View previous topic :: View next topic |
Author |
Message |
Gerald
Joined: 25 Mar 2014 Posts: 1196 Location: Brazil
|
Posted: Sun Jun 22, 2014 3:32 pm Post subject: The Photographic Lens |
|
|
Gerald wrote:
This article discusses the basics of photographic lenses, and the technological evolution of optical glasses and lens design. It was published in Scientific American in August 1976. The author, William H. Price was the successor of Rudolf Kingslake as manager of the Optical Design Department at Kodak.
The author gives Otto H. Schade the merits for the first-time use of the concept of Optical Transfer Function, OTA, and Modulation Transfer Function, MTF. Schade was an electrical engineer who worked for Radio Corporation of America in the 1950s in the development of television systems. In fact, the concept of frequency response has been widely used by radio engineers from the early twentieth century to analyze and design amplifiers, filters, antennas, etc. The analogue television systems, NTSC, PAL and SECAM, are based on image scan, which ultimately transform the spatial variations of the image into an electrical signal with temporal variations. This relation between electrical signal and image led Schade and other researchers to think the image as a combination of sinusoidal variations, which is the foundation of the analysis based on OTF and MTF.
_________________ If raindrops were perfect lenses, the rainbow did not exist. |
|
Back to top |
|
|
Oldhand
Joined: 01 Apr 2013 Posts: 6000 Location: Mid North Coast NSW - Australia
|
Posted: Mon Jun 23, 2014 3:01 am Post subject: |
|
|
Oldhand wrote:
Thank you for a fascinating read.
OH |
|
Back to top |
|
|
calvin83
Joined: 12 Apr 2009 Posts: 7555 Location: Hong Kong
|
Posted: Mon Jun 23, 2014 4:45 am Post subject: |
|
|
calvin83 wrote:
We will have much cheaper photography lenses if we can make cheap aspherical GRIN lens. _________________ https://lensfever.com/
https://www.instagram.com/_lens_fever/
The best lens is the one you have with you. |
|
Back to top |
|
|
kenetik
Joined: 02 Mar 2014 Posts: 521
|
Posted: Mon Jun 23, 2014 5:17 am Post subject: |
|
|
kenetik wrote:
Having curved sensors would simplify lens design considerably, as Kodak achieved bending film plane in the Stretch 35 disposable. Sony is currently making a lot of progress and has manufactured curved sensors in the lab. Of course, if Sony gets there first for consumers, these simplified lenses will cost a boatload of cash initially and be proprietary. _________________ ________________________________________________________
I buy lots of lenses and sometimes I even use them... |
|
Back to top |
|
|
Gerald
Joined: 25 Mar 2014 Posts: 1196 Location: Brazil
|
Posted: Mon Jun 23, 2014 3:19 pm Post subject: |
|
|
Gerald wrote:
calvin83 wrote: |
We will have much cheaper photography lenses if we can make cheap aspherical GRIN lens. |
Aspherical lenses of high quality and low cost are already being done routinely. For example, the zoom lens of the new Panasonic FZ1000 has five aspherical lenses with eight aspheric surfaces. Panasonic has even developed a method of polishing the aspheric molds that eliminates the "onions-rings" typical of the images produced by spherical lenses made by the precision molding method.
http://www.imaging-resource.com/news/2014/05/02/the-end-of-onion-ring-bokeh-panasonic-beats-the-curse-of-aspheric-lenses _________________ If raindrops were perfect lenses, the rainbow did not exist. |
|
Back to top |
|
|
calvin83
Joined: 12 Apr 2009 Posts: 7555 Location: Hong Kong
|
Posted: Mon Jun 23, 2014 4:09 pm Post subject: |
|
|
calvin83 wrote:
Gerald wrote: |
calvin83 wrote: |
We will have much cheaper photography lenses if we can make cheap aspherical GRIN lens. |
Aspherical lenses of high quality and low cost are already being done routinely. For example, the zoom lens of the new Panasonic FZ1000 has five aspherical lenses with eight aspheric surfaces. Panasonic has even developed a method of polishing the aspheric molds that eliminates the "onions-rings" typical of the images produced by spherical lenses made by the precision molding method.
http://www.imaging-resource.com/news/2014/05/02/the-end-of-onion-ring-bokeh-panasonic-beats-the-curse-of-aspheric-lenses |
Sorry. It should be achromatic GRIN lens.
