Commons:Featured picture candidates/File:Gates-Hillman Complex at Carnegie Mellon University 3.jpg

From Wikimedia Commons, the free media repository
Jump to navigation Jump to search

File:Gates-Hillman Complex at Carnegie Mellon University 3.jpg[edit]

Voting period is over. Please don't add any new votes.Voting period ends on 25 Oct 2015 at 23:49:33 (UTC)
Visit the nomination page to add or modify image notes.

The Gates-Hillman Complex, a unique building at Carnegie Mellon University, housing the prestigious School of Computer Science, consistently ranked amongst the best computer science programs in the world.
  •  Comment White balance looks fine to me. It is a daylight white balance for a daylight photo. If we inspect the concrete pillars underneath the building, they are actually yellowish. What exactly is too bluish? The windows are naturally supposed to be blue/green tinted. The grey tiles on the exterior of the building are not exactly purely desaturated in real life either. dllu (t,c) 14:14, 17 October 2015 (UTC)[reply]
  • I am shocked to hear you complain about sharpness for a 142 megapixel image. Let me remind you that this was shot using the Sony Zeiss T* Sonnar 55mm f/1.8 lens on the Sony A7R undisputedly one of the sharpest optical systems you can buy for $3000. Moreover this was shot at f/8, at ISO 100, at 1/300 s shutter speed, which all photographers will agree are the optimal settings for sharpness. The reason for softness is that this is a rectilinear projection (to preserve straightness of lines) which naturally stretches the image in the sides (not unlike perspective correction) -- of course there is going to be loss of sharpness at regions other than the center. But I chose to upload the highest resolution to preserve detail in the center. The next time I choose to contribute a photo to Wikimedia Commons, I will have two options. Either I can upload the full 142 megapixel image, or I can downsample it by a factor of 16 to only 8.8 megapixels. Many photos -- even of architecture! fewer than 8 megapixels sail happily through the FP process without issue, simply because the resolution is low enough to hide any aberrations. When downsampled to 8 megapixels, all the stitching errors you see will become subpixel in magnitude and nobody will notice them. At 8.8 megapixels, everything will be tack sharp and there will be no noise. Look at the picture below, which do you prefer?
Please view this at full resolution. On the bottom left we see a crop from the version downsampled to 3958 by 2235 pixels. On the right is the original. The minor stitching error is all but gone in the left, but at what cost?
When future photographers read Commons:FPC and see people complaining about sharpness in a sharp 142 megapixel image, do you think anyone will still upload the full resolution? Please reflect carefully on the harm that your words may cause. dllu (t,c) 14:14, 17 October 2015 (UTC)[reply]