Graphics Tablet Firmware

[Zekromtor]

I'm interested in potentially taking apart a wacom tablet and attaching it to a teensy or whatever programmable controller would be necessary and powerful enough to function properly to have complete control over the firmware. Whereas there was plenty of information floating around online about how to do something like this for my keyboard, I can find virtually none on graphics tablets. Anyone have any info/tips on where to start on such an endeavor?

The end goal here is a tablet that functions more like a mouse. Ultra simple, no tilt, pressure, etc, but fast, ie 1000hz polling and using relative motion.

[jacobolus]

You’ll want to get a logic analyzer, so that you can reverse engineer the communication between the device and the computer.

Good luck!

[dorkvader]

wacom is complex as it uses passive (no battery) pens and can sense pressure, etc. It's a lot easier if you don't need all that to just get a cheap resistive touchscreen. Not sure about relative motion, however.

What do you mean by "relative motion" ?

[jacobolus]

What do you mean by "relative motion" ?

He means it would work like a mouse where movements on the pad make movements relative to the current cursor position, instead of like a graphics tablet’s usual mode which is to associate positions on the pad with absolute positions on the screen.

FWIW, Wacom tablets already support a “relative motion” mode.

[dorkvader]

What do you mean by "relative motion" ?

He means it would work like a mouse where movements on the pad make movements relative to the current cursor position, instead of like a graphics tablet’s usual mode which is to associate positions on the pad with absolute positions on the screen.

FWIW, Wacom tablets already support a “relative motion” mode.

Oh Ok, relative like a trackpad, vs absolute. I see.

So first of all it does depend on the wacom table tin question. I'm only familiar with the digitizers they use in tablet PCs (both pen & touch) which they keep separate, as opposed to other manufacturers like n trig who have the same digitizer for both. Unfortunately the "normal" wacom digitizer is not transparent: it is mounted behind the screen. Not sure if this is a good or bad thing for your usage. I don't know of the normal wacom computer drivers (I think they are PS/2 devices, but they newer ones might be USB) support relative motion for the pen digitizer part.

These are actually relatively cheap. You can get replacement parts from a computer repair shop. In the case of Fujitsu, it will come with a controller mounted on it, but it does depend on what model fujitsu it's for. Lenovo glues their wacom digitizer to the screen so it's super expensive. Dell uses n trig for the XT2 and XT3, and Hp changes manufacturers all the time.

Your best bet is likely to just get a used wacom tablet online. I know there is osme DIY support for the older serial models, and USB ones shouldn't be *that* expensive. I don't know anything about their desktop models though: only about the embedded device in a tablet PC.

[Zekromtor]

Got a wacom coming that I'll at least be able to test relative motion mode with. It may be completely adequate as is, but if there's any noticeable latency it's going to drive me nuts. Says it's 200hz (pps) but that of course doesn't tell the whole story. The fact that more people aren't already using tablets as mouse replacements leads me to believe there's something more than a lack of trying relative mode to it.

[jacobolus]

A few reasons:

(1) it’s tricky to use a pen without direct visual feedback (i.e. at the pen tip); personally I find a Wacom tablet (in both absolute and relative modes) very difficult to use to navigate around the standard desktop UI. One attached to a screen so you can directly tap stuff is probably a lot easier for novices but I haven’t really used one of those much.

(2) Wacom tablets are expensive and marketed exclusively to artists and illustrators

(3) mice are well known technology everyone is familiar with, they are cheap and robust and widely available

(4) it’s relatively easy to switch back and forth between mouse and keyboard (or touchpad or trackball and keyboard) whereas with a tablet you need to pick up the pen which is slightly slower

(5) software is designed with the capabilities of the mouse in mind, except for a few applications aimed at artists.

[tufty]

I know "a bit" about wacom tablets.

You don't want to **** with the firmware on the tablet itself.  Frankly, you won't be able to do anything with it; it's tied to the hardware, and the hardware is tied to a number of very specific RC tank circuits.  You can forget trying to get any faster sampling than Wacom can, because the hardware simply can't do it.

If you want to use a wacom in "simple" mode with no pressure or tilt and relative positioning, it's easy to do.  Simply turn off pressure, tilt, and absolute positioning in Wacom's driver.  If you've got a "professional" wacom (i.e. one of the intuos line), you should be able to do this on a per-tool basis.  Here's a hint, though - pens are awful to use in relative mode.

The reason wacoms haven't taken off as a general input device is nothing to do with them being "hard to use" (they aren't, they are exceptionally natural to use for most purposes), and everything to do with them being hideously expensive.

If you want cheap, and don't care about tilt, I can line you up with someone who has a bunch of bare "penabled" sensor boards.  All you'd need is level converters (they talk USB at 3v3) and a suitable pen (penabled or ultrapad).

But really, forget messing with the device firmware.

[jacobolus]

The reason wacoms haven't taken off as a general input device is nothing to do with them being "hard to use" (they aren't, they are exceptionally natural to use for most purposes), and everything to do with them being hideously expensive.

