<html> <head> <meta http-equiv="Content-Type" content="text/html; charset=ISO-8859-15"> </head> <body> <div class="moz-cite-prefix">On 19/06/2020 08:23, Hilaire Fernandes via Cuis-dev wrote: </div> <blockquote type="cite" cite="mid:15a1558e-224d-1510-3362-43a6c8b1125b@drgeo.eu"> <meta http-equiv="Content-Type" content="text/html; charset=ISO-8859-15"> I try to imagine a dynabook with two A4 screens, where the information is layout in a vectorial canvas for each screen whose contents are texts flowing around rectangular like Morphs with dynamic content and an aditionnal layer of vectorial hand writtings.</blockquote> I have wanted something like this, with both touch and (pressure sensitive) pen support ever since I saw a speculative video showing an ebook reader in this style over 20 years ago. <blockquote type="cite" cite="mid:15a1558e-224d-1510-3362-43a6c8b1125b@drgeo.eu"> In that scenario, can the start up cost of OpenCL be mitigated? </blockquote> I have to wonder how much the display model of the VM restricts this. Maybe directly using something hardware accelerated like OpenGL, Vulcan or even SDL2 would help, VM side? </body> </html>