Tokina 11-16 + D7000 = Moire

I think this sensor has a whole new set of issues since under same conditions, I can't get my D300s to show Moire.
 
I shot 28 without any issues, but found the same problem with a 14. I'll try to do a more detailed test this weekend, adjusting can to subject distance to compensate for varying focal lengths.
 
Sinapps and Macgregor,

The lens is not the issue, really.

It looks like there is a fine horizontal pattern in the air vents on those monitors in your samples. That "interferes" with the pattern of your (line-skipped) sensor and causes moire. I's not really the lens, but the subject matter, and the size of it as projected by the lens and camera distance. The reason you only see moire over part of that vent area is that as the lines get farther away the change in scale ("frequency") and interfere more or less with the sensor.
 
Yeah, its not the lens, although using wides often show this since so much of the scene is taken in and mostly in the same focus plane. The moire often shows when repeating patterns are at the right size and angle to trigger it.

The samples shown by macgregor just barely or shall we say luckily passed the threshold. But the test proves it has same moire and aliasing characteristics as most dslrs except ghXX cameras.
 
Sinapps,

I conducted tests on caprock, Ziess softar and kenko softon. The softar and kenko are similar, and I think its like the hoya as well, a clear glass with small "lenslets" for lack of a better term.
Conclusions, the softar and kenko DO NOT WORK. Only the caprock effectively removed moire without aggressive diffusion. Now the caprock's varying power must be matched with the appropriate lenses.
 
Thanks Ted,

This is not good news. Caprock screw on is too small for the 77mm front on the 11-16 and their 4x4s are not available in the #1 which I think would be required for the widest angle on that lens.
 
polfilmblog said:
Can the moire be seen while filming or does it bite you afterward when you look at the footage?
Click to expand...

Yes, it can be seen when filming. Since I never saw it on my D300s, I assumed it was a peculiarity of the camera lcd until I got it to a big monitor.
 
sinapps said:
Thanks Ted,

This is not good news. Caprock screw on is too small for the 77mm front on the 11-16 and their 4x4s are not available in the #1 which I think would be required for the widest angle on that lens.
Click to expand...

Sinapps,

I have the largest ring size which is only 67mm, I use a 77-67 step down ring on the tokina and you can still use it at 15mm without vignette, 14.2mm with very negligible vignette. I have also used it on my 20-40mm sigma at 20mm.

If you want to squeeze out every mm, you can take out the step down ring, and carefully tape the caprock's rim to the tokina's filter thread and bayonet mount and you can have a usable image at 13.5mm. -just tested this now for you. ;-)

-oops -just added this, if you use a nikon, it could be a wee bit wider considering the "point one" (.1) difference in crop factor bet nikon and canon cropped sensors. :-)
 
Last edited:
Here's the footage:




Downsampled to 720p. No grading. Video updated with additional steps to reduce Moire.
 
Last edited:
Finally found a workable formula. Top part, straight from camera, bottom part manipulated with following steps:

1. Select color from non moire area (blue) and fill layer
2. Use moire color (red) as mask for layer above
3. Apply Chroma Blur
4. Apply NeatVideo

Untitled-1.jpg
 
You must log in or register to reply here.
Back
Top