Removing Mouse Acceleration OSX

R C-R wrote:

A width of 1 what? One screen? How do you divide that into distance units?

static_cast<float>(1) or static_cast<double>(1)

Every "count" from a mouse is "equal" - http://computer.howstuffworks.com/mouse2.htm (ball mouse, same principle has carried on to optical / laser) - paragraph 3.

foxx1337 wrote:

static_cast<float>(1) or static_cast<double>(1)

That gives you a pair of raw numbers. It does not explain how they get converted into distance (or velocity) units.

By the way, reminds me of the presentation for Lion this year when the Apple guy hooked a mouse into the computer and kept missing the buttons on the screen. It's awesome as it is!

And to continue your point, it's "due to the vectorial nature of the UI". Uh-uh.

That would have nothing to do with it being a new UI scheme and how some core principals have been changed with Lion, and it has everything to do with the fact that the mouse acceleration has remained the same across all versions of OSX and is not likely to change.

My only advice to you is learn how to move your cursor better.

foxx1337 wrote:

By the way, reminds me of the presentation for Lion this year when the Apple guy hooked a mouse into the computer and kept missing the buttons on the screen. It's awesome as it is!

And to continue your point, it's "due to the vectorial nature of the UI". Uh-uh.

Can you point out at what time that was? I watched it again and didn't see anything that looked like what you describe. I figured it would be incredibly obvious, so I wasn't paying much attention. Or was it another Lion presentation other than the keynote?

I did see a guy trying to manipulate a computer while talking to an audience and trying to keep his eyes on both.

There was one point that might have been interpretted as missing the target, but he was trying to use the two finger zoom out and it didn't register two fingers, so it moved the cursor instead of zooming.

It's intersting to note that some long time OS X users think this long standing issue with the pointer physicis is one of people being ingrained to the Windows way vs OS X way. Mac OS X used the same algorithm for pointer physics as Windows from version 10.0 to 10.3. The new pointer physics model that so many are having issues with was introduced in Mac OS X 10.4. Many users didn't notice and others were able to adapt quickly. Some users, such as myself however, are simply unable to adjust to it. None of the third party tools and hacks are able to provide a pointer physics model that is identical to pre OS X 10.4.

Apple needs to provide better support for generic USB HID devices (many third party mice jitter on OS X (also, if you've ever used the Hid Manager API...)) and an option box in the mouse prefrence applet to switch pointer models. This simple issue has probably cost them millions in sales and needlessly aggrivated untold numbers of users.

koft wrote:

Mac OS X used the same algorithm for pointer physics as Windows from version 10.0 to 10.3.

What exactly do you mean by that? Windows hasn't even used the same pointer ballistics algorithms in all its many versions, nor have OS X & Windows ever used the same transfer functions. They never have even used the same method to draw the pointer or other graphics elements. About the only thing they have in common is support for variable precision at different mouse velocities, commonly called acceleration.

Apple needs to provide better support for generic USB HID devices...

Again, what specifically do you mean by "better"? Apple complies with the industry-wide USB.org standards & is a member of USB-IF. Those standards specify among other things how much support for USB HID devices should be built into generic HID class drivers provided with an OS & how much should be provided by device-specific drivers provided by the device maker. Are you suggesting Apple should ship USB drivers that support something over & above what complies with those standards, or something else?

Don't you dare talk uphill to the Apple fans up here. They've been using Apple mice for 80+ years. Plus, your point is invalid.

It's all "ballistics" and "vectorial", it's up on a pedestal and there are other terms for it that haven't even been coined yet, ok?

The company has only been in existence for 35 years and mice have barely been around for perhaps 25 of that ... so where is this 80+ years of Apple Mice ?????

....Couldn't resist.

koft wrote:

It's intersting to note that some long time OS X users think this long standing issue with the pointer physicis is one of people being ingrained to the Windows way vs OS X way. Mac OS X used the same algorithm for pointer physics as Windows from version 10.0 to 10.3. The new pointer physics model that so many are having issues with was introduced in Mac OS X 10.4. Many users didn't notice and others were able to adapt quickly. Some users, such as myself however, are simply unable to adjust to it. None of the third party tools and hacks are able to provide a pointer physics model that is identical to pre OS X 10.4.

