Image talk:Quality comparison jpg vs saveforweb.jpg

From Wikipedia, the free encyclopedia

[edit] Constructive criticism

While I understand concerns of the image size, as with other people in the FP nomination (I was going to comment there but carried out some tests and the picture had been delisted by the time I cam back) I don't think it makes sense to have a lossly compressed JPG even at level 12 and even if the picture seems to show the quality difference is minimal because you can't rule out additional compression artifacts as well as an interaction between compression artifacts introduced by the second compression step. Also, for a proper comparison you ideally should have a RAW source (that was never lossly/JPEG compressed) and you should include this as the baseline (using JPEG maximum quality as the baseline is insufficient IMHO). Indeed currently it's somewhat unclear to me what the source was. If the source was a JPEG this is a major confounding factor in the comparisons. Note that size wise, it isn't too bad, a PNG is about 19.6mb (18.7mb with PNGOUT) so less then double. A JPEG could be included additionally but a losslessly compressed file should be the main source. Also, do note that PNGs aren't uncompressed they are losslessly compressed. Big difference... Nil Einne 11:30, 8 October 2007 (UTC)