|
On January 11 2012 06:56 3nigmatiq wrote:
I want to thank you Bumblebees. Those are eight brilliant kilobytes.
I tested your program on a 24" iMac running Mac OS X Snow Leopard with a Razer DeathAdder 3500. It works flawlessly. I use a setting of 3500dpi with 5/10 sensitivity and no acceleration.[...]
Im not 100% sure, but if i remember correctly, 3500 DPI @ 5/10 is not linear. I think it even says so in the manual that comes with the deathadder - 10/10 sensitivity is the only linear setting.
So having 3500 DPI @ 5/10 means a loss in precision. 10/10 in combination with the a DPI setting that is comfortably fast for you should be the way to go.
And if my memory tricks me in this case, just forget all the bullshit that I've written xD (I'm fairly certain though that what I just said actually is the case)
|
|
|
I'd like to see something to back that statement up, haven't seen anything about a fixed 32ms lag until now.
(the link to the Logitech forums does only talk about the polling rate issue on Leopard and below)
The partly untrue statement about mouse accel on OS X doesn't make this article very believable.
And a little add-on I just had to think about:
The article says that the OS X input lag is thus twice as big as the lag Windows has.
Thus: OS X 32ms, Win 7 16ms.
Now think about the polling rate of mice. NO ONE ever complained about an input lag on Windows when 125mhz mice were standard. So let's compare the lag of a 125hz mouse on Windows to a modern 1000hz mouse on OS X:
125hz=8ms lag + 16ms Windows lag gives us a total of 24ms input lag.
1000hz=1ms lag + 32ms OS X lag gives us a total of 33ms input lag.
That's (assuming this article is true) 9ms of an additional lag on OS X compared to what was considered perfectly playable. I can live with that.
The point I want to make with this is not what OS X is doing regarding mouse ballistics is cool. Nah uh! I hate the bad coding on the OS X front and Windows definitely is superior, be it the input lag or mouse accel. At most though, if you're using the same polling rate, you get 16ms addition delay on OS X.
I can't' see how this renders a properly configured OS X in combination with a good mouse "useless".
|
Bumblebees you are a hero!
|
To add my voice to the chorus, I used this utility after trying out Decelerator and it works wonderfully. Decelerator worked well for making the acceleration more consistent in the OS in general, but this gives the mouse a much more natural feel in SC2. Awesome!
|
On January 14 2012 04:04 MrBazinga wrote:Show nested quote +On January 11 2012 06:56 3nigmatiq wrote:
I want to thank you Bumblebees. Those are eight brilliant kilobytes.
I tested your program on a 24" iMac running Mac OS X Snow Leopard with a Razer DeathAdder 3500. It works flawlessly. I use a setting of 3500dpi with 5/10 sensitivity and no acceleration.[...] Im not 100% sure, but if i remember correctly, 3500 DPI @ 5/10 is not linear. I think it even says so in the manual that comes with the deathadder - 10/10 sensitivity is the only linear setting. So having 3500 DPI @ 5/10 means a loss in precision. 10/10 in combination with the a DPI setting that is comfortably fast for you should be the way to go. And if my memory tricks me in this case, just forget all the bullshit that I've written xD (I'm fairly certain though that what I just said actually is the case)
In the manual, I did not find information backing up your statement. The closest on the subject is the following.
Razer DeathAdder Master Guide (PDF, page 8)
Sensitivity
Mouse sensitivity is an internal filter of the driver. The maximum sensitivity setting of 10 is unfiltered. Each incremental stop or setting is a 10% decrease, i.e. 1 setting will decrease mouse pointer input to 10% (i.e. a linear graph).
From the above, I understand that 3500dpi with 5/10 sensitivity is approximately equal to half that dpi with twice the sensitivity, or 1750dpi with 10/10 sensitivity. Therefore, when acceleration is turned off in the Razer's driver app, the sensitivity slider does not affect acceleration, since there is no acceleration (thanks to MouseFixer2). But most importantly, it certainly feels that way in practice. I do not notice any change in the distance travelled by the pointer as a function of mouse speed.
|
By the way, as my previous setting was a bit too fast, I am now using an even lower sensitivity (3500dpi at 2.5/10 sensitivity) and I still see no sign of pointer acceleration. I guess 3500*25% is about 1800 * 50%, i.e. what Bumblebees suggested for a Windows-like mouse setting.
|
Thanks you, Bumblebees. This little fix definitely works, and it is great for editing and work besides gaming.
|
On January 18 2012 23:30 3nigmatiq wrote:Show nested quote +Razer DeathAdder Master Guide (PDF, page 8)
Sensitivity
Mouse sensitivity is an internal filter of the driver. The maximum sensitivity setting of 10 is unfiltered. Each incremental stop or setting is a 10% decrease, i.e. 1 setting will decrease mouse pointer input to 10% (i.e. a linear graph).
