Interesting new technology - diffractive optics

[bobatkins]

While this isn't an equipment forum and I normally don't like postings

and speculation about unreleased products, I thought I'd make an exception in this

case. Moderators can bend the rules a little - especially for

themselves! This is really more about technolgy than a specific lens.

<p>

Canon have announced

a prototype<a href="http://www.canon.com/do-info/index.html"> diffractive optics

400/4L IS lens </a>to be shown for the first time at Photokina in the

fall. The lens won't be avaialbe for sale until 2001. I think this is one of the most interesting new technologies to

hit nature photography since autofocus. If it works out, those big

heavy 600/4 lenses may be a thing of the past, to be replaced by

smaller, lighter 600/4 lens - and, of course, significantly ligher

wallets for those who like to chase technology (and cheap 2nd hand 600/4 IS

lenses for the rest of us...).

<p>

This also might answer the "crystal ball" question of why Canon don't make

a 400/4, which has been asked here more than once (and usually deleted!). Now we

know.

[cliff_lesergent]

Oh, great... just when I thought it was safe to thaw out my credit card...

[andrew_schank]

I've been around cameras and photography nearly my whole life, and never once heard anything about this type of optical devise. Makes you wonder what else is possible. I can imagine that those lenses will redefine the word expensive.

[bobatkins]

At this point speculation on cost is brobably a waste of time. They could be cheaper or more expensive to produce, it's hard to tell. The

materials may be cheaper than fluorite for example, but the time, effort and expense of producing them may be greater. Of course being the only game in town for a while may mean they'll sell at a premium that doesn't refelect their intrinsic cost.

 

Diffractive optics are quite well known, but usually they have

been used at longer wavelengths (e.g. infrared, microwaves).

[Former P.N Member]

Adding fuel to the fire regarding cost of the 400mm, DO, IS, USM the following is submitted for consideration (it's a posting to the EOS thread on eGroups):

 

"I hate empty rumours, just curious whether abybody else heard any thing about a 400/4.0 IS lens to be released shortly. A friend of mine (converting to Canon) spoke to Canon S-Africa yesterday, asking for advice on achieving 800mm while retaining AF. They replied that 'tight now, options are limited to the 600/4', but phoned him back later indicating that 'in approx a month's time a new 400/4.0 IS will be released'. Cost was indicated as approx R25000, or US$3500."

[dan_smith]

And where is Nikon while this stuff is going on? Watching from the cheap seats during the news conference, furiously writing down the information so in another 5 years they can introduce this 'latest breakthrough' as their own?

Canon was 5 years ahead of Nikon with big, fast AF glass. They beat the Nikon by a couple years with IS lenses. Now, will Nikon get off their collective duff & actually produce something like this? Or will they buy Canon lenses & a few cans of black spray paint & then re-badge the damn things & say that they were actually first with them? And, as many know I shoot Nikon, but somehow the "worlds greatest picture taking company" seems to be eating Canon's dust lately.

But, who can afford to change an entire system? Also, the older manual bodies I can still use in subzero or 130 degree heat keep me with the "leader in following Canon technology".

O'well, maybe someday Nikon will once again make something before Canon does.

[Former P.N Member]

Dan,

 

This is also from an eGroups thread:

 

"check out

http://www.klt.co.jp/Nikon/New_Products/index.html and scroll down to look

what they've done with four of their lenses. I don't know if "Imitation is

the best form of flattery", or "If you can't lick 'em, join 'em" applies, or

it's just a shameless rip-off. Obviouly though, Nikon is getting tired of

all the white lenses on the sidelines of sporting events among other things."

[peter_wagemans]

Why do they use the term "diffractive"? Canon's own footnote describes

what they mean by this:

 

<P>

 

> Diffraction refers to the tendency of light to spread around the

edge of an opaque obstruction.

 

<P>

 

Given this description, does anyone understand why Canon uses this

term to describe the new optical element? To me the picture looks like

a double Fresnel lens construction, i.e., refractive like normal

lenses: bending the light by passing it through slanted surfaces.

 

<P>

 

Also I would be worried about light scattering on the ridges,

especially for slanted light rays (at an angle to the optical axis),

to which the ridges represent a larger part of the total area. So will

the flare indeed be minimal if you have strong lights off center? I

suppose we'll have to wait for actual tests. Or can any optics experts

shed an early light on this?

 

<P>

 

For the same reason I suspect it will be difficult to use these new

elements with shorter focal lengths: much more light at large angles

to the optical axis.