This is a single GRIN lens design.
http://www.opticsinfobase.org/oe/fulltext.cfm?uri=oe-21-4-4970&id=249430
Olympus had try to make a camera lens with GRIN lens in the last 1980s.
http://www.maitani-fan.com/maitani_fan/camera_technologies.html _________________ https://lensfever.com/
https://www.instagram.com/_lens_fever/
The best lens is the one you have with you. |
|
Back to top |
|
|
pd1987
Joined: 08 Dec 2013 Posts: 74 Location: Poland
|
Posted: Mon Jun 23, 2014 4:28 pm Post subject: |
|
|
pd1987 wrote:
That is a magnificent article! Thanks for sharing! _________________ Samples and subjective tests of manual lenses here:
LensCraze.blogspot.com |
|
Back to top |
|
|
Gerald
Joined: 25 Mar 2014 Posts: 1196 Location: Brazil
|
Posted: Mon Jun 23, 2014 6:56 pm Post subject: |
|
|
Gerald wrote:
Oldhand wrote: |
Thank you for a fascinating read.
OH |
pd1987 wrote: |
That is a magnificent article! Thanks for sharing! |
I'm glad you liked it.
Very interesting. In recent decades, the field of optics has evolved gradually, but certainly there are many new ideas being researched that could produce a technological breakthrough.
My intuition says that progress in the next years will come by combining software with hardware. Today, lateral chromatic aberration, distortion and light falloff are already corrected efficiently by software. There are also many researches being made to correct other aberrations (spherical, astigmatism, coma, etc.) through deconvolution.
A difficulty of correcting aberrations by software is that a huge processing power is required. However, when you see that a compact and relatively inexpensive camera like the Panasonic FZ200 can generate high-speed 1280x720 HD video at 240fps, it is clear that the processing capacity of digital cameras has reached spectacularly high levels. Expect this increasing processing capacity will be used more and more to improve the final performance of the photographic lenses. _________________ If raindrops were perfect lenses, the rainbow did not exist. |
|
Back to top |
|
|
calvin83
Joined: 12 Apr 2009 Posts: 7555 Location: Hong Kong
|
Posted: Tue Jun 24, 2014 3:36 am Post subject: |
|
|
calvin83 wrote:
Gerald wrote: |
Oldhand wrote: |
Thank you for a fascinating read.
OH |
pd1987 wrote: |
That is a magnificent article! Thanks for sharing! |
I'm glad you liked it.
Very interesting. In recent decades, the field of optics has evolved gradually, but certainly there are many new ideas being researched that could produce a technological breakthrough.
My intuition says that progress in the next years will come by combining software with hardware. Today, lateral chromatic aberration, distortion and light falloff are already corrected efficiently by software. There are also many researches being made to correct other aberrations (spherical, astigmatism, coma, etc.) through deconvolution.
A difficulty of correcting aberrations by software is that a huge processing power is required. However, when you see that a compact and relatively inexpensive camera like the Panasonic FZ200 can generate high-speed 1280x720 HD video at 240fps, it is clear that the processing capacity of digital cameras has reached spectacularly high levels. Expect this increasing processing capacity will be used more and more to improve the final performance of the photographic lenses. |
You are right. New material can always makes a better lens. The negative-index metamaterials with is able to bend the light in other direction. However, software correction is more easy to be achieved with the increasing processing power of the camera.
I guess we can see what is the performance of the curved sensor and the four element lens from Sony. If Sony can sell the camera well at good price, we can expect the curved sensor become more common in the near future. _________________ https://lensfever.com/
https://www.instagram.com/_lens_fever/
The best lens is the one you have with you. |
|
Back to top |
|
|
JohnBar
Joined: 21 Jun 2012 Posts: 581 Location: Liverpool
|
Posted: Tue Jun 24, 2014 12:12 pm Post subject: |
|
|
JohnBar wrote:
Thanks for posting this Gerald, the article deserves to be a Wiki page by itself.
However recent advances in tech have drawn to the conclusion that the difference between a top end Schneider or Zeiss and a -made in nowhere- lens has always been down to the tolerances (GDT) in manufacture and assembly and the glass used. The materials science and fabrication processes required to produce a specific index to less than 0.1% nominal variance are immense.
Strangely despite all the tech, lenses are using more elements than before, when I think it should be the other way around. I think the dreaded need for AF has brought about this. Perhaps we should now be thinking of moving the sensor to focus a lens as well as controlling shake. _________________ Rectilux 3FF Series single focus anamorphic attachments
http://www.transferconvert.co.uk/cinemania/rectilux-3ff.html
Regular News on https://www.facebook.com/pages/Rectilux/704770636267200 |
|
Back to top |
|
|
ZoneV
Joined: 09 Nov 2009 Posts: 1632 Location: Germany
Expire: 2011-12-02
|
Posted: Tue Jun 24, 2014 2:02 pm Post subject: |
|
|
ZoneV wrote:
JohnBar wrote: |
...However recent advances in tech have drawn to the conclusion that the difference between a top end Schneider or Zeiss and a -made in nowhere- lens has always been down to the tolerances (GDT) in manufacture and assembly and the glass used. The materials science and fabrication processes required to produce a specific index to less than 0.1% nominal variance are immense... |
The way I learnt is melt data fitting.
