How to set up a Wacom Tablet

How to do things, solutions, recipes, tutorials
Message
Author
User avatar
mikeb
Posts: 11297
Joined: Thu 23 Nov 2006, 13:56

#61 Post by mikeb »

The eraser works that way as well, however, the eraser tool isn't automatically selected in Gimp when you use the pen eraser. In other words, it doesn't select the background color, but continues in the selected color, with the paintbrush cursor. Maybe that's how Gimp works? Or maybe it's a matter of setting the extended input device?
You set the function for each end and then it gets remembered..ie use the eraser then choose eraser tool then it should stay that way.....handy as the mouse can be set to a different tool to the pen.

http://www.murga-linux.com/puppy/viewto ... 55&t=36179
xournal ^ works nicely with tablets too :)
Thank you again Pizzasgood for all your work with this. You've made a lot of people happy in this thread!
Here here...tis true :)

regards

mike

User avatar
Pizzasgood
Posts: 6183
Joined: Wed 04 May 2005, 20:28
Location: Knoxville, TN, USA

#62 Post by Pizzasgood »

I added a package to the first page that includes just the xorg driver, without the kernel module, so that people who would prefer not to (or shouldn't) change the kernel module don't need to.
[size=75]Between depriving a man of one hour from his life and depriving him of his life there exists only a difference of degree. --Muad'Dib[/size]
[img]http://www.browserloadofcoolness.com/sig.png[/img]

User avatar
vtpup
Posts: 1420
Joined: Thu 16 Oct 2008, 01:42
Location: Republic of Vermont
Contact:

#63 Post by vtpup »

Are you sure about that, pizzasgood?

Maybe you tested it against a 412retro, and in that case apologies for this foolish note, but if not, and you went on what I did....I am not proof of it, since I DID add and load your wacom driver module, as well as the recompiled GTK, in my own system when it finally started working.

That's not to say that it wouldn't have worked without adding the driver module -- and have been able to use the pre-existing Wacom module that ships with 4.1.2. I just never tried it that way.

At present my system always has the wacom driver working, but does not show a wacom driver anywhere in the Puppy Bootmanager. It is not addable there, or removable there. It does not exist in any of the Bootmanager's lists.

It did formerly.

I'm not complaining -- it works fine, and I can't think of a reason I'd ever want to remove it.

@MikeB thanks for the advice -- it works correctly as an eraser in Gimp, now that I know how to set it up. Thanks also for that cool program xournal! Very nice.

I haven't been able to figure out how to get the eraser end to work distinctly from the point end in that program, but this is a pretty trivial nitpicking observation. I mention it only in case you happen to know of the way to do it there as well. It's not very important.

Also, if you could tell me how to improve my handwriting......

User avatar
mikeb
Posts: 11297
Joined: Thu 23 Nov 2006, 13:56

#64 Post by mikeb »

Options in xournal........handwriting...sorry no hope I'm afraid...

mike

User avatar
vtpup
Posts: 1420
Joined: Thu 16 Oct 2008, 01:42
Location: Republic of Vermont
Contact:

#65 Post by vtpup »

Went to Options when I first tried the program, but "Eraser Tip" makes no difference one way or another. The eraser end of the stylus works whether checked or unchecked, and does exactly the same thing as the pointy end.

No big deal, really. I just thought maybe there was another setting or technique (as in Gimp) to get it to work separately.

The problem may just be the fact that I have an older pad....

re. handwriting -- I was joking of course. But thinking about it, there is a serious component:

1.) The scaling of movement on the pad is not adjustable to the screen movement. If I write my signature on the pad I must restrict the size of it to a tiny signature. This is blown up on the screen, and so inaccuracies of movement are exaggerated. Zooming the page onscreen does not change the relative scaling of pen movement to screen, it only changes the relative size of the page onscreen. Therefore to get a closer approximation you would have to zoom IN. This is like trying to write with a magnifying glass which only allows you to see a small window of the paper, and blows up your movements.

and much more important:

2.) In general in Linux implementation for the drivers:

As I recall in the Windows drivers, the pad used absolute coordinates for positioning and this related directly to the screen. So if you placed the stylus from far away to any position on the pad, it would appear in the same relative position on the screen rectangle.

