P&S versus DSLR -- actual photos for comparison

LOL, you think you are telling anyone something new?

We know you have got the labels back to front anyway....

> I invite you to look at the pictures and draw your own conclusions. You
> can find them here:
>
> http://www.pbase.com/ark/ps_versus_dslr
>
 >> Stay informed about: P&S versus DSLR -- actual photos for comparison 

"tony cooper" wrote in message


> Peering into the gift horse's mouth...it would be easier to make a
> determination if the P&S images and the dslr images- at the same ISO
> setting - were viewable on the same screen at the same time.

Open the images you want to compare in two separate windows and move them
where you like.
 >> Stay informed about: P&S versus DSLR -- actual photos for comparison 

"Pete D" wrote in message


> LOL, you think you are telling anyone something new?

Well, probably not; but I was curious to see the images for myself, and once
I had them, I figured I might as well post them.
 >> Stay informed about: P&S versus DSLR -- actual photos for comparison 

On Fri, 05 Dec 2008 18:23:45 GMT, "Andrew Koenig" wrote
Re P&S versus DSLR -- actual photos for comparison:

>I invite you to look at the pictures and draw your own conclusions. You can
>find them here:
>
> http://www.pbase.com/ark/ps_versus_dslr

Well, the P&S seems to have a lot more noise noticeable in the ISO800
center blow-ups. It seems particularly noticeable in the Civilization
& Capitalism series. My guess is that it wouldn't be a problem in a
4x6 print if the image was captured at ISO200. Larger prints and
higher ISOs might be disappointing.

As usual, the selection of the tool depends on what you want for an
end result. Thanks for doing the test and reporting the results.

BTW, very nice library.
 >> Stay informed about: P&S versus DSLR -- actual photos for comparison 

This message is not archived
 >> Stay informed about: P&S versus DSLR -- actual photos for comparison 

wrote in message


> So what SW and specific steps did you use to do this? There are many
> scaling methods. And since these are jpgs with no exif it appears these
> may
> possibly have been resaved.

I used a program called "JPEG Wizard" to do the cropping. After that, I
uploaded the original JPEG files and the cropped output files from JPEG
Wizard onto Pbase. Any further transformations were done by Pbase as part
of putting the images into their galleries. I see no obvious visual
difference between the cropped images on Pbase and corresponding sections of
the original JPEG images on my machine.
 >> Stay informed about: P&S versus DSLR -- actual photos for comparison 

Andrew Koenig wrote:
> It is said that a picture is worth 1,000 words. Accordingly, I set up a
> tripod and took a bunch of pictures of the same subject with a P+S camera
> and a DSLR. The cameras I used were the ones I happened to have handy: A
> Canon SD800 (7.1 megapixels) and a Nikon D700 (12.1 megapixels).
>
> In all cases, I set the camera to delay for a few seconds before taking the
> picture, to allow any vibration from my hands to settle down. I used a 50mm
> f/1.4 lens with the Nikon; I set the lens to f/11 for all pictures because I
> think that is close to the optimum image quality. This particular P&S does
> not allow manual aperture adjustments, so I had no choice but to let the
> camera pick the aperture.
>
> The DSLR offers an ISO range from 200 through 6400, so I took pictures at
> 200, 400, 800, 1600, 3200, and 6400. The P&S offers an ISO range from 80
> through 1600, so I took pictures at 80, 100, 200, 400, 800, and 1600.
>
> For each picture, I have posted a scaled-down version of the picture and a
> full-size crop from the center portion of it. Aside from scaling and
> cropping, the pictures are exactly as they came from the cameras; I have not
> applied sharpening, additional noise reduction, or any other
> image-processing algorithms.
>
> I invite you to look at the pictures and draw your own conclusions. You can
> find them here:
>
> http://www.pbase.com/ark/ps_versus_dslr
>
>
>
Hi,

Canon SD 800 is not a very good choice for comparison. In order to get a
28mm equivalent at the wide end, the quality of this zoom lens is
compromised. Why not compare with a P & S with a better lens?

Morton
 >> Stay informed about: P&S versus DSLR -- actual photos for comparison 

"Morton" wrote in message


> Canon SD 800 is not a very good choice for comparison. In order to get a
> 28mm equivalent at the wide end, the quality of this zoom lens is
> compromised. Why not compare with a P & S with a better lens?

