Roland Karlsson
Veteran Member
Exactly!
--
Roland
support http://www.openraw.org/
(Sleeping - so the need to support it is even higher)
X3F tools : http://www.proxel.se/x3f.html
-
Welcome to the new forums! Please read this first. For known issues we are working to resolve, click here.
Updated product pages: Why not pixel density ?
- Thread starter h2k
- Start date
--
Roland
support http://www.openraw.org/
(Sleeping - so the need to support it is even higher)
X3F tools : http://www.proxel.se/x3f.html
Roland Karlsson
Veteran Member
Of course. But we were talking about sensor image quality. And sensor image quality usually go up with number of pixels, contrary to what many believe.
--
Roland
support http://www.openraw.org/
(Sleeping - so the need to support it is even higher)
X3F tools : http://www.proxel.se/x3f.html
tom sugnet
Veteran Member
For A3 prints, clean ISO6400 is more important than astronomic sharpness.
Clients are not really interesting in seeing tiny skin pores.
But a noise-free low light no flash portrait is what many wedding and performance photographers are after.
--
Feel free to visit my photo sites:
http://tom.st , http://www.foto.tom.st
Clients are not really interesting in seeing tiny skin pores.
But a noise-free low light no flash portrait is what many wedding and performance photographers are after.
--
Feel free to visit my photo sites:
http://tom.st , http://www.foto.tom.st
bobn2
Veteran Member
Great Bustard
Veteran Member
- Messages
- 45,961
- Solutions
- 17
- Reaction score
- 34,046
...of assuming that more pixels is at odds with cleaner photos. Let's say we have two sensors with the same efficiency (same QE and same read noise / area). One sensor is 10 MP, the other is 40 MP.
We take a pic of the same scene with the same settings. Let's say, f/2.8 1/100 ISO 6400.
Now, at 100% view, some will say that the 40 MP photo is more noisy. Others will say the 10 MP photo is more blurry:
http://forums.dpreview.com/forums/read.asp?forum=1018&message=35068712
Either way, this is because at 100% view, you are viewing the scene from the 40 MP sensor with double the enlargement as from the 10 MP sensor.
There are a couple of things we can do. We can just print the photos. If the person vewing the photo is unable to resolve the higher detail of the 40 MP photo (due to viewing distance, print size, visual acuity, paper, etc.), then the 40 MP photo won't look any more noisy.
If, however, they can see the extra detail from the 40 MP photo, and they would prefer a cleaner photo to a more detailed photo, we could simply apply NR (noise reduction) to the 40 MP photo, and create a photo that has the same detail as the 10 MP photo, but with less noise.
In other words, more pixels simply give you more IQ options than fewer pixels for equally efficient sensors.
Nor are sensors with larger pixels necessarily more efficient. The Canon G12 has larger pixels than the Canon G10, and the G12 sensor is more efficient. However, the Canon 5D2 has smaller pixels than the Canon 5D, but the 5D2 sensor is more efficient.
In both cases, it's the newer generation sensor that is more efficient, more pixels or not. So, the more pixels you can put on the sensor, the better, in terms of IQ.
Roland Karlsson
Veteran Member
You are making the assumption that less pixels makes the A3 print visible cleaner?
--
Roland
support http://www.openraw.org/
(Sleeping - so the need to support it is even higher)
X3F tools : http://www.proxel.se/x3f.html
Le Kilt
Veteran Member
No, there is nothing abstract about pixel density and it is exactly what I was looking for, if DPReview used it for a time there was a reason.
But worry not, you are free to look up and use whichever measurements you feel more comfortable with ;-)
robert1955
Veteran Member
- Messages
- 8,149
- Solutions
- 10
- Reaction score
- 4,959
The reasons DPR used it have been amply discussed above. You want to use it for a completely different reason.
If I understand you correctly you want to know a number for 'how many pixels on the duck'. That was at least the way someone tried to show it years ago: if you photograph a sitting duck. When shooting such a subject, a crop camera not only has the advantage of the smaller sensor [same focal length gives a smaller angle of view] but mostly [at least in the same generation] smaller pixels than the FF. This gives a more opportunity to crop if you use the result for screen display or small prints only.
