How can I calculate my CL_mouseaccel?

[lucky7]

hi~~
Now I'm using 400DPI.
sensitivity in game: 4.4
cl_mouseaccel: 0.05
Other commands related to senitivity is default.

I'll change 400DPI to 800 or 1600DPI.(same mouse)
So..
Now 400DPI
sensitivity: 4.4
cl_mouseaccel: 0.05

800DPI
sensitivity: 2.2
cl_mouseaccel: ??

1600DPI
sensitivity: 1.1
cl_mouseaccel: ??

I want to know how to set cl_mouseaccel?
Is there somthing fomular for "cl_mouseaccel"?

[JustLOL]

new_accel = old_accel * ((old_cpi/new_cpi)^2)

(400/800)^2 = (800/1600)^2 = 0,25, so when going from 400 to 800, just divide your accel by 4, and that's it.

400: 0.05
800: 0.0125
1600: 0.003125

For accel offset, it's: new_offset = old_offset * (new_cpi/old_cpi), so if you had cl_mouseAccelOffset 5 at cpi 400, just double it for 800, and double it once more for 1600.

[lucky7]

JustLOL~ Thanks for your help.
For accel offset
400 CPI: 5
800 CPI: 10
1600 CPI: 20
Is that right?

Now I'm using sensitivity 4.4, cl_mouseaccel 0.05,400 CPI, but other commands are below.
CL_mouseaccelDebug 0
CL_mouseacceloffset 0
cl_mouseaccelpower 2
cl_mousesenscap 0

all default setting.
I do not know above commands yet.
Should I know those commands to play better than now?

[Lorfa]

For accel offset
400 CPI: 5
800 CPI: 10
1600 CPI: 20
Is that right?

Yes.

Assuming you have m_cpi set to 0 which it is by default.

other commands are below.
CL_mouseaccelDebug 0
CL_mouseacceloffset 0
cl_mouseaccelpower 2
cl_mousesenscap 0

all default setting.
I do not know above commands yet.
Should I know those commands to play better than now?

They aren't needed.

Debug makes a log file with mouse information, so no need for that.

Offset you were just asking about, so you must know something about it :-)
Positive values just say "don't start mouse accel until the mouse is moving at this velocity".

MouseAccelPower is the exponent that sets how the sensitivity curves up as velocity increases. It is subtracted by 1 in the code, so the default is a linear progression (straight line). The vast majority of players just leave this at default, and certainly there is nothing wrong with the default value of 2.

MouseSensCap just stops your sensitivity from increasing past a certain point. So if you find that you like the way your current mouse accel feels at regular velocities, but during fast turns it seems to sky rocket way too much, you could try setting this value to prevent it from escalating past a certain point.

So for example cl_MouseSensCap 7 will prevent mouseaccel from increasing the sensitivity past sensitivity 7. At the fairly low value of mouse acceleration you are using I highly doubt you'll have any use for cl_MouseSensCap.

[jigglyboobs]

I'd actually recommend retuning both sense and mouse accel. Neither scales perfectly with dpi. All one can say is that 800 is about 2x more sensitive than 400, etc.

[Lorfa]

I'd actually recommend retuning both sense and mouse accel. Neither scales perfectly with dpi. All one can say is that 800 is about 2x more sensitive than 400, etc.

I find this misleading.

As far as the game code is concerned it is just reading counts, and performs operations on them that lead to consistent results. The only rounding that's done is to an absolutely superfluous precision. If the counts are correct you can be sure that the scaling is correct.

We cannot be absolutely certain of the CPI as this would require a lot of testing at various speeds. However there isn't really any reason to believe they are "way off" or anything.

What is undeniable though is that a change in CPI changes the way the mouse feels. So that even when everything is scaled perfectly, it is of course not going to feel exactly the same. The coherence has changed, as has the way the mouse handles quick changes in velocity.

[jigglyboobs]

I find this misleading.

As far as the game code is concerned it is just reading counts, and performs operations on them that lead to consistent results. The only rounding that's done is to an absolutely superfluous precision. If the counts are correct you can be sure that the scaling is correct.

We cannot be absolutely certain of the CPI as this would require a lot of testing at various speeds. However there isn't really any reason to believe they are "way off" or anything.

What is undeniable though is that a change in CPI changes the way the mouse feels. So that even when everything is scaled perfectly, it is of course not going to feel exactly the same. The coherence has changed, as has the way the mouse handles quick changes in velocity.

Exactly. It's not the game code. It's the myriad of factors sitting between your mouse, the pc, the drivers, windows, and finally your game that could make scaling noticeably different than what you expect to feel.

My recommendation again, is to recalibrate around the extrapolated values. Make sure to turn on all the optimizations you want. Make sure you're using a consistent gaming posture you're comfy with. Make sure you know exactly what the in_mouse is set to, etc.

Since the OP was still using 400 dpi I assume he is new. I recommend the following calibration procedure:

Set

in_mouse "2"
in_restart

take a note of m_pitch and m_yaw

m_yaw "0.022" is the default value. What ever it is, keep your existing value.
m_pitch "-.022" or "0.022" is the default depending on whether you invert your mouse. You might want to change this one, but first just write it down.

