[Xpert]Proposal for mouse speed/acceleration settings improvement
I propose adding an option to the XF86Config mouse settings that lets a user modify the mouse speed (not acceleration) dynamically (with xset rather than XF86Config). Right now, xset m acceleration threshold tells X to make the mouse go acceleration times faster when it's going beyond speed threshold, unless threshold is 0, in which case it makes the mouse go at a polynomial mouse_speed^acceleration instead of mouse_speed^1. This last feature (with threshold=0) is undocumented. The right way to do things is to have smooth acceleration curve (like polynomial) and to let the user control both the acceleration and a constant multiplier dynamically (not in XF86Config). For example, the mouse should behave something like a * speed^b where a and b can be set through xset. Currently, one can fake a constant multiplier by setting a Resolution parameter in XF86Config. But this feature requires root access, a server restart, a lot of figuring out (how does resolution correspond to speed?), is buried in the documentation (it's not on XF86Config's man page), and isn't supported for all mice. Proposal: Add a boolean option to XF86Config called UseSmoothMouseAccel that changes the behavior of xset. If this variable is set to true, the command xset m VALUE_A VALUE_B will be interpreted as set the mouse function to VALUE_B * mouse_speed^VALUE_A. This way, nobody has to modify xset. The server would also need some way to communicate the state of UseSmoothMouseAccel to clients, in order to show the right UI to the user settings mouse settings. Michael ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
[Xpert]Proposal for mouse speed acceleration settings
I propose adding an option to the XF86Config mouse settings that lets a user modify the mouse speed dynamically with xset, rather than with the Resolution option in XF86Config. Right now, xset m acceleration threshold means: Move the cursor at speed `acceleration*raw_mouse_speed' whenever `raw_mouse_speed' is greater than `threshold' -- unless `threshold' is 0, in which case the cursor will move at `raw_mouse_speed^acceleration'. This last feature (with threshold=0) is completely undocumented. In addition, there's no way to speed up the mouse by a constant multiplier (independent of acceleration) without changing the Resolution option in XF86Config. This requires a server restart, root access, isn't documented in XF86Config's man page, isn't supported for all mice, can't be customized per-user, and doesn't really make sense (Resolution is inversely proportional to MouseSpeed). The right way to do mouse acceleration is to have a smooth acceleration curve (like the polynomial you get with threshold=0) and to let the user control both the acceleration and a constant multiplier dynamically (not in XF86Config). For example, it'd be great if xset m [A] [B] made the cursor move according to the formula [A] * raw_mouse_speed^[B]. Proposal: Add a boolean option to XF86Config called UseSmoothMouseAccel that changes the behavior of xset. If this variable is set to true, the command `xset m [A] [B]' will mean set the cursor movement function to `[A] * raw_mouse_speed^[B]'. This will also require some way for client X applications (like gnome's mouse settings applet) to determine the value of UseSmoothMouseAccel, etc. Comments? There has been scattered discussion of possible ways to improve XFree86's mouse acceleration in the past, but there haven't been many practical solutions suggested. I'd really like to get this cobweb cleaned up. Michael ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Proposal for mouse speed acceleration settings
Michael Toomim [EMAIL PROTECTED] writes:
Proposal: Add a boolean option to XF86Config called
UseSmoothMouseAccel that changes the behavior of xset. If this
variable is set to true, the command `xset m [A] [B]' will mean set the
cursor movement function to `[A] * raw_mouse_speed^[B]'.
Back in May this year I sent the mail below to [EMAIL PROTECTED], but
I never got any response.
Søren
From: Soeren Sandmann [EMAIL PROTECTED]
Subject: Smoother mouse acceleration
To: [EMAIL PROTECTED]
The current mouse acceleration algorithm when the user defined
threshold is non-zero is essentially this:
if (d t)
d' = A * d
else
d' = d;
where d' is the new traveled distance, and d the distqance reported by
the mouse. The constant t is the user defined threshold, and A
(= num/den) is the user defined amount of acceleration.
