View previous topic :: View next topic |
Author |
Message |
BlueByte Tux's lil' helper
Joined: 08 Apr 2004 Posts: 87
|
Posted: Tue Oct 19, 2004 9:58 pm Post subject: |
|
|
i did like they did but i got:
Code: |
root@Pingu:~# logitech_applet --get-res
002/002 046D/C50D M-RAA89 Cordless Click! Plus Optical Mouse
Error getting resolution from device : error sending control message: Broken pipe
Resolution: (Unexpected result:-65) |
aww crap
...
ok looks like my crappy mouse doesn't support 800 dpi ... stupid wireless shit |
|
Back to top |
|
|
Jagasian n00b
Joined: 04 Dec 2004 Posts: 7
|
Posted: Sat Dec 04, 2004 9:33 pm Post subject: USB Mouse Rate Reader |
|
|
The Quake 1 and more specifically the Quakeworld community (http://www.quakeworld.nu/) has been using the logitech_applet for 800 dpi mousing and the polling rate hack for 500hz mice for a long time now.
In this thread :
http://www.fuhquake.net/forum/viewtopic.php?t=1623
...there is a post detailing code and the steps required to build a simple "read" executable that can be used to experimentally measure the actual time between separate mouse movement events. Running the utility requires you to pass the event device to the program, and this differs from system to system. On mine, it is "/dev/input/event2". Here is what I get WITHOUT the kernel hack:
Code: | [root@girard Rate Reader]# ./read /dev/input/event2
62
125
125
125
125
125
125
125
125
|
The program simply waits for a mouse movement event, and calculates the difference in time since the last event. Obviously, if you move your mouse really slowly and create an event and then wait a few seconds and move it again to create the next event, the measured time difference will lead to a polling rate calculation that is really low, so you get the best measurements when you move your mouse really fast. Hence the numbers the program spits out will always be lower than or equal to your actual mouse rate. Varying numbers do not mean that your mouse is unstable. However, with the kernel hack and polling rate set to "2", running the program and moving your mouse in a consistent and fast motion should result in a bunch of 500s printed to the screen.
I have never seen a mouse that could create new mouse events faster than 500hz.
Also, some people have claimed that the fact that your game FPS is limited to say 125 fps means that a mouse rate above 125 hz is useless. This couldn't be more wrong! Your mouse is not synchronized to your framerate. Hence if your FPS is 125 and your mouse rate is 125hz, then the maximum amount of latency (time delay) from when you actually last moved your mouse and when the game actually reads the movement is just under 8 miliseconds. That might not seem like much lag, but the average gamer can notice small amounts of lag in mouse movement.
That is why a 500hz mouse rate is a huge improvement! With a 500hz rate and a 125fps, the maximum amount of latency from when your mouse last reports movement and and when the game reacts to that movement is 2 miliseconds. That means your mouse movement has up to 8 times as much lag without the hack compared to with the hack set to "2". Since mouse movement is all client side in games like Quake and Quake 3, and because mouse movement consists of many small move events, a little difference in lag makes a big difference.
The rate hack also improves other things than games. I hate surfing the web without it |
|
Back to top |
|
|
rel Tux's lil' helper
Joined: 20 Apr 2002 Posts: 137
|
Posted: Fri Dec 17, 2004 8:37 am Post subject: |
|
|
Hello,
this mouse patch must go into mainstream!
I've used it for a month, and after updating a kernel
and forgetting patching it with this chmp patch.
I thought, whats wrong with my mouse. Until I remembered the patch.
It makes a big diff not only for games.
greets |
|
Back to top |
|
|
krejler Tux's lil' helper
Joined: 10 Nov 2003 Posts: 142 Location: Denmark
|
Posted: Mon Dec 20, 2004 3:14 pm Post subject: |
|
|
Great!
rel wrote: | Hello,
this mouse patch must go into mainstream!
I've used it for a month, and after updating a kernel
and forgetting patching it with this chmp patch.
I thought, whats wrong with my mouse. Until I remembered the patch.
It makes a big diff not only for games.
greets |
Hehe, I have posted a quite slimmed down version to LKML. If you want it in, you better respond on there.
Here's a link:
http://lkml.org/lkml/2004/12/18/112
This new and trimmed down version almost works the same way as the old one. Although I've removed the Kconfig entry (Greg KH thought some kind of sysfs entry would be better), so I added a module parameter (just like the 'old' one) but with a sysfs entry.
To set the polling interval, you do:
Code: | modprobe/insmod usbhid mousepoll=yourinterval | (You can probably add parameters to /etc/module.conf or something, I'm not too sure)
If you decide that the set polling interval is too low/high, you can just:
Code: | echo your_interval > /sys/module/usbhid/parameters/mousepoll |
...and replug your mouse.
Oh, and make sure you don't echo a string to it, echo an integer!