In Linux drivers, if I place the stylus on the pad from far away, the results are mixed. It may show up in a few different places. This seems to happen most when coming in from the lower right edge to middle right edge.

The subsequent moves work in good relative proportion once the cursor is established onscreen, but I find you can run off of the pad before reaching the border of a window or screen while drawing.

This means I have to reposition. That is difficult because I can't bring the cursor in necessarily where you want it. I find myself waving my hand around trying to find a place to come in that will allow me to reach a spot. Usually the upper left is a problem area to reach.

The positioning of the Windows drivers is a lot better than the somewhat arbitrary relative positioning I seem to get with the linux drivers. In Windows, the cursor was always in the same relative position on the screen as on the pad.

I wonder if anyone else has noticed this?

It reminds me a little of what happens when you get xorg or xvesa wrong in setup and find that the screen contents are visible but partially wrapped around to the other side.

In the case of the pad, perhaps this is going on with the bottom of the pad area.

This problem may also be specific to monitor resolution (since a translation has to be calculated by the pad driver to locate the cursor) or my older pad may be incorrectly supported by the linux drivers. So maybe this doesn't happen for everyone. My monitor is set to 1024 x 768 pixels x 24bit xorg

User avatar
vtpup
Posts: 1420
Joined: Thu 16 Oct 2008, 01:42
Location: Republic of Vermont
Contact:

#66 Post by vtpup »

More observations:

I think I understand the mechanism of pad to cursor location a little better now. I'm intuiting the following from behavior-- I don't know what the code looks like

In the linux driver the pad is seen as a much larger coordinate plane (in pixels) than my screen (1024 x 768). Instead of calculating the proportional position of the cursor onscreen, a simple one-for-one pixel map is generated. This pixel map is moveable across the pad according to certain rules.

If a cursor is moved to one edge of the screen and stylus movement continues on the pad in the same direction, the pixel map is moved in the direction of the stylus. Backing the stylus up immediately moves the cursor back into the pixel map and the cursor moves with it.

Continuing in that direction, the cursor will eventually reach the other screen edge, well before the stylus has reached the edge of the pad. There will appear to be a "dead" area" of stylus movement. If movement continues, the pixel map will shift in the direction of stylus movement.

Removing the stylus from the pad will result in a re-location for the pixel map to an indeterminate position -- it may be affected by a default location, the last cursor exit position, or other pointing devices also mapped to the system. I don't know which, or if all are involved.

I'm going to say this is definitely a function of the drivers -- the basic algorithm seems to have avoided using absolute coordinate calculation for positioning the screen cursor, and relies on a much simpler moveable pixel map.

Some users with screen resolutions that nearly match the pad resolution (or possibly even multiples) will probably have near absolute positioning, while others with less fortunate proportions might find more of a problem.

The problem will be less of a hangup for users who write or sketch, or use it to do mouse pointing. It will be a definite problem for CAD and technical illustrators (if they are used to better stylus action in Windows).

The Wacom Intuit3 large pads have an available digitizing mouse with a crosshair, and I'm willing to bet this won't work properly, unless these pads have better Linux drivers.

User avatar
vtpup
Posts: 1420
Joined: Thu 16 Oct 2008, 01:42
Location: Republic of Vermont
Contact:

#67 Post by vtpup »

Correction;

Well, after all of that investigation and explanation, I found that my system had reverted to the old mode. Apparently if I don't have the pad plugged in at boot, the old driver dominates.

The new drivers do seem to have absolute positioning, so my above two posts all apply to the old wacom drivers. Even the comments about the eraser tip in Xournal, and the relative handwriting size! :oops: sorry!

The pad can be started properly only if plugged in at boot, or by restarting the X-server with it plugged in. It doesn't work to enter "modprobe wacom" in the terminal with the pad connected.

It would be nice to be able to avoid program or OS resets to get the pad going. Or keep the driver loaded even after unplugging, so that re-plugging would work.

