Rodolfo Paiz
Offline
Image Upload: On
|
brainiac wrote:
Some people can see why this is true, and others can't.
Absent clear evidence of obtuse behavior, it is best to assume that all reasonable people can see something if it's true. Statements like the one above give me the impression that anyone who disagrees will be brushed off as being "one of those who can't." Emperor's new clothes, and all that.
brainiac wrote:
Imagine you print a picture at 16x24 and at 24x36. You then cut out a square inch of print from each, and hand the two crops to a judge. She will say that the cropped square inch from the more enlarged print has worse (more visually intrusive) noise than the crop from the smaller enlargement. The noise will be bigger. The less enlarged crop will clearly be the noise winner. But it's the same file.
If you're trying to make any kind of scientific and fact-based analysis, numbers become essential. Don't just bandy about generalities, it confuses the issue. So in order to understand you better, and to ensure that I am looking at the same "facts" as you are, let me see if I can put numbers to this example of yours.
For ease of use, I'm going to use round numbers. So I'll assume that your picture above is actually 4800x7200 pixels in size, so that the 16x24 print is at 300 dpi and the 24x36 print is at 200 dpi. No interpolation or software manipulation involved. If that is indeed what you meant to say, then of course the smaller print is going to look better, since it's got better resolution (defined as number of pixels per unit of distance, in this case PPI). Any noise will be reduced in size, and every line, curve, and detail captured will seem to be of higher quality as well, because for that print it really is.
If, however, you make prints of similar images from two different cameras, and print both at 300 dpi without any image manipulation, then I think cutting that one-square-inch piece from each print is entirely fair. The camera with more MP has the ability to make larger prints without image resizing, and that is an advantage in and of itself. But it is separate from the camera's noise signature, which is inherently pixel-based.
At this point, I am of the opinion that either your theory is wrong, or your explanation is woefully inadequate. A picture of "the Queen's head on a postage stamp", as you put it, will favor the camera with the most MP (not the least, as you said) because the Queen's head was photographed with more pixels and is thus printed at higher resolution if it's printed at the same size. And, the laws of physics still say that a larger photosite will more accurately capture light, which is why the D3 has better noise performance than the D300: equal number of pixels (roughly), but each pixel is bigger and fatter and therefore does better.
brainiac wrote:
Meanwhile, I am criticised for being unscientific, or biased.
The good thing is that you're looking to validate your views based on empirical evidence and facts. Honestly, though, the bad thing is that it soulds like you're way too convinced that your method is correct and valid... and sometimes the problem is not in the answer, but in how we asked the question. Do some constructive self-examination... can all those people really be totally wrong, and you're the only guy who sees the light? Or is there a flaw in your paradigm somewhere, and there really is something to criticize in your methodology or philosophy? Note that I'm not saying "maybe you're wrong," because this is not that simple. It's almost impossible to be 100% wrong or 100% right, which makes ego-less self-evaluation all the more critical.
Please feel free to continue the discussion and explain your reasoning further... I'm listening. At this point, what you've said gives me the impression that your reasoning is not valid and your conclusions are therefore incorrect. But I'm only telling you what I think now; again, if you feel that I have misunderstood or that there is more to say, I'm listening.
|
