F ratio - again!

[Ags]

13 hours ago, ollypenrice said:

Who wins?

Telescope mass increases as the cube of aperture, so Noah (focal reducer) wins, for me anyway.

[ollypenrice]

3 minutes ago, Ags said:

Telescope mass increases as the cube of aperture, so Noah (focal reducer) wins, for me anyway.

Get down to the gym!!!

lly

[johnfosteruk]

1 minute ago, ollypenrice said:

Get down to the gym!!!

lly

Aha, the Gym owner is the winner.

[Ags]

14 hours ago, ollypenrice said:

an invention of her own, an Aperture Increaser

Aperture increaser... better known as a Credit Card?

[DaveS]

No, second mortgage lol

[Adam J]

On 05/05/2021 at 21:49, ollypenrice said:

OK, I'm a former teacher so 'how to teach things' is a bit of an obsession. How about this:

Noah has an 8 inch F10 telescope and wants to speed up his image capture time. He buys a 0.5x focal reducer to bring his telescope to 8 inch F5.

His sister Jemima has the same scope but she's opinionated and ignores the 0.5 focal reducer option in favour of an invention of her own, an Aperture Increaser , which also takes her scope to F5.

Who wins?

👹lly

You have not defined your performance measures.

However they will both achieve the same SNR in the same amount of time. 

But Jemima may not fit the object into the FOV of her sensor assuming the same sensor. One the other hand if the object does fit into the FOV of Jemima's scope Noahs may be under sampled by comparison depending on seeing. If Jemima is oversampled then she may bin in HW or SW to increase SNR or Perceived SNR. 

If they both have the same FOV and optimal sampling then Jemima will win. 

But one thing is for certain Olly and that is that Jemima will need significantly deeper pockets that NOAH and so theory aside most will be better off...(literally) going with Noah's solution and accepting the potential trade off in terms of resolved detail. In a nut shell that is why most people use a reducer. 

You say above that you don't increase her focal length. Fine but you did reduce Noah's So if they are both to be F5 Jemima must have a longer effective focal length than Noah.

The other big issue with this is that while you can get scopes with almost any focal length you want you can't get cameras with any pixel size and sensor size you want. Hence you can't optimise your FOV and image scale to fully match your scope. If you could and kept image scale and FOV the same between Jemima's and Noah's imaging setups she would always win....but you can't and more often than not the camera is the same camera as it's one of only very few available, more so if mono. 

Adam

 

Edited May 6, 2021 by Adam J

[Alien 13]

It depends somewhat on the type of scope too, if its a frac then changing the f/ratio by increasing aperture would be disastrous without also increasing focal length too and by more than double, you might be able to do this with a mirror based scope though.

Alan

[PadrePeace]

27 minutes ago, Adam J said:

You have not defined your performance measures.

However they will both achieve the same SNR in the same amount of time. 

But Jemima may not fit the object into the FOV of her sensor assuming the same sensor. One the other hand if the object does fit into the FOV of Jemima's scope NOAH may be under sampled by comparison depending on seeing. If Jemima is oversampled then she may bin in HW or SW to increase SNR or Perceived SNR. 

If they both have the same FOV and optimal sampling then Jemima will win. 

But one thing is for certain Olly and that is that Jemima will need significantly deeper pockets that NOAH and so theory aside most will be better off...(literally) going with NO AS solution and accepting the potential trade off in terms of resolved detail. In a nut shell that is why most people use a reducer. 

You say above that you don't increase her focal length. Fine but you did reduce Noah's So if they are both to be F5 Jemima must have a longer effective focal length than Noah.

The other big issue with this is that while you can get scopes with almost any focal length you want you can't get cameras with and pixel size and FOV you want. Hence you can't optimise your FOV and image scale to fully match your scope. If you could and kept image scale and FOV the same between Jemima's and Noah's imaging setups she would always win....but you can't and more often than not the camera is the same camera. 

Adam

 

What he just said! Yes… absolutely what he just said 😵‍💫 

 

[vlaiv]

It depends on cameras used and criteria of winning.

