Go Back   Steve's Digicams Forums > Digicam Help > What Camera Should I Buy?

Reply
 
Thread Tools Search this Thread
Old Dec 25, 2016, 9:10 AM   #1
Senior Member
 
Join Date: Jan 2013
Posts: 231
Default image resolution

This is probably a dumb question, but here goes.
The camera I'm currently playing with (a small point and shoot Casio EX Z800) has 14M, 10M, 6M and 3M) image size settings. One might assume that the 14M size is the 'best' setting, quality wise; but I've read that some small sensors (this one is "1/2.3-inch square pixel CCD") actually introduce distortion by trying to pack too much (ie 14M) into such a small array.
My questions are:
1. How are the image sizes determined by the camera? Is it just a software setting?
2. Is it possible that a lower setting, say 10M might produce higher quality images?
Thanks,
...... john
P.S. Merry Christmas!!

Last edited by Shinnen; Dec 25, 2016 at 9:11 AM. Reason: add some text
Shinnen is offline   Reply With Quote
Sponsored Links
Old Dec 25, 2016, 5:06 PM   #2
Senior Member
 
Join Date: Jan 2011
Location: Australia, New South Wales central coast
Posts: 2,849
Default

G'day John

As I understand it, most cameras that offer the Large - Medium - Small options for pixel use simply divide the sensor into 100% of the sensor ... 50% ... & 25% - ie: the central 25% of the pixels [rather than the top left corner as if you were folding a sheet of paper]

Something like this ... [ see attached image below ]

Several of my earlier superzoom cameras also altered the JPEG compression as the image was reduced in size, going from 'normal JPG mode' to 'superfine JPG mode', thus providing an image that was remarkably sharp even though the pixel count was reduced

As to your camera - I suspect that you need to do some experiments and print the results as until they're printed you don't really see what's going on

Hope this helps
Phil
Attached Images
 
__________________
Has Fuji & Lumix superzoom cameras and loves their amazing capabilities
Google me at Travelling School of Photography Australia
Recent images at http://www.flickr.com/photos/ozzie_traveller/sets/
Ozzie_Traveller is online now   Reply With Quote
Old Dec 25, 2016, 7:00 PM   #3
Senior Member
 
Join Date: Jan 2013
Posts: 231
Default

Hi Phil,
I took 4 pictures at 14M and 4 pictures at 10M (same object, same conditions) and then resized the 10Ms to the size of the 14Ms, i.e. from 3648X2736 to 4320X3240; and 3 of the 4 10Ms were clearer than all of the 14Ms. (The remaining resized 10M was about the same as the best 14M). Does that prove anything?
.... john
P.S. Your sensor diagram completely screws up my assumption about how sensors are built. I thought that they would spatially approximate the layout of a picture, i.e. the center of the sensor array being representative of the center of the picture, etc.
Shinnen is offline   Reply With Quote
Old Dec 25, 2016, 9:47 PM   #4
Senior Member
 
VTphotog's Avatar
 
Join Date: Mar 2005
Location: Extreme Northeastern Vermont, USA
Posts: 4,198
Default

I believe that Phil's explanation would be valid if the effective zoom increased when you chose the smaller sizes, but if the framing is the same, it would mean that the camera is using the full sensor for each resolution, and resizing it internally. Or, to put it another way, the diagram shows the result if the camera was cropping the photo, which would look as if it was zoomed in.
I have made some test shots with my Fuji, and it creates the smaller resolution by resizing the full frame. Printing each at the same dimensions seems to give about the same results for small sized prints, anyway. Larger sizes may show a difference.
I suspect that, viewing each at 100% on a monitor would show reduced noise, as well as reduced detail for the smaller resolutions, but for most viewing, there would be little difference. The biggest advantage to the smaller sizes is the number of shots you can fit on a memory card. (I can't see myself ever being able to fill a card, though without multiple battery recharges)
VTphotog is offline   Reply With Quote
Old Dec 26, 2016, 5:25 AM   #5
Senior Member
 
TCav's Avatar
 
Join Date: Sep 2005
Location: Washington, DC, Metro Area, Maryland
Posts: 13,478
Default

All digital cameras provide the option of recording lower resolution images. In doing so, the camera uses the entire image sensor to record the image, thus preserving the perspective, and simply downsamples the recorded image to the selected resolution. With greater efforts at downsampling come a loss of detail. If you upsample a lower resolution image to a greater resolution, You can't recreate detail that has been lost in the downsampling process. I suspect that what you're seeing isn't more detail, but greater acutance, or an increased contrast at the edges of objects in the image. This is frequently a step in the upsampling process, in an attempt to make the resulting image appear sharper.

Distortion is a technical term with a specific meaning, and nothing described in your original post or in any of the associated replies. conforms to the actual definition of the word.

