lenicolas Posted February 28, 2014 Share #21 Posted February 28, 2014 Advertisement (gone after registration) What about the aperture? If the lens is longer then maybe the aperture is actually faster than 1,4! You think you have a f/1,37 lens? What does your meter say? "Envoyé depuis mon HTC One X avec Tapatalk" Link to post Share on other sites More sharing options...
Advertisement Posted February 28, 2014 Posted February 28, 2014 Hi lenicolas, Take a look here Resolution not 6000 x 4000?. I'm sure you'll find what you were looking for!
farnz Posted February 28, 2014 Share #22 Posted February 28, 2014 What about the aperture? If the lens is longer then maybe the aperture is actually faster than 1,4!You think you have a f/1,37 lens?... It wouldn't have any effect on the speed of the lens or its aperture. If you think about it, lenses don't get faster on cropped sensors even though the effective field of view is decreased. Pete. Link to post Share on other sites More sharing options...
Nick_S Posted March 1, 2014 Share #23 Posted March 1, 2014 No one has explained the technical reason why the DNG and JPEG files are different sizes. If I take a raw or TIFF file and save it with lossy compression to JPEG using computer software, the number of pixels never changes (which I appreciate as sometimes I target an end number of pixels such as for web display, or for printing to an exact size to avoid up- or down-sampling by the printer). I have noticed on the Monochrom that there is a slight change in the displayed images as the initial JPEG image is replaced by a rendering of the DNG file. This I took to be deliberate so that one could tell which was being displayed when viewing with the colour clipping indicators on. Could there be some analogous reason for the differences in the full size files in the present case? Nick Link to post Share on other sites More sharing options...
CheshireCat Posted March 1, 2014 Share #24 Posted March 1, 2014 No one has explained the technical reason why the DNG and JPEG files are different sizes. The JPEG size has the standard aspect ratio of 3/2 == 1.5 Link to post Share on other sites More sharing options...
pop Posted March 1, 2014 Share #25 Posted March 1, 2014 It wouldn't have any effect on the speed of the lens or its aperture. If you think about it, lenses don't get faster on cropped sensors even though the effective field of view is decreased. Pete. The aperture is expressed as the ratio between the diameter of the diaphragm and the focal length. Making the lens shorter makes it brighter, much the same as with people, I hear. Link to post Share on other sites More sharing options...
luigi bertolotti Posted March 1, 2014 Share #26 Posted March 1, 2014 Btw, the effective focal length varies with focus distance too, and this also affects the effective aperture, significant only at very short focusing distances, which results in the "exposure factors" often engraved on bellows Link to post Share on other sites More sharing options...
lenicolas Posted March 1, 2014 Share #27 Posted March 1, 2014 Advertisement (gone after registration) It wouldn't have any effect on the speed of the lens or its aperture. If you think about it, lenses don't get faster on cropped sensors even though the effective field of view is decreased. Changing the focal length does change the relative aperture (f/). As does changing the magnification (the reason why bellows need compensation) Never heard anyone claim that the field of view had any effect on aperture. Link to post Share on other sites More sharing options...
jaapv Posted March 1, 2014 Share #28 Posted March 1, 2014 As field of view has to do with distance which has to do with focus distance which has to do with relative focal length, yes field of view does relate to aperture. I guess you mean angle of view, which is something else. Link to post Share on other sites More sharing options...
lenicolas Posted March 1, 2014 Share #29 Posted March 1, 2014 The comment I was responding to was talking about the changes in field of view induced by cropping, surely then you'll agree with us that cropping a photograph never did anything to relative aperture, right? Link to post Share on other sites More sharing options...
jaapv Posted March 1, 2014 Share #30 Posted March 1, 2014 No,you are right there but neither has cropping to do with field of view, unless you specify a distance. Link to post Share on other sites More sharing options...
r7photo Posted March 1, 2014 Share #31 Posted March 1, 2014 Shoot uncompressed Link to post Share on other sites More sharing options...
