Anonymous user
Talk:Pinstripe/Printer: Difference between revisions
no edit summary
m (We can always knock it back to draft again) |
No edit summary |
||
Line 16:
:::: The task was drafted, people improved it. I waited a month for feedback. No more came, I saw no problems, and we had some implementations, so I undrafted it. We can always knock it back to draft again, if there is a problem.
:::: It would be an acceptable solution to use the smallest marks that the language provides, rather than the printer provides, if this makes a difference. I do want to go too far down the road of "subpixel rendering" here, so true resolution is also an acceptable solution here. [[User:Markhobley|Markhobley]] 09:09, 9 July 2011 (UTC)
::::: There's no "smallest marks" provided by the language, at least not in postscript. You can draw a rectangle as small as the floating number precision the VM allows, and the language spec has something about how a renderer is supposed to rasterize it (if size is not zero, ''something'' should be drawn). If you have a PS printer, you could draw a line that's as thin as possible in some sense (0 setlinewidth), but you can't tell how much to move forward to draw the next line so that it's exactly "1 pixel" apart -- and why would you ever want to do that? For a printout, you always care more about how big it is physically in terms of inches and milimeters, and that's why PS and PCL (and PDF) are vector based page description. On a modern printer, you don't quite control if it does "subpixel rendering", and it normally doesn't need to: you have to sqint pretty hard to see a 600dpi dot, it's pointless. Really, this task serves what purpose exactly? Could you outline it here? --[[User:Ledrug|Ledrug]] 02:13, 10 July 2011 (UTC)
| |||