Pages: [1]   Go Down

Author Topic: New article on the future of cameras  (Read 4628 times)

Philip Weber

  • Full Member
  • ***
  • Offline Offline
  • Posts: 185
New article on the future of cameras
« on: March 16, 2009, 01:04:17 pm »

This was in today's New Scientist:

What's next after the megapixel wars?

And an older article from the same site, in case anyone missed it:

The Future of Photography


Interesting stuff!

Phil

Logged

Roger Calixto

  • Full Member
  • ***
  • Offline Offline
  • Posts: 141
New article on the future of cameras
« Reply #1 on: March 16, 2009, 06:22:29 pm »

I liked the part on the gigapod was news to me.

Cool. Tks.
Logged
--------------
If my day job wasn't so cool, I'd quit and be a photographer =)

dalethorn

  • Guest
New article on the future of cameras
« Reply #2 on: March 16, 2009, 06:24:15 pm »

Quote from: Philip Weber
This was in today's New Scientist:
What's next after the megapixel wars?
And an older article from the same site, in case anyone missed it:
The Future of Photography
Interesting stuff!
Phil

All of the techniques described in these articles are promising and encouraging, but despite the euphemisms and hyperbole, the bottom line is real resolution - whatever it takes to get there. You either capture the detail or you don't, and if you don't, people will learn to see through the smoke and mirrors.
Logged

250swb

  • Full Member
  • ***
  • Offline Offline
  • Posts: 220
New article on the future of cameras
« Reply #3 on: March 17, 2009, 04:44:23 am »

Quote from: dalethorn
All of the techniques described in these articles are promising and encouraging, but despite the euphemisms and hyperbole, the bottom line is real resolution - whatever it takes to get there. You either capture the detail or you don't, and if you don't, people will learn to see through the smoke and mirrors.


I agree with you that resolution is important. But I also think that the pixel race is a lot to do with smoke and mirrors itself.

For instance there really are a lot of photographers out there who still think 24mp is 100% 'better' than 12mp, when in reality (given like for like) you need four times the number of pixels to double the size of a print at the same resolution. So in terms of print size and resolution 24mp isn't such a big jump over 12mp after all. And it is possible that an average lens lens on a 24mp camera isn't going to beat a good lens on a 12mp camera for print resolution (all things being equal).

Which is the point of the misquoted Mr Watanabe in the main article. He didn't actually say that 12mp was 'enough, no more ever again', but he was making the point that innovation was around the corner that would make the pixel race redundant, and that the innovations in sensor technology would not be about the largest number of pixels. There is a cross over point coming when the increasingly ponderous nature of more pixels will cause the majority of DSLR companies to run out of steam if they stay on the course they are on. Not to mention the need to re-design lenses to meet the increasing pixel count. So perhaps we will all need tomorrow things that are called 'smoke and mirrors' today. The early days of photography saw a similar disbelief from critics, and they were won over.

Steve

Nick Rains

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 705
    • http://www.nickrains.com
New article on the future of cameras
« Reply #4 on: March 17, 2009, 06:18:52 am »

Quote from: 250swb
For instance there really are a lot of photographers out there who still think 24mp is 100% 'better' than 12mp, when in reality (given like for like) you need four times the number of pixels to double the size of a print at the same resolution. So in terms of print size and resolution 24mp isn't such a big jump over 12mp after all. And it is possible that an average lens lens on a 24mp camera isn't going to beat a good lens on a 12mp camera for print resolution (all things being equal).
Steve

Just to point out that 24Mp does indeed give you twice the size of print. You are talking about a print 4 time the size. Double the area is twice as big, double both dimensions is 4 times as big. Double the pixels allows a print twice as big in area, for example A4 to A3.

I'd call being able to run a double page spread in a magazine a big step forward from being limited to a single page.

Please forgive me for being pedantic but it's a common confusion of terms - 'size' could be area or length.
Logged
Nick Rains
Australian Photographer Leica

dalethorn

  • Guest
New article on the future of cameras
« Reply #5 on: March 17, 2009, 08:15:50 am »

Quote from: 250swb
Which is the point of the misquoted Mr Watanabe in the main article. He didn't actually say that 12mp was 'enough, no more ever again', but he was making the point that innovation was around the corner that would make the pixel race redundant, and that the innovations in sensor technology would not be about the largest number of pixels.

I'm all for innovations of all kinds. But since this is the digital (not film) era, pixels are *all* we have at the end of the day. Repeat: Pixels are *all* that you have to make an image. We've already seen every possible way to massage pixels to make them "better". Are you suggesting that there's a fundamental way to improve resolution coming that doesn't require more pixels? Or are you saying that the breakthrough is mental - to make me *believe* that my compact or small-DSLR images look as detailed as LF/MF? Or make me not care?
Logged

Gemmtech

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 526
New article on the future of cameras
« Reply #6 on: March 17, 2009, 03:21:08 pm »

Quote from: Nick Rains
Just to point out that 24Mp does indeed give you twice the size of print. You are talking about a print 4 time the size. Double the area is twice as big, double both dimensions is 4 times as big. Double the pixels allows a print twice as big in area, for example A4 to A3.

