On Mon, Oct 18, 2010 at 8:24 PM,  <ch...@cnpbagwell.com> wrote:
> From: Chris Bagwell <ch...@cnpbagwell.com>
>
> I would apreciate if people could test these patches in real
> drawing applications and see if they notice any negative affects.
> Also, I'd appreciate any review comments over them.
>
> These are currently just to test some basic ideas to improve
> filtering logic.
>
> Patch #3 does not store raw samples if nothing changed since
> last store.  This is to help stop cursor movements occuring
> when user isn't moving and only changing pressure.  This could
> occur when drawing fast stoke and then lifting pen.
>
> Patch #4 shrinks sliding window average from RawSamples to only 1
> when time since last sample extends.  This seems to resolve an offset
> issue caused during fast strokes on Bamboo's that is visible when
> you go out-of-prox at edge of tablet.

Chris,

Patches 3 and 4 make sense to me. I have a question for patch 1
though. When you say:

"2) wcmCheckSuppress() is called at all times.  There used to be 1
very small window were it was skipped on initial startup but I can't
see any harm in calling it always."

do you mean wcmCheckSuppress is called even for the first in-prox
event? I don't think we want to do that since the pLast could, and
most likely, be an invalid data at that time.

Testers (anyone in the list who is willing to test the patchset),

I don't have time to test the patchset. I hope someone can test the
patchset on a Cintiq or Intuos4, especially for the corner cases: such
as in and out prox; moving super fast or slow; moving around the
corners and edges of the tablet, etc. When testing the patchset,
please share your detailed observation with us.

If you'd like to get the driver as a package instead of apply the
patchset yourself, please email me or Chris directly. We'll send the
package to you.

Thank you in advance for your support.

Ping

> I've not fully tested.  I could be fooling myself and math error
> reduces to window size of 1 always.
>
>
> Chris Bagwell (4):
>  push storing of raw samples into wcmFilter file
>  Remove unused filter variables.
>  Do not buffer raw samples when nothing changes
>  Use dynamic window size for filter averaging
>
>  src/wcmCommon.c     |   80 ++++++----------------------
>  src/wcmFilter.c     |  143 
> +++++++++++++++++++++++++++++++++++++++++++++++----
>  src/wcmFilter.h     |    1 +
>  src/xf86WacomDefs.h |    3 +-
>  4 files changed, 151 insertions(+), 76 deletions(-)
>
> --
> 1.7.3.1
>
>
> ------------------------------------------------------------------------------
> Download new Adobe(R) Flash(R) Builder(TM) 4
> The new Adobe(R) Flex(R) 4 and Flash(R) Builder(TM) 4 (formerly
> Flex(R) Builder(TM)) enable the development of rich applications that run
> across multiple browsers and platforms. Download your free trials today!
> http://p.sf.net/sfu/adobe-dev2dev
> _______________________________________________
> Linuxwacom-devel mailing list
> Linuxwacom-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/linuxwacom-devel
>

------------------------------------------------------------------------------
Nokia and AT&T present the 2010 Calling All Innovators-North America contest
Create new apps & games for the Nokia N8 for consumers in  U.S. and Canada
$10 million total in prizes - $4M cash, 500 devices, nearly $6M in marketing
Develop with Nokia Qt SDK, Web Runtime, or Java and Publish to Ovi Store 
http://p.sf.net/sfu/nokia-dev2dev
_______________________________________________
Linuxwacom-devel mailing list
Linuxwacom-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linuxwacom-devel

Reply via email to