There are some problems with this:
- if you set the threshold high, the mouse cursor feels
sticky. If you move the mouse at just the right speed, you
will se jerky cursor movements as the traveled distance
oscillates around the threshold.
- if you set the threshold low and the acceleration high, you
lose precision as the cursor moves too fast even on small
movements.
- if you set the acceleration low, it gets difficult to reach
to corners of the screen.
So, while this isn't terribly bad, it not really good either
Here is a patch that modifies the mouse acceleration code to use the
formula
d' = c * d^alpha + d
where d' is the new distance and d is the distance reported by the
mouse. The constants c and alpha are based on the acceleration
settings, specifically
c = 1 / (threshold / 4) ^alpha
which ensures that when the travelled distance is less then
threshold/4, the distance is not modified by more than 1. This ensures
that the threshold set by the user still corresponds to the precision
of the device.
The constant alpha is just a linear function of the user set
acceleration.
The patch also changes the default acceleration to 8/3 3, which in my
opinion is a more useful setting than 2/1 4, which is the current
default.
Søren
--- ../../../xcoriginal/programs/Xserver/hw/xfree86/common/xf86Xinput.c Tue May 21 01:50:43 2002
+++ ./hw/xfree86/common/xf86Xinput.c Tue May 21 01:49:00 2002
@@ -905,15 +905,42 @@
/*
* Accelerate
*/
- if (device-ptrfeed device-ptrfeed-ctrl.num) {
- /* modeled from xf86Events.c */
- if (device-ptrfeed-ctrl.threshold) {
- if ((abs(dx) + abs(dy)) = device-ptrfeed-ctrl.threshold) {
- valuator[0] = (dx * device-ptrfeed-ctrl.num) /
- device-ptrfeed-ctrl.den;
- valuator[1] = (dy * device-ptrfeed-ctrl.num) /
- device-ptrfeed-ctrl.den;
+ if (device-ptrfeed device-ptrfeed-ctrl.num device-ptrfeed-ctrl.den) {
+ if (device-ptrfeed-ctrl.num = device-ptrfeed-ctrl.den) {
+ valuator[0] = (dx * device-ptrfeed-ctrl.num) / device-ptrfeed-ctrl.den;
+ valuator[1] = (dy * device-ptrfeed-ctrl.num) / device-ptrfeed-ctrl.den;
+ }
+ else if (device-ptrfeed-ctrl.threshold) {
+ static int cached_num = -1;
+ static int cached_den = -1;
+ static int cached_threshold = -1;
+
+ static double c = -1;
+ static double alpha = -1;
+
+ double k;
+
+ if (cached_num != device-ptrfeed-ctrl.num ||
+ cached_den != device-ptrfeed-ctrl.den ||
+ cached_threshold != device-ptrfeed-ctrl.threshold) {
+
+ double A, t;
+
+ cached_num = device-ptrfeed-ctrl.num;
+ cached_den = device-ptrfeed-ctrl.den;
+ cached_threshold = device-ptrfeed-ctrl.threshold;
+
+ A = (double)cached_num / cached_den;
+ t = (double)cached_threshold;
+
+ alpha = A/3.0 + 2.0/3.0;
+ c = 1 / pow (t/4.0, alpha);
}
+
+ k = c * pow (dx*dx + dy*dy, 0.5 * (alpha - 1)) + 1;
+
+ valuator[0] = k * dx;
+ valuator[1] = k * dy;
}
else if (dx || dy) {
mult = pow((float)(dx*dx+dy*dy),
--- ../../../xcoriginal/programs/Xserver/include/site.h Tue May 21 01:50:43 2002
+++ ./include/site.h Tue May 21 01:17:48 2002
@@ -113,9 +113,9 @@
#define DEFAULT_INT_MAX_VALUE 100
#define DEFAULT_INT_DISPLAYED 0
-#define DEFAULT_PTR_NUMERATOR 2
-#define DEFAULT_PTR_DENOMINATOR 1
-#define DEFAULT_PTR_THRESHOLD 4
+#define DEFAULT_PTR_NUMERATOR 8
+#define DEFAULT_PTR_DENOMINATOR 3
+#define DEFAULT_PTR_THRESHOLD 3
#define DEFAULT_SCREEN_SAVER_TIME (10 * (60 * 1000))
#define DEFAULT_SCREEN_SAVER_INTERVAL (10 * (60 * 1000))
Re: [Xpert]Proposal for mouse speed acceleration settings
Soeren Sandmann wrote: Back in May this year I sent the mail below to [EMAIL PROTECTED], but I never got any response. That sucks! Would it be rude to send it again? I like the idea of changing the acceleration formula completely (as you have done) rather than introducing a new XF86Config option to change it (as I had suggested). ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Proposal for mouse speed acceleration settings
