Fast telescope

[vlaiv]

25 minutes ago, Louis D said:

So, about the photographic equivalent of moving to a larger format sensor and increasing the ISO at the same time while maintaining similar SNR.  For this to actually work, though, the f/10 would need to have a significantly larger image circle than the f/5 if the f/5 were already maxing out the fully illuminated image circle.  I'm thinking of a 4" refractor versus a 6" SCT where this probably wouldn't hold true due to the the rear baffles.

Don't mix ISO into it - it's just numeric conversion factor that will not impact signal levels or SNR.

Above was true even if we made FOV and sampling resolution the same. If we don't take that into account things can get even crazier.

You don't necessarily have to have very large illuminated/corrected circle on F/10 scope for this to be true. This is because one can get dedicated astronomy camera with rather small sensor these days. Sensors of 8-10mm in diagonal are not uncommon.

Couple that with variety of pixel sizes, and you can see that novice astro photographer can find them selves underwhelmed by their result from their very fast F/4.8 scope in comparison to someone else's F/7 or F/8 slow imaging rig using APS-C or full format sensor.

Or you can think of it even like this:

4" F/5 refractor with camera that has 2.4um pixels vs 6" F/9 RC with pixel size 12um size pixels - F/9 is going to "rule" F/5 scope in terms of speed as long as both scopes can fit target in their FOV (but sampling will be different).

 

[Louis D]

44 minutes ago, vlaiv said:

Don't mix ISO into it - it's just numeric conversion factor that will not impact signal levels or SNR.

Above was true even if we made FOV and sampling resolution the same. If we don't take that into account things can get even crazier.

You don't necessarily have to have very large illuminated/corrected circle on F/10 scope for this to be true. This is because one can get dedicated astronomy camera with rather small sensor these days. Sensors of 8-10mm in diagonal are not uncommon.

Couple that with variety of pixel sizes, and you can see that novice astro photographer can find them selves underwhelmed by their result from their very fast F/4.8 scope in comparison to someone else's F/7 or F/8 slow imaging rig using APS-C or full format sensor.

Or you can think of it even like this:

4" F/5 refractor with camera that has 2.4um pixels vs 6" F/9 RC with pixel size 12um size pixels - F/9 is going to "rule" F/5 scope in terms of speed as long as both scopes can fit target in their FOV (but sampling will be different).

 

But, if the same imaging sensor was used in both scopes, would there be any way for the f/10 to rule the f/5?  I guess throughout my thought process, I assumed that the same imaging sensor would be used to eliminate that variable from the equation.  After all, what happens if you stuck that larger sensor from the f/10 example on the f/5 system?  I would assume it would create like-density images more quickly but at a different magnification.

[vlaiv]

2 minutes ago, Louis D said:

But, if the same imaging sensor was used in both scopes, would there be any way for the f/10 to rule the f/5?  I guess throughout my thought process, I assumed that the same imaging sensor would be used to eliminate that variable from the equation.  After all, what happens if you stuck that larger sensor from the f/10 example on the f/5 system?  I would assume it would create like-density images more quickly but at a different magnification.

Yes there is a way for that to happen as well.

It is in part to processing - so it is not down to scope as such - but to the way it is used.

Imagine using same sensor in my above example on 6" F/10 and 4" F/5 scopes. Since we use the same sensor - it is obvious that FOV is going to be smaller on F/10 scope, but let's for argument sake make target fit smaller FOV - it will then fit larger FOV as well, so both scopes will image what we are interested in with single frame (no need for mosaics).

Now we want to match sampling resolution and we decide to bin x3 pixels when using sensor with F/10 scope. That will make pixels x3 larger and paired with x3 longer FL - we will have same resolution in arc seconds per pixel.

What really determines speed of imaging setup is - "aperture at resolution". We again have same resolution but F/10 scope has larger aperture (and x2.25 light gathering surface - so x2.25 more signal).

If you want general "speed number" of the system it would be pixel surface * aperture surface or (pixel_size * aperture)^2 - that will give you comparable numbers whose ratio will give you roughly total integration ratio to get same SNR.

Even if you use large sensor on small fast scope - you can still be faster by using even smaller sensor on slow larger scope - provided that in both cases intended target fits FOV, so difference in FOV size is not important - again it would be matter of aperture at resolution or above given  (pixel_size * aperture)^2.

 

[RobertI]

9 hours ago, ollypenrice said:

The first problem arises when focal length is not constant between optics being compared. The Hyperstar would be an obvious example. Point it at a small target first at F10, then at F2.  Unlike the first situation we are now getting precisely the same number of object photons from the target. So is it faster at F2 than at F10 when there is the same amount of light in each case? Not in any way that's very useful. All that's happening is that the same amount of light is placed onto fewer pixels at F2, producing a tinier but brighter image in the same time.

