Jump to content

Narrowband

Care to check my F ratio visual aids?


ollypenrice

Recommended Posts

This comes up so often, and I always cover it when helping beginner guests with AP, that I thought I ought to sit down and do a bit of 'worksheet' production. (Done a bit of that in my time as a teacher.  :rolleyes: ) So be free with the criticism and if I've got anything wrong jump in and tell me. Any suggestions welcome.

The first one shows why fast F ratios are fast and why F5 is four times faster than F10, for instance.

F%20RATIO%20AT%20FIXED%20FOCAL%20LENGTH%

The second looks at focal reducers and why they don't work for small targets which will fit on the chip at native focal length. (The dreaded and controvertial F Ratio Myth.)

THE%20F%20RATIO%20MYTH-L.jpg

The third shows what focal reducers do offer when the imager wants everything that fits into the widened field of view.

reducers%20used%20properly-L.jpg

These are not meant to stand alone as explanations but accompany a verbal explanation simply as visual aids.

Cheers,

Olly

Link to comment
Share on other sites

I'm probably being incredibly dense, but isn't the whole point of scenario 2 that you are getting more M33 photons per pixel, resulting in a brighter image, faster? True, this is at the expense of resolution, but no more so than scenario 3 surely. [Though it would not be beneficial when the object is very small (sub-pixel) in size (e.g. trying to image a very faint "asteroid" - or whatever we're supposed to call them nowadays!). In that case you would already have all the photons on one pixel (theoretically!).]

Link to comment
Share on other sites

I'm probably being incredibly dense, but isn't the whole point of scenario 2 that you are getting more M33 photons per pixel, resulting in a brighter image, faster? True, this is at the expense of resolution, but no more so than scenario 3 surely. [Though it would not be beneficial when the object is very small (sub-pixel) in size (e.g. trying to image a very faint "asteroid" - or whatever we're supposed to call them nowadays!). In that case you would already have all the photons on one pixel (theoretically!).]

Yes, a reducer puts the same number of M33 onto fewer pixels and 'fills' them faster to give a final M33 of reduced size/resolution but improved S/N ratio. However, won't you get the same effect if you downsize the unreduced image in software so that M33 is the same size as the one from the reducer? Some call this 'software binning.'

It seems to me that the M33 information received on the unreduced chip is of a higher order than that received on the reduced chip because, if you like, the reduced chip is collecting unwanted information from the outlying sky as well. This information will be discarded by cropping. So there is more M33 information arriving unreduced. (This is information we would call resolution, I guess?) However, the increased information is also more highly contaminated by noise due to the relative 'photon starvation' of each pixel.

Olly

Link to comment
Share on other sites

Yes, a reducer puts the same number of M33 onto fewer pixels and 'fills' them faster to give a final M33 of reduced size/resolution but improved S/N ratio. However, won't you get the same effect if you downsize the unreduced image in software so that M33 is the same size as the one from the reducer? Some call this 'software binning.'

So does this mean that if I (1) collect 5hrs of data at F5, and then discard the outer 50% of the image, and (2) collect 5hrs of data at F10, and then reduce the full image to 50% of its size, that I would end up with two identical photos (subject to non-system related vagaries, like "seeing")?

If that is the case, then I can certainly see there could be advantages to removing 26mm (not to mention lots of additional glass surfaces) from the optical path.

Link to comment
Share on other sites

So does this mean that if I (1) collect 5hrs of data at F5, and then discard the outer 50% of the image, and (2) collect 5hrs of data at F10, and then reduce the full image to 50% of its size, that I would end up with two identical photos (subject to non-system related vagaries, like "seeing")?

If that is the case, then I can certainly see there could be advantages to removing 26mm (not to mention lots of additional glass surfaces) from the optical path.

That's about it so far as I can see. The same information comes from M33 in my example but the reducer discards some of the information on resolution. However, it also overwhelms some of the noise. The precise details of how this will work out will be governed by the level of uncalibrated noise and the effects of the additional glass, both of which are of a practical nature and need to be measured.