User avatar
mikeb
Posts: 11297
Joined: Thu 23 Nov 2006, 13:56

#68 Post by mikeb »

Yes.........

pad cannot be hotplugged with X and without the X driver performance is not really suitable. ctrl-alt-f3 then ctrl-alt-f1 is a quicker way to reset X.
We leave ours plugged in all the time and i think the X driver loads the kernel driver automatically.

mike

User avatar
Pizzasgood
Posts: 6183
Joined: Wed 04 May 2005, 20:28
Location: Knoxville, TN, USA

#69 Post by Pizzasgood »

Are you sure about that, pizzasgood?

Maybe you tested it against a 412retro, and in that case apologies for this foolish note, but if not, and you went on what I did....I am not proof of it, since I DID add and load your wacom driver module, as well as the recompiled GTK, in my own system when it finally started working.
Not sure what was going on on your end. On mine, the packaged kernel module didn't work in 4.12retro, but the built-in kernel module did. So all I needed was the Xorg driver.
[size=75]Between depriving a man of one hour from his life and depriving him of his life there exists only a difference of degree. --Muad'Dib[/size]
[img]http://www.browserloadofcoolness.com/sig.png[/img]

User avatar
vtpup
Posts: 1420
Joined: Thu 16 Oct 2008, 01:42
Location: Republic of Vermont
Contact:

#70 Post by vtpup »

Not sure what was going on on your end. On mine, the packaged kernel module didn't work in 4.12retro, but the built-in kernel module did. So all I needed was the Xorg driver.
I'm not entirely sure what's going on at my end either. Not just in relation to this. I'm thinking it might be clean install time. This one was a "downgrade" about a month ago from a non-retro 4.1.2 to the retro version. Did it by pfix=purge.

Probably not the best idea. And I've done a lot of experimenting since. So maybe it's time for a fresh clean start. So I'm on a level playing field.

The pad is working fine. So well, in fact that I'm springing for the big time. Intuos3 12"x19" That cost will set me back more than any computer I've ever owned -- by a factor of two. But I'm going to start designing again -- this whole thing with Puppy Linux and your work, Pizzasgood with the pad, has re-energized me. I have a lot of things I want to draw now, and that will allow it at full scale, with drafting curves. I can also digitize off of old drawings I have with it. So thanks again for all you did.

RJARRRPCGP
Posts: 97
Joined: Tue 09 Dec 2008, 06:10
Location: USA (Springfield, Vermont)

#71 Post by RJARRRPCGP »

Anyone here with a Bamboo? I just got one in December.

User avatar
mikeb
Posts: 11297
Joined: Thu 23 Nov 2006, 13:56

#72 Post by mikeb »

Anyone here with a Bamboo? I just got one in December.
yes...works fine...use the latest drivers..check out my posts :)

have fun

mike

jayz
Posts: 9
Joined: Tue 03 Jun 2008, 15:28
Contact:

bamboo fun

#73 Post by jayz »

Here is how I configured Puppy Linux 4.12 to use my Wacom Bamboo Fun tablet:

plug in pad
install wacom_k2.6.25.16.pet (for kernel 2.6.25.16)
in terminal: modprobe wacom
add to xorg.conf:
(appropriate lines from linux wacom project input dev and srv layout)
NOTE: add these to appropiate sections of xorg.conf. Sticking them down at the bottom doesn't work.


do cat /proc/bus/input/devices
to identify event number
alter xorg entries above to correct event number if necessary
if necessary make new nodes:

I did need to make a new node - event number 5.

Now it acts like a mouse.

install gtk+-2.12.1-xinput-debug.pet

Now Gimp recognizes it as an input device.
Configure Gimp.
*
Now it's a whole beautiful groove thing!
*
Thanks to Pizzasgood and everyone else here.

*
It does appear that the pad must be plugged in before puppy boots. Also, I'm getting an error message about udevd at shutdown. I can't quite catch what it says. Is this related to the tablet? Is it ignorable?
Learn to laugh at your troubles and you'll never run out of things to laugh at. [Walt Kelly]

User avatar
Pizzasgood
Posts: 6183
Joined: Wed 04 May 2005, 20:28
Location: Knoxville, TN, USA

#74 Post by Pizzasgood »

I don't think I've seen any udevd errors, but I haven't been paying much attention to the shutdown process. Usually I just click "poweroff" and then climb into bed.