Farewell! |
|
Back to top |
|
|
servo888 Apprentice
Joined: 22 Feb 2004 Posts: 293
|
Posted: Sun Dec 26, 2004 4:54 pm Post subject: |
|
|
omg... It's like giving my mouse an orgasm... Thanks for the tip! My dual optical logitech is running flawlessly and smoothly! |
|
Back to top |
|
|
neenee Veteran
Joined: 20 Jul 2003 Posts: 1786
|
Posted: Fri Jan 07, 2005 1:58 pm Post subject: |
|
|
hm. i just tried this latest patch on 2.6.10-cko3, and it does work,
but the problem which i had many months ago exists in this ver-
sion as well:
selecting text in aterm is very choppy when this patch is enabled.
in xterm there's no problem, but ofcourse, i want aterm.
on the bright side - movement of the cursor is very smooth. |
|
Back to top |
|
|
krejler Tux's lil' helper
Joined: 10 Nov 2003 Posts: 142 Location: Denmark
|
Posted: Sat Jan 08, 2005 4:40 pm Post subject: |
|
|
neenee wrote: | hm. i just tried this latest patch on 2.6.10-cko3, and it does work,
but the problem which i had many months ago exists in this ver-
sion as well:
selecting text in aterm is very choppy when this patch is enabled.
in xterm there's no problem, but ofcourse, i want aterm.
on the bright side - movement of the cursor is very smooth. |
Hmm? Did you patch your cko kernel?
AFAIK it's already included in cko, but I think the one in cko is a bit old.
Anyway, the aterm problems are a bit sad. And oh, Eterm is also affected.
I guess it's due to some wrong assumptions in the code somewhere, but what do I know.
But while we're at it, I have an example of a program that has the exact reversed symptoms. Namely, openbox.
It feels very jerky, when, for instance, moving windows, or going through menus. (EDIT: Without the patch applied. )
And oh, since the latest version haven't been noticed in this thread yet, I might as well do it:
Latest: http://omfg.linux.dk/pub/chmp/archive/chmp-r5-FULL.patch.
Split out patches can be grabbed from http://omfg.linux.dk/pub/chmp/.
Farewell.
P.S.
I decided to include the Kconfig entry again, since it's depature seemed to cause a bit of confusion.
Last edited by krejler on Thu Jan 13, 2005 9:17 pm; edited 2 times in total |
|
Back to top |
|
|
neenee Veteran
Joined: 20 Jul 2003 Posts: 1786
|
Posted: Sat Jan 08, 2005 8:46 pm Post subject: |
|
|
indeed. i patched cko, which did not seem to contain the patch.
i'll probably go back to the logitech applet, though it does not
give the same feeling as this patch.
good luck with your patch
*update* i moved to urxvt, setup the same look as i had with
aterm and selection is silky smooth. hurray |
|
Back to top |
|
|
Assgier Tux's lil' helper
Joined: 28 May 2003 Posts: 127 Location: Europe -> Netherlands -> Utrecht -> Amersfoort
|
Posted: Sun Jan 09, 2005 4:45 am Post subject: |
|
|
that patch should defenitely get included in gentoo-dev-sources
i'm now using the logitech_applet on my MX500:
Quote: |
Jezus root # logitech_applet
003/003 046D/C025 M-BP81A MX500 Optical Mouse
Resolution: 800cpi
Cruise Control / Smart Scroll: Enabled
|
Tough i find myself (*cough*) an experiënced GNU/Linux user, i keep discovering new cool things that can be done with it
The program is great _________________
|
|
Back to top |
|
|
krejler Tux's lil' helper
Joined: 10 Nov 2003 Posts: 142 Location: Denmark
|
Posted: Thu Jan 13, 2005 9:14 pm Post subject: |
|
|
Assgier wrote: | that patch should defenitely get included in gentoo-dev-sources |
If you say so.
However, I don't really think it's these kind of things gentoo-dev-sources are aiming for.
Also, the patch is, as the title of this thread says, a hack.
It really should be done in a more clean manner.
For example, I think that a per-device polling-interval, adjustable by a sysfs-entry would be the best way to implement this. Unfortunately, I haven't really had the time to play with the idea, yet.
neenee wrote: | *update* i moved to urxvt, setup the same look as i had with
aterm and selection is silky smooth. hurray |
Yay. |
|
Back to top |
|
|
bennettp Guru
Joined: 22 Feb 2004 Posts: 335 Location: on my back and tumbling
|
Posted: Fri Jan 14, 2005 3:06 pm Post subject: |
|
|
You should start a new thread to tell people this. This is an essential patch for any gamers with a usb mouse! Get the word out!
