Sensor Breakthrough--Where Are We Headed?

[fotografr]

Advertisement (gone after registration)

http://www.blogcdn.com/www.engadget.com/media/2008/08/iphone_telescope_mtlsc003300_01_l.jpg - $19 and available now!

 

That looks pretty cool.

[ho_co]

Hey, guys, there are a lot of good points above, but I think you're missing the main point: Sensuality.

 

Bits Blog speaks of "a technology called BSI, or Back Side Illumination."

 

And did you see the Deutsche Welle CES article a few days back (CES - das Gadget-Paradies in Las Vegas | Wirtschaft | Deutsche Welle | 07.01.2010):

Wenn es nach der britischen Firma Light Blue Optics geht, wird man in Zukunft weder Bildschirm noch Tastatur brauchen, wenn man unterwegs in Internet surfen will. Das Unternehmen stellt in Las Vegas einen Beamer vor (neudeutsch für Projektor), der jede beliebige glatte Oberfläche in eine zehn Zoll große Touchscreen verwandelt. Mithilfe der berührungsempfindlichen Oberfläche kann der Benutzer alle Multimedia-Inhalte ansteuern, die das Internet bietet.

 

„Jede beliebige glatte Oberfläche“ aber die muss auch berührungsempfindlich sein?

 

You can project on "any flat surface," but it must be "touch-sensitive"?

 

I think it's clear where they're going.

[jaapv]

1950: "Nothing can exceed the speed of light."

 

2007: "Oops, maybe it is possible to exceed the speed of light."

'We have broken speed of light' - Telegraph

SF buffs like me have always known that

But you cannot increase the number of photons hitting your sensor.

[adan]

Hi, Brent - you are correct about lens resolution being the ultimate limit - but which will get limited first, the number of resolved lines one can squeeze into a sensor 5mm x 7mm, or the number of resolved lines one can squeeze into a 24 x 36 (or 40 x 54) sensor?

 

Diffraction doesn't pay attention to how large the image area is, just how long the wavelengths of light are and how wide the aperture is.

[theendlesshouse]

The Great Horse-Manure Crisis of 1894

[ho_co]

The Great Horse-Manure Crisis of 1894

 

Gee, was it that long ago? I was sure I remembered it.

[Riley]

Advertisement (gone after registration)

With any sensor technology - a larger sensor will have "more of it" than a small sensor. If a small sensor can do ISO 10,000, a large sensor with the same technology can do ISO 500,000. If a small sensor can do 20 Mpixels, a large sensor with the same technology can do 200 Mpixels.

 

If 110-format Kodachrome looked good - 120-format Kodachome looked 10 times better. That pattern will never end.

 

well 110 was troubled by getting the neg flat in the carrier, resulting in OOF areas on a print, but that aside.

 

While I agree that there will always be separation in performance attributes of various crops and FF, at some point the need for more ISO must begin to fall off. So if/when we get to ISO 500,000 with FF, crops will still be behind, but will it matter as much, me thinks not.

 

Lets say more immediate technology sees an improvement of 4 stops within a few years. This has the potential to produce marketable frames for FF at ISO 25,600, and at ISO6400 for the deepest crop 4/3rds. Given that, the range of available stops would have increased from ISO100 to 3200 for FF (5 stops), ISO100 to 800 on 4/3rds (3 stops) for 4/3rds.

 

If a future proposition would see FF at ISO 25,600 (8 stops), and 4/3rds at ISO6400 (6 stops) the ratio of usable stops has increased for 4/3rds from 3 to 6 stops, which is double, as opposed to FF which only went from 5 to 8 stops.

 

Add to that, the higher the ISO capacity goes, the less pressure there is for competition in this sphere, the less needy the marketplace will be for the desire for more ISO performance.

[griffster]

If phone cams will get so much better, think of how much better the high-end equipment will get too. End of story.

 

ps: Let's not forget the gunk your phone collects by residing in your trousers most of the time

[Manolo Laguillo]

Hi,

with photography it's always the same mix between technique and "metier", "Beruf", knowledge, know-how... A friend of mine, also teacher of photography, and a very good photographer himself, found a very nice answer to the usual saying "no wonder your photos are so good: your's is a very good camera as well...". He says: "Yes, indeed, but the strange thing is that my camera knows very well when I am the one photographing with her, and when not...!"

Minor White did say that the time a photographer needs to develop fully is more or less 15 years, the same amount a musician needs to learn the craft.

Regards,

Manolo Laguillo

[regedit]

nice topic! I'm a tech enthusiastic as well as photo enthusiastic and this debate seems interesting.

 

It's a pre-conception that only large sensors can generate quality images (that's true only today). if we retrospect the digital history will see that sensor producer manage to increase the number of pixels in a very small form factor. now they need to deal with interference that generates noise. When first generation of digital sensor appear even on big ones a ISO 800 was very noisy or N/A. Now with digital post-processing even a relatively small sensor can deliver acceptable results at ISO 800.

 

A second thought is that digital post-processing is accelerating in terms of versatile and in-camera adjustments. Look at Avatar movie (my impression is that some DOF was not optically but digitally controlled in order to increase 3D experience - as example I saw a scene where main character was in focus, the secondary was 1meter behind and was in a very well defined blur (but heavy) and the background was in less blur than character - do you know any lens that naturally produce a image like that?!) maybe in near future will have a full controlled DOF recreated digitally. By heavy increase ISO digital get some good stops and make usable slow lens even in low light. It's just a metter of time to digitally select an aperture (why not to 0.95) from an original F4 lens and recreate the scene.