Because it's the one I had available.

You're welcome to take similar pictures with whatever camera you like and
post them. If you want to try for a similar scale, the bookshelf in the
photo is 40 inches wide.
 >> Stay informed about: P&S versus DSLR -- actual photos for comparison 

Andrew Koenig wrote:
> It is said that a picture is worth 1,000 words. Accordingly, I set up a
> tripod and took a bunch of pictures of the same subject with a P+S camera
> and a DSLR. The cameras I used were the ones I happened to have handy: A
> Canon SD800 (7.1 megapixels) and a Nikon D700 (12.1 megapixels).
>
> In all cases, I set the camera to delay for a few seconds before taking the
> picture, to allow any vibration from my hands to settle down. I used a 50mm
> f/1.4 lens with the Nikon; I set the lens to f/11 for all pictures because I
> think that is close to the optimum image quality. This particular P&S does
> not allow manual aperture adjustments, so I had no choice but to let the
> camera pick the aperture.
>
> The DSLR offers an ISO range from 200 through 6400, so I took pictures at
> 200, 400, 800, 1600, 3200, and 6400. The P&S offers an ISO range from 80
> through 1600, so I took pictures at 80, 100, 200, 400, 800, and 1600.
>
> For each picture, I have posted a scaled-down version of the picture and a
> full-size crop from the center portion of it. Aside from scaling and
> cropping, the pictures are exactly as they came from the cameras; I have not
> applied sharpening, additional noise reduction, or any other
> image-processing algorithms.
>
> I invite you to look at the pictures and draw your own conclusions. You can
> find them here:
>
> http://www.pbase.com/ark/ps_versus_dslr
>
>
>
Nice comparison, although putting a D700 up against an SD800 was pretty
much a foregone conclusion.
A few points are worth noting:
The P&S at 80 ISO and DSLR at 3200 ISO, give roughly equivalent noise
levels. The p&s also has at least one hot pixel..
The low P&S lens quality is evident in the corners of the image, even at
those highly reduced sizes, and there's a lot of barreling.
The P&S seems to be suffering from a lot of flare/discoloration at left
- was there a lighting issue?
 >> Stay informed about: P&S versus DSLR -- actual photos for comparison 

"Mark Thomas" wrote in message


> Nice comparison, although putting a D700 up against an SD800 was pretty
> much a foregone conclusion.

Thanks. As I remarked in another post, I used what I had.

> A few points are worth noting:
> The P&S at 80 ISO and DSLR at 3200 ISO, give roughly equivalent noise
> levels. The p&s also has at least one hot pixel..
> The low P&S lens quality is evident in the corners of the image, even at
> those highly reduced sizes, and there's a lot of barreling.

Actually, I thought the center crops were surprisingly good at low ISO. And
of course, distortion is easily corrected when needed.

I don't immediately see where the hot pixel is, so I would appreciate it if
you could tell me.

> The P&S seems to be suffering from a lot of flare/discoloration at left -
> was there a lighting issue?

Yes -- there's a window right next to the bookshelf, just outside the image
area
 >> Stay informed about: P&S versus DSLR -- actual photos for comparison 

Andrew Koenig wrote:
> "Mark Thomas" wrote in message
>
>
>> Nice comparison, although putting a D700 up against an SD800 was pretty
>> much a foregone conclusion.
>
> Thanks. As I remarked in another post, I used what I had.
>
>> A few points are worth noting:
>> The P&S at 80 ISO and DSLR at 3200 ISO, give roughly equivalent noise
>> levels. The p&s also has at least one hot pixel..
>> The low P&S lens quality is evident in the corners of the image, even at
>> those highly reduced sizes, and there's a lot of barreling.
>
> Actually, I thought the center crops were surprisingly good at low ISO. And
> of course, distortion is easily corrected when needed.
>
> I don't immediately see where the hot pixel is, so I would appreciate it if
> you could tell me.