Apple needs to provide better support for generic USB HID devices (many third party mice jitter on OS X (also, if you've ever used the Hid Manager API...)) and an option box in the mouse prefrence applet to switch pointer models. This simple issue has probably cost them millions in sales and needlessly aggrivated untold numbers of users.

if it helps, you could email me a shipping address and I could send you my 10.0-10.3 discs.

foxx1337 wrote:

It's all "ballistics" and "vectorial", it's up on a pedestal and there are other terms for it that haven't even been coined yet, ok?

Whatever you want to call it, two OS's with such fundamental "under the hood" differences never will have exactly the same pointer characteristics.

If you just want to gripe about that pointlessly, it's your choice, but if you are interested in what Apple feasibly could change to make it more to your liking it is hard to see how using vague, overly simplistic terminology benefits anyone.

God, i hoped they would remove mouse acceleration or at least let me decide to remove it in the settings panel but no luck with that. Even after 2 years now using a mac, taking a mouse on a windows machine in my hand feels SO much better.

R C-R wrote:

koft wrote:

Mac OS X used the same algorithm for pointer physics as Windows from version 10.0 to 10.3.

What exactly do you mean by that? Windows hasn't even used the same pointer ballistics algorithms in all its many versions, nor have OS X & Windows ever used the same transfer functions. They never have even used the same method to draw the pointer or other graphics elements. About the only thing they have in common is support for variable precision at different mouse velocities, commonly called acceleration.

The graphics calls and other platform specific implementation detials are irrelevant to this subject. They have no bearing on the concept of receiving an event from a mouse, converting it to a screen coordinate and then updating a sprite. The formulas used on Windows never resulted in hordes of users complaining that fine hand movements are like dragging the cursor through mud and then when speeding it up results in the cursor flying off the screen. It didn't work that way on the NeXT machines and it didn't work that way on the Mac until OS X 10.4 was rolled out. Put a mouse in any users hand infront of a Windows 3.1, Windows 95, Windows 2000, XP or Windows 7 machine and then have them compare the movements on any Mac OS X machine running 10.0 to 10.3 and they won't notice the difference. The algorithm converting the mouse movements to screen coords may not be *identical* but they'll be so close you can't tell. I don't see the point of getting pendantic about this, the way the mouse movements are translated to screen coordinates in OS X 10.4 and greater is drastically different than it was in previous incarnations and it's causing lots of users significant irritation.

R C-R wrote:

Apple needs to provide better support for generic USB HID devices...

Again, what specifically do you mean by "better"? Apple complies with the industry-wide USB.org standards & is a member of USB-IF. Those standards specify among other things how much support for USB HID devices should be built into generic HID class drivers provided with an OS & how much should be provided by device-specific drivers provided by the device maker. Are you suggesting Apple should ship USB drivers that support something over & above what complies with those standards, or something else?

Apple has put the bare minimum into supporting HID devices in their OS. This is why so many non-Apple branded mice using the generic HID driver skip when moving the cursor. It's so bad you can't draw a straight diagonal with on a large number of USB mice. This is especially frustrating for gamers playing FPS titles on the mac where using Apple branded mice are unsuitable.

Also, have you ever worked on a project where you needed to support steering wheels, game pads and joysticks? Well, if you're writing a Mac app you'll have to familiarize yourself with the inner workings of the USB protocol and slog through the HID Manager API. You don't have to do this on Windows, Linux or even Classic Mac OS. It blows so hard most people just call it a day and simply support one controller, the xbox 360 controller. Apple should add a game controller class to Cocoa and roll out a more robust USB HID mouse driver.

R C-R wrote:

Whatever you want to call it, two OS's with such fundamental "under the hood" differences never will have exactly the same pointer characteristics.

That's an absurd statement that's patently wrong. There are several thrid party USB mouse drivers for OS X that use a different algorithm for updating the pointer on the screen. The problem with all of them is that they can't take advantage of the proprietary features of Apple branded mice. You can "fix" the cursor problem but you'll loose all the fancy touch and scrolling features, not to mention the skipping issues with some mice.

There is nothing fundamentally different about how OS X is architected that ***** it to have suboptimal mouse support. The folks at Apple who wrote the mouse driver can make it do whatever they want. Here's to hoping they'll add a simple option in the mouse prefs to switch the pointer model for all those out there who can't adapt to the current one.

As a business owner who makes most of his money writing software for the Mac, I cringe every time I hear somebody complaing about the mouse. There are people who won't buy a mac because of this simple issue, and that's sad not to mention pathetic (on Apple's part).