 

classic photography have some clear physical limitations, while digital one - only imagination is the limit!

[fotografr]

The Great Horse-Manure Crisis of 1894

 

Thank you for your most excrement and manure response to the discussion.

[h00ligan]

@regedit - you make a very valid point. We know that so far Moore's Law has generally held true. This may be an interesting read to some of the contributors to this post. Moore's Law Corollary: Pixel power

 

I think one important thing to note is not just the increase in transistors or functions of correction via increased processing power, but that as the number of those transistors doubles, triples, etc - it allows us to attain current levels of power in a much smaller package as we see through the die process in cpu's. They get smaller, run faster, and stay cooler. Eventually we will see a widespread trend of stellar pocket cams that are cell phones.. there are already some available - although not as commonly in the US marketplace.

 

If we can get an dslr quality sensor (with newer much smaller design - which is inevitable according to Moore's Law) into say, an iPhone - the limitation then becomes one of glass design - for a while. At some point as stated in the second linked article, we will be using smaller size sensors that correct glass issues (as we see now in jpg processing).

 

Whilst purists will prefer one form factor or another - the time of a convergent device is upon us, and it will only improve. Some will prefer the tradition of manual controls, etc - but ultimately the quality WILL be there, although it will likely be controlled in a foreign method to traditional photography - just as lcd's are being accepted by many for composition - or evf's are used for those who want a mirrorless system - etc.

 

Here's an 8mp camera phone - new to the US, but not as good as some outside. Yes mp isn't everything, as we all agree - but again, as the processors and sensors grow in power and shrink in size, glass will matter less and less. http://reviews.cnet.com/cell-phones/samsung-memoir-t929-black/4505-6454_7-33516609.html?tag=contentMain;contentBody

[fotografr]

You make several good points, Edward. Considering the technological breakthroughs that have occurred just in my 65 year lifetime, it surprises me that so few people realize physics is a fluid science and that sometimes what is thought to be impossible today will become commonplace tomorrow. Without the possibility of discovery, there wouldn't be very many people wanting to become physicists. It seems clear that in the very near future nano technology is going to impact many aspects of our daily lives (including photography) in ways we can't even imagine now.

 

I loved the Nathan Myhrvold piece, Moore's Law Corollary: Pixel Power that you linked.

[earleygallery]

As digital technology becomes ever better and ever cheaper, so a 'phone cam' will take a photo as good in quality as today's FF DSLR's, people looking for Pro photographers (for weddings, portraits, fine art) will demand film photography - they won't want their wedding or family portrait captured on something that looks like it was taken on their phone cam, and they won't know how to use a film camera anymore.

 

The future is already here!

[h00ligan]

Brent: Glad you enjoyed, I teetered on posting links to the technical stuff, and ended up doing so hoping it would possibly illustrate the point more clearly. I agree that article was a fun and informative read - without being overly verbose as I am about to... I also like the fact it is a few years old!

 

My father was born in 1919 (passed in 07) (put in perspective I'm 33) and was ALWAYS on the cutting edge with video. I remember him asking me what that little boxy 'walkman' I had was in 2002, regarding the first gen ipod. He was fortunate enough to have THREE technologists in the family (both of my nephews - one older/younger and myself) and he actually embraced the tech more as age went on..we video conferenced a few times a week starting in 02 as well. My point is, he never held onto a traditionalist view. He didn't reject Mozart cd's citing records as better. He recognized the resolution was better, and that they sounded a bit different. He didn't poo poo videocassette switching from 8mm.. the point is, I think his fluidity and willingness to always learn really enhanced his life - in terms of never tiring of subjects and making the most of the technology available. As a side note, my appreciation of all the film and video he shot is one reason I decided to take the plunge into photography.

 

Earley - Film grain imitation could become a science, who knows.

 

Discounting technological advance or fighting the flowing trend for the sake of perceived exclusivity or a demonstration of history in the art to me makes little sense...but then, I'm a technologist.

[chkphoto]

Open-source camera could revolutionize photography (w/ Video)

[h00ligan]

I'm going to have to get a coffee and read up on this - cursory glance this is going to be VERY interesting! Good post!

[Speenth]

... It's not difficult to extrapolate from the information in this article that in 10 years the camera market, as we now know it, might be completely dead ...

 

It is always difficult to extrapolate where future technology is concerned:

 

"I think there is a world market for maybe five computers".

- Thomas Watson, chairman of IBM, 1943

 

We may not be able to imagine the cameras of the future, but I feel sure there'll still be some kit that'll put IQ first and user control of that IQ a very close second. It'll probably be made by Leica and the chances are it'll still have an M mount ... or at least, I hope so, otherwise what'll I do with all this precious glass lying around my house?!

[fotografr]

Open-source camera could revolutionize photography (w/ Video)

 

I'd be interested in open-source lenses so I could teach a cheap lens to behave like a new 0.95 Noctilux.

[Advertisement]

Hallo,

Look here: Interesting Leica gear

[h00ligan]

I'd be interested in open-source lenses so I could teach a cheap lens to behave like a new 0.95 Noctilux.

hah, you'd end up having to buy one every 2 weeks - due to branching.