Brilliantly clear and concise explanation as always Olly.

[ollypenrice]

Woo Hoo, I can see a whole new set of variables appearing here! Just what we need! 

On the left, posh Japanese 4 inch F5. On the right, budget Chinese 4 inch F5. Which is faster? Now we have to look at the image circle (say 80mm on the left, 40mm on the right) and the sensor size. The imager with the small sensor says, 'They are the same.' The imager with the giant sensor says, 'The one on the left is twice as fast.' They are both right if the target needs a 2-panel with the smaller circle and smaller sensor. After all, a larger sensor collects more light, does it not?

It just goes to show that no one number, in isolation from the rest, can tell you what you need to know. (Even Vlaiv's 'aperture at resolution' which is the best candidate so far, in my view.)

Olly

[vlaiv]

27 minutes ago, ollypenrice said:

Woo Hoo, I can see a whole new set of variables appearing here! Just what we need! 

On the left, posh Japanese 4 inch F5. On the right, budget Chinese 4 inch F5. Which is faster? Now we have to look at the image circle (say 80mm on the left, 40mm on the right) and the sensor size. The imager with the small sensor says, 'They are the same.' The imager with the giant sensor says, 'The one on the left is twice as fast.' They are both right if the target needs a 2-panel with the smaller circle and smaller sensor. After all, a larger sensor collects more light, does it not?

It just goes to show that no one number, in isolation from the rest, can tell you what you need to know. (Even Vlaiv's 'aperture at resolution' which is the best candidate so far, in my view.)

Olly

Yes, way to too complicated to formulate a single number, and there will always be that edge case.

Let's say two completely same setups - one at dark site of mag21 and one in LP skies of mag19. Is there a single number that can specify the difference? No - it will depend on chosen target - fainter the target, larger the difference.

Edited December 24, 2019 by vlaiv
typo ...

[Peter Drew]

So many "cans of worms" in AP it's surprising that it doesn't come under Biology.      😄

[Mr Spock]

AP voodoo - even the experts find it confusing... 

[Science562h]

I'm entering the debate a bit late, checking on something.

My simple question: So, is the Canon 50mm f/1.8 lens better for widefield astrophotography than the standard 18-55mm f/5.6 zoom lens? I have been led to believe, that it is, due to, here it goes, "faster" focal ratio? 🐸

It's been recommended, along with stepping it down to 2.8. Reply would be appreciated. 

[vlaiv]

3 minutes ago, Science562h said:

I'm entering the debate a bit late, checking on something.

My simple question: So, is the Canon 50mm f/1.8 lens better for widefield astrophotography than the standard 18-55mm f/5.6 zoom lens? I have been led to believe, that it is, due to, here it goes, "faster" focal ratio? 🐸

It's been recommended, along with stepping it down to 2.8. Reply would be appreciated. 

If you have both lens at 50mm and use them with same camera, then yes - 50mm F/1.8 lens will be faster even when stopped down to F/2.8.

[wookie1965]

On 21/12/2019 at 13:31, Neil H said:

Ok so in lay man terms a fast scope draw light in faster to the mirror then  eyepiece , so this now makes me ask this 

If you have a fast telescope the eyepiece needs to be very good as well ?

BST Starguiders are inexpensive around £50 they will perform great in a F5 scope I know because I had one and still have a couple of BST'S. 

The celestron xcels are the same great eyepieces a little bit dearer but look 9n the second hand market for them. 

[Science562h]

22 hours ago, vlaiv said:

If you have both lens at 50mm and use them with same camera, then yes - 50mm F/1.8 lens will be faster even when stopped down to F/2.8.

Thank's. 

[alacant]

On 21/12/2019 at 10:27, Mr Spock said:

Speed has nothing to do with aperture

Mmm. Not sure. So an 80mm Skywatcher refractor f7 is faster than the 2000mm f10 telescope at La Palma?

Maybe this is a case of the need to specify what we mean by speed.

Cheers and clear skies.

Edited February 9, 2020 by alacant

[johninderby]

deleted

 

Edited February 9, 2020 by johninderby

[Frank the Troll]

On 22/12/2019 at 07:46, Stu said:

 

 

 

Edited February 9, 2020 by Frank the Troll

[Barry-W-Fenner]

I have tried to understand this thread, but ultimately, it has given me a major brain fart...

