If you have GAS, you'd better sit down.

[J A C S]

Or this might knock you down.

All Canon 400/2.8L II's.

http://www.symmetrymagazine.org/article/99-percent-invisible

There will be a phone that can do that some day.

You mean Canon will install phone modules on those big white lenses?

 

[Szumi]

I hope they don't have to manual focus each of those.

 

[stevo23]

Or this might knock you down.

All Canon 400/2.8L II's.

http://www.symmetrymagazine.org/article/99-percent-invisible

--
Lee Jay

Excellent. Thanks for sharing.

 

[Chimere]

There will be a phone that can do that some day.

Yes, for sure. And we also have individuals who believe the earth is flat.

https://en.wikipedia.org/wiki/Modern_flat_Earth_societies

Chimere

 

[mike703]

Interesting!

So what would be the optical parameters of this beast if you consider it as a single lens (which is effectively is)? The focal length remains 400 mm however many individual lenses you add, but I guess with 24 individual lenses its light gathering power is 24x greater than that of an individual lens... around 4.5 stops. So what does that make its f-number?

Let's see... 400mm / 2.8 = 143mm effective aperture diameter for each lens. WIth 24x the light gathering power that means the (effective) aperture has 24x the area, so the diameter of the aperture would increse by SQRT(24), which makes it 143 x 4.9 = 700mm. That makes the f-number 400/700, or around 0.57. Is that meaningful? And would the depth of field be reduced in the way that you would expect for a wider aperture on an individual lens?

Best wishes

 

[Bill Ferris]

Or this might knock you down.

All Canon 400/2.8L II's.

http://www.symmetrymagazine.org/article/99-percent-invisible

--
Lee Jay

Here's a link to the ApJ Letters article describing the Dragonfly Telephoto Array: http://inspirehep.net/record/1278457

In a nutshell, each 400mm f/2.8 IS II lens has an SBIG (Santa Barbara Instruments Group) CCD camera mounted to it. The multiple camera/lens cluster is affixed to a common mount and the individual images are integrated. According to the authors, the nanocrystal coatings reduce image ghosting by an order of magnitude (~10x) in comparison with the best research-grade reflecting telescopes. As a result, the array is able to image low surface brightness objects as faint as 32 magnitudes per square arcsecond.

To put that in some context, the faintest point sources (stars) routinely visible to the naked eye are about 6th magnitude or about 25 billion times brighter. The Andromeda galaxy (M31), which is among the most distant extended objects (galaxies, nebulae, etc.) visible to the naked eye, is a 3.2 magnitude object with a surface brightness of 22.2 magnitudes per square arcsecond (mags/arcs^2); about 10,000 times brighter than Dragonfly's limit. The lowest surface brightness galaxies which are visually observable with a telescope have surface brightnesses of about 24 mags/arcsec^2 (1600 times brighter) and the faintest extended objects routinely imaged by the most powerful research telescopes have surface brightnesses of about 29-30 mags/arcsec^2.

In short, Dragonfly is able to look where we've not previously been able to look and they're finding stuff not previously found.

--
Bill Ferris Photography
Flagstaff, AZ
http://www.billferris.photoshelter.com

 

[Patrick Thrush]

Aside from the lens, the optical heart of this beast is the image sensor used. It is a Kodak KAF-8300 CCD, which is approximately 18 x 14mm and has a diagonal measurement of about 22 millimeters. This translates to a 3326x2504 active pixel format of around 8.3 Mpx. See the tech details here:

http://www.kodak.com/ek/uploadedFil...lutions/Datasheets(pdfs)/KAF-8300LongSpec.pdf

So, the sensor is about the size of the Olympus 'Four-Thirds' system. The rig these astronomers have built is indeed cooled--each SBIG sensor box is brought down to 14F (or -10C if you prefer ;-) ) to drop the noise. Interestingly, they are using an approach very similar to the 'dust mapping' on a DSLR. Every so often, they blind the sensor box aperature and take a "dark shot" which reveals any noise pattern--then using that map to correct the final image.

How I would have liked to roll up with one of these rigs some night back in the day at a Local Group of Deep Sky Observers star party...

--
Papa Tango
"You are always exactly where you are at, unless of course you are somewhere else... quoted April 18, 2031

 

[Ellis Vener]

Actually it all riding on the backs of fout giant elephants and they are standing on a giant turtle.

