James Cafferty
Veteran Member
Excellent!!!!! Clear a spot on the wall for this one.
--
All the best,
Jim
Photographers take pictures; the camera is only a tool.
Follow along with the video below to see how to install our site as a web app on your home screen.
Note: This feature may not be available in some browsers.
Excellent!!!!! Clear a spot on the wall for this one.
Fortunately there are other numeric deciding factors as well - the more the better:There is every sign to me that users still obsess about megapixel count as it's one of the main deciding factors new buyers seem to focus on.
Not to mention buyers obsessed with less pixels, a.k.a. large pixel huggers. It is a growing religion and the manufacturers can very well cash on it by offering niche models.Fortunately there are other numeric deciding factors as well - the more the better:There is every sign to me that users still obsess about megapixel count as it's one of the main deciding factors new buyers seem to focus on.
megazoom factor,
highest ISO number,
the number of detection modes: (face, smile, eye blink, pet...)
etc.
Well, a simple downsampling will do just that (like we can see in DxO's 8mp 'print' figures), so it can't be completely wrong.Since when was increasing resolution and reducing the visibility of digital artifacts through increasing the number of pixels a bad thing? The belief that reducing the number of pixels somehow increases the dynamic range and reduces noise is a pervasive one but still wrong.
More megapixels means better resolution, fewer megapixels means less resolution - there are no advantages in going forward to the past.
As for me, I can't wait to see images from the rumoured Canon 1D Mark IV with 56 megapixels and no AA filter.
Downsampling can only degrade an image, not improve it.Well, a simple downsampling will do just that (like we can see in DxO's 8mp 'print' figures), so it can't be completely wrong.Since when was increasing resolution and reducing the visibility of digital artifacts through increasing the number of pixels a bad thing? The belief that reducing the number of pixels somehow increases the dynamic range and reduces noise is a pervasive one but still wrong.
Not even an ISO 51200 image shot with a 200mp APS-C camera? ;-)Downsampling can only degrade an image, not improve it.Well, a simple downsampling will do just that (like we can see in DxO's 8mp 'print' figures), so it can't be completely wrong.Since when was increasing resolution and reducing the visibility of digital artifacts through increasing the number of pixels a bad thing? The belief that reducing the number of pixels somehow increases the dynamic range and reduces noise is a pervasive one but still wrong.
Not even then. You can only lose data. But I can see that you cannot get the 100% zoom comparisons of different enlargements out of your head.Not even an ISO 51200 image shot with a 200mp APS-C camera? ;-)Downsampling can only degrade an image, not improve it.Well, a simple downsampling will do just that (like we can see in DxO's 8mp 'print' figures), so it can't be completely wrong.Since when was increasing resolution and reducing the visibility of digital artifacts through increasing the number of pixels a bad thing? The belief that reducing the number of pixels somehow increases the dynamic range and reduces noise is a pervasive one but still wrong.
The downsampling does NOT do it. The downsampling is the simplest and easiest way of cashing in on the sensor's DR in a primitive world where monitors have fixed, coarse pixels.Well, a simple downsampling will do just that (like we can see in DxO's 8mp 'print' figures), so it can't be completely wrong.Since when was increasing resolution and reducing the visibility of digital artifacts through increasing the number of pixels a bad thing? The belief that reducing the number of pixels somehow increases the dynamic range and reduces noise is a pervasive one but still wrong.
On a 1000 DPI monitor?Not even an ISO 51200 image shot with a 200mp APS-C camera? ;-)
Yes, it makes the real relationship between signal and noise more easy to observe. With large monitors you might be able to demonstrate the same thing by simply backing away and viewing images from a distance.The downsampling does NOT do it. The downsampling is the simplest and easiest way of cashing in on the sensor's DR in a primitive world where monitors have fixed, coarse pixels.
One thing I can't understand with the more megapixels means better resolution line of thought that I hope you can explain.Chris59 wrote:
More megapixels means better resolution, fewer megapixels means less resolution - there are no advantages in going forward to the past.
As for me, I can't wait to see images from the rumoured Canon 1D Mark IV with 56 megapixels and no AA filter.
To clarify:skyglider wrote:
One thing I can't understand with the more megapixels means better resolution line of thought that I hope you can explain.
If the viewing medium is 4x6 or 5x7 inch prints, or on typical computer monitors, or on 1920x768 HD TV sets, how is the better resolution provided by more than 10 MP perceptible?
Sky
It wouldn't be. If those are your targets for viewing, then 3Mp would probably suffice just fine.If the viewing medium is 4x6 or 5x7 inch prints, or on typical computer monitors, or on 1920x768 HD TV sets, how is the better resolution provided by more than 10 MP perceptible when viewing the entire picture?