I do know that udev should be configurable to allow us to skip fiddling with the event devices and just use something like /dev/wacom. I have been meaning to look into that for over a month now, but I've been pretty swamped. I'll try to do it next time I have some free time.
[size=75]Between depriving a man of one hour from his life and depriving him of his life there exists only a difference of degree. --Muad'Dib[/size]
[img]http://www.browserloadofcoolness.com/sig.png[/img]

User avatar
mikeb
Posts: 11297
Joined: Thu 23 Nov 2006, 13:56

#75 Post by mikeb »

It does appear that the pad must be plugged in before puppy boots.
X does not do hotplugging....there is a workaround mentioned earlier in this thread.
The latest test version of X does but will be a while before puppy uses it.

Ignore udevd ...it's the only way :D

regards

mike

turbodig
Posts: 4
Joined: Fri 20 Feb 2009, 19:05

#76 Post by turbodig »

Resurrecting this thread a bit...

I've got a HP tc1100 tablet that I'm trying to get running under Puppy 4.12
(2.6.25.16 kernel)
There's a few places that have gotten this working under other distros:

http://math.bu.edu/people/kayeats/computers/tc1100.html

http://groundstate.ca/TC1100

They all appar to be using standard wacom drivers.

Anyway, modprobe seems to load the driver ok, but the links above seem to refer to a wacom_acpi module, which I don't have. Does anybody happen to have a prebuilt one for 2.6.25.16?
(I did install the proper .pets listed in the first post)

A 'cat /proc/bus/input/devices' lists no wacom or tablet devices of any sort.

My gut feel is that the wacom_acpi won't fix this, but I gotta start somewhere. Plan D is to recompile from the wacom project sources, but the puplet I'm running doesn't have sources or the dev stuff, so I'm kinda saving that as the last resort.

User avatar
Argolance
Posts: 3767
Joined: Sun 06 Jan 2008, 22:57
Location: PORT-BRILLET (Mayenne - France)
Contact:

How to (hopefully) set up a Wacom Tablet...

#77 Post by Argolance »

Bonjour,
Resurrecting this thread a bit..
... It must be so, necessary !

I just installed Puppy Linux 4.2 (full install) : wonderful but... As I did it with Puppy Linux 4.12 with success, I followed this topic to install my Wacom Graphire3 : unfortunately, it doesn't work anymore ! What's going wrong Doctor ?
Thank you for your kind attention
______________________________________________
cat /proc/bus/input/devices

.I: Bus=0003 Vendor=056a Product=0013 Version=0314
N: Name="Wacom Graphire3"
P: Phys=
S: Sysfs=/devices/pci0000:00/0000:00:10.1/usb6/6-2/6-2:1.0/input/input4
U: Uniq=
H: Handlers=mouse0 event4
B: EV=f
B: KEY=1c43 0 70000 0 0 0 0 0 0 0 0
B: REL=100
B: ABS=100 3000003
______________________________________________

User avatar
Pizzasgood
Posts: 6183
Joined: Wed 04 May 2005, 20:28
Location: Knoxville, TN, USA

#78 Post by Pizzasgood »

I haven't tried my tablet with 4.2 yet*. Maybe something got left out or modified. I'll try to test it this weekend. Maybe I'll also finally get around to taking a look at how to make udev set up a /dev/wacom to eliminate all those event number searches (very annoying - I use Gimp to do some of my VLSI homework, because it must be submitted electronically and I don't have a scanner, but my numbers keep changing on me, forcing a modification of xorg.conf and subsequent restart of X before I can get started. Grrrr).

