Stange Pattern Testing Lenses

[Carlo Rho]

We were testing the new Panchro Lenses by Cooke and while shooting this Esser Chart we got this strange pattern. It appears only when pefectly focused. This chart could be shoot a bit out of focus but what is it? Anyone has ideas?


Our published M-X test could be downloaded here.

 

[Jake Bastian]

Maybe you need a higher resolution chart?

 

[Dan Hudgins]

It appears only when pefectly focused.

Does the chart have a half tone screen used to make the shades of gray?

If the chart was made using some kind of raster device that can also make a pattern that can moire with the sensor.

The OLPF in RED ONE (tm) is said to have a blur radius between the spacing of the green/green and red/blue pixels, so when you get moire it tends to be more as a color pattern than one of brightness.

You can also get moire effects from resize and compression of images, so you need to check the full size de-Bayer as a 48bpp TIF file to see if you can make out any pattern in the original shot full size.

If you do a gauss blur before you downsize the images the moire may be reduced.

 

[Vico Martin]

Very strange Carlo, nunca he rodado con panchro pero por aqui la gente no ha reportado nada parecido, todo son alagos.
A mi no me parece moire cromatico, es demasiado ancho; me imagino que habrás echo varias pruebas, a mi me parece tema software de la cámara o el transcode.

keep us informed, please.

Vico.

 

[Dan Hudgins]

To wide?

To wide?

Quote: [via babel fish: I have never rolled with panchro but people have this way not reported anything seemed, everything are alagos. moire does not seem to me chromatic, is too wide; I imagine that there will be I throw several tests, seems to me software subject of the camera or transcode.]

When the dot pattern on the chart is close to the sensor pixel size (or in integer ratios) you can get wide moire patterns, the tell tale yellowish/cyan tint of the pattern looking like OLPF, but it is very regular and the only time I have seen such a regular pattern before was some sun set images where the OLPF was making a pattern of dots that looked like they were on a grid pattern. The spacing of those dots was a little larger maybe. With very close pattern sizes the moire can be half the width of the frame. One wat to tell that is to use a zoom lens and zoom in a little slowly, if the pattern changes (when you look at full size TIF frames) then it is probably something in front of the camera like a dot pattern on the chart. If the cooke lens is rectilinear then you might get a regular pattern, if you use a lens with barrel distortion (fish eye) then you would get another pattern, so you can try that as well.

Moire is always chromatic mixed with luma moire, they are different sizes because most of the luma moire comes from the green pixels and most of the chroma moire from the red/blue pixels.

Can you make a extreme close up of the chart to see if it has a dot pattern on it?

 

[Vico Martin]

Carlo, you printed the chart yourshelf or is a professional one??
what Dan says makes sense. Got the same with other metrics, or only one of them?

 

[Parra]

We use a transparency Esser test chart on a LV5 spherical illuminator. I had the same problem with Sony F3 and sony prime lens. So I don´t think it is a lens problem but OLPF filter.

Alfonso Parra AEC
www.alfonsoparra.com

 

[Carlo Rho]

Thanks Vico & Dan.
As Alfonso wrote is a transparency professional chart without any visible dot pattern, so I don't think it's a moire coused by the resolution of the chart. I never saw a moire with a color solid chart as this one. Generally moire in resolution charts "move" a bit as you pull the focus between appearing and disappearing, this one appear only when perfectly in focus. May be I can't make enough little focus movements but it feels like something different, like a reflection.
Any other idea?

 

[Graeme Nattress]

Carlo, if you want to send me an R3D I'll take a look and we can talk on email about this. I'd like to help you get to the bottom of it.

Graeme