Dichroic/Microscope Florescence Filters

petrochemist wrote:

Matty W wrote:

I haven't tried the other enhancer filters, but I suspect you're right.

What I thought would be really interesting is a narrowband (I don't know the terminology) band pass filter around each chromaticity, if that makes any sense. I used to shoot a lot of Velvia and loved the saturation and remember that its spectral acceptance curves (wrong term again, I'm sure, but it's been a while since I looked at the white paper) are very narrow, whereas bayer filters seem pretty broad to improve low light performance, and I know some even prefer the color of early dSLRs to newer ones...

Phase One/Sony's trichromatic back seems to aim at something similar, but I was wondering why not a filter that has narrow acceptance peaks around R, G, and B, and blocks light elsewhere? It would essentially work as an ND filter, but would also increase perceived color saturation dramatically.

Then again, maybe something similar is possible in post:

https://www.thebrim.pictures/vivid.html

I don't think the Bayer transmission has been made wide for low light performance. The bands chosen are chosen more for accurately determining colour. The Bayer colours having considerably less overlap than the cones in the human eye.

It you use narrowband filters you'll end up with some colours not detected at all.

My Dichroic filters are relatively broad band ones. My blue dichroic filter transmits 400-470nm while the average human apparently perceives 455-492nm as blue. So my filter will pass light seen to humans as violet, but not the bit from 470-492 humans see as blue.

Likewise my green dichroic filter transmits 515-570nm, while the average human sees 492-577 as green. Incoming green light from doubly ionised oxygen (common viewed in astronomy for OIII nebulae) at 496 & 501nm would not be seen at all.

With blackbody type illumination the absence of short ranges of wavelengths may not be too significant, but with elemental light sources (sodium lights, neon lights etc.) the light can all come in a very narrow band. Old type sodium lights gave well over 95% of their lighting between 588 & 590nm and many fluorescent lights have distinct spikes in their spectral power distribution.

Yeah, that makes a lot of sense. I suspect narrower transmission peaks would result in a punchier, more saturated look at the cost of sensitivity: comparisons of the Phase One trichromatic back with the normal Phase One back indicate as much, and Velvia has infamously narrow acceptance curves. So it makes me curious about more carefully engineered enhancer filters. I suspect the DXL OLPF on the DXL Red is designed with similar goals in mind and it seems to work well.

Of course, something like that would probably result in a less accurate image, too, and would probably have to be tailored to a specific sensor. But for certain landscapes it might be interesting. I miss shooting Velvia mostly.