First lens design is made with the nominal values. With the exact measured data of the glass melt one buys for this lenses one does a second design of the lens, and produce the lens.
So if another batch of lenses is produced with a new melt, the design has to be a bit optimized with the new melt data.
Regarding the new lenses: AF lenses are better internal or rear element focussed, not helicoid focussed. Helicoid focus would need to move too much mass, so slower. Image stabilisation in the lens need some extra for this optimized elements.
And for the microlenses it is better to have a as much as possible telecentric image side - so another reason for a more complex design than for film. _________________ Camera modification, repair and DIY - some links to look through: http://www.4photos.de/camera-diy/index-en.html
I AM A LENS NERD!
Epis, Elmaron, Emerald, Ernostar, Helioplan and Heidosmat.
Epiotar, Kameraobjektiv, Anastigmat, Epis, Meganast, Magnagon, Quinar, Culmigon, Novotrinast, Novflexar, Colorplan, Sekor, Kinon, Talon, Telemegor, Xenon, Xenar, Ultra, Ultra Star. Tessar, Janar, Visionar, Kiptar, Kipronar and Rotelar.
|
|
Back to top |
|
|
beomagi
Joined: 14 May 2010 Posts: 46
|
Posted: Tue Jun 24, 2014 4:26 pm Post subject: |
|
|
beomagi wrote:
Thanks! Made it into a cbr for easy reading |
|
Back to top |
|
|
JohnBar
Joined: 21 Jun 2012 Posts: 581 Location: Liverpool
|
Posted: Tue Jun 24, 2014 4:36 pm Post subject: |
|
|
JohnBar wrote:
ZoneV wrote: |
The way I learnt is melt data fitting.
First lens design is made with the nominal values. With the exact measured data of the glass melt one buys for this lenses one does a second design of the lens, and produce the lens.
So if another batch of lenses is produced with a new melt, the design has to be a bit optimized with the new melt data.
|
Yes Markus, this is true and a popular design method for optical companies who do not have access to foundry facilities and rely upon glass catalogs.
There are many optical parameters in glass and the need to find a perfect balance between these properties including index, curvature and inter-lens distance, a multi variable problem, is sometimes hampered by this approach. Also sometimes two elements of catalog glass are used instead of one of custom glass because of cost strangely enough. _________________ Rectilux 3FF Series single focus anamorphic attachments
http://www.transferconvert.co.uk/cinemania/rectilux-3ff.html
Regular News on https://www.facebook.com/pages/Rectilux/704770636267200 |
|
Back to top |
|
|
Gerald
Joined: 25 Mar 2014 Posts: 1196 Location: Brazil
|
Posted: Tue Jun 24, 2014 6:57 pm Post subject: |
|
|
Gerald wrote:
JohnBar wrote: |
Strangely despite all the tech, lenses are using more elements than before, when I think it should be the other way around. I think the dreaded need for AF has brought about this. Perhaps we should now be thinking of moving the sensor to focus a lens as well as controlling shake. |
The really good modern lenses are better corrected than the lenses of 30, 40 or 50 years ago. This explains why a Zeiss or Sigma Otus Art have 12 or 13 elements, ie a much larger number than the usual 7 elements of the old 50mm F1.4 normal lenses.
One way to reduce the number of elements required for a given level of correction of aberrations is to use aspherical elements. However, for lenses that are produced in small quantities, such as Zeiss Otus, Sigma Art and Leica Summilux, the number of aspherical elements must be limited to just one or two due to the high cost of the precision mold need to make an aspherical lens. In contrast, relatively inexpensive lenses, such as that for compact and M43 cameras, use up to 8 aspherical surfaces. When a lens is mass-produced, the cost of the molds for producing precision aspherical elements is diluted by the large number of lenses.
Finally, another reason for the higher number of optical elements in modern lenses is the use of internal focusing and image stabilization that require additional optical elements.
beomagi wrote: |
Thanks! Made it into a cbr for easy reading |
What is a cbr? _________________ If raindrops were perfect lenses, the rainbow did not exist. |
|
Back to top |
|
|
|
|
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum
|