Starship Asterisk*

Chris Peterson wrote: ↑Tue Mar 14, 2023 8:28 pm

johnnydeep wrote: ↑Tue Mar 14, 2023 8:13 pm

Fred the Cat wrote: ↑Tue Mar 14, 2023 7:23 pm How to get useful work out of a rainbow. No pot of gold but fun.

Future meteorologist or astrophysicist in the making perhaps. I notice the dad said "when the water hits the sunlight", as opposed to "when the sunlight hits the water", but I guess it doesn't make much of a difference, though both are necessary in this case

And then there's this. No future meteorologists or astrophysicists in this family!
_

Click to play embedded YouTube video.

johnnydeep wrote: ↑Tue Mar 14, 2023 8:13 pm

Fred the Cat wrote: ↑Tue Mar 14, 2023 7:23 pm How to get useful work out of a rainbow. No pot of gold but fun. :ssmile:

Future meteorologist or astrophysicist in the making perhaps. I notice the dad said "when the water hits the sunlight", as opposed to "when the sunlight hits the water", but I guess it doesn't make much of a difference, though both are necessary in this case

Fred the Cat wrote: ↑Tue Mar 14, 2023 7:23 pm How to get useful work out of a rainbow. No pot of gold but fun.

Chris Peterson wrote: ↑Tue Mar 14, 2023 7:18 pm

johnnydeep wrote: ↑Tue Mar 14, 2023 7:13 pm

Chris Peterson wrote: ↑Tue Mar 14, 2023 4:40 pm
That's a form of tangent error, a simple consequence of geometrical optics. But no lens system can uniformly map its spherical input to a planar output without some kind of distortion. That's a kind of mapping error.

Ok, so what I tried to depict is not "pixel scaling"?

"Pixel scale" just refers to what I said: the angle subtended by a pair of adjacent pixels. It is not uniform across any image for a variety of reasons, including the one you depicted.

johnnydeep wrote: ↑Tue Mar 14, 2023 7:13 pm

Chris Peterson wrote: ↑Tue Mar 14, 2023 4:40 pm

johnnydeep wrote: ↑Tue Mar 14, 2023 3:00 pm

So, I guess you mean something like the effect that I try to show in this image of red rays of light from a lens hitting a blue detector, where the same angle covers more of the detector the further off center it is?


pixel scale.png

That's a form of tangent error, a simple consequence of geometrical optics. But no lens system can uniformly map its spherical input to a planar output without some kind of distortion. That's a kind of mapping error.

Ok, so what I tried to depict is not "pixel scaling"?

Chris Peterson wrote: ↑Tue Mar 14, 2023 4:40 pm

johnnydeep wrote: ↑Tue Mar 14, 2023 3:00 pm

Chris Peterson wrote: ↑Mon Mar 13, 2023 10:35 pm
Pixel scale refers to the angle subtended by a pixel (or between the centers of two adjacent pixels). It isn't the same across an entire image. One reason is because of geometric aberration in the lens system, and another is because you are mapping a spherical world onto a flat image. This is visually extreme in fisheye images (which this isn't, of course, so the effect is more subtle).

So, I guess you mean something like the effect that I try to show in this image of red rays of light from a lens hitting a blue detector, where the same angle covers more of the detector the further off center it is?


pixel scale.png

That's a form of tangent error, a simple consequence of geometrical optics. But no lens system can uniformly map its spherical input to a planar output without some kind of distortion. That's a kind of mapping error.

johnnydeep wrote: ↑Tue Mar 14, 2023 3:00 pm

Chris Peterson wrote: ↑Mon Mar 13, 2023 10:35 pm

johnnydeep wrote: ↑Mon Mar 13, 2023 7:40 pm

Pixel scale? You can't mean the size of the pixels in a CCD detector, since those are surely uniform throughout, right? Does your statement apply to both film and CCD images? Is it a "feature" of lenses? (yeah, I still suck at optics...) ...Oh, maybe a lens always spreads light out at the edges so that whatever is detecting it - film or CCD - will detect a smaller area of the actual thing being imaged?