Stan Moore discusses the F ratio myth at length in the first chapter of Lessons From the Masters. He agrees that, optically, it is indeed a myth but suggests that real world camera effects can sometimes lend it a little genuine substance. In the end I think that the way we would all want to test this would be by comparing a downsized, unreduced imaged with a reduced one. (Reversing that test - upsizing the reduced image to the size of the unreduced one - would be asking the reduced one to reveal details it didn't record in the first place. That is quite a telling difference in its own right...)

My purpose in putting together this handout was to show that, detailed adjustments aside, F ratio should not be considered in a context in which focal length doesn't appear. For example, Starizona's website says that the Hyperstar makes imaging up to 28 times faster than imaging at F10.  (Their emphasis.) This isn't a lie, it's perfectly true, provided we are clear that F10 and F2 don't image the same thing.

I've incorporated my graphic efforts into a Word Doc with text should anyone like a copy. Just PM me.

I may have some M42 data reduced and not reduced. If so I'll work on a comparison. Or if anyone else has similar data and would like to share it that would be instructive.

Olly

Link to comment
Share on other sites

My purpose in putting together this handout was to show that, detailed adjustments aside, F ratio should not be considered in a context in which focal length doesn't appear. For example, Starizona's website says that the Hyperstar makes imaging up to 28 times faster than imaging at F10.  (Their emphasis.) This isn't a lie, it's perfectly true, provided we are clear that F10 and F2 don't image the same thing.

Olly

I'm paraphrasing here so I may have read this out of context but coming from a photography background (amateur hobbyist that got into Astro later), from my understanding F2 and F10 will image the same thing, providing you are using a fixed focal length "lens" or telescope. I understand that fitting a reducer will lower the focal length and increase aperture value, or F stop number. However from the statement quoted, it appears that you are saying F2 and F10 will inherently image different things. They wont. F2 will just capture 28 times more light in a set time limit than f10 will.

That's my understanding anyway so I'm happy to be corrected! :)

Phil

Link to comment
Share on other sites

Phil

No, the aperture of the telescope will always remain the same.

The primary mirror of an SCT is (for example) f/2, but the secondary mirror alters the angle the light travels, so that it is lengthened by a factor of (for example) 5, making f/10. The hyperstar replaces the secondary mirror and collects the photons at the f/2 focal point.

A focal reducer operates similarly in bending the cone of light, making it wider, before entering the camera.

So same aperture, smaller f#, more photons collected (faster), but from a wider area.

Link to comment
Share on other sites

I'm paraphrasing here so I may have read this out of context but coming from a photography background (amateur hobbyist that got into Astro later), from my understanding F2 and F10 will image the same thing, providing you are using a fixed focal length "lens" or telescope. I understand that fitting a reducer will lower the focal length and increase aperture value, or F stop number. However from the statement quoted, it appears that you are saying F2 and F10 will inherently image different things. They wont. F2 will just capture 28 times more light in a set time limit than f10 will.

That's my understanding anyway so I'm happy to be corrected! :)

Phil

I knew as soon as I'd posted this that I'd allowed that ambiguity to creep in! When I said, 'provided we are clear that F10 and F2 don't image the same thing  ' it was in a sentence about the Hyperstar conversion which does change the focal length and so the image taken. An F10 C11 does not take the same picture as an F2 C11. Sorry about that. In the case of camera lenses then you are quite right, the F stop controls the aperture and not the focal length. That's the job of a zoom lens.

DP, a useful term to come back to is 'object photons.' If the object in question will fit on the chip unreduced then there are no new object photons delivered via the reducer.

Olly

Link to comment
Share on other sites

I really like the pictures, this is one of those subjects that can be very confusing especially when faced with a bunch of calculations.

I would say there is nothing wrong with using a reducer for reasons other than F ratio though, if you are over sampled it could actually be very beneficial. Hmm maybe we need nice pictures for pixel scale too [emoji6]

Sent from my iPad using Tapatalk

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue. By using this site, you agree to our Terms of Use.