[edit]: on second thoughts I'll do it. _________________ Registered Linux User #363420 |
|
Back to top |
|
|
krejler Tux's lil' helper
Joined: 10 Nov 2003 Posts: 142 Location: Denmark
|
Posted: Fri Jan 14, 2005 9:19 pm Post subject: |
|
|
bennettp wrote: |
You should start a new thread to tell people this. This is an essential patch for any gamers with a usb mouse! Get the word out!
[edit]: on second thoughts I'll do it. |
Thanks. Too bad it got slayed, though.
Anyway, thank you, tomk, for adding the reference to the first post! |
|
Back to top |
|
|
R!tman Veteran
Joined: 18 Dec 2003 Posts: 1303 Location: Zurich, Switzerland
|
Posted: Fri Jan 14, 2005 11:56 pm Post subject: |
|
|
I just noticed this very nice thing. I could not test I out, I am with my notebook right now. I will have to try it with armyops . |
|
Back to top |
|
|
gaminggeek Apprentice
Joined: 13 Nov 2003 Posts: 231 Location: New Zealand
|
Posted: Sat Jan 15, 2005 10:40 pm Post subject: |
|
|
anyone tryed this on a genius mouse/a non-Logitech mouse? |
|
Back to top |
|
|
krejler Tux's lil' helper
Joined: 10 Nov 2003 Posts: 142 Location: Denmark
|
Posted: Sat Jan 15, 2005 11:18 pm Post subject: |
|
|
gaminggeek wrote: | anyone tryed this on a genius mouse/a non-Logitech mouse? |
To my knowledge it hasn't been tested on anything other than Logitech and Microsoft mice. (And OEM re-branded Logitech/Microsoft hardware.)
There is absolutely no reason that other mice wouldn't work. Except, of course, if the hardware cannot handle a lowered polling interval.
Feel free to try it, and please report back, if you do. |
|
Back to top |
|
|
gaminggeek Apprentice
Joined: 13 Nov 2003 Posts: 231 Location: New Zealand
|
Posted: Sat Jan 15, 2005 11:37 pm Post subject: |
|
|
OK I will try
*edit* I can't patch this file
Code: | root@cbb linux # cat /home/caleb/chmp-r5-FULL.patch | patch --dry-run -p1
patching file drivers/usb/input/hid-core.c
Hunk #2 succeeded at 1647 (offset -17 lines).
Hunk #3 succeeded at 1873 (offset -43 lines).
patching file drivers/usb/input/Kconfig
Hunk #1 succeeded at 23 with fuzz 2 (offset -1 lines).
patching file drivers/usb/input/hid-core.c
Hunk #1 FAILED at 42.
1 out of 1 hunk FAILED -- saving rejects to file drivers/usb/input/hid-core.c.rej
root@cbb linux # |
|
|
Back to top |
|
|
krejler Tux's lil' helper
Joined: 10 Nov 2003 Posts: 142 Location: Denmark
|
Posted: Sat Jan 15, 2005 11:55 pm Post subject: |
|
|
gaminggeek wrote: | OK I will try
*edit* I can't patch this file
Code: | root@cbb linux # cat /home/caleb/chmp-r5-FULL.patch | patch --dry-run -p1
patching file drivers/usb/input/hid-core.c
Hunk #2 succeeded at 1647 (offset -17 lines).
Hunk #3 succeeded at 1873 (offset -43 lines).
patching file drivers/usb/input/Kconfig
Hunk #1 succeeded at 23 with fuzz 2 (offset -1 lines).
patching file drivers/usb/input/hid-core.c
Hunk #1 FAILED at 42.
1 out of 1 hunk FAILED -- saving rejects to file drivers/usb/input/hid-core.c.rej
root@cbb linux # |
|
What kernel version are you trying to apply it to?
They're diffed against a clean 2.6.10.
Yes, it know it doesn't say so anywhere, blame me.
EDIT: I've now added a notice to the page.
EDIT 2:
Actually, you should be all fine.
I just noticed you're using dry-run to test it. But actually, you can't do that. On the FULL-version, that is.
The reason for that is simple. The FULL version is simply all three patches concatentated into one.
What makes it fail, is, that "chmp-r5-enable-kconfig-value-in-driver.patch" depends on a line added by "chmp-r5-add-modparam.patch". Since it cannot find the line, because you're using dry-run, it spits an error at you.
I'm sure it will apply cleanly, if you try without dry-run.
