18783 2012-09-05 19:42:09 +0000 imageSmothingEnabled should not be associated strictly with image scaling 2013-02-04 20:58:02 +0000 1 1 1 Unclassified WHATWG HTML unspecified Other other RESOLVED FIXED http://www.whatwg.org/specs/web-apps/current-work/#image-smoothing P3 normal Unsorted 1 contributor ian ian junov mike contributor oldest_to_newest 73256 0 contributor 2012-09-05 19:42:09 +0000 Specification: http://www.whatwg.org/specs/web-apps/current-work/multipage/the-canvas-element.html Multipage: http://www.whatwg.org/C#image-smoothing Complete: http://www.whatwg.org/c#image-smoothing Comment: imageSmothingEnabled should not be associated strictly with image scaling Posted from: 2620:0:100c:0:cd41:eadb:836c:786d by junov@chromium.org User agent: Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.1 (KHTML, like Gecko) Chrome/21.0.1180.83 Safari/537.1 73257 1 junov 2012-09-05 19:49:42 +0000 From the current description of imageSmoothingEnabled: Returns whether pattern fills and the drawImage() method will attempt to smooth images if they have to rescale them (as opposed to just rendering the images with "big pixels" The term "rescale" should probably be replaced with "resampled". Image smoothing is pertinent for for rotations, sub-pixel translations, and other transforms. Basically any type of transform that is not pixel-aligned. Also, the comment in parenthis assumes magnification. Perhaps it should be more general: "as opposed to nearest neighbour filtering" 78005 2 ian 2012-11-07 19:32:00 +0000 Justin: The text in question here is non-normative and is intended for authors who may not be graphics experts. Do you still think it should be changed? 78029 3 junov 2012-11-08 03:41:33 +0000 (In reply to comment #2) > Justin: The text in question here is non-normative and is intended for > authors who may not be graphics experts. Do you still think it should be > changed? I get your point. "resample" is a little too expert-ish. Nonetheless, the current text falsely suggests that smoothing is strictly related to scale. Even though it is non-normative, the description could be more accurate. I agree that the most prevalent use case for this feature is likely to be the "big pixel" look, but there is more to it than that. Here is an alternate suggestion which I think would be clear to non graphics experts, but at the same time gives accurate information to implementors and to authors who are graphics experts: Returns whether pattern fills and the drawImage() method will attempt to smooth images in cases where source image pixels don't line up one-for-one with canvas pixels. (as opposed to, for example, rendering enlarged images with "big pixels") 82544 4 ian 2013-02-04 20:57:47 +0000 I've tried to make the text clearer, and dropped the mention of "big pixels". Let me know (and reopen this bug) if the text is inadequate now. 82545 5 contributor 2013-02-04 20:58:02 +0000 Checked in as WHATWG revision r7691. Check-in comment: Precision in the non-normative text. Hopefully people still understand what it means. http://html5.org/tools/web-apps-tracker?from=7690&to=7691