H. G. Muller wrote on Mon, Oct 2, 2017 08:37 AM UTC:
The only difference I see is that the weight of the text font now seems a bit better. The font for the headers is still awful. And the text font is still way too big. It completely wrecks pages with floating images on my 1366px-wide screen, e.g. the Chu Shogi article.
> As I understand the meaning of vw, this would make the size of the cells dependent on the size of the window of the client, while the piece images have a fixed size.
No, piece images shrink when the width narrows. In the CSS, all images have max-width set to 100% and height set to auto.
Well, that is very bad. These are raster images, not vecor images, and thus lose a lot of quality on resizing. Especially if they are not anti-aliased, such as Alfaerie. I don't provide separate xboard50 and xboard33 piece sets for no reason, even though the XBoard piecesare anti-aliased. I want the author to decide how much resolution a piece image needs, and the browser not too mess with it. Have you seen what happens in your Chess-and-a-half test diagram with vw units when you resize the window? The board cells get of unequal width, and unregularly jump as you increase the window width.
But this has become a moot point, as the non-compliant behavior seems to be caused by too small a font size, and is cured by specifying a very large font size in the <td> cells, which do not contain any text anyway.
The only difference I see is that the weight of the text font now seems a bit better. The font for the headers is still awful. And the text font is still way too big. It completely wrecks pages with floating images on my 1366px-wide screen, e.g. the Chu Shogi article.
Well, that is very bad. These are raster images, not vecor images, and thus lose a lot of quality on resizing. Especially if they are not anti-aliased, such as Alfaerie. I don't provide separate xboard50 and xboard33 piece sets for no reason, even though the XBoard piecesare anti-aliased. I want the author to decide how much resolution a piece image needs, and the browser not too mess with it. Have you seen what happens in your Chess-and-a-half test diagram with vw units when you resize the window? The board cells get of unequal width, and unregularly jump as you increase the window width.
But this has become a moot point, as the non-compliant behavior seems to be caused by too small a font size, and is cured by specifying a very large font size in the <td> cells, which do not contain any text anyway.