How much drawing experience do you have?

My friends who are illustrators and sketchers all love Wacom tablets, and found them completely natural right from the start.

When my friends who don’t have extensive drawing experience (including myself) picked up a Wacom tablet, we all found it very difficult and confusing, and quickly ran back to the mouse.

If you take someone without drawing skill, and without any prior mousing experience, and hand them a Wacom tablet and a mouse, I suspect that the majority would greatly prefer the mouse. Since most everyone furthermore has mousing experience these days, the Wacom is a non-starter for most people in comparison.

YMMV.

(On the other hand, I had no problem when I tried the Cintiq, since the tablet is also the screen, so you can directly look at what you’re pointing it with the pen.)

[dorkvader]

(On the other hand, I had no problem when I tried the Cintiq, since the tablet is also the screen, so you can directly look at what you’re pointing it with the pen.)

No problems here with the wacom convertible tablets for the same reason. Sadly almost all of what I do is easier to have it in 'Laptop" mode which makes the use of the wacom part of the thing much harder. And since everything I do on a computer can also be done easily with a mouse / trackball, I rarely used the tablet functionality.

[Coreda]

Just put it in 'Mouse' mode and voila, acts just like a mouse. In that mode it's really not much different to a regular mouse, apart from the obvious benefits in image editing applications (and comfort for some). Could never get used to Pen mode, the mapping is just too unpredictable for small, precise actions.

The fact that more people aren't already using tablets as mouse replacements leads me to believe there's something more than a lack of trying relative mode to it.

Yea, they're relatively expensive  And most don't need the sensitivity, a mouse is perfectly adequate.

[tufty]

The reason wacoms haven't taken off as a general input device is nothing to do with them being "hard to use" (they aren't, they are exceptionally natural to use for most purposes), and everything to do with them being hideously expensive.

How much drawing experience do you have?

I'm not a graphic designer or anything like that - hell, I was rubbish at drawing even at school.  I'm a coder by trade, and the reason I got into graphics tablets was because mice make my hands and wrists hurt like hell.  Friend of mine (who *is* a graphic designer) let me borrow his tablet (summagraphics A3, IIRC) for a week or so to digitise some of my grandfather's technical drawings, and I fell in love with it as a general pointing device.  That was around 1990, and it took me a month or so to scrape together enough money to get my first Wacom.

I suspect that a lot of people's "bad experiences" with tablets come down to:

- "I'm not used to it, so I don't like it".  They *are* different to mice, some adaptation is necessary
- Postage-stamp sized tablets (anything less than ~A5 is pointless).
- Cheap non-name-brand tablets with batteries in the pens, horrible jitter and awful sampling speed 

Tablet's aren't the be all and end all of pointing devices.  Particularly for gaming, they're awful, but that's what USB game controllers are for, amirite?  Personally, I have no cognitive issue with "move hand over here, pointer moves over there"; indeed, to my eyes a tablet is more "direct" than a mouse in that respect, as pointing to *this* spot with your hand makes the cursor move to *that* spot on the screen. YMMV of course.

What I can tell you is this:

When I use my tablet, my hands don't hurt.  My trackball, if used intensively, hurts the index, middle and ring fingers of my right hand.  Trackpads are worse, I don't get on with trackpoint, and mice make my entire arm scream for mercy.

[Zekromtor]

Thanks for the info on the firmware Tufty.

I ended up just doing some windows API level stuff to hook input functions and modify the functionality of my tablet to suit my needs. I use the tablet more like a mouse, but I use my keyboard as the mouse buttons. When the tablet point is making contact with the surface my S D and F keys function and normal mouse keys, and if I hold my A button and move the pen it simulates a vertical and horizontal mouse wheel.

What I've found from this Intuos 5 is: It's polling at 133 hertz in spite of the box saying 200. If there's a way to switch it to 200, I don't know of it, and I fear doing so even if there is. They probably have it defaulting to 133 for a reason.

There is no perceptible lag between my motion of the pen and the cursor motion on the screen. Obviously there is 1000/133 ms at the very least, but that doesn't end up being enough for me to notice. I have all of the touchscreen features turned off (on the tablet and in the windows services). My understanding is this is important in order to avoid extra overhead which can cause input lag.

One thing that bugs the **** out of me is that in relative motion mode the cursor will move in 2 pixel increments. This is a quirk in the firmware, not a hardware limitation, as in absolute mode there is no such problem. You can sort of get around this by ticking your windows mouse sensitivity down from the standard 1:1 ratio 6/11 to whichever the 2:1 is, i think 4/11, but then you're moving pretty damn slow, even with the wacom speed maxed out.

Using the pen is both more intuitive and more ergonomic than a mouse. I'd highly recommend it to anyone who has pain in their mouse hand. The biggest advantage a mouse has over the pen, I've found, is the ability to use the attached buttons with virtually 0 effect on the position of the mouse. With the pen you are forced to use a finger that is also used to move the pen, and you always end up moving the cursor a slight amount unless you're firmly planted or using a high friction nib. Since I prefer low friction nibs (I actually put a piece of teflon over my tablet to have even less friction) the buttons on the pen itself are rarely if ever used.