Pixel scale refers to the angle subtended by a pixel (or between the centers of two adjacent pixels). It isn't the same across an entire image. One reason is because of geometric aberration in the lens system, and another is because you are mapping a spherical world onto a flat image. This is visually extreme in fisheye images (which this isn't, of course, so the effect is more subtle).

So, I guess you mean something like the effect that I try to show in this image of red rays of light from a lens hitting a blue detector, where the same angle covers more of the detector the further off center it is?


pixel scale.png

Chris Peterson wrote: ↑Mon Mar 13, 2023 10:35 pm

johnnydeep wrote: ↑Mon Mar 13, 2023 7:40 pm

Chris Peterson wrote: ↑Mon Mar 13, 2023 4:33 pm

I'd guess we're just seeing lens aberration and mapping issues. Typically the pixel scale is larger around the edges of a field than at the center, which makes things look somewhat compressed as you near edges.

Pixel scale? You can't mean the size of the pixels in a CCD detector, since those are surely uniform throughout, right? Does your statement apply to both film and CCD images? Is it a "feature" of lenses? (yeah, I still suck at optics...) ...Oh, maybe a lens always spreads light out at the edges so that whatever is detecting it - film or CCD - will detect a smaller area of the actual thing being imaged?

Pixel scale refers to the angle subtended by a pixel (or between the centers of two adjacent pixels). It isn't the same across an entire image. One reason is because of geometric aberration in the lens system, and another is because you are mapping a spherical world onto a flat image. This is visually extreme in fisheye images (which this isn't, of course, so the effect is more subtle).

hypermetabolic wrote: ↑Tue Mar 14, 2023 1:55 am And what is that fainter rainbow just inside the main one, most visible near the horizon?

johnnydeep wrote: ↑Mon Mar 13, 2023 7:40 pm

Chris Peterson wrote: ↑Mon Mar 13, 2023 4:33 pm

Don M wrote: ↑Mon Mar 13, 2023 3:40 pm Why is the rainbow’s spectral dispersion (i.e., width) larger near the horizon?

I'd guess we're just seeing lens aberration and mapping issues. Typically the pixel scale is larger around the edges of a field than at the center, which makes things look somewhat compressed as you near edges.

Pixel scale? You can't mean the size of the pixels in a CCD detector, since those are surely uniform throughout, right? Does your statement apply to both film and CCD images? Is it a "feature" of lenses? (yeah, I still suck at optics...) ...Oh, maybe a lens always spreads light out at the edges so that whatever is detecting it - film or CCD - will detect a smaller area of the actual thing being imaged?

eljayhalflife wrote: ↑Mon Mar 13, 2023 9:24 pm I drove through a rainbow once, but no one believes me. I was on I-5 in Redding, CA crossing a highway bridge over the Sacramento River, very low to the water. I was on the phone to my niece and was flabbergasted that the colors actually beamed across my arms and the passenger seat as I went through. I saw the rainbow, of course, but I always thought it impossible to go through one. And before you ask, I was on my way home from working all day as a nuclear medicine technologist, so I wasn't under the influence of anything but fatigue.

Chris Peterson wrote: ↑Mon Mar 13, 2023 4:33 pm

Don M wrote: ↑Mon Mar 13, 2023 3:40 pm Why is the rainbow’s spectral dispersion (i.e., width) larger near the horizon?

I'd guess we're just seeing lens aberration and mapping issues. Typically the pixel scale is larger around the edges of a field than at the center, which makes things look somewhat compressed as you near edges.

Don M wrote: ↑Mon Mar 13, 2023 3:40 pm Why is the rainbow’s spectral dispersion (i.e., width) larger near the horizon?

APOD Robot wrote:
What lies at the end of a rainbow? Something different for everyone. For the photographer taking this picture, for example, one end of the rainbow ended at a tree. Others nearby, though, would likely see the rainbow end somewhere else. The reason is because a rainbow's position depends on the observer.