I'd call being able to run a double page spread in a magazine a big step forward from being limited to a single page.

Please forgive me for being pedantic but it's a common confusion of terms - 'size' could be area or length.


FINALLY, somebody gets it!  I stated this before on another thread and am so sick of reading it.  One person writes on the web that 24mp isn't double 12mp and then it gets propagated ad infinitum.  If I have a room that is 10' wide x 10' length that's 100 sq.ft.  If I want 200 sq.ft then I need a room that is 10' x 20' or  12' x 16.5' approx. or 13' x 15.5' approx.  If I double the original width and length then I have a room that is 20' x 20' or 400 sq.ft. is somebody going to tell me that 400sq.ft. isn't 4x bigger than 100sq.ft?  I know what comes next, Well, the jump from 12mp to 24mp isn't that great when viewing an 8x10 print or 12x18 print etc. It doesn't become apparent until printing larger than 12x18 (or whatever size) Guess what, if I just want to put a couch, table and two people in a room then 10' x 10' would be fine!  Subjective viewing of print sizes doesn't change the fact that 24mp is double 12mp, is it really that difficult to understand?
Logged

spidermike

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 535
New article on the future of cameras
« Reply #7 on: March 25, 2009, 01:32:51 pm »

Quote from: Gemmtech
FINALLY, somebody gets it!  I stated this before on another thread and am so sick of reading it.  One person writes on the web that 24mp isn't double 12mp and then it gets propagated ad infinitum.  If I have a room that is 10' wide x 10' length that's 100 sq.ft.  If I want 200 sq.ft then I need a room that is 10' x 20' or  12' x 16.5' approx. or 13' x 15.5' approx.  If I double the original width and length then I have a room that is 20' x 20' or 400 sq.ft. is somebody going to tell me that 400sq.ft. isn't 4x bigger than 100sq.ft?  I know what comes next, Well, the jump from 12mp to 24mp isn't that great when viewing an 8x10 print or 12x18 print etc. It doesn't become apparent until printing larger than 12x18 (or whatever size) Guess what, if I just want to put a couch, table and two people in a room then 10' x 10' would be fine!  Subjective viewing of print sizes doesn't change the fact that 24mp is double 12mp, is it really that difficult to understand?

I agree. But the fact remains that if you say to most people that they can print 'twice as big', they will take this to mean you are going from 10x12 to 20x24 (ie linear dimensions). But as you rightly say it means going from 10x12 to 14x16 (approximately). But you certainly could not go from A4 to A3 as that would rely on doubling one dimension.
And you do see people extolling that a 24MP has twice the resolution of a 12MP and I think that certainly is wrong (you wouldn't be able to resolve twice the density of marks on a ruler for instance).

Logged

250swb

  • Full Member
  • ***
  • Offline Offline
  • Posts: 220
New article on the future of cameras
« Reply #8 on: March 25, 2009, 06:10:29 pm »

Its interesting that all of a sudden 'area' is the 'in' word used to defend pixel count. I think in terms of my writing a quick note and using 'double' in the context of the conversation means a 10x10 room becomes a 20x20 room in terms of 'doubling' dimensions. If you wish to downgrade this to 'area' to make lesser point then you have, but it still doesn't answer the main question of what the mass of people understand is the difference between 12mp and 24mp. I know we all want our proverbial Canon 5d MkII to have as many cubic inches under the hood as possible to justify its existence, and statistics are a great way to massage the facts depending on the answer you want.  But don't you think playing games with words by taking them out of context is part of the pixel count problem, and not a long term solution to it?

Steve

BernardLanguillier

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 13983
    • http://www.flickr.com/photos/bernardlanguillier/sets/
New article on the future of cameras
« Reply #9 on: March 25, 2009, 08:09:32 pm »

Quote from: spidermike
I agree. But the fact remains that if you say to most people that they can print 'twice as big', they will take this to mean you are going from 10x12 to 20x24 (ie linear dimensions). But as you rightly say it means going from 10x12 to 14x16 (approximately). But you certainly could not go from A4 to A3 as that would rely on doubling one dimension.
And you do see people extolling that a 24MP has twice the resolution of a 12MP and I think that certainly is wrong (you wouldn't be able to resolve twice the density of marks on a ruler for instance).

Going from a 12 to a 24 MP camera will enable you to print an A3 (29.7 x 42 cm) at the exact same dpi as would have previously had on A4 (29.7 x 21 cm).

Cheers,
Bernard

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
New article on the future of cameras
« Reply #10 on: March 26, 2009, 01:34:02 am »

Hi Bernard,

I have some issues with that:

1) DPI is a setting in the printer it has little to do with image resolution, what you send to the printer is always an image with certain size (in PPI)
2) The resolution of the camera says what the camera is capable of. The amount of detail depends what you put in front of it.
3) Going much higher than 25 MP diffraction would limit resolution for any usable aperure.
4) Depth of focus (or field) is very small, if sharpness is asked for. Focus merging you are doing obviously helps.