At 02\11\01 11:52 -0800 Friday, Michael Toomim wrote: Soeren Sandmann wrote: Back in May this year I sent the mail below to [EMAIL PROTECTED], but I never got any response. That sucks! Would it be rude to send it again? I like the idea of changing the acceleration formula completely (as you have done) rather than introducing a new XF86Config option to change it (as I had suggested). The XFree86 mouse deceleration function gets dx,dy integers that are small integers in between -1 and +1 quite often. Whatever the function (formula) was, it could be replaced with these two without much change: dx' = k1 * dx * (+1 if dx 0 else -1) dx' = k2 * dx [dx is a C int, dx' is real that is added to a real sum and then later converted into a screen pixel position] Mr David Dawes was replying to some e-mail on the topic of the mouse deceleration algorithm. He told me about Opensource projects. What could be done is this: * xf86PostMotionEvent() is repeatedly called. EAch time it is called, the dx,dy and time values are saved. * an algorithm averages the last k dx (and dy) values. The function that calculates k is not too simple. Certainly k is not a constant since that can lead to a slow mouse pointer appearing to have inertia. I presume patches aiming to tweak functions ought be filed/etc away while there is not some substantially better averaging algorithm in the XFree86 source code. xf86PostMotionEvent(): http://cvsweb.xfree86.org/cvsweb/xc/programs/Xserver/hw/xfree86/common/xf86Xinput.c Craig Carey ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
[Xpert]Proposal for mouse speed acceleration settings
I propose adding an option to the XF86Config mouse settings that lets a user modify the mouse speed dynamically with xset, rather than with the Resolution option in XF86Config. Right now, xset m acceleration threshold means: Move the cursor at speed `acceleration*raw_mouse_speed' whenever `raw_mouse_speed' is greater than `threshold' -- unless `threshold' is 0, in which case the cursor will move at `raw_mouse_speed^acceleration'. This last feature (with threshold=0) is completely undocumented. In addition, there's no way to speed up the mouse by a constant multiplier (independent of acceleration) without changing the Resolution option in XF86Config. This requires a server restart, root access, isn't documented in XF86Config's man page, isn't supported for all mice, can't be customized per-user, and doesn't really make sense (Resolution is inversely proportional to MouseSpeed). The right way to do mouse acceleration is to have a smooth acceleration curve (like the polynomial you get with threshold=0) and to let the user control both the acceleration and a constant multiplier dynamically (not in XF86Config). For example, it'd be great if xset m [A] [B] made the cursor move according to the formula [A] * raw_mouse_speed^[B]. Proposal: Add a boolean option to XF86Config called UseSmoothMouseAccel that changes the behavior of xset. If this variable is set to true, the command `xset m [A] [B]' will mean set the cursor movement function to `[A] * raw_mouse_speed^[B]'. This will also require some way for client X applications (like gnome's mouse settings applet) to determine the value of UseSmoothMouseAccel, etc. Comments? There has been scattered discussion of possible ways to improve XFree86's mouse acceleration in the past, but there haven't been many practical solutions suggested. I'd really like to get this cobweb cleaned up. Michael ___ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
Re: [Xpert]Proposal for mouse speed acceleration settings
I have a webpage on the topic of the mouse in XFree86: http://www.ijs.co.nz/linux-orbiting-mouse.htm At 2002\10\31 18:20 -0800 Thursday, Michael Toomim wrote: I propose adding an option to the XF86Config mouse settings that lets a user modify the mouse speed dynamically with xset, rather than with the Resolution option in XF86Config. Right now, xset m acceleration threshold means: Move the cursor at speed `acceleration*raw_mouse_speed' whenever `raw_mouse_speed' is greater than `threshold' -- unless `threshold' is 0, in which case the cursor will move at `raw_mouse_speed^acceleration'. This last feature (with threshold=0) is completely undocumented. In It is also not available to KDE users (who can't get the threshold down to 0, if I recall correctly). Anyway, the unavailable