JPEG images are all compressed to some extent. The different settings for image quality (as opposed to resolution, often referred to simply as size) is the amount of compression applied to the JPEG file. This is a "lossy" compression, which results in a loss of detail, and that detail can not be regained. That is, if your camera saves an image with a quality setting of "Normal", and in post processing, you use a setting of "Fine" when saving the image, you won't get back the original detail that was eliminated by the camera.

I suspect that the authors of the original material you read, and that prompted you to post your questions here, weren't fully aware of what they were describing, and the conclusions they drew were flawed.
__________________
  • 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.

Last edited by TCav; Dec 26, 2016 at 5:28 AM.
TCav is offline   Reply With Quote
Old Dec 26, 2016, 11:55 AM   #6
Senior Member
 
Join Date: Jan 2013
Posts: 231
Default

Thanks VTphotog/TCav,
I knew I was getting into the reeds when I posted this question. To be honest, I really don't know much about the technical aspects of IQ.
And I'm afraid I broke my own rule - "Don't use words that you can't define" i.e. love, truth, and ... distortion
;- )
So, I've taken some more pictures; and I have an idea about what's going on here. The original pictures were taken at auto iso (which in this case mean 800). When I realized this I took another set at iso 50; and again resized them to 14M (this time 10 to 14, 6 to 14, and 3 to 14). What I noticed is that all the pictures taken at iso 800, show lots of artifacts (black and white dots, and wavy lines at the interface of objects). I think these are caused by overprocessing; and they are more noticeable at 14M. The other 3 sizes (10M, 6M, and 3M) resized to 14M, look pretty much the same. By the way, it's hard to seem much difference until one gets to 100%. (At 150% they jump right out.)
The other set I took was at iso 50. This is a much different story. These are nice shots; but again, 14M did show more artifacts than 10M resized .... over processing?? The other 2 (6M and 3M) were generally a little fuzzier (6 fuzzier than 10, and 3 fuzzier than 6), which is what I expected, and did not see at iso 800.
So, my conclusion is that 10M may be the optimal size for this camera, and that iso 800 is not a good idea.
..... john
Shinnen is offline   Reply With Quote
Old Dec 26, 2016, 5:49 PM   #7
Senior Member
 
Join Date: Jan 2011
Location: Australia, New South Wales central coast
Posts: 2,849
Default

Quote:
Originally Posted by VTphotog View Post
I believe that Phil's explanation would be valid if the effective zoom increased when you chose the smaller sizes, but if the framing is the same, it would mean that the camera is using the full sensor for each resolution, and resizing it internally. Or, to put it another way, the diagram shows the result if the camera was cropping the photo, which would look as if it was zoomed in.
G'day VT

Yeah mate ... that's what I came up with to self-explain how my superzoom cameras gave the impression of increased focal lengths as one downsized the pixels in use. It may not hold water with all cameras

Maybe - the other side of the equation is 'reverse interpolation' - instead of inserting pixels via interpolation in order to boost the pixel count, maybe the camera deletes pixels in order to reduce the pixel count while still seeing the same angles of view during the recording process ???

Phil
__________________
Has Fuji & Lumix superzoom cameras and loves their amazing capabilities
Google me at Travelling School of Photography Australia
Recent images at http://www.flickr.com/photos/ozzie_traveller/sets/

Last edited by Ozzie_Traveller; Dec 26, 2016 at 5:55 PM.
Ozzie_Traveller is online now   Reply With Quote
Old Dec 26, 2016, 6:08 PM   #8
Senior Member
 
Join Date: Sep 2011
Location: Bangor,North Wales
Posts: 3,720
Default

Shinnen- a question you might ask yourself is "what size do I NEED my images to be ?"
If you rarely print large images and just web-share small images, perhaps 3m setting will suffice. Firstly you'll get far more images on your memory card and secondly, a downsized image can quite often reduce the visible impact of high-iso image noise.
SIMON40 is offline   Reply With Quote
Old Dec 26, 2016, 8:35 PM   #9
Senior Member
 
TCav's Avatar
 
Join Date: Sep 2005
Location: Washington, DC, Metro Area, Maryland
Posts: 13,478
Default

Quote:
Originally Posted by Ozzie_Traveller View Post
Maybe - the other side of the equation is 'reverse interpolation' - instead of inserting pixels via interpolation in order to boost the pixel count, maybe the camera deletes pixels in order to reduce the pixel count while still seeing the same angles of view during the recording process ???
That's downsampling.
__________________
  • 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 Dec 27, 2016, 6:49 PM   #10
Senior Member
 
Join Date: Jan 2013
Posts: 231
Default

Hi Simon,
Yes, you have a good point. I often go for more than is really necessary.
... john
Shinnen 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 5:16 PM.