From the above, I understand that 3500dpi with 5/10 sensitivity is approximately equal to half that dpi with twice the sensitivity, or 1750dpi with 10/10 sensitivity. Therefore, when acceleration is turned off in the Razer's driver app, the sensitivity slider does not affect acceleration, since there is no acceleration (thanks to MouseFixer2). But most importantly, it certainly feels that way in practice. I do not notice any change in the distance travelled by the pointer as a function of mouse speed.
Sorry, I didn't mean acceleration but accuracy.
Let's say u move the mouse 15px to the right.
Sensitivity 10/10: Mouse moves 15px to the right
Sensitivity 05/10: Mouse moves 7,5px to the right
Since there is no half-pixel, the software then either sends 7 or 8 pixel as movement.
So yeah, this seems to be really nitpicky and I would probably never ever be able to tell the difference between 1800dpi @ 5/10 and 900dpi @ 10/10.
BUT: If you go more extreme like an 80% reduction, that would actually mean that you move your mouse 2 pixel yet the cursor doesn't move at all. Now that doesn't seem right to me.
And maybe I'm mistaken and someone please correct me if I'm wrong - but i think there is no advantage whatsoever having higher DPI at lower sensibility. (compared to using your DPI 1 to 1)
|
On January 19 2012 07:51 MrBazinga wrote:
Sorry, I didn't mean acceleration but accuracy.
Let's say u move the mouse 15px to the right.
Sensitivity 10/10: Mouse moves 15px to the right
Sensitivity 05/10: Mouse moves 7,5px to the right
Since there is no half-pixel, the software then either sends 7 or 8 pixel as movement.
So yeah, this seems to be really nitpicky and I would probably never ever be able to tell the difference between 1800dpi @ 5/10 and 900dpi @ 10/10.
BUT: If you go more extreme like an 80% reduction, that would actually mean that you move your mouse 2 pixel yet the cursor doesn't move at all. Now that doesn't seem right to me.
And maybe I'm mistaken and someone please correct me if I'm wrong - but i think there is no advantage whatsoever having higher DPI at lower sensibility. (compared to using your DPI 1 to 1)
This makes more sense. However, your argument stands only as long as we assume enough dexterity at practical mouse settings (unlike the extreme case in my example below), such that the hand will "know" the difference between signalling the mouse to travel a distance of d pixels and a distance of d-c pixels, where c will be much smaller than d in practice.
For example, take a setting of 3500dpi, 10/10 sensitivity. This translates into 3500 pixels (dots) per inch, or 137.8 pixels per millimeter. The mouse distance to move only 1 pixel in either direction (x or y coordinates) is the reciprocal of that value, or 0.007257 millimeters per pixel.
Let's decrease sensitivity down to an extreme 1/10 (which is approximately equal to an effective 350dpi setting, just below that of an ordinary low precision mouse). Keep in mind that such a setting without acceleration will be unbearably slow, so it is not as realistic a choice as say, a sensitivity of 2/10 at 3500dpi. But I am choosing it on purpose to test your argument. Using these values, we translate them into 13.78 pixels per millimeter, which means we must move the mouse 0.07257 millimeters to move the pointer one pixel on the screen in either x or y coordinates. This is one mouse step.
Now, assuming the engineers at Razer weren't retarded and didn't just truncate the pixel values or take a ceiling function, they would have implemented a function in the internal driver (before the mouse sends the information to the operating system) that rounds the number of mouse steps to the closest integer number (as you said, there is no half-pixel). If we want to move the pointer half of that minimum value in either the horizontal or the vertical direction on the screen, but we get a pointer movement of either 0 or 1 pixels, this leaves us with a maximum error, in either coordinate, of half of 0.07257mm/px, that is, 0.03629mm/px. Taking this worst case scenario to a 2-dimensional coordinate-system, the maximum error would be that same value, multiplied by 1.414 (the square root of 2), which yields a maximum error of 0.1026mm/px diagonally (x and y coordinates)
Can the hand-brain relationship tell the difference in that case? Certainly, 0.1mm is well within the physical limits of perception of human dexterity. However, this is really a worst case scenario, as it is highly impractical to use a mouse at 350 effective dpi without any acceleration, because it is so slow for any general purpose. This is also assuming we want to move the mouse by a single pixel, in most cases this will be irrelevant, as we will want to move the mouse a greater number of pixels, and as you will recall above, the distance d will be significantly greater than the error c.
About your last statement, it is hard to say for sure, but it certainly seems like you are right about that though: there seems to be no advantage to using 1800dpi at 2.5/10 sensitivity versus 450dpi at 10/10 sensitivity. I say it is hard to say for sure because, if I had engineered such a mouse and if it had 4 distinct sensor settings (the sensor behaviour changes physically - I don't know if that is the case with the Razer), I would have tried to allow the internal mouse driver to use the higher dpi settings' higher resolution in order to better resolve the issue of travelling less than an integer number of pixels. However it is highly likely that the mouse's 4 dpi settings are simply sensitivity filters themselves (with respect to the maximum dpi value of 3500), as I assume such a feature is cheaper to implement in software than in hardware.
|
Great app, thank you! First I got hyped, hoping to have some real solving for these mouse problems. Then I found out, Razer has no OSX drivers for the Abyssus. Fuck 'em 
|
On January 20 2012 00:31 3nigmatiq wrote:
[blablablablabla] I would have tried to allow the internal mouse driver to use the higher dpi settings' higher resolution in order to better resolve the issue of travelling less than an integer number of pixels. However it is highly likely that the mouse's 4 dpi settings are simply sensitivity filters themselves (with respect to the maximum dpi value of 3500), as I assume such a feature is cheaper to implement in software than in hardware.