algorithm maybe is no better (or if it is, then comparing the two is going to produce indecision over which is better). addition, there's no way to speed up the mouse by a constant multiplier (independent of acceleration) without changing the Resolution option in XF86Config. This requires a server restart, root access, isn't documented in XF86Config's man page, isn't supported for all mice, can't be customized per-user, and doesn't really make sense (Resolution is inversely proportional to MouseSpeed). The right way to do mouse acceleration is to have a smooth acceleration curve (like the polynomial you get with threshold=0) and to let the user That is not correct because the data going into the functions. is mainly (dx,dy) taking these values: (0,0),(1,0),(-1,0),(0,1),(0,-1). It might be different if some high resolution expensive mouse gets around the problem. Such things are not needed when Windows 2000 is used. To get the digits +/-2 to show up, when there is a plain Microsoft PS/2 mouse, then a speed that is often not reached, is needed. control both the acceleration and a constant multiplier dynamically (not in XF86Config). For example, it'd be great if xset m [A] [B] made the cursor move according to the formula [A] * raw_mouse_speed^[B]. It is a 2-D curve. Also it has a discontinuity in it which is a bug. The effect of the bug would be to make the mouse orbit around that stopping point. I tried removing the bug but it was not any better. The function ought be replaced eventually even though it seems to me that most of the problem is due to the missing code to do averaging of dx,dy integers. Proposal: Add a boolean option to XF86Config called UseSmoothMouseAccel that changes the behavior of xset. If this variable is set to true, the command `xset m [A] [B]' will mean set the cursor movement function to `[A] * raw_mouse_speed^[B]'. While the bad problem of an absence of code to average mouse data (for a plain XFree86 configuration), is not remedied, then other problems can't easily be fixed, due to reviewers being unable to decide if the change made the mouse better or worse. After that there could be a problem of too many possible algorithms. It could be nice to shift disputes over retaining the meaning of the sliders in GNOME and rewriting the parameters, out to the desktop bug tracking systems, so it does not add disputes unnecessarily to the topic of whether copying the mouse algorithm of Windows 2000/Timeout is best. The problems are inside of the xf86PostMotionEvent() procedure, which can be viewed at this URL: http://cvsweb.xfree86.org/cvsweb/xc/programs/Xserver/hw/xfree86/common/xf86Xinput.c This will also require some way for client X applications (like gnome's mouse settings applet) to determine the value of UseSmoothMouseAccel, etc. Comments? There has been scattered discussion of possible ways to improve XFree86's mouse acceleration in the past, but there haven't been many practical solutions suggested. I'd really like to get this cobweb cleaned up. I am guessing people would prefer hundreds of iterations recompiling. A quickest way could be to start out compiling some Windows program. Tapping into Windows 2000 mouse data and then inferring from an analysis of the data should allow the Microsoft algorithm to be precisely copied. They have a worse mouse deceleration algorithm in Windows 98 SE, and possibly the best effort of someone could produce something as bad, Anyway, for many, having the mouse be about identical to a Windows 2000 mouse is aim independent of one of having the behaviour be unimprovable. --- To fix the bad design of the sliders in GNOME and KDE could justify a preprocessing subroutine/procedure that alters the parameters that then pass into the mouse pointer deceleration algorithm. Possibly they might want to control the rewriting of the parameter integers but not provide their own new mouse deceleration code. Would someone say how to get debugging data piped out of XFree86 and put into a window in XFree86 in any OS. Any sample code to copy?. Craig Carey [EMAIL PROTECTED] Auckland, New Zealand Ada 95 mailing lists;
