Go Back   Steve's Digicams Forums > Digicam Help > General Discussion

Reply
 
Thread Tools Search this Thread
Old Sep 18, 2009, 6:20 PM   #11
Senior Member
 
Join Date: Jun 2004
Posts: 477
Default

I think that the practical limit (seemingly due to LENS RESOLUTION) is around 25-30Mpixel for FF, and maybe APS-C probably comes in at around 18-23Mpixel, and 12-15Mpixel for 4/3.

P&S cameras SHOULD come in at around 7-8Mpixel, in my opinion!!!!
dnas is offline   Reply With Quote
Old Sep 18, 2009, 7:55 PM   #12
Senior Member
 
Join Date: May 2006
Posts: 181
Default

I get some nice pictures from my Panasonic DMC-LZ3 at 5MP, but its image processing engine is not very good. It has a fairly good telephoto (216mm equivalent in my pocket), so if I know what I want to crop, I don't need to crop out much.

You could use a 20MP camera to take the same photo at 108mm equiv lens and crop out, but it still wouldn't have the same quality as a 5MP picture taken through a 216mm lens using optical zoom.

For many pictures I take, 640 x 480 is all I need, however, if I use the VGA setting on the camera, I get a poor quality image with jagged edges that does not come close to the quality of the resampled image made from a full resolution image.

My older 3.3MP camera doesn't suffer as bad when using VGA mode.

I think 1/1.8 to 1/2.5 miniature sensor gives a good balance of high depth of focus, fast lens and reasonable zoom range in a small package. A larger sensor would need a larger lens to achieve the same F stop, have lesser depth of field and require a longer lens to achieve the wide end.
itsme000 is offline   Reply With Quote
Old Sep 18, 2009, 8:01 PM   #13
Senior Member
 
TCav's Avatar
 
Join Date: Sep 2005
Location: Washington, DC, Metro Area, Maryland
Posts: 13,544
Default

Quote:
Originally Posted by dnas View Post
I think that the practical limit (seemingly due to LENS RESOLUTION) is around 25-30Mpixel for FF, and maybe APS-C probably comes in at around 18-23Mpixel, and 12-15Mpixel for 4/3.

P&S cameras SHOULD come in at around 7-8Mpixel, in my opinion!!!!
In order for the diffraction limited f-stop of a Full Frame dSLR to be f/11, the image sensor would need to be 73MP! We still have a way to go before the resolving power of a lens is the reason to stop increasing the pixel count.
__________________
  • The lens is the thing.
  • 'Full Frame' is the new 'Medium Format'.
  • "One good test is worth a thousand expert opinions." - Tex Johnston, Boeing 707 test pilot.
TCav is offline   Reply With Quote
Old Sep 18, 2009, 8:32 PM   #14
Senior Member
 
Join Date: May 2006
Posts: 181
Default

Quote:
Originally Posted by TCav View Post
In order for the diffraction limited f-stop of a Full Frame dSLR to be f/11, the image sensor would need to be 73MP! We still have a way to go before the resolving power of a lens is the reason to stop increasing the pixel count.
What about the limitation of resolving power of the sensor caused by cross sensitization of adjacent cells?
itsme000 is offline   Reply With Quote
Old Sep 19, 2009, 12:31 AM   #15
Super Moderator
 
peripatetic's Avatar
 
Join Date: Nov 2004
Posts: 3,599
Default

Diffraction limits are not the whole story either.

The overall system resolution can be improved by improving any part of the chain. In other words, with a given lens, a higher resolving sensor will always help to improve overall quality. A 100% (whatever that means) improvement in sensor resolution will not result in a 100% improvement in the overall chain unless every other part of the chain is much better than the sensor; instead you will see a fraction of that 100% improvement, 10,25,50% for example - depending on the other elements in the chain.

See Erwin Puts' website for more details.