Farewell! |
|
Back to top |
|
|
gaminggeek Apprentice
Joined: 13 Nov 2003 Posts: 231 Location: New Zealand
|
Posted: Sun Jan 16, 2005 12:12 am Post subject: |
|
|
I was using 2.6.9
I ran it with out dryrun and it worked fine thats
going to try it out now |
|
Back to top |
|
|
krejler Tux's lil' helper
Joined: 10 Nov 2003 Posts: 142 Location: Denmark
|
Posted: Sun Jan 16, 2005 12:37 am Post subject: |
|
|
gaminggeek wrote: | I was using 2.6.9
I ran it with out dryrun and it worked fine thats
going to try it out now |
Good luck. |
|
Back to top |
|
|
gaminggeek Apprentice
Joined: 13 Nov 2003 Posts: 231 Location: New Zealand
|
Posted: Sun Jan 16, 2005 12:41 am Post subject: |
|
|
I cannot change the number in menuconfg
and I added that line to the file that you said to and then rebooted to check and Code: | T: Bus=03 Lev=01 Prnt=01 Port=00 Cnt=01 Dev#= 2 Spd=1.5 MxCh= 0
D: Ver= 1.10 Cls=00(>ifc ) Sub=00 Prot=00 MxPS= 8 #Cfgs= 1
P: Vendor=0458 ProdID=001a Rev= 0.00
S: Manufacturer=KYE
S: Product=PowerScroll EYE
C:* #Ifs= 1 Cfg#= 1 Atr=a0 MxPwr=100mA
I: If#= 0 Alt= 0 #EPs= 1 Cls=03(HID ) Sub=01 Prot=02 Driver=usbhid
E: Ad=81(I) Atr=03(Int.) MxPS= 5 Ivl=10ms
|
|
|
Back to top |
|
|
krejler Tux's lil' helper
Joined: 10 Nov 2003 Posts: 142 Location: Denmark
|
Posted: Sun Jan 16, 2005 12:52 am Post subject: |
|
|
gaminggeek wrote: | I cannot change the number in menuconfg
and I added that line to the file that you said to and then rebooted to check and Code: | T: Bus=03 Lev=01 Prnt=01 Port=00 Cnt=01 Dev#= 2 Spd=1.5 MxCh= 0
D: Ver= 1.10 Cls=00(>ifc ) Sub=00 Prot=00 MxPS= 8 #Cfgs= 1
P: Vendor=0458 ProdID=001a Rev= 0.00
S: Manufacturer=KYE
S: Product=PowerScroll EYE
C:* #Ifs= 1 Cfg#= 1 Atr=a0 MxPwr=100mA
I: If#= 0 Alt= 0 #EPs= 1 Cls=03(HID ) Sub=01 Prot=02 Driver=usbhid
E: Ad=81(I) Atr=03(Int.) MxPS= 5 Ivl=10ms
|
|
Try a:
Code: | cat /sys/module/usbhid/parameters/mousepoll |
It should return the value you specified. Just to check if it actually received it.
Also, you should also be able to just pass the parameter "mousepoll" to the module when loading it. I'm not sure if this module is loaded automatically on your box, either by hot/coldplug or some /etc/modules.autoload.d/-foo.
If it is, you should rmmod it and insert it into the kernel again:
Code: | modprobe usbhid mousepoll=<your_desired_interval> |
After that, check "/proc/bus/usb/devices" and "/sys/module/usbhid/parameters/mousepoll" again.
You should also be able to re-set the value without removing the module. I think I've described it in a post earlier, but you should simply do something like this:
Code: | echo [your_desired_interval] > /sys/module/usbhid/parameters/mousepoll |
And replug your mouse afterwards.
If you're still stuck, please keep me updated.
Thanks.
NOTE: "the module", or "it", is usbhid. |
|
Back to top |
|
|
gaminggeek Apprentice
Joined: 13 Nov 2003 Posts: 231 Location: New Zealand
|
Posted: Sun Jan 16, 2005 1:27 am Post subject: |
|
|
root@cbb linux # echo 2 > /sys/module/usbhid/parameters/mousepoll bash: /sys/module/usbhid/parameters/mousepoll: No such file or directory
root@cbb linux #
|
|
Back to top |
|
|
krejler Tux's lil' helper
Joined: 10 Nov 2003 Posts: 142 Location: Denmark
|
Posted: Sun Jan 16, 2005 1:30 am Post subject: |
|
|
gaminggeek wrote: | root@cbb linux # echo 2 > /sys/module/usbhid/parameters/mousepoll bash: /sys/module/usbhid/parameters/mousepoll: No such file or directory
root@cbb linux #
|
Do you have sysfs mounted?
Do you have usbhid loaded? Is it built directly into your kernel, or as a module? |
|
Back to top |
|
|
gaminggeek Apprentice
Joined: 13 Nov 2003 Posts: 231 Location: New Zealand
|
Posted: Sun Jan 16, 2005 1:36 am Post subject: |
|
|
syfs?
yes I have the module loaded and doom3 seems easyer to control |
|
Back to top |
|
|
krejler Tux's lil' helper
Joined: 10 Nov 2003 Posts: 142 Location: Denmark
|
Posted: Sun Jan 16, 2005 1:57 am Post subject: |
|
|
gaminggeek wrote: | syfs?
yes I have the module loaded and doom3 seems easyer to control |
So... It is working? |
|
Back to top |
|
|
|