 

Baz

[Mr Spock]

1 hour ago, alacant said:

Mmm. Not sure. So an 80mm Skywatcher refractor f7 is faster than the 2000mm f10 telescope at La Palma?

f7 is faster than f10. Don't confuse the speed of an optical system with it's ability to put photons on a sensor - that process was explained earlier in the thread. Speed has only one meaning - no need to be more specific than that.

[vlaiv]

54 minutes ago, Mr Spock said:

f7 is faster than f10. Don't confuse the speed of an optical system with it's ability to put photons on a sensor - that process was explained earlier in the thread. Speed has only one meaning - no need to be more specific than that.

I sort of agree with you, but given that it is called speed - a term that, as far as I know, originated from world of photography, and had meaning "speed at which image is formed on a film during image capture" - I think it would be good to point out that although we are talking about F/ratio and we use term "faster" instead of lower or higher - one ought not use it to compare speed at which image can be acquired with said telescopes.

[ollypenrice]

1 hour ago, Mr Spock said:

f7 is faster than f10. Don't confuse the speed of an optical system with it's ability to put photons on a sensor - that process was explained earlier in the thread. Speed has only one meaning - no need to be more specific than that.

But is it? If it were, an f8 telescope of any aperture would be slower than an F5 telescope of any aperture. But my neighbour's F8 telescope saturates his chip in anything over a couple of minutes.

It looks like this:

It's an 80cm F8.

It's fast.

Olly

Edit: More on Marc's project here: http://www.obs-bp.com/album-1774976.html

 

 

Edited February 9, 2020 by ollypenrice

[Captain Scarlet]

My understandings are:

- “fast” or “slow” comes from daylight photography, and I always interpreted it as bigger aperture => “faster” shutter speed and vice versa. Never meant to be anything more than a colloquialism.

- f-number is a useful ratio that in most cases can act as a simplified 2-D substitute for the “solid angle” ingredient of Étendue, which is actually solid angle x an area (either pixel area or optical aperture), and which also is a fundamental property of an optical system.

- Signal to Noise Ratio is ultimately determined by Étendue x sqrt(exposure) time (I think) of the smallest-etendue portion of the optical train.

- for a given aperture and a given object magnification at the eyepiece, a “faster” scope will show exactly the same level of brightness as a slow scope. It is “aperture” that determines object brightness for a given object magnification, not f-ratio.

Happy to be corrected...

Magnus

 

Edited February 9, 2020 by Captain Magenta

[vlaiv]

Ok, I'm ultimately confused here.

I do understand that people use term - telescope "speed" to specify ratio of aperture to focal length and in that sense - there is only one understanding of the term "speed" or better call it F/ratio and it is in fact significant in many aspects of operation of telescope - for visual and for photography.

Only problem that I see with this is using term speed, faster, slower as that implies something that is simply not true, and in that sense yes, many people will confuse it with scope's ability to put photons on the sensor.

But if we agree that we don't call those terms - speed, faster and slower and instead use terms F/ratio, lower and higher, that don't have other significant meaning in this context - then I'm perfectly fine.

Are we on the same page here and we talk about F/ratio and related terms lower and higher and we are not discussing fast telescopes in terms of imaging speed (speed to reach certain SNR at given target resolution) but rather just discussing the fact that there are scopes which F/ratio is low in number - like F/5 and such scopes have certain properties and there are scopes that have F/ratio in for example 6-9 range and there are scopes that have F/15 - focal ratio. These scopes behave differently optically so yes, there is real distinction - just lets not call them fast and slow (I can't really think of reason to call F/5 scope fast - other than in photography context - what, fast to setup? Fast to show aberrations at low cost eyepiece? Fast as fast spending of money on more expensive eyepieces to get decent field )

@Mr Spock - what use of term speed did you have in mind?

[bingevader]

5 hours ago, Barry-W-Fenner said:

I have tried to understand this thread, but ultimately, it has given me a major brain fart...

 

Baz

Me too for the AP stuff!

My very layman understanding from a purely visual perspective is that the f number describes how "bendy" the primary is, whether it be a lens or mirror.

A low f number means that the light has been "bent" more by the primary in order to reach the secondary or eyepiece (such as in short Newtonians with very large mirrors).

A bendier mirror or lens requires more manufacture and so is more expensive.

Light from the outside of the mirror/lens will travel a different distance to the secondary or eyepiece than light from the centre of the mirror/lens and so requires more correction by the eyepiece to focus all the light.

That's why lower f number 'scopes need better corrected and so more expensive eyepieces.

[John]

Fast telescope = one with a cost sufficient that it requires the owner to move very rapidly when their other half finds out

[johninderby]

Now this is a fast telescope. 😁

Edited February 9, 2020 by johninderby

[Mr Spock]

If people stop using a word to mean something different to it's original meaning there isn't an issue