And this interesting post by Ctein:

http://theonlinephotographer.typepad...be-enough.html
__________________
My gallery
My X100 blog
peripatetic is offline   Reply With Quote
Old Sep 19, 2009, 10:26 AM   #16
Senior Member
 
TCav's Avatar
 
Join Date: Sep 2005
Location: Washington, DC, Metro Area, Maryland
Posts: 13,544
Default

I know there are other issues to be considered. I was addressing a single issue raised by dnas:
Quote:
Originally Posted by dnas View Post
I think that the practical limit (seemingly due to LENS RESOLUTION) ...
__________________
  • The lens is the thing.
  • 'Full Frame' is the new 'Medium Format'.
  • "One good test is worth a thousand expert opinions." - Tex Johnston, Boeing 707 test pilot.
TCav is offline   Reply With Quote
Old Sep 19, 2009, 3:31 PM   #17
Junior Member
 
Join Date: Sep 2009
Posts: 17
Default

As a long term (well actually who really knows when it comes to tech stuff like this) view, there must be a point where pixel count is thwarted by lens quality. As an example; pixel numbers on sensor seem to be rising almost exponentially. So at some point when I take a picture of the moon, if I then zoom in enough later I will be able to see the footprints of Armstrong, etc. However, lens technology is a lot slower moving than electronics and to create a lens that could accurately channel that light to the lens is a long way off.

So another question to ponder is; when does the quality of today's glass lenses infringe on the rising pixel count?

Update: sorry I had not realised that my question had proved so interesting and that the posts had reached to 2 pages!, so some of the above has already been answered, and thanks for that.

So instead : extra points for anybody who can suggest ways of refining lenses or the even the chain as mentioned by Peripatetic. E.g. I know in Rifle sight optics, they often have inert gases between the lenses for improved clarity, etc.

TM

Last edited by Taxed Mind; Sep 19, 2009 at 3:37 PM.
Taxed Mind is offline   Reply With Quote
Old Sep 19, 2009, 6:44 PM   #18
Senior Member
 
TCav's Avatar
 
Join Date: Sep 2005
Location: Washington, DC, Metro Area, Maryland
Posts: 13,544
Default

True. At some point, the path taken by individual photons starts to limit the resolving power of a system. (The diffraction limited f-stop is one part of that.) The simpler the lens design (the fewer polished surfaces), the greater the potential for a lens' resolving power. And the less refracting going on, the less the potential for aberrations. Those are some of the reasons why it's easier to make a good catadioptric lens than a refractive lens. And those are some of the reasons why it's easier to make a good prime lens than a good zoom lens. And those are some of the reasons why it's easier to make a good P&S lens than a good SLR lens.

Have you opened a can of worms? You betcha!
__________________
  • The lens is the thing.
  • 'Full Frame' is the new 'Medium Format'.
  • "One good test is worth a thousand expert opinions." - Tex Johnston, Boeing 707 test pilot.
TCav is offline   Reply With Quote
Old Sep 19, 2009, 8:45 PM   #19
Senior Member
 
VTphotog's Avatar
 
Join Date: Mar 2005
Location: Extreme Northeastern Vermont, USA
Posts: 4,214
Default

The inert gas in riflescopes (usually dry Nitrogen) is there to prevent fogging due to the ingress of moisture as the temperature changes. I am not aware of any improvement in light transmission from this.
One possible way to improve optics, is to eliminate glass, and use optical grade polycarbonate (Lexan), which does not suffer from the dispersion difficulties of glass, and has a higher refractive index. It also is a pretty good UV filter. As of now, though, the technology for coating it is not as good as with glass, so chromatic aberration is still a problem. It is also easily scratched.

brian
VTphotog is offline   Reply With Quote
Old Sep 19, 2009, 9:09 PM   #20
Senior Member
 
Join Date: Jun 2004
Posts: 477
Default

Quote:
Originally Posted by TCav View Post
In order for the diffraction limited f-stop of a Full Frame dSLR to be f/11, the image sensor would need to be 73MP! We still have a way to go before the resolving power of a lens is the reason to stop increasing the pixel count.
A lens, diffraction limited (a lens with such a high resolution, that the resolution is only limited by diffraction not by lens imperfections) to F4, will will provide an equivalent sensor resolution of around 120Mpixel.
The only lens I know of with this diffraction limiting is the Leica 280 mm f/4 Apo-Telyt-R.
At F5.6, diffraction limiting is half that, at F8 half again (around 30Mpixel)

But we are talking about CONSUMER lenses, not 99% perfect lenses.

When you take a REALISTIC good quality lens, diffraction limited at say F11, full frame sensor, then the lens resolution will limit the equivalent sensor resolution to less than 20Mpixel.
dnas is offline   Reply With Quote
 
Reply


Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off



All times are GMT -5. The time now is 9:45 PM.