WM_INPUT support for SDL1.2 | High-definition mouse support for Windows

Have I understood correctly that 1.2.x does not have high definition
mouse support on Windows?

(according to this MS article, WM_INPUT must be used
http://msdn.microsoft.com/en-us/library/ee418864(v=VS.85).aspx

I notice WM_MOUSEMOVE is used instead and I guess DirectInput too.
According to MS[on the above document too] they are both 'laggy’
compared to WM_INPUT)

If that’s true it took me by a negative surprise since I assumed
ioquake3 had that covered all these years.

I notice 1.3 has code including WM_INPUT.

So my questions are

  1. Are the above true? (is really high-definition not supported on 1.2?)

  2. Is it possible to use WM_INPUT on 1.2? I don’t control ioquake3 but
    I’d like to use it there, it also sounds a bit hard to include it. SDL
    isn’t the most simple piece of code.

I’d love it if someone had something like that hacked already. [1.2
with WM_INPUT]

what’s the right maling list to ask about this?On Sat, Apr 10, 2010 at 1:49 AM, Michael Menegakis <@Michael_Menegakis> wrote:

Have I understood correctly that 1.2.x does not have high definition
mouse support on Windows?

(according to this MS article, WM_INPUT must be used
http://msdn.microsoft.com/en-us/library/ee418864(v=VS.85).aspx

I notice WM_MOUSEMOVE is used instead and I guess DirectInput too.
According to MS[on the above document too] they are both 'laggy’
compared to WM_INPUT)

If that’s true it took me by a negative surprise since I assumed
ioquake3 had that covered all these years.

I notice 1.3 has code including WM_INPUT.

So my questions are

  1. Are the above true? (is really high-definition not supported on 1.2?)

  2. Is it possible to use WM_INPUT on 1.2? I don’t control ioquake3 but
    I’d like to use it there, it also sounds a bit hard to include it. SDL
    isn’t the most simple piece of code.

I’d love it if someone had something like that hacked already. [1.2
with WM_INPUT]

Yes, but it’s usually quite slow over the weekend - you’ll probably get a
response MondayOn Sun, Apr 11, 2010 at 12:09 PM, Michael Menegakis wrote:

what’s the right maling list to ask about this?

On Sat, Apr 10, 2010 at 1:49 AM, Michael Menegakis wrote:

Have I understood correctly that 1.2.x does not have high definition
mouse support on Windows?

(according to this MS article, WM_INPUT must be used
http://msdn.microsoft.com/en-us/library/ee418864(v=VS.85).aspx

I notice WM_MOUSEMOVE is used instead and I guess DirectInput too.
According to MS[on the above document too] they are both 'laggy’
compared to WM_INPUT)

If that’s true it took me by a negative surprise since I assumed
ioquake3 had that covered all these years.

I notice 1.3 has code including WM_INPUT.

So my questions are

  1. Are the above true? (is really high-definition not supported on 1.2?)

  2. Is it possible to use WM_INPUT on 1.2? I don’t control ioquake3 but
    I’d like to use it there, it also sounds a bit hard to include it. SDL
    isn’t the most simple piece of code.

I’d love it if someone had something like that hacked already. [1.2
with WM_INPUT]


SDL mailing list
SDL at lists.libsdl.org
http://lists.libsdl.org/listinfo.cgi/sdl-libsdl.org

Michael Menegakis wrote:

what’s the right maling list to ask about this?

You’re in the right place.

We probably won’t add it to 1.2, as we’re moving forward with 1.3. But
if a patch shows up, I’m sure people would appreciate it.

–ryan.

By the way; I noticed 1.3 uses getcursorpos() at a point in WM_INPUT.
I won’t pretend I’m an expert but wouldn’t that have to be avoided and
directly use a RAWINPUT* var?

per the microsoft article:
http://msdn.microsoft.com/en-us/library/ee418864(VS.85).aspx

Just suspecting getcursorpos() negates the ‘high def’ aspect, but not
sure if it does.On Mon, Apr 12, 2010 at 8:52 AM, Ryan C. Gordon wrote:

Michael Menegakis wrote:

what’s the right maling list to ask about this?

You’re in the right place.

We probably won’t add it to 1.2, as we’re moving forward with 1.3. But if a
patch shows up, I’m sure people would appreciate it.

–ryan.

By the way; I noticed 1.3 uses getcursorpos() at a point in WM_INPUT.
I won’t pretend I’m an expert but wouldn’t that have to be avoided and
directly use a RAWINPUT* var?

A lot of this code is being rewritten soon.

–ryan.