1. Synaptics touchpad (other touchpads are not supported, sorry!)
2. Synaptics COM libraries (if you installed the touchpad drivers, you likely have these)
3. Visual C++ 2015 runtime (x86 version) (https://www.microsoft.com/en-us/download/details.aspx?id=48145)

1. The emulation is done completely in user-mode, meaning that it will not work in raw-input mode, so disable that, obviously ;-)
2. If you are not playing in fullscreen on your native resolution, you will have to specify some options (see below)
3. You cannot actually click with your touchpad while the program is enabled, so you will have to use the keyboard.

1. -t <x1,y1,x2,y2>: sets the mapped touchpad region. (x1,y1) is the top-left corner, (x2,y2) is the bottom-right corner.
2. -s <x1,y1,x2,y2>: sets the mapped screen region. See above.
3. -m: enables the touchpad on start, disables it on exit (for people like me who normally have their touchpad disabled).
4. -w <weight>: sets the touch smoothing factor (0~1). Lower = less input lag, higher = less cursor vibration.
5. -d: enables debug mode (useful for determining your touchpad coordinates).

BetaStar wrote:

Cool project you got there!

Too bad my laptop uses ELAN touchpad instead of Synaptics version touchpads
due to this problem my fingerprint is unharmed

Hopefully you'll release an update for more support for additional touchpads!

abraker wrote:

>.> which version?

SPOILER

abraker wrote:

Currently it clips to the top left of the touchpad, is it possible to make it so the offset can be changed with the -x and -y arguments? For example, 0.5 would set it to the middle of the touchpad

Also the cursor shakes a lot if you keep it in one spot. I am guess rounding errors? If not, consider implementing cursor stabilization.

crossbowffs wrote:

That's weird, the code should be detecting the dimensions of your touchpad automatically. I'm not too sure on what you mean by clipping to the top left, did you try using the -width and -height parameters? (If you don't explicitly provide them, they are calculated from your screen resolution when you start AbsoluteTouch)

crossbowffs wrote:

As for the shaking, I believe it's due to the high resolution of the touchpad (for example, mine has a resolution of about 6000x3000 "units"), though it shouldn't be more than a few pixels, unless you have some kind of ultra-HD screen. The problem with "stabilization" algorithms is that they would cause input lag, which IMO is not worth the trade-off if you're using it in osu!

abraker wrote:

I mean I don't want to reach all the way to the top of my touchpad. Instead want to play on the bottom right corner. Allowing to move the mapped area anywhere on the touchpad.

abraker wrote:

Is it possible to allow it use a lower resolution by skipping every 2 units or so (not making it smaller, just skip units it reads)?

crossbowffs wrote:

I'll have to find out what's causing the shaking first. If your display resolution is really high, an alternative solution would be to lower your screen resolution.

AbsoluteTouch-1.1.exe -t 2480,1855,4576,3583

AbsoluteTouch-1.1.exe -d

Igoorx wrote:

I never thought this was possible, it is very cool :v
Thanks for showing me this and by rewritten it in C++

Tatsuya Kaname wrote:

THIS IS AWESOME! Now I'm able to play osu! in higher difficulties! Thanks a lot!

const double weight = 0.4;
coordsAvg.x = weight * (double)coords.x + (1.0 - weight) * (double)coordsAvg.x;
coordsAvg.y = weight * (double)coords.y + (1.0 - weight) * (double)coordsAvg.y;

abraker wrote:

Mapping a smaller area of the touchpad to my screen really makes it impossible to be precise. I suggest to implement an option to set how much averaging the program would do. I did some research and an exponential averaging technique looks best for this. I played around with the code and a weight between 0.3 and 0.4 is a decent range such that the latency and cursor instability are at equal trade-off.

const double weight = 0.4;
coordsAvg.x = weight * (double)coords.x + (1.0 - weight) * (double)coordsAvg.x;
coordsAvg.y = weight * (double)coords.y + (1.0 - weight) * (double)coordsAvg.y;

abraker wrote:

For some reason I can't get it to work in osu! anymore. Made sure raw input is off and tried stable to make sure it's not of an update or something.