If they use the same camera and they use same processing method (ie - same pixel binning if any) and criteria for winning is SNR achieved in set amount of time or time spent to reach set SNR on the same target - provided that both capture intended target completely and don't need mosaics due to difference in FOV, then:

It is a tie!

[JeremyS]

21 hours ago, ollypenrice said:

OK, I'm a former teacher so 'how to teach things' is a bit of an obsession. How about this:

Noah has an 8 inch F10 telescope and wants to speed up his image capture time. He buys a 0.5x focal reducer to bring his telescope to 8 inch F5.

His sister Jemima has the same scope but she's opinionated and ignores the 0.5 focal reducer option in favour of an invention of her own, an Aperture Increaser , which also takes her scope to F5.

Who wins?

👹lly

They are both losers. They should have used a small refractor. Preferably a Tak.

(My assumption is that Noah, like most people, couldn’t afford an 8 inch OG)

 

😊

[Alien 13]

13 minutes ago, JeremyS said:

They are both losers. They should have used a small refractor. Preferably a Tak.

(My assumption is that Noah, like most people, couldn’t afford an 8 inch OG)

 

😊

Didn't think small Taks were any good for imaging without replacing the focusser and using a tiny sensor 😀

Alan

[wimvb]

The light gathering ability (efficiency) of an imaging system is proportional to (p/F)^2, where p is pixel size, and F is the F-number. This can be rearranged to (r*D)^2, where r is pixel scale (arcsecs / pixel) and D is aperture.

When @gorann and I processed images from the Liverpool telescope, we didn't need lots of data because a slow system , F=10 was combined with large pixels (30 um). When Göran switched to a RASA, he used a more common pixel size with a fast system. Again, he didn't need much integration time.

The optical resolution of the system is an entirely different matter.

[Whirlwind]

1 hour ago, wimvb said:

The light gathering ability (efficiency) of an imaging system is proportional to (p/F)^2, where p is pixel size, and F is the F-number. This can be rearranged to (r*D)^2, where r is pixel scale (arcsecs / pixel) and D is aperture.

When @gorann and I processed images from the Liverpool telescope, we didn't need lots of data because a slow system , F=10 was combined with large pixels (30 um). When Göran switched to a RASA, he used a more common pixel size with a fast system. Again, he didn't need much integration time.

The optical resolution of the system is an entirely different matter.

I'd agree.  I've always thought that an imaging area per "/pixel is a more useful representation as to how 'fast' a system (as above) as there are other ways to increase the speed of your system through simple binning (excusing any noise differences from doing so)

[Chris]

This is such an open ended puzzle Olly has put forward 😆 Who wins? I guess that depends on what assumptions we make! I assumed it was DSO imaging because this is Olly we're talking about here, and I also assumed the camera was the same, so the same arcseconds per pixel and same binning etc. 

 

[wimvb]

11 minutes ago, Chris said:

This is such an open ended puzzle Olly has put forward 😆 Who wins? I guess that depends on what assumptions we make! I assumed it was DSO imaging because this is Olly we're talking about here, and I also assumed the camera was the same, so the same arcseconds per pixel and same binning etc. 

 

If the camera is the same, it's only pixel size in micrometers that is the same, hence (p/F)^2. Pixel scale still changes with focal length.

The interpretation of the formulas is that if you change aperture, you change the collecting opening of your system. Whereas, if you change the focal length, you change the sky area that is covered by each pixel. Shorter focal length means that each pixel collects light from a larger patch of sky. Larger aperture means that the photons coming from a patch of sky are collected by a larger funnel. Either way, you get more photons on each pixel. And also, either way, you decreased the F-number of your system.

Edited May 8, 2021 by wimvb

[Mr Spock]

See how much simpler visual is? Versus an f10 scope, an f5 of the same aperture means shorter tube, lower magnifications, wider fov - just use 2 as the divisor/multiplier 😜

[wimvb]

 

4 minutes ago, Mr Spock said:

See how much simpler visual is? Versus an f10 scope, an f5 of the same aperture means shorter tube, lower magnifications, wider fov - just use 2 as the divisor/multiplier 😜

What if you swap the eyepiece? 