 

[Brad Bohland]

Or this might knock you down.

All Canon 400/2.8L II's.

http://www.symmetrymagazine.org/article/99-percent-invisible

There will be a phone that can do that some day.

You mean Canon will install phone modules on those big white lenses?

Maybe. I was thinking about a phone shot of the multi-lens camera image and then posting it on Faceplace.

 

[glowingshutter]

Aside from the lens, the optical heart of this beast is the image sensor used. It is a Kodak KAF-8300 CCD, which is approximately 18 x 14mm and has a diagonal measurement of about 22 millimeters. This translates to a 3326x2504 active pixel format of around 8.3 Mpx. See the tech details here:

http://www.kodak.com/ek/uploadedFil...lutions/Datasheets(pdfs)/KAF-8300LongSpec.pdf

So, the sensor is about the size of the Olympus 'Four-Thirds' system. The rig these astronomers have built is indeed cooled--each SBIG sensor box is brought down to 14F (or -10C if you prefer ;-) ) to drop the noise. Interestingly, they are using an approach very similar to the 'dust mapping' on a DSLR. Every so often, they blind the sensor box aperature and take a "dark shot" which reveals any noise pattern--then using that map to correct the final image.

How I would have liked to roll up with one of these rigs some night back in the day at a Local Group of Deep Sky Observers star party...

--
Papa Tango
"You are always exactly where you are at, unless of course you are somewhere else... quoted April 18, 2031

I wonder if this is really the most efficient use of resources possible. These lenses are $12,000 each, for a total of more than half a million.

The nano crystal coating isn't exactly unique to Canon. I use Micro Four Thirds, and we have a lot of lenses that are nano coated. Some of them are pretty cheap actually. Nikon does something similar. While I do believe that there is a positive effect to it, I wouldn't call it groundbreaking in the overall result, and I doubt it plays a huge role in the cost of these lenses.

Perhaps these Canon lenses have more surfaces coated with with the nano particles than a conventional lens, although I don't know.

Also, if the sensor for each lens is MFT sized, then a lot of potential of the lens goes unused in terms of light gathering, because they're designed for FF sensors. Not to mention that CMOS sensors are probably more efficient now in terms of light gathering. I wonder if there's some reason to be using these specific sensors that I'm not aware of.

The thing that I'm really curious about is if they've compared these modern lenses to some much cheaper legacy lenses to see if there is indeed a major benefit. Or perhaps compare them to cheaper modern lenses with different coatings.

Perhaps they did all of that and indeed these are the best but, damn, that's some serious dough if they haven't tried cheaper lenses first.

 

[Lee Jay]

Aside from the lens, the optical heart of this beast is the image sensor used. It is a Kodak KAF-8300 CCD, which is approximately 18 x 14mm and has a diagonal measurement of about 22 millimeters. This translates to a 3326x2504 active pixel format of around 8.3 Mpx. See the tech details here:

http://www.kodak.com/ek/uploadedFil...lutions/Datasheets(pdfs)/KAF-8300LongSpec.pdf

So, the sensor is about the size of the Olympus 'Four-Thirds' system. The rig these astronomers have built is indeed cooled--each SBIG sensor box is brought down to 14F (or -10C if you prefer ;-) ) to drop the noise. Interestingly, they are using an approach very similar to the 'dust mapping' on a DSLR. Every so often, they blind the sensor box aperature and take a "dark shot" which reveals any noise pattern--then using that map to correct the final image.

How I would have liked to roll up with one of these rigs some night back in the day at a Local Group of Deep Sky Observers star party...

 

[glowingshutter]

Aside from the lens, the optical heart of this beast is the image sensor used. It is a Kodak KAF-8300 CCD, which is approximately 18 x 14mm and has a diagonal measurement of about 22 millimeters. This translates to a 3326x2504 active pixel format of around 8.3 Mpx. See the tech details here:

http://www.kodak.com/ek/uploadedFil...lutions/Datasheets(pdfs)/KAF-8300LongSpec.pdf

So, the sensor is about the size of the Olympus 'Four-Thirds' system. The rig these astronomers have built is indeed cooled--each SBIG sensor box is brought down to 14F (or -10C if you prefer ;-) ) to drop the noise. Interestingly, they are using an approach very similar to the 'dust mapping' on a DSLR. Every so often, they blind the sensor box aperature and take a "dark shot" which reveals any noise pattern--then using that map to correct the final image.