[jeff_drew4]

Considering the things "they" can do directly for our eyes to correct and improve things, why not logically change lenses? Astronomy is seeing decades of advances whittled down to years due to advances in seeing. Awesome!

[bobatkins]

I was a little confused by the Canon diagram, which does look like a Fresnel lens (not srictly a diffractive optics, but sort of similar). True diffractive optics operate via diffraction and interference and require the use of wave optics to describe their properties. Frsenel lenses, though looking somewhat similar, can be analyzed with ray optics. My guess is that the Canon diagram is an attempt to explain things in a simple way to a general audience. Concepts of diffraction, interference and wave theory are more difficult to explain on a simple level.

 

As for speculation about flare, again a bit pointless without any data. I presume Canon wouldn't produce an L series telephoto lens with any significant flare problems.

[phil_armitage1]

From a technical perspective I think that this is quite impressive -

diffractive optics have rarely been used in broadband applications,

let alone consumer products (even rather expensive ones).

 

As others have said, common sense suggests that this lens won't be

marketed unless its flare characteristics are very good. Still,

it will be interesting to see if any differences with conventional

lenses are discernable, since in principle one might expect rather

different problems from this kind of optics.

[ted dean]

About five years ago there was a press relase from a defense company

touting liquid lens optics that were supposedly far cheaper and better

for long teles. Supposedly $150.00 worth of material cost vs $1500.00

in a typical 300 2.8. It was said to be much sharper and lighter since

one of these elements replaced many ED or Fluorite pieces. If I

remember corrrectly the technology was "liberated" from Russia. Does

anybody know anymore a bout this? It seemed to have similar benefits

as this new Canon techno

[peter_wagemans]

Thanks Bob. I also realised last night that the diagram from Canon is

much too coarse. It got me thinking in the wrong direction. I suppose

that in fact it is a very fine grating.

 

<P>

 

Rather loosely speaking, I suppose the steps are angled so that each

individual small step bends the light in approximately the right

direction to a beam with a broad peak (diffraction), and the

interference between those beams narrows it down to a single small

spot.

[craig_andrew_yuill]

Oh, the horror! I conceived about using fresnel-type lens elements as a way to reduce camera-lens size back in the early 80s. Now someone has come up with a practical way to do this. And, like so many other things, I get no credit for the concept. Blast! ;-)

[leo s.]

<<Astronomy is seeing decades of advances whittled down to years due to advances in seeing. Awesome!>>

 

Another new technology called "adaptive optics" which has been used in some of the largest and latest optical telescopes is currently being developed for the consumer market.In the large telescopes its used to correct aberrations caused by looking through the atmosphere,but its been suggested that the same technology could be used to design a device which surpasses 20:20 vision and could be worn like eye-glasses....Im sure this same technology could be used to radically advance lenses...though we might have to wait a while before its out of the R&D stage,and on the store shelves.

[bill_meyer]

This is a dream come true for the nature photographer. A medium telephoto with a fast aperture, that is both shorter, lighter and more compact than anything else available in it's class. Add IS technology on top of this and could you have a 400mm lens that is capable of outstanding image quality while being hand held? What about with a 1.4x tc, could someone feasably hand hold a shot at 560mm and f/5.6, sure you lose a stop of light but at 7 lbs lighter than the 600 f/4 IS? Of course, as always the best image quality will come with a tripod, but the possibilities are exciting. The lens is only a 1/2" longer than the current 300mm f/4 IS and only 26 oz. heavier which is almost 2 lbs. lighter than the 300mm f/2.8 IS! If the one poster's information is true, that the lens will cost around $3500, this could revolutionize the whole telephoto lens market; considering the current 300mm f/2.8 IS runs about $5800. I for one am very excited about this lens and will be looking for it next year.

 

Also, Canon compares this lens to a standard 400mm f/4 with only refractive elements. Interesting, Canon doesn't make a 400 f/4 lens, could it be that Canon will offer both lenses next year? And if so, does that mean the DO lens will actually be quite a bit more expensive? I guess we'll all have to wait and find out.

[pixseal]

Some info about liquid lens technology: <A HREF="http://beta.atc.lmco.com/opticalsciences/UnderstandingLLT.html">click here</A><P>

In 1994, Lockheed entered into an exclusive licensing agreement with Panavision International for use of their liquid lens technology in cinematographic and High Definition Television camera and projector lenses. I have no knowledge of whether such lenses made into commercial use.

[edward_de_bruyn]

A question about all this is that these improvements may be a little late, why:

 

The digital world is coming with improvements at a high speed

 

Newer CMOS or CCS now 3.3 Mega pixels future maybe 6.0 , 10.0 Mp. or more and why not

with even better contrast balance ( already now they seem to be better than a 100 ASA XXchrome).