*I haven't yet switched to 4.2 from 4.1.2 either. I won't be making the switch until the week after next, when I start prepping my laptop for the summer (normally use my desktop, but it is in storage over summer as I'm out of state for three months).
[size=75]Between depriving a man of one hour from his life and depriving him of his life there exists only a difference of degree. --Muad'Dib[/size]
[img]http://www.browserloadofcoolness.com/sig.png[/img]

User avatar
Argolance
Posts: 3767
Joined: Sun 06 Jan 2008, 22:57
Location: PORT-BRILLET (Mayenne - France)
Contact:

#79 Post by Argolance »

Maybe I'll also finally get around to taking a look at how to make udev set up a /dev/wacom to eliminate all those event number searches
... very annoying indeed : This should really be great !
I'll be waiting for your test : thanks a lot...
(What a pity to have to reconfigure devices every time we upgrade...)
Best regards.

My xorg.conf :
_____________________________________________
# nvidia-xconfig: X configuration file generated by nvidia-xconfig
# nvidia-xconfig: version 1.0 (buildmeister@builder63) Tue Nov 4 14:07:17 PST 2008

#Special base config file used in Puppy Linux.
# **********************************************************************
# Module section -- this section is used to specify
# which dynamically loadable modules to load.
# **********************************************************************
#
# **********************************************************************
# Files section. This allows default font and rgb paths to be set
# **********************************************************************
# **********************************************************************
# Server flags section.
# **********************************************************************
#everything past here is auto-generated by Puppy's Xorg Wizard...
#PuppyHardwareProfile=NVIDIAMB24W

Section "ServerLayout"
Identifier "X.org Configured"
Screen 0 "Screen0" 0 0
InputDevice "Mouse0" "CorePointer"
InputDevice "Keyboard0" "CoreKeyboard"
InputDevice "stylus" "SendCoreEvents"
InputDevice "eraser" "SendCoreEvents"
InputDevice "cursor" "SendCoreEvents" # For non-LCD tablets only
EndSection

Section "Files"

# The location of the RGB database. Note, this is the name of the
# file minus the extension (like ".txt" or ".db"). There is normally
# no need to change the default.
# Multiple FontPath entries are allowed (which are concatenated together),
# as well as specifying multiple comma-separated entries in one FontPath
# command (or a combination of both methods)
RgbPath "/usr/X11R7/lib/X11/rgb"
FontPath "/usr/X11R7/lib/X11/fonts/misc/"
FontPath "/usr/X11R7/lib/X11/fonts/Type1/"
FontPath "/usr/X11R7/lib/X11/fonts/TTF/"
EndSection

Section "Module"

# This loads the DBE extension module.
# This loads the miscellaneous extensions module, and disables
# initialisation of the XFree86-DGA extension within that module.
# This loads the font modules
Load "dbe" # Double buffer extension
SubSection "extmod"
Option "omit xfree86-dga" # don't initialise the DGA extension
EndSubSection
Load "type1"
Load "freetype"
# This loads xtrap extension, used by xrandr
Load "xtrap"
# This loads the GLX module (if present)
Load "glx"
# This loads dri module (if present)
Load "wacom"
# This loads Wacom module (if present) added by me
EndSection

Section "ServerFlags"

# Uncomment this to disable the <Crtl><Alt><Fn> VT switch sequence
# (where n is 1 through 12). This allows clients to receive these key
# events.
# Option "DontVTSwitch"
# Enables mode switching with xrandr
# There is a report that this can cause Xorg not to work on some
# video hardware, so default is commented-out...
# but i want to use it in xorgwizard so leave on...
Option "RandR" "on"
EndSection

Section "InputDevice"
Identifier "Keyboard0"
Driver "kbd"
Option "XkbRules" "xorg"
Option "XkbModel" "pc102"
Option "XkbLayout" "fr" #xkeymap0
EndSection

Section "InputDevice"

#Option "Emulate3Timeout" "50"
Identifier "Mouse0"
Driver "mouse"
Option "Protocol" "IMPS/2" #mouse0protocol
Option "Device" "/dev/mouse"
#Option "Emulate3Buttons"
Option "ZAxisMapping" "4 5" #scrollwheel
EndSection

Section "Modes"

#modes0modeline0
Identifier "Modes0"
EndSection

Section "Monitor"

#UseModes "Modes0" #monitor0usemodes
Identifier "Monitor0"
VendorName "Monitor Vendor"
ModelName "Monitor Model"
HorizSync 30.0 - 74.0
VertRefresh 50.0 - 61.0
Option "PreferredMode" "1920x1200"
EndSection