CheshireCat Posted March 1, 2014 Share #32 Posted March 1, 2014 No,you are right there but neither has cropping to do with field of view, unless you specify a distance. brb getting popcorn... Link to post Share on other sites More sharing options...
jaapv Posted March 1, 2014 Share #33 Posted March 1, 2014 The rendering of popcorn has to do with resolution… Link to post Share on other sites More sharing options...
Rob Roskopp Posted March 1, 2014 Share #34 Posted March 1, 2014 No one has explained the technical reason why the DNG and JPEG files are different sizes. If I take a raw or TIFF file and save it with lossy compression to JPEG using computer software, the number of pixels never changes (which I appreciate as sometimes I target an end number of pixels such as for web display, or for printing to an exact size to avoid up- or down-sampling by the printer). I have noticed on the Monochrom that there is a slight change in the displayed images as the initial JPEG image is replaced by a rendering of the DNG file. This I took to be deliberate so that one could tell which was being displayed when viewing with the colour clipping indicators on. Could there be some analogous reason for the differences in the full size files in the present case? Nick Maybe the camera requires some extra pixels around the borders/corners for lens/distortion corrections to be applied. The corrections might be present in the JPEGs, thus the slightly smaller resolution. Link to post Share on other sites More sharing options...
pop Posted March 1, 2014 Share #35 Posted March 1, 2014 The rendering of popcorn has to do with resolution… The rendering of some kinds of corn also has to do with my resolution. Link to post Share on other sites More sharing options...
luigi bertolotti Posted March 1, 2014 Share #36 Posted March 1, 2014 Maybe the camera requires some extra pixels around the borders/corners for lens/distortion corrections to be applied. The corrections might be present in the JPEGs, thus the slightly smaller resolution. Surely jpg processing takes into account pixels around... but I think the different pixel count is a design choice : being a bitmap obtained by processing, they could surely make it, if they liked, of the same count of DNG... maybe it is related indeed to the exact 3/2 ratio (it's so "round" that looks a specific choice, keeped also in some of the lower res. jpgs) , or to the LCD bitmap management, in which a jpg image is used. Link to post Share on other sites More sharing options...
CheshireCat Posted March 1, 2014 Share #37 Posted March 1, 2014 Maybe the camera requires some extra pixels around the borders/corners for lens/distortion corrections to be applied. The M does not correct distortion. And even if it did, those extra pixels would not be enough. It simply is just an aspect ratio matter. While it is possible to have a JPEG file with the native sensor resolution, no one would want to postprocess it (time and further quality loss) just to match the standard print aspect ratio. Link to post Share on other sites More sharing options...
luigi bertolotti Posted March 1, 2014 Share #38 Posted March 1, 2014 The M does not correct distortion. And even if it did, those extra pixels would not be enough. It simply is just an aspect ratio matter. While it is possible to have a JPEG file with the native sensor resolution, no one would want to postprocess it (time and further quality loss) just to match the standard print aspect ratio. Indeed one of my hipotesis (I didn't express it completely in the above post) is that the pixel count of M hi res jpg is what they obtained with the best (and quickest) algorhitm, given the constraint to have a 3/2 aspect ratio. Link to post Share on other sites More sharing options...
Nick_S Posted March 2, 2014 Share #39 Posted March 2, 2014 So basically from the responses given, it seems that the original sensor resolution is not quite 3:2, with some pixels being used for other purposes than just recording light at a given location. So the image is cropped to the exact 3:2 ratio to make JPEGs more convenient for immediate use with minimal post-processing. Nick Link to post Share on other sites More sharing options...
pop Posted March 2, 2014 Share #40 Posted March 2, 2014 When demosaicing and - I believe - converting to JPEG, the camera has to take into account for the calculation of each pixel the values of the pixels surrounding it. As the pixels right at the border of the image area do not have the same number of neighbouring pixels as those within the image area, their values would not be calculated to the same acccuracy. Hence, it seems reasonable to use their values for the reckoning of those pixels surrounded by them but not including them into the image. Hence, there would not be pixels missing from the JPEG but extra sites added to the raw pixel array, in order to arrive at the targeted image size. Link to post Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.