That said it seems that there are some advantages to oversampling:

1) No risk of moire effects even without AA-filtering
2) Oversampling may be better than interpolation

My guess is that we should not go beyond f/8 on FX format with good lenses at the present size of pixels for optimal sharpness even if image detoriation will not be very visible before f/16. Would we have lenses that perform optimally at f/4 we could quadruple the resolution but have half the depth of focus.

Best regards
Erik

Quote from: BernardLanguillier
Going from a 12 to a 24 MP camera will enable you to print an A3 (29.7 x 42 cm) at the exact same dpi as would have previously had on A4 (29.7 x 21 cm).

Cheers,
Bernard
« Last Edit: March 26, 2009, 01:36:20 am by ErikKaffehr »
Logged
Erik Kaffehr
 

Gemmtech

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 526
New article on the future of cameras
« Reply #11 on: March 26, 2009, 10:06:34 am »

Quote from: 250swb
Its interesting that all of a sudden 'area' is the 'in' word used to defend pixel count. I think in terms of my writing a quick note and using 'double' in the context of the conversation means a 10x10 room becomes a 20x20 room in terms of 'doubling' dimensions. If you wish to downgrade this to 'area' to make lesser point then you have, but it still doesn't answer the main question of what the mass of people understand is the difference between 12mp and 24mp. I know we all want our proverbial Canon 5d MkII to have as many cubic inches under the hood as possible to justify its existence, and statistics are a great way to massage the facts depending on the answer you want.  But don't you think playing games with words by taking them out of context is part of the pixel count problem, and not a long term solution to it?

Steve

I have been building & renovating homes as well as buying them for over 30 years and have never, not one single time heard anybody state that a 10' x 10' room doubles if it's 20' x 20'.  Doubling dimensions of the room is not the same as doubling the size of the room, don't you see the difference?   If I have a kitchen that is 10' x 10' and my neighbor's kitchen is 10' x 20' we would say his kitchen is twice as big as mine.  Granted we could also say that his kitchen is the same width as mine but double the length, but it's still twice as large.  I don't dispute that the perceived increase isn't necessarily doubled, just like in some rooms or houses that are compartmentalized, a house four times larger than another house might only feel like 2 times larger.  "Area" is size right?  Aren't you talking about area when you state wxh?  Don't ALL manufacturers calculate the total number of pixels by multiplying the horizontal x vertical?  Aren't there different format sensors?   As one person stated correctly, open a magazine and look at the left side, now look at the right, the two pages together are double the size of the one and when you are advertising you pay as such (sometimes they give you a "double page" discount.
Logged

250swb

  • Full Member
  • ***
  • Offline Offline
  • Posts: 220
New article on the future of cameras
« Reply #12 on: March 26, 2009, 12:47:31 pm »

The fact that I did not originally use the word 'area' seems to have passed you by Gemmtech. I simply said a dimension doubles if it goes from, say, 10' to 20' (or whatever). If you want to talk about 'area' please do so, but I'm baffled as to why you want to quote me as if I started using the word? There are other words I didn't use as well, so you have pretty well the whole dictionary to go at if you want.

Steve

j-land

  • Newbie
  • *
  • Offline Offline
  • Posts: 35
New article on the future of cameras
« Reply #13 on: March 27, 2009, 01:45:52 am »

Traditionally, photographic enlargement is expressed in terms of magnification; that is, the linear dimensions of the print. So an 8x10 is twice as big as a 4x5. When you blow your 4x5 neg up to a 16x20 print, it's a 4x enlargement, not a 16x enlargement. And the formulas to figure out the increase in exposure when you rack the enlarger head up the column to increase the print size are based on magnification. When it comes to film size, however, many photographers compare formats by the increase in imaging area one would get by going up a size, so 8x10 is 4x as big as 4x5. But when it comes to talking about prints, an 8x10 negative is twice as "enlargeable" as 4x5 negative from the same film stock. So really both ways of thinking about sensor size are "correct". Another variable that can influence enlarge-ability with digital cameras, of course, is pixel density, a whole other can of worms - maybe the rough equivalent to a high resolving film vs a low resolving one. I think it's fine to use either definition, as long as one realizes it's not so cut and dried, and specify which one when trying to communicate with one's forum buddies   . If you go to a store to buy a cardboard box and you ask for one that's twice as big as a 10"x10"x10", would you be more likely to get a 20"x20"x20" or a 12.6"x12.6"x12.6"? Hmm...
Logged

Daniel Browning

  • Full Member
  • ***
  • Offline Offline
  • Posts: 142
New article on the future of cameras
« Reply #14 on: March 27, 2009, 02:27:41 am »

I hope that someone comes up with a way to increase dynamic range by simulating larger full well capacities without any negative impact on image quality (such as lower sensitivity or higher read noise than current designs).

Some day I imagine being able to use ISO 5 or ISO 10 and getting a tremendous amount of dynamic range a low noise, from which I can choose any tone curve imaginable, while I fly to work in my hovercraft.
Logged
--Daniel

sgietler

  • Newbie
  • *
  • Offline Offline
  • Posts: 11
    • http://www.scottgietler.com
New article on the future of cameras
« Reply #15 on: April 02, 2009, 09:00:29 pm »

thanks for the Article, the Gigapan was interesting to read about.
Pages: [1]   Go Up