Pixel density gets you information on 'cropability' only partially, as it can't take the final step into account.
If I understand you correctly you want to know a number for 'how many pixels on the duck'. That was at least the way someone tried to show it years ago: if you photograph a sitting duck. When shooting such a subject, a crop camera not only has the advantage of the smaller sensor [same focal length gives a smaller angle of view] but mostly [at least in the same generation] smaller pixels than the FF. This gives a more opportunity to crop if you use the result for screen display or small prints only.
Pixel density gets you information on 'cropability' only partially, as it can't take the final step into account.
Le Kilt
Veteran Member
Several reasons actually ;-)
I'm not quite sure of the point you're trying to make, but it's a question of whether a piece of information is useful or interesting, and the less obvious implications do need an understanding of other factors affecting image quality
bobn2
Veteran Member
As has been discussed before, the pixel density is not 'primary' information. There are two basic bits of information, the sensor size (x and y) and the pixel count (x and y), from these can be derived other things such as the pixel pitch and pixel density. The question is, if more than the two basic figures are to be given, which and why. There has been a case made for pixel pitch. Only one case has been made for pixel density, that is that it helps make 'reach' calculations. This is a bit bogus, because pixe frequency (reciprocal of pixel pitch) would be directly applicable to that need, so even that case is weak. So now it would seem if we are to have figures apart from the basic size and pixel count, that two (pitch and frequency) would be more usefult than 'density'. What the proponents of 'density' are having a hard time doing is actually coming up with some concrete reason why it's useful - w3e are given vague reasons or just 'I like it'.Several reasons actually ;-)
I'm not quite sure of the point you're trying to make, but it's a question of whether a piece of information is useful or interesting, and the less obvious implications do need an understanding of other factors affecting image quality
--
Bob
Le Kilt
Veteran Member
Is the reason for it being withdrawn not due to its misuse?As has been discussed before, the pixel density is not 'primary' information. There are two basic bits of information, the sensor size (x and y) and the pixel count (x and y), from these can be derived other things such as the pixel pitch and pixel density. The question is, if more than the two basic figures are to be given, which and why. There has been a case made for pixel pitch. Only one case has been made for pixel density, that is that it helps make 'reach' calculations. This is a bit bogus, because pixe frequency (reciprocal of pixel pitch) would be directly applicable to that need, so even that case is weak. So now it would seem if we are to have figures apart from the basic size and pixel count, that two (pitch and frequency) would be more usefult than 'density'. What the proponents of 'density' are having a hard time doing is actually coming up with some concrete reason why it's useful - w3e are given vague reasons or just 'I like it'.Several reasons actually ;-)
I'm not quite sure of the point you're trying to make, but it's a question of whether a piece of information is useful or interesting, and the less obvious implications do need an understanding of other factors affecting image quality
--
Were you not also able to deduce pixel pitch and frequency from pixel density?
A piece of interesting information is taken away from the front page, and we're being told you shouldn't have been looking at that, I like this alternative that you don't get either...
I do apologise for "liking it", it's like an old friend who's been put down, sniff... :-(
tom sugnet
Veteran Member
It would take some really exotic and expensive glass to be able to keep up with the pixel densities you're dreaming of (point&shootish ones in a full frame).
--
Feel free to visit my photo sites:
http://tom.st , http://www.foto.tom.st
--
Feel free to visit my photo sites:
http://tom.st , http://www.foto.tom.st
Roland Karlsson
Veteran Member
Hmmmm .... so when you buy milk, vegetable oil and other liquids you would want to have the density specified? Isnt volume and/or weight somewhat more useful? Except in very special cases of course.
No ... not really. I cannot look up whats not there. And now there is neither pixel density nor pixel pitch.
And actually I would not be bothered if any of those were added.
The main thing that bothered me regarding pixel density was that DPReview used it as a weapon in the anti mega pixel race crusade.