Next, turn OFF cl_mouseaccel
seta cl_mouseaccel "0"

Using your scaled sensitivity, go to an empty server and start a map, eg, dm6
Now, take your crosshair and point at the wall.
Adjust your sensitivity until you can, without even thinking about it, snap your crosshair to any point within a 10 degree angle around your cross hair.
Do this first for the horizontal direction. Then change m_pitch in .001 increments until you can do the same comfortably in the vertical direction as well. Then go back and jiggle your horizontal sensitivity again.
Do this well and you should be able to trace the outlines of the doorways on DM6 with your crosshair without too much trouble. You should also be able to do things like rocket jump. For me, I have to pick a m_pitch that's a compromise between accurate aim and movement.

OK, now, you are ready to tune your cl_mouseaccel. Set your cl_mouseaccel so that you can snap your crosshair to 90 degrees from center, consistently, again and again.

Your new settings should feel comfy and natural, both aiming and moving around the map. You should be able to simultaneously have the ability to make minute adjustments for railing from far away, and also be able to move around easily and fight close in.

For me, I use sens 2.58, m_pitch -.034 and cl_mouseaccel .127 on 800 dpi on in_mouse 2.

[Lorfa]

Exactly. It's not the game code. It's the myriad of factors sitting between your mouse, the pc, the drivers, windows, and finally your game that could make scaling noticeably different than what you expect to feel.

Ok, but at the very least, most of these would be consistent across CPI values since same mouse/pc/OS/pad/drivers.

This is assuming that the pc can change the sens somehow, the only way I can think of is malfunctioning usb ports.

in_mouse 2 does a good job of eliminating driver influence imo.

Since the OP was still using 400 dpi I assume he is new.

??

[jigglyboobs]

Ok, but at the very least, most of these would be consistent across CPI values since same mouse/pc/OS/pad/drivers.

This is assuming that the pc can change the sens somehow, the only way I can think of is malfunctioning usb ports.

in_mouse 2 does a good job of eliminating driver influence imo.

Think of the shear complexity in running processes and the diversity of devices your PC have to deal with. It's not at all obvious that USB devices would respond consistently from PC to PC, or even with different setting changes. They are not going to be drastically different, but they are not going to be perfectly consistent either.

In fact, there's a whole science to getting a complex device like a PC to act consistently every time, and operating systems specifically designed for that purpose. Windows is not one of them:

http://en.wikipedia.org/wiki/Real-time_operating_system

Every second, your USB mouse can send up to 500 packets to your PC (but some times it doesn't) while sampling the mousing surface at an even higher rate. All of those little inputs leads to "jitters", little tiny errors that can build up to make your movement behave in a way that you don't expect over the course of a fight. Good enough for casual gaming but annoying if you are trying to push your aim to the next level. Best thing to do is to recalibrate.

??

I assume most experienced players would have switched to higher dpi already, but actually lucky7 seem pretty good already, so what do I know.

[Lorfa]

Think of the shear complexity in running processes and the diversity of devices your PC have to deal with. It's not at all obvious that USB devices would respond consistently from PC to PC, or even with different setting changes. They are not going to be drastically different, but they are not going to be perfectly consistent either.

I think this is unnecessary FUD (Fear, uncertainty, doubt). USB is a specification, one which must be followed by hardware manufacturers if they wish to use the USB name. It has also been around a long time, so that any issues that would have cropped up would have been ironed out by now.

It is infinitely more likely that you'd have problems coming from device drivers, but even there in the vast majority of cases you wouldn't have a problem for something as ubiquitous as usb.

There was a guy having trouble adjusting the polling rate for his mouse, but he was on one of those special acer laptops probably with a proprietary motherboard which I think was fine, but perhaps the device drivers on it were messed. That was the only case I've ever heard of that happening though.

In fact, there's a whole science to getting a complex device like a PC to act consistently every time, and operating systems specifically designed for that purpose. Windows is not one of them:

http://en.wikipedia.org/wiki/Real-time_operating_system

Yes I know what an RTOS is. We don't need an RTOS to get consistent device behavior though!

Every second, your USB mouse can send up to 500 packets to your PC (but some times it doesn't) while sampling the mousing surface at an even higher rate. All of those little inputs leads to "jitters", little tiny errors that can build up to make your movement behave in a way that you don't expect over the course of a fight. Good enough for casual gaming but annoying if you are trying to push your aim to the next level. Best thing to do is to recalibrate.

Each packet can contain more than one mickey though eh? Any errors would occur and be fixed at the 2 ms polling rate if set to 500 Hz. The errors would be more likely to come from the sensor than the usb. Maybe it would produce errors if you had some ultra strong fields sitting on the cord somewhere but that would be very rare.

I assume most experienced players would have switched to higher dpi already, but actually lucky7 seem pretty good already, so what do I know.

See this post:

http://www.quakelive.com/forum/showt...l=1#post230056

You'll see that 400/450 is practically the most common setting amongst top players!