G
- enricosavazzi
- Posts: 1474
- Joined: Sat Nov 21, 2009 2:41 pm
- Location: Västerås, Sweden
- Contact:
Thank you for the images.ohdeeremee wrote:For anyone that wants to pixel peep, here is a link to OneDrive that has a sampling of 13 pictures.
https://1drv.ms/f/s!AvUX6lOYSxvraU-FAhs_h5dlX-E
Appears that OneDrive lets you zoom in to about 140%
Let me know if this works OK, or if you have issues using this method.
Thanks,
G
So it would appear that the lens IQ is OK-ish on APS-C, compared with known-good macro lenses and microscope objectives. Even at 12 Mpixel, the fine detail one expects to see is simply not there. The comparison images with the Nikon microscope objective contain unquestionably more detail, although the sensor should be able to give even more detail (albeit the physical anti-aliasing filter and the 11-year old de-bayering algorithm may be showing their limitations here).
In the image with the Mitakon lens at 1x there is also a visible flare in the center of the image, which significantly lowers contrast in this region. It might be avoided by using a narrow lens shade, which however is going to reduce the working distance (not a big problem with this lens, though).
The Mitakon lens is going to be blurrier on Micro 4/3 because of the smaller pixels, so I would not recommend this combination.
On full frame, the Mitakon lens is going to perform visually better in the center unless one uses a sensor with 40 Mpixel or more. How the lens performs in the corners on full frame remains to be seen.
You can get a similar subject coverage at 5x on full frame and 2.5x on Micro 4/3, and at this magnification on Micro 4/3 there are several better choices for a lens. It might be interesting to compare the Mitakon lens at 5x on full frame and 2.5x on Micro 4/3, though.
Where the Mitakon lens is unbeatable (or rather, only has one competitor, i.e. the Canon MPE 65, which is probably optically better but is only usable on some cameras) is the ease of changing magnification between 1x and 5x without switching lenses. Unless the highest IQ is necessary, this Mitakon lens is therefore ideal for working quickly, and is more portable (e.g. for work on location) than a set of separate lenses, helicoids etc.
--ES
-
- Posts: 36
- Joined: Mon Nov 24, 2014 9:50 am
- Location: Colorado
You are welcome.
My intent with this lens was to use outside in the field, mostly 1x-2x range, on a D500. The D300s setup is on the Macro rig, bench top only.
The flare is the first thing I noticed, and I did try a black shade I made from construction paper, about 6" long, same diameter as the outside of the lens, thinking that I was getting unwanted external light. Did not seem to help at all, suspect the flare is from reflection internal to the lens body. Using Lightroom to measure black values (as shot) there is about a 10% difference comparing the center black letters with the ones in the corners.
I do not have a full frame to test on, just the APS-C. Suspect there will be some differences on FF. The IQ looks good here corner to corner, but will probably suffer some on FF. Maybe even the flare issue will degrade on FF.
For the price of USD500, I can not complain. It is the only one I found that has a variable magnification AND a large working distance. You just have to take the good with the bad.
G
My intent with this lens was to use outside in the field, mostly 1x-2x range, on a D500. The D300s setup is on the Macro rig, bench top only.
The flare is the first thing I noticed, and I did try a black shade I made from construction paper, about 6" long, same diameter as the outside of the lens, thinking that I was getting unwanted external light. Did not seem to help at all, suspect the flare is from reflection internal to the lens body. Using Lightroom to measure black values (as shot) there is about a 10% difference comparing the center black letters with the ones in the corners.
I do not have a full frame to test on, just the APS-C. Suspect there will be some differences on FF. The IQ looks good here corner to corner, but will probably suffer some on FF. Maybe even the flare issue will degrade on FF.
For the price of USD500, I can not complain. It is the only one I found that has a variable magnification AND a large working distance. You just have to take the good with the bad.
G