On 5/17/06, Roderick Colenbrander <[EMAIL PROTECTED]> wrote:
Does the attached version work? I tried a few test apps and it works fine here
but I don't have games that directly call DirectInput8Create.
Yes, that one works.
> On 5/15/06, Roderick Colenbrander <[EMAIL PROTECTED]> wrote:
> >
> > I think the issue is that CoInitialize isn't called from
> DirectInput8Create
> > main. I think I missed this because my test app did this for me. Try to
> add
> > a CoInitialize(NULL); and see if it helps. (and a CoUninitialize
On 5/15/06, Roderick Colenbrander <[EMAIL PROTECTED]> wrote:
I think the issue is that CoInitialize isn't called from DirectInput8Create
main. I think I missed this because my test app did this for me. Try to add
a CoInitialize(NULL); and see if it helps. (and a CoUninitialize)
Ok, I did tha
> On 5/15/06, Roderick Colenbrander <[EMAIL PROTECTED]> wrote:
>
> >
> > Yes, that's the reason. I forgot that I still need to send the patch for
> the
> > registry settings. Basicly what you need to add is this:
>
>
> What about bf798030-483a-4da2-aa99-5d64ed369700?
>
> trace:ole:DllMain 0x7f3
> On 5/15/06, Roderick Colenbrander <[EMAIL PROTECTED]> wrote:
> >
> > Are you sure it is this patch that causes the regression and not
> Vitaly's
> > latest one? This patch changes the way dinput8 objects are created and
> it
> > does this similar to windows. Could you figure out what part breaks
On 5/14/06, Roderick Colenbrander <[EMAIL PROTECTED]> wrote:
Hi,
This patch lets DirectInput8Create work similar to the native version which
according to MSDN uses CoCreateInstance to create a dinput8 object. This
rewrite should fix the dll count related issues.
This patch needs linking again