How I would have liked to roll up with one of these rigs some night back in the day at a Local Group of Deep Sky Observers star party...

--
Papa Tango
"You are always exactly where you are at, unless of course you are somewhere else... quoted April 18, 2031

I wonder if this is really the most efficient use of resources possible. These lenses are $12,000 each, for a total of more than half a million.

$12,000 * 24 = $288,000

$12,000 * 48 = $576,000

--
Lee Jay

 

[Gary Martin]

We're gonna need a bigger lens strap.

 

[Lee Jay]

Aside from the lens, the optical heart of this beast is the image sensor used. It is a Kodak KAF-8300 CCD, which is approximately 18 x 14mm and has a diagonal measurement of about 22 millimeters. This translates to a 3326x2504 active pixel format of around 8.3 Mpx. See the tech details here:

http://www.kodak.com/ek/uploadedFil...lutions/Datasheets(pdfs)/KAF-8300LongSpec.pdf

So, the sensor is about the size of the Olympus 'Four-Thirds' system. The rig these astronomers have built is indeed cooled--each SBIG sensor box is brought down to 14F (or -10C if you prefer ;-) ) to drop the noise. Interestingly, they are using an approach very similar to the 'dust mapping' on a DSLR. Every so often, they blind the sensor box aperature and take a "dark shot" which reveals any noise pattern--then using that map to correct the final image.

How I would have liked to roll up with one of these rigs some night back in the day at a Local Group of Deep Sky Observers star party...

 

[glowingshutter]

Aside from the lens, the optical heart of this beast is the image sensor used. It is a Kodak KAF-8300 CCD, which is approximately 18 x 14mm and has a diagonal measurement of about 22 millimeters. This translates to a 3326x2504 active pixel format of around 8.3 Mpx. See the tech details here:

http://www.kodak.com/ek/uploadedFil...lutions/Datasheets(pdfs)/KAF-8300LongSpec.pdf

So, the sensor is about the size of the Olympus 'Four-Thirds' system. The rig these astronomers have built is indeed cooled--each SBIG sensor box is brought down to 14F (or -10C if you prefer ;-) ) to drop the noise. Interestingly, they are using an approach very similar to the 'dust mapping' on a DSLR. Every so often, they blind the sensor box aperature and take a "dark shot" which reveals any noise pattern--then using that map to correct the final image.

How I would have liked to roll up with one of these rigs some night back in the day at a Local Group of Deep Sky Observers star party...

--
Papa Tango
"You are always exactly where you are at, unless of course you are somewhere else... quoted April 18, 2031

I wonder if this is really the most efficient use of resources possible. These lenses are $12,000 each, for a total of more than half a million.

$12,000 * 24 = $288,000

$12,000 * 48 = $576,000

There are only 24 lenses shown in the OP.

Yes, but there's also an article in the OP which states 48. If it were me, I would have read the article or researched it further after someone makes two posts stating that there are 48 (or probably 50 by now), rather than rely on a picture that may not represent the current state of development.

I'm not sure why you're not reading anything about it but rather looking at a picture. ;-)

--
Lee Jay

 

[Lee Jay]

Aside from the lens, the optical heart of this beast is the image sensor used. It is a Kodak KAF-8300 CCD, which is approximately 18 x 14mm and has a diagonal measurement of about 22 millimeters. This translates to a 3326x2504 active pixel format of around 8.3 Mpx. See the tech details here:

http://www.kodak.com/ek/uploadedFil...lutions/Datasheets(pdfs)/KAF-8300LongSpec.pdf

So, the sensor is about the size of the Olympus 'Four-Thirds' system. The rig these astronomers have built is indeed cooled--each SBIG sensor box is brought down to 14F (or -10C if you prefer ;-) ) to drop the noise. Interestingly, they are using an approach very similar to the 'dust mapping' on a DSLR. Every so often, they blind the sensor box aperature and take a "dark shot" which reveals any noise pattern--then using that map to correct the final image.

How I would have liked to roll up with one of these rigs some night back in the day at a Local Group of Deep Sky Observers star party...