On Thu, Apr 14, 2011 at 5:20 AM, Andrzej Giniewicz <[email protected]> wrote: > Some more tests done. First of all, I made tests for 2.6.38.2 and it > seems to work just as well as on 2.6.37.5 which I had before. > > Anyway, when doing more tests I noticed few new odd things in > gestures, things that as far as I remember worked better in 0.10.11 > (better as in better than now, but not necessarily 100% correctly) so > I believe this stuff can be labelled as regressions: > > First of all it's very very easy to generate multiple (usually 2 or 3) > RMB events when tapping second finger. It happens almost every time > for me. Is there some parameter to control it? LMB tapping works just > well with my setup. (I'm doing simple tests with: "xev | grep > Button"), maybe RMB events aren't respecting some parameters? Or > maybe, after successful tap and RMB generated, it should wait another > TapTime before accepting next event like that? > > Second thing is that when I align fingers correctly for scroll and > start moving them, for some time (approx half, maybe quarter of > second) the cursor moves and after that time, it starts to generate > normal scroll events like it should. > > There is also one thing that was reported for earlier versions too, > i.e. the jumps. Lets say I label fingers as A and B, If I put fingers > on the tablet in order A then B I have to remove them in exactly > reverse order, like B then A, or the cursor will jump by value that is > I believe difference between place where finger A started gesture to > on tablet where it is put now. Simple experiment: > > 1) Put finger A in top-right of tablet > 2) Put finger B right to it > 3) remove finger A from tablet > 4) remove finger B from tablet (actually, no gesture or events have to > be generated during this as far as I see) > 5) put finger in bottom-left corner, you should observe very big > absolute jump. The xev reports something like: > > MotionNotify event, serial 40, synthetic NO, window 0x1800001, > root 0x15d, subw 0x0, time 1918061, (1654,22), root:(1654,43), > state 0x0, is_hint 0, same_screen YES > > MotionNotify event, serial 40, synthetic NO, window 0x1800001, > root 0x15d, subw 0x0, time 1919525, (1004,386), root:(1004,407), > state 0x0, is_hint 0, same_screen YES > > MotionNotify event, serial 40, synthetic NO, window 0x1800001, > root 0x15d, subw 0x0, time 1919561, (1004,386), root:(1004,407), > state 0x0, is_hint 0, same_screen YES > > (those are 3 events, just before jump, the one generating jump and one > after the jump. There are no other events in between) > > I will play with it more, but those are all odd things I found for now > > Cheers, > Andrzej. >
I'm interested in knowing if you're seeing these behaviors with my RFC gesture patch in place or not (and if so, whether you've remapped the scroll direction). Its the only overt change to gesture code I can remember off the top of my head, but wasn't included in the 0.10.99.2 release. Jason --- Day xee-nee-svsh duu-'ushtlh-ts'it; nuu-wee-ya' duu-xan' 'vm-nvshtlh-ts'it. Huu-chan xuu naa~-gha. ------------------------------------------------------------------------------ Benefiting from Server Virtualization: Beyond Initial Workload Consolidation -- Increasing the use of server virtualization is a top priority.Virtualization can reduce costs, simplify management, and improve application availability and disaster protection. Learn more about boosting the value of server virtualization. http://p.sf.net/sfu/vmware-sfdev2dev _______________________________________________ Linuxwacom-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/linuxwacom-devel
