Thank you, very good info. Manymouse looks like very useful for my needs. I am prototyping bitmap editor now, probably it will grow into something serious, I have very good ideas in this direction. But on the other hand would be nice to stay with pygame, writing in python is just a pleasure, so to this time I almost forgot all my adventures with C. BTW, I have tested now the pygame's mouse a bit more and I take my words back about the loop frequency and cursor speed - so it is all OK, the movements are precise and equal at all FPS, but yes, indeed the movement speed calculated out of get_rel() looks exactly the same as system cursor speed. It is not really bad actually, though no chance to overcome OS-level setups.
On Friday, January 16, 2015 at 1:46:53 AM UTC+1, Weeble wrote: > > Oops, I meant to reply to the list. Resending: > > Maybe you could use something like manymouse[1] and interface to it > with ctypes? It's a cross-platform C library for low-level mouse > input. On Windows it uses raw input[2] which gets raw movement data > with no acceleration applied[3]. Of course, you could write directly > for the Windows API, but I think manymouse looks a bit easier and will > be handy if you want to use other platforms. Using ctypes can be a bit > tricky if you're new to it, but this API looks extremely > straightforward, so I don't think it should be too painful. > > [1] - http://icculus.org/manymouse/ > [2] - > http://msdn.microsoft.com/en-us/library/windows/desktop/ms645543(v=vs.85).aspx > > [3] - > http://msdn.microsoft.com/en-us/library/windows/desktop/ee418864%28v=vs.85%29.aspx > > > On 16 January 2015 at 00:01, Mikhail V <mikha...@gmail.com <javascript:>> > wrote: > > Wow that works! Thanks a lot, you made my day. It is however not truely > > low-level, since the values depend on my loop frequency, but anyway > solves > > my tasks at 100%. And yes the trick is not obvious at ALL. It is very > good > > reason to add this info to pygame official documentation since its a > very > > common task. > > > > On Friday, January 16, 2015 at 12:26:18 AM UTC+1, Weeble wrote: > >> > >> It is possible, just not entirely obvious. I believe you need to: > >> > >> 1. Grab mouse focus with pygame.event.set_grab() > >> 2. Hide the mouse cursor with pygame.mouse.set_visible() > >> 3. Read mouse events, taking the rel.x and rel.y properties to find > >> the relative mouse motion. > >> > >> Make sure to have a way to release the grab, otherwise your app will > >> be difficult to get out of! > >> > >> I haven't tried this lately, but I'm sure I've done it in the past. > >> Here's the documentation for SDL, which Pygame is built on top of, and > >> which describes the mouse motion events in a bit more detail: > >> > >> > http://www.libsdl.org/release/SDL-1.2.15/docs/html/sdlmousemotionevent.html > >> > >> On 15 January 2015 at 23:05, Mikhail V <mikha...@gmail.com> wrote: > >> > Hello pygame devs and users, > >> > > >> > I have a simple task - I want to read input from the mouse - NOT the > >> > cursor > >> > position inside my window, but the actual input data from the device. > >> > So currently I suppose it is not possible with Pygame. Is there a > >> > possiblity > >> > to add such capabilities to Pygame? I know that depends on the device > >> > type > >> > and connection so it is impossible to implement tools for all > devices. > >> > But > >> > having tools to read low-level input at least for the mouse is > >> > absolutely a > >> > must for developing interactive applications. > >> > > >> > If there is no chance with pygame then what would you recommend, say, > >> > for > >> > Windows and C/C++ to read from the mouse at low level? > >> > > >> > Mikhail > >> > > >> > >