More pixels is in general good. Up to some limit where it simply is more than you can use or actually starts to degrade the image.
--
Roland
support http://www.openraw.org/
(Sleeping - so the need to support it is even higher)
X3F tools : http://www.proxel.se/x3f.html
Roland Karlsson
Veteran Member
I can sympathize with this motivation.
Losing things is never fun. Its against the human nature to want to lose anything at all. Just take a look in your homes guys and girls .... arnt there lots of things you really dont need and never is going to use? Still you keep them.
One point to you!
--
Roland
support http://www.openraw.org/
(Sleeping - so the need to support it is even higher)
X3F tools : http://www.proxel.se/x3f.html
bobn2
Veteran Member
I think it is an unwelcome reminder of a failed and misguided campaign.Is the reason for it being withdrawn not due to its misuse?As has been discussed before, the pixel density is not 'primary' information. There are two basic bits of information, the sensor size (x and y) and the pixel count (x and y), from these can be derived other things such as the pixel pitch and pixel density. The question is, if more than the two basic figures are to be given, which and why. There has been a case made for pixel pitch. Only one case has been made for pixel density, that is that it helps make 'reach' calculations. This is a bit bogus, because pixe frequency (reciprocal of pixel pitch) would be directly applicable to that need, so even that case is weak. So now it would seem if we are to have figures apart from the basic size and pixel count, that two (pitch and frequency) would be more usefult than 'density'. What the proponents of 'density' are having a hard time doing is actually coming up with some concrete reason why it's useful - w3e are given vague reasons or just 'I like it'.Several reasons actually ;-)
I'm not quite sure of the point you're trying to make, but it's a question of whether a piece of information is useful or interesting, and the less obvious implications do need an understanding of other factors affecting image quality
--
Not directly, you need to know the sensor size and/or pixel count as well, and if you have both those things you can calculate all of pixel pitch, pixel frequency and pixel density. The question is, why should pixel density have favoured status over pixel pitch and pixel frequency when both of those are more directly useful in concrete ways.
And I can't see why you're getting emotional about it, since you've been unable to say what use it actually was to you.
--
Bob
bobn2
Veteran Member
The point is not to have the glass keep up with it. The point is to extract the maximum out of every piece of glass.
--
Bob
Roland Karlsson
Veteran Member
Well put.
The main problem for DPReview (and indirectly us) is that the camera manufacturers dont generally give the actual sensor size. Its only given for system cameras.
So neither pixel pitch nor pixel density can be computed.
Not only was the pixel density DPReview presented of questionable value, it was also wrong, based on assumptions how large e.g. a 1/2.3 inch sensor really was.
As soon as you can remove the lens you can get the actual size, either by the manufacturer giving it or by you measuring it.
--
Roland
support http://www.openraw.org/
(Sleeping - so the need to support it is even higher)
X3F tools : http://www.proxel.se/x3f.html
ejmartin
Veteran Member
Easy enough to do using simple geometry. They take photos of calibration targets; all you need to know is the lines per mm of the target, the focal length, and the distance from the camera to the target. Simple geometry gets you the pixel pitch once you measure the number of pixels per line. I've done this with some of my cameras and a photo of a ruler, it's reasonably accurate.
--
emil
--
http://theory.uchicago.edu/~ejm/pix/20d/
Roland Karlsson
Veteran Member
I understand how to do it.
But ... to be able to do it you need to know the focal length of the lens. How accurate focal length do you think you can get from a 4 - 11 mm zoom?
And even if you are able to get the focal length you have to take into consideration that most compact cameras have huge distortion in the lenses.
--
Roland
support http://www.openraw.org/
(Sleeping - so the need to support it is even higher)
X3F tools : http://www.proxel.se/x3f.html
Steen Bay
Veteran Member
Well, just noticed that DPR's database tells us that the "lens focal length" for e.g. the G12 is "28-140mm". Maybe a little 'equivalent' would be appropriate here? ;-)
Keyboard shortcuts
- f
- Forum