Haha, that just made me try and realize how ridiculously small of an movement I have to make in order to get my 10/10@1800CPI DA to move the cursor one pixel. Fascinating ^^
Well, yeah, it seems more like a theoretical disadvantage that won't really be noticeable in real life - especially for everyone who is not a 10mio APM progamer - I totally agree.
On January 20 2012 01:30 SkyTheUnknown wrote:Great app, thank you! First I got hyped, hoping to have some real solving for these mouse problems. Then I found out, Razer has no OSX drivers for the Abyssus. Fuck 'em
I wouldn't give up that fast. I'm using my DA without the mac drivers, simply because I don't like running stuff that isn't really necessary for me (it's only useful if u switch your mouse settings regularly)
The Deathadder internally saves 1 profile that will become the default setting if u connect it to a machine with no razor driver. So if I configure the DA in Windows to have 1000hz polling, 1800dpi - then this will be exactly the setting that it'll use on a mac.
Now, I'm not 100% sure the Abyssus works exactly the same way, but since they already share the same sensor - if I were u, I would definitely check that option out!
|
This mousefixer is the only thing I've found that can actually get rid of acceleration in Lion, USB overdrive can't seem to do it anymore. Unfortunately I can't use UO to adjust the mouse speed either, when MF is installed, so I'm stuck on the default speed. Luckily it's not too bad.
As for the OSX latency I think it's definitely noticable, it's not the end of the world but you just can't get the same "crisp" feeling as in Windows.
|
Hello, I have a macbook (late 2008) and just received a logitech M510 mouse as gift.
I noticed some problems with it (especially while playing starcraft) so I downloaded this program and tried to run it.
Unfortunately nothing changed.
Is open the program, get this screen, and restart my laptop all I have to do or is there something I am missing?
Please let me know 
EDIT: I tried to run the program again, restart and now the acceleration is gone. Unfortunately often in SC2 the cursor gets stuck on a corner or starts acting weird for few seconds. Is there a way to solve this issue?
EDIT2: I realized the cursor gets stuck and acts weird mostly while i play multiplayer...if i play single player it almost never happens. I don't have any lag at all with the rest of the game though.
|
Help...
I have a MacBook Pro with Mac OS X Version 10.7.3
and a
Razer Mamba Mouse
I tried to install the Fixer and i noticed change but when i tried to play CounterStrike Source the Mouse Cursor won't move.
Now i wanted to unistall everything but i can't find the "com.bumblebees.mousefixer.plist" in my LaunchAgents Folder... Deleting the Application doesnt work ofcourse.
Also checket the activity monitor. But there is no Mousefixer entry too ....
Can't Play CounterStrike anymore since then...
Please help me....
Thanks for any reply!
Regards,
M47 from Germany
|
Anyone else having problems with this in Lion?
(tl;dr of the paragraph below is that I now beach ball on startup whenever I install mousefixer)
I just finally upgraded, and everything went smoothly except that my mouse became skittish like it had been before I installed mousefixer. Tried uninstalling/reinstalling to no effect. Upgraded my deathadder firmware, no effect. Upgraded the deathadder driver, no effect. Came back to this thread and got the updated 2nd version of mouse fixer, uninstalled the old version, installed the new one, and any time I try and install the new one I beach ball infinitely on the restart afterwards. I get to my desktop but it's completely bricked. I can move my mouse around but it's just a beach ball and clicking on things has no effect. After a hard reset mousefixer 2 seems to be installed but the mouse operates as if it's not.
Just wondering if anyone has experienced this or knows what I should do. Up until now mousefixer has been a godsend
|
|
|
|
|
Sorry for bumping this old thread but the tool you provided worked great when I tried it some time ago with my logitech G5. In the meantime i got 10.7 and a G700 (today) and was expecting to work the same way. But after a quick test, nothing changed at all. But i noticed, during the login screen of OS X, the mouse behavior was perfect. But after logging in, it was terrible again... Any ideas?
|
On January 20 2012 01:30 SkyTheUnknown wrote:Great app, thank you! First I got hyped, hoping to have some real solving for these mouse problems. Then I found out, Razer has no OSX drivers for the Abyssus. Fuck 'em
I'm rolling with this bump because I have the same problem as Sky. The mouse fixer does remove acceleration, but I can't adjust mouse speed, even within SC2. 1800 is too high for me (1366*768 monitor) and 450 is too low, so I'm going with acceleration for now.
|
|
|
|
|
|
EPT