 

Storage ( Mavica ) on CD-r 150 Mb and later on DVD ( very soon ) or hard disc ( see IBM microdrive 1 Mb )

 

So maybe in few months, years we will have a Canon Dxxx.. (Nikon.., Sony.., Kodak.., Fuji.. or whatever ) with storage CMOS or CCD with 10 megapixels and be able to shoot sharp images at 800 ASA ( a 500 F5.6 and even F8 will be OK)

 

See website : http://www.dpreview.com/

[konrad_beck1]

"The digital world is coming with improvements at a high speed ... shoot sharp images at 800 ASA"

 

and thus no lenses but only pinholes will be necessary?

[edward_de_bruyn]

My contribution was only a reflection of question which came up to me, sorry

And Konrad a 500 mm F5.6 is a lens ore not ???

 

What I mean is that a 300 mm F4 will do the job on a digital body : 300 x 1.5 = 450 mm lens !!

The Mavica1000 has a 40-400 mm Carl Zeis zoom with IS but it is only a 2 megapixel

The new improvements are maybe already in use in this type of combinations

Anyway I�m still using a 500 F4 mm 4Kg lens and a reduction of 36% + 10 cm reduction would be great but to late for me.

[cliff_lesergent]

Canon Canada had meetings with our local Canon dealers here in Calgary yesterday, and no mention was made of either of the new 400mm f4 lenses. When someone asked about it, the marketing manager apparently turned pale and didn't say a word. I guess they were told to keep it secret, and were shocked that someone had found out about it. They were even more shocked to find out that it had been released on their own corporate web site three days earlier! So much for finding out prices.

[alex_lofquist]

Until they tell us more, it looks like a "blazed" zonal plate.

[struan_gray]

It is always fun to speculate when research technologies will crop up

in photography. My money was on gradient index lenses, which do many

of the same things as diffractive elements, but which are more tightly

tied up by a few key patent holders. Rodenstock signed exclusive

agreements to put them into their graphics arts and scanning lenses a

couple of years back, but I haven't heard anything since.

 

The difference between a diffractive optic and a fresnel lens is that

the grooves are of the order of a wavelength in size in the DO

element. The light rays are thus bent by diffraction rather than

refraction (you don't say :-), although when used in transmission the

total light-shaping effect is a combination of both. Unlike a zone

plate a DO element modulates the phase of the light rather than its

amplitude, and so in principle it doesn't reduce the intensity.

 

Existing photographic super-teles are already very good - so good that

film and operator error almost always limit optical performance.

Although diffractive elements can give an improvement in pure

technical image quality, I suspect the advantages to photographers

will lie elsewhere.

 

The Canon web page shows the reduced size and weight compared to a

normal refractor. Diffractive elements let you bend light through a

bigger angle without getting bad chromatic aberrations, and so you can

reduce the overall length of the lens. DOs are also typically thinner

than the equivalent set of refractive elements, so they reduce the

weight too. Sadly, the lens' diameter is set by the maximum aperture

and the desired resolution at the focal plane, so there's not a lot of

room for improvements there.

 

The second major advantage I see is that diffractive optics reduce the

total number of elements which have to be ground, assembled and

aligned. The lens is easier to make, and so super-teles with DO

elements may actually be cheaper than existing designs. Gradient

index technology has brought down the price of high-quality microscope

objectives in this way.

 

It is interesting that Canon do not state whether the DO element is

all-glass, or if it is a hybrid resin-glass design. The former would

be a technical tour-de-force, so I suspect the latter, in which case

one possible downside will be the resin's long term optical and

mechanical stability. It might also add to the stated difficulty of

controlling scatter and flare - a known downside of optical resins.

 

I don't plan to buy a long tele in 35 mm any time soon, but if I did

Canon really does seem to have the technical edge. Kudos to them.

[cliff_lesergent]

Olympus is, or was, one of the companies working on gradient index lenses. They built a prototype 85mm f1.4 GRIN lens back in the early '80's.

[craig_andrew_yuill]

I'm beginning to doubt whether long-term durability of a lens' elements is such a big deal anymore, especially since the camera manufacturers seem to be making their products obsolete every few years or so. Nikon recently released a bunch of AF-S lenses, and now it looks like they will have to replace them with VR lenses so that they don't get snowed under by Canon. Canon recently introduced IS lenses. Canon photographers will undoubtedly want to switch over to the much smaller and lighter DOE lenses. Will this never end?