Well, you could use the same line of thought when you argue against quality lenses, a large sensor, quality filters and good technique - no one is going to notice the improvement in resolution so why bother. This also contradicts the "more megapixels equals more noise/less dynamic range" camp - if no one is going to notice with small images, what is the point of having less noise or greater dynamic range? Indeed, we don't really need anything more than a point and shoot with 2 megapixels for full HD images on even the biggest screen, so why do we bother?One thing I can't understand with the more megapixels means better resolution line of thought that I hope you can explain.Chris59 wrote:
More megapixels means better resolution, fewer megapixels means less resolution - there are no advantages in going forward to the past.
As for me, I can't wait to see images from the rumoured Canon 1D Mark IV with 56 megapixels and no AA filter.
If the viewing medium is 4x6 or 5x7 inch prints, or on typical computer monitors, or on 1920x768 HD TV sets, how is the better resolution provided by more than 10 MP perceptible?
Sky
You hit a nerve here. As my beloved Canon A610 developed a stuck shutter beyond economic repair, I picked a slightly used second hand G11, and I have been happy with it ever since despite its having meager 10 Mpx.Not to mention buyers obsessed with less pixels, a.k.a. large pixel huggers.
Well, my monitor is 1920x1200 and it is not even large by today's standard. To answer your question: a weaker AA filter; demosaicing requiring less sharpening, and therefore, less noise; more mp allow for more intelligent NR; less loss of quality when rotating, fixing distortions and CA, cropping, more data for pp, etc. I can see the difference between 8mp and 15mp.To clarify:skyglider wrote:
One thing I can't understand with the more megapixels means better resolution line of thought that I hope you can explain.
If the viewing medium is 4x6 or 5x7 inch prints, or on typical computer monitors, or on 1920x768 HD TV sets, how is the better resolution provided by more than 10 MP perceptible?
Sky
If the viewing medium is 4x6 or 5x7 inch prints, or on typical computer monitors, or on 1920x768 HD TV sets, how is the better resolution provided by more than 10 MP perceptible when viewing the entire picture?
Yep, the 1920x768 should have been 1920x1080 as you said. After all, TV resolution is usually referred to as 720 (always p), 1080i or 1080p. Gotta get my brain out of 1st gearIt wouldn't be. If those are your targets for viewing, then 3Mp would probably suffice just fine.If the viewing medium is 4x6 or 5x7 inch prints, or on typical computer monitors, or on 1920x768 HD TV sets, how is the better resolution provided by more than 10 MP perceptible when viewing the entire picture?
BTW, the higher resolution HD is 1920 x 1080 or just shy of 2Mp.
Yes, you have a point there. But "faster", lower quality lenses do make a difference in low light image quality.Chris59 wrote:
Well, you could use the same line of thought when you argue against quality lenses,
with low pixel density will reduce low light noise. The problem with small, high pixel density sensors is how the manufacturers use excessive noise reduction to try to cover up the noise. That does result in a perceptible drop of low light images in the viewing mediums I mentioned.a large sensor,
Dynamic range limitations are very visible in all of the viewing mediums I mentioned. Blown out skies or clipped shadows are very visible indeed.quality filters and good technique - no one is going to notice the improvement in resolution so why bother.
This also contradicts the "more megapixels equals more noise/less dynamic range" camp - if no one is going to notice with small images, what is the point of having less noise or greater dynamic range?
A 5x7 inch print at 300 dpi requires 3.15MP so that's the bare minimum. Allow for cropping for imperfect framing, straightening out camera tilt, and an occasional 8x10 inch print, a 7MP sensor is fine. But manufacturers will never go back to single digit size sensors, so the smallest size sensor in reality from now on will never be less than 10MP.Indeed, we don't really need anything more than a point and shoot with 2 megapixels for full HD images on even the biggest screen, so why do we bother?
Yes, that's a very valid reason. For the masses though, my guess is that 95% of the small sensor camera buyers never make prints larger than 8x10 and very rarely, if ever, for even that size. 10MP cameras are all they need.For myself, I just like to look at big detailed photographs and I never know ahead of time which of my images I will want to enlarge.
But that's not really true. The noise per pixel may be lower, but the overall sensor noise isn't necessarily greater. In fact, the general trend last I looked was that the pixels got smaller and the noise per pixel was either reduced or increased non-proportionally so that the per area noise was reduced.with low pixel density will reduce low light noise. The problem with small, high pixel density sensors is how the manufacturers use excessive noise reduction to try to cover up the noise. That does result in a perceptible drop of low light images in the viewing mediums I mentioned.