Centre crops, look at the shelf below the "Wheels of Commerce", directly
under the 'r&' of Harper&Row


>> The P&S seems to be suffering from a lot of flare/discoloration at left -
>> was there a lighting issue?
>
> Yes -- there's a window right next to the bookshelf, just outside the image
> area

Did the 700 have a lens hood on, perhaps? Or is that just another cross
against the p&s?
 >> Stay informed about: P&S versus DSLR -- actual photos for comparison 

"Mark Thomas" wrote in message


>> I don't immediately see where the hot pixel is, so I would appreciate it
>> if you could tell me.

> Centre crops, look at the shelf below the "Wheels of Commerce", directly
> under the 'r&' of Harper&Row

Got it; thanks.

>>> The P&S seems to be suffering from a lot of flare/discoloration at
>>> left - was there a lighting issue?

>> Yes -- there's a window right next to the bookshelf, just outside the
>> image area

> Did the 700 have a lens hood on, perhaps? Or is that just another cross
> against the p&s?

Yes, I always use a hood if the lens has a way to attach one.
 >> Stay informed about: P&S versus DSLR -- actual photos for comparison 

On Fri, 05 Dec 2008 18:23:45 GMT, "Andrew Koenig" wrote:

>It is said that a picture is worth 1,000 words. Accordingly, I set up a
>tripod and took a bunch of pictures of the same subject with a P+S camera
>and a DSLR. The cameras I used were the ones I happened to have handy: A
>Canon SD800 (7.1 megapixels) and a Nikon D700 (12.1 megapixels).
>
>In all cases, I set the camera to delay for a few seconds before taking the
>picture, to allow any vibration from my hands to settle down. I used a 50mm
>f/1.4 lens with the Nikon; I set the lens to f/11 for all pictures because I
>think that is close to the optimum image quality. This particular P&S does
>not allow manual aperture adjustments, so I had no choice but to let the
>camera pick the aperture.
>
>The DSLR offers an ISO range from 200 through 6400, so I took pictures at
>200, 400, 800, 1600, 3200, and 6400. The P&S offers an ISO range from 80
>through 1600, so I took pictures at 80, 100, 200, 400, 800, and 1600.
>
>For each picture, I have posted a scaled-down version of the picture and a
>full-size crop from the center portion of it. Aside from scaling and
>cropping, the pictures are exactly as they came from the cameras; I have not
>applied sharpening, additional noise reduction, or any other
>image-processing algorithms.
>
>I invite you to look at the pictures and draw your own conclusions. You can
>find them here:
>
> http://www.pbase.com/ark/ps_versus_dslr
>
>

Conclusion 1: When compared for relative image resolution and viewed at true
pixel levels (not viewed with bicubic interpolation) there is no difference
between resolution on an inexpensive P&S zoom lens and expensive fixed-focus
DSLR lens on the resulting image. If taking sensor size (photosite sizes) into
account then the cheap P&S lens is vastly out-resolving the expensive DSLR lens.

The interesting thing is that this is happening on one of the worst P&S lenses
I've ever had the misfortune to see. The edge and corner softness and geometric
distortion in that camera would be unacceptable for purchase were I doing the
shopping. None of the P&S cameras I've ever bought show any signs of that kind
of poor edge performance and geometric distortion.

Conclusion 2: You need to learn to shop better. Also learn how to turn down
sharpening and use less JPG compression in your P&S camera. What JPG compression
did you use for image quality on that P&S, "economy" mode? I don't see JPG
artifacts like that unless I'm re-compressing images for the web with 70%
settings (where 100%=none). If adjusting those things is not possible then may
your next P&S purchase be more intelligently decided on (that sort of goes
without saying). A great majority of P&S cameras have full manual controls, JPG
processing controls, and much better lenses. Learn to find them.

Conclusion 3: Putting a lens-hood on the DSLR and not on the P&S camera when it
is known that there's a strong light to the side reveals the poster
intentionally trying to sway the test result.

Conclusion 4:

Nikon D700 + 50mm f/1.4 lens = $3,000 + $400 = $3,400

Canon SD800 = $300

A higher-quality P&S camera without all those edge and geometry defects, and
even sharper optics = $250-$350

Conclusion 5: Someone's been sorely "had". Not only by their expensive DSLR
purchase but by their inability to choose much better P&S cameras for the same
price.
 >> Stay informed about: P&S versus DSLR -- actual photos for comparison