Section "Device"

### Available Driver options are:-
### Values: <i>: integer, <f>: float, <bool>: "True"/"False",
### <string>: "String", <freq>: "<f> Hz/kHz/MHz"
### [arg]: arg optional
#Option "ShadowFB" # [<bool>]
#Option "DefaultRefresh" # [<bool>]
#Option "ModeSetClearScreen" # [<bool>]
Driver "vesa" #card0driver
Identifier "Card0"
Driver "nvidia"
VendorName "nVidia Corporation"
BoardName "Unknown Board"
EndSection

Section "Screen"
Identifier "Screen0"
Device "Card0"
Monitor "Monitor0"
DefaultDepth 24
Option "RenderAccel" "True"
Option "AllowGLXWithComposite" "True"
Option "AddARGBGLXVisuals" "True"
Option "DisableGLXRootClipping" "True"
SubSection "Display"
Depth 24
Modes "1920x1200"
EndSubSection
EndSection

Section "Extensions"
Option "Composite" "Enable"
EndSection

Section "InputDevice"
Driver "wacom"
Identifier "stylus"
# Option "Device" "/dev/ttyS0" # SERIAL ONLY
Option "Device" "/dev/input/event4" # USB ONLY
Option "Type" "stylus"
Option "Mode" "absolute" # Position sur la tablette
Option "USB" "on" # USB ONLY
# Option "ForceDevice" "ISDV4" # Serial Tablet PC ONLY
EndSection

Section "InputDevice"
Driver "wacom"
Identifier "eraser"
# Option "Device" "/dev/ttyS0" # SERIAL ONLY
Option "Device" "/dev/input/event4" # USB ONLY
Option "Type" "eraser"
Option "Mode" "absolute" # Position sur la tablette
Option "USB" "on" # USB ONLY
# Option "ForceDevice" "ISDV4" # Serial Tablet PC ONLY
EndSection

Section "InputDevice"
Driver "wacom"
Identifier "cursor"
# Option "Device" "/dev/ttyS0" # SERIAL ONLY
Option "Device" "/dev/input/event4" # USB ONLY
Option "Type" "cursor"
Option "Mode" "absolute" # Position sur la tablette
Option "USB" "on" # USB ONLY
# Option "ForceDevice" "ISDV4" # Serial Tablet PC ONLY
EndSection
__________________________________________

User avatar
Pizzasgood
Posts: 6183
Joined: Wed 04 May 2005, 20:28
Location: Knoxville, TN, USA

#80 Post by Pizzasgood »

Well, turns out the whole udev thing was absurdly simple. I wish I'd looked into it ages ago. All I had to do was place a fie in /etc/udev/rules.d named '65-wacom.rules', containing this:

Code: Select all

# udev rule for wacom tablets.

KERNEL=="event[0-9]*", SYSFS{idVendor}=="056a", SYMLINK+="wacom"
Unplugged my tablet, plugged it in, and presto: there was a /dev/wacom symlinked to /dev/event8. Changed my xorg.conf file to point to /dev/wacom, restarted X, and sat back and enjoyed a happy little glow for a while.


Hmm... I do wonder what happens if you don't have enough /dev/event* numbers... I also tested this in a fresh 4.2 without adding new events, and it worked, but I didn't think to look and see what it did about the missing /dev/event8 file. I suppose it just created one on the fly or something. I'll have to check and see next time I boot it.


Anyway, I've compiled the latest version of the Wacom driver (for 2.6.25.16 kernels), and added the udev rule. I also made a package with just the xorg driver and the udev rule. I'll do other kernel versions on request.

wacom-0.8.2-2_xorg_driver.pet
wacom-0.8.2-2-k2.6.25.16.pet <-- includes the xorg_driver



I want to hold off on updating the first post until I've had some feedback.

As for 4.2, I had no issues. It worked fine. Not sure what went wrong on your end. Try it again with this one.
[size=75]Between depriving a man of one hour from his life and depriving him of his life there exists only a difference of degree. --Muad'Dib[/size]
[img]http://www.browserloadofcoolness.com/sig.png[/img]

Post Reply