For gaming I believe the tablet has the potential to supersede the mouse, but there are huge caveats. Mouselook games are not going to work well, even in relative motion mode which solves much of your problems, the larger issue is the loss of the mousebuttons. With your dominant fingers of your keyboard hand already occupied for your four motion keys, it makes this type of game unsuited as most would suspect. But for other games such as RTS, LoL style games (hate calling them MOBA), virtually any game where you're actually using a cursor, OSU! especially, the tablet has the potential to outperform mice. In absolute mode you have 0 worries about acceleration or mouse rotation and snap motion becomes far easier and faster than even a light mouse can offer.

TLDR: I really like my Intuos 5, and use it instead of a mouse now for everything except first person mouselooking games. If I get crazy enough to try to use footpedals for mousebuttons, however, I may even be able to do those as well.

[tufty]

What I've found from this Intuos 5 is: It's polling at 133 hertz in spite of the box saying 200. If there's a way to switch it to 200, I don't know of it, and I fear doing so even if there is. They probably have it defaulting to 133 for a reason.

I would doubt very much that there's any issue going to 200ps.  My Intuos *1* gives a solid 200 pps, and that's talking down ADB (to a teensy, which then converts to USB).  My guess would be that the 133 samples/sec thing is a windows issue, ad not anything Wacom side; certainly at a low level, when Wacom say 200 samples/sec, they *mean* 200 samples/sec, and nothing less.  You wouldn't believe the hoops they jumped through to get 200 samples/sec down ADB.

One thing that bugs the **** out of me is that in relative motion mode the cursor will move in 2 pixel increments. This is a quirk in the firmware, not a hardware limitation, as in absolute mode there is no such problem. You can sort of get around this by ticking your windows mouse sensitivity down from the standard 1:1 ratio 6/11 to whichever the 2:1 is, i think 4/11, but then you're moving pretty damn slow, even with the wacom speed maxed out.

That's not firmware, that's either a driver issue or, again, Windows.  The tablet doesn't deal in pixels at all, it deals in 100ths of a millimeter increments, and it doesn't know or care whether you're treating the packets it sends as absolute co-ordinates, relative movements, or some combination of the above..  BTW, WRT 100th of a millimeter - yes, they *really* are that accurate. 

FWIW, if I shove my tablet into relative mode on the Mac, I don't see the 2 pixel issue at all.  My money's on it being a Windows issue. Annoying, yes, but almost certainly not Wacom's fault.

[Zekromtor]

What I've found from this Intuos 5 is: It's polling at 133 hertz in spite of the box saying 200. If there's a way to switch it to 200, I don't know of it, and I fear doing so even if there is. They probably have it defaulting to 133 for a reason.

I would doubt very much that there's any issue going to 200ps.  My Intuos *1* gives a solid 200 pps, and that's talking down ADB (to a teensy, which then converts to USB).  My guess would be that the 133 samples/sec thing is a windows issue, ad not anything Wacom side; certainly at a low level, when Wacom say 200 samples/sec, they *mean* 200 samples/sec, and nothing less.  You wouldn't believe the hoops they jumped through to get 200 samples/sec down ADB.

One thing that bugs the **** out of me is that in relative motion mode the cursor will move in 2 pixel increments. This is a quirk in the firmware, not a hardware limitation, as in absolute mode there is no such problem. You can sort of get around this by ticking your windows mouse sensitivity down from the standard 1:1 ratio 6/11 to whichever the 2:1 is, i think 4/11, but then you're moving pretty damn slow, even with the wacom speed maxed out.

That's not firmware, that's either a driver issue or, again, Windows.  The tablet doesn't deal in pixels at all, it deals in 100ths of a millimeter increments, and it doesn't know or care whether you're treating the packets it sends as absolute co-ordinates, relative movements, or some combination of the above..  BTW, WRT 100th of a millimeter - yes, they *really* are that accurate. 

FWIW, if I shove my tablet into relative mode on the Mac, I don't see the 2 pixel issue at all.  My money's on it being a Windows issue. Annoying, yes, but almost certainly not Wacom's fault.

You're right the 2 pixel issue could be in the driver and not the firmware. It could also be Windows.

I'm going to write to them and see if they have a solution for the polling speed or the pixel precision.

[tufty]

I'm going to write to them and see if they have a solution for the polling speed or the pixel precision.

From a quick bit of googling, it would seem that the pixel precision thing is a windows issue and although you might be able to alleviate it a bit by messing with settings, you'll probably ever make it fully go away.

Sampling speed might be helped by putting the driver into "recognition mode", if that makes any sense to you.

[Zekromtor]

Recognition mode did not mean anything to me since I had never seen it in any of the options but after a bunch of searching I found a comment where someone mentioned you have to double click the intuos icon in the options window to bring up another window that allows you to toggle modes. I would have never found it otherwise.

I am now polling at 200hz! Thank you.