Edit: unintended doubling of the magnification

Edited May 8, 2021 by wimvb

[Mr Spock]

Yes, 200mm f10, 10mm eyepiece = x200; 200mm f5, 5mm eyepiece = x200. My maths are so awesome!

I suppose now we need to explain to imagers what an eyepiece is... 😜

[andrew s]

For what's worth the pros use etendue to define their telescopes/instruments.

Roughly, this is aperture ^2 x field of view for imaging instruments.  They are normally intent on getting as wide a field as deep as possible in the shortest time. They the make a detector to match. Focal length comes into play for field size but also corrected field size depending on optical design. For example a corrected Newtonian compared to a standard one of the same aperture and focal length.

So it's not just aperture but also the size of the corrected field  and then the detector to match seeing and field size.

You can disregard the elements that don't  apply to your imaging aims e.g. small field for planets and most galaxies. 

Regards Andrew 

PS Remind me what is an eyepiece? 

PPS @ollypenrice your a very naughty boy.

[wimvb]

40 minutes ago, Mr Spock said:

I suppose now we need to explain to imagers what an eyepiece is... 😜

No not at all, those are the things we take off a telescope when we put a camera in place.

Oh wait a minute, those are focuser caps.

No, no, now I got it. It's those things you put in the holes in that plate under the mount.

(Now you see why imagers want a pier.)

[gorann]

24 minutes ago, wimvb said:

No not at all, those are the things we take off a telescope when we put a camera in place.

Oh wait a minute, those are focuser caps.

No, no, now I got it. It's those things you put in the holes in that plate under the mount.

(Now you see why imagers want a pier.)

Yes, I never understood why those things needed to plug the holes in the plate have to be so expensive. And after you dropped one on the ground it tends to make an irritating rattling sound. A pier is much tidier.

[Chris]

2 hours ago, wimvb said:

If the camera is the same, it's only pixel size in micrometers that is the same, hence (p/F)^2. Pixel scale still changes with focal length.

The interpretation of the formulas is that if you change aperture, you change the collecting opening of your system. Whereas, if you change the focal length, you change the sky area that is covered by each pixel. Shorter focal length means that each pixel collects light from a larger patch of sky. Larger aperture means that the photons coming from a patch of sky are collected by a larger funnel. Either way, you get more photons on each pixel. And also, either way, you decreased the F-number of your system.

Yes indeed, pixel scale is a function of focal length (which is why we need match cameras to telescopes)I did have this in mind, only I thought they were playing with the same focal length. Having looked at the question again, Noah has 1000mm (2000mm x 0.5 with the reducer) and Jemima has the native 2000mm

I'm still not sure who the winner is? is it the person who's arcseconds per pixel best matches their seeing conditions 🤔

If it's Jemima, I hope she wins a massive mount and obsy for that telescope! 😃 

[wimvb]

1 hour ago, Chris said:

I'm still not sure who the winner is?

Jemima. She doesn't lose pixel scale or field of view. Noah does. But they both can get away with shorter exposure times because of the gain in F-ratio.

[vlaiv]

34 minutes ago, wimvb said:

Jemima. She doesn't lose pixel scale or field of view. Noah does. But they both can get away with shorter exposure times because of the gain in F-ratio.

Aha! F-ratio myth!

What if there is a third person with the same scope and same camera that uses neither focal reducer nor "aperture amplifier" and simply bins their pixels x2.

All three of them now are a tie in terms of time to set SNR and all three gain over stock setup - but only two of them due to gain in F-ratio. How can we explain third person's improvement with original F-ratio if F-ratio is what determines speed?

[wimvb]

3 hours ago, vlaiv said:

Aha! F-ratio myth!

What if there is a third person with the same scope and same camera that uses neither focal reducer nor "aperture amplifier" and simply bins their pixels x2.

All three of them now are a tie in terms of time to set SNR and all three gain over stock setup - but only two of them due to gain in F-ratio. How can we explain third person's improvement with original F-ratio if F-ratio is what determines speed?

Yes,

but (p/F)^2 is still the same for all three. Noah and Jemima adjusted the F ratio, each in their own way. The third person adjusted the second parameter: pixel size.

Edited May 8, 2021 by wimvb