Re: [DIGImend-users] Experiencing lag (with Huion 1060)

[Nikolai K. <sp...@gm...>]

On 03/03/2015 07:23 AM, Sarah Owens wrote:
> I sent this, and didn't notice it bounce because of my large attachment, so
> here it is without the tests (but let me know if you'd like to see them)

I've got them, you got the bounce from the maillist. Which is actually a good
thing, because you seem to have added some other files into the archive
accidentally.

> Ok, I think I understand what you were looking for now.  The SYN_REPORTs
> from evntest happened about 245 times in a second (every 4ms?), and the
> reports from usbhid were about the same (235 times / second).
> I attached the two files, just in case I misunderstood again, haha.  I
> didn't see a version number for the driver, but I got the most recent
> version just a few days ago-- plus the one-line change from Dan Roberts.
>
> In the evtest, I clicked the side-button of the pen 3 times ("BTN_0"
> registering 1 and 0 on push and release, respectively) and during the time
> that button was held down, the events looked normal-- just registering x and
> y position.
>
> I started scribbling at 1425314198.986, and every event while the pen was
> pressed down registered 2 different BTN_TOUCHes, value 1 and then 0 (on then
> off, I think).
>
> BTN_TOUCH being toggled every time would explain the sort of lines I was
> seeing.  I attached a png of 3 lines I drew-- the first line on the left is
> drawn with my laptop trackpad, the middle one is drawn with the tablet by
> clicking the button on the body of the pen, and the rightmost one was drawn
> with the pen tip pressing on the tablet.  Something I also found interesting
> was the trackpad line can be undone as a single action (as expected), but
> the other two will only undo  a single star at a time even though my stroke
> was continuous.

I think I managed to reproduce this locally and it seems to be a problem with
driver rebinding script (hid-rebind). I'll work on a fix.

Meanwhile you can rebind the drivers manually via sysfs, if you can figure out
how. Basically, all the interfaces should be handled by hid-uclogic now.

Thanks, Sarah!

Nick