You have run an XP mouse fix before yes? Also you have enhance pointer precision unchecked right?

If so do this:

set setting 6/11

Open up a paint program where it shows you the coordinates of a picture. Make sure the picture is larger than the screen. Measure a specific distance between two points on your pad, like say exactly 10 cm.

Move the mouse exactly 10 cm and have the paint program tell you the starting and ending points.

6/11 is 1:1, that is one pixel / count, so we should be able to get an idea of how many CPI the ms ball mouse uses. I'll put in a guess of 250 cpi.

Once we have that we should be able to calculate the "accel" from 8/11 sens.

or he could tell how much cm he needs to do a 360 turn while playing CS:S

qr: do these steps:

1) go to cs:s, load any map and find a starting point from where you will do a 360 degree turn and put your mouse to the edge of your mousepad, so it will be easier to measure the distance needed to do a 360 turn
2) once you found it, turn slowly 360 until you point at the starting point again
3) measure the distance that your mouse traveled
4) post results

Isn't windows accel disabled in quake live?

only with in_mouse 2 i think

Or in_mouse 1 which is just as direct, but handled by a separate thread.

Only in_mouse -1 uses cursor polling and is affected by Windows accel and MarkC fix.

6/11 Adobe 10cm was ~1.65cm

CS:S 360 going slow, 9 sensitivity, 1.1 acceleration was ~11.4cm

eduguy, CPMA/Defrag

Thanks dudes.

What do you mean by 10 cm was 1.65 cm?

I opened up a window in adobe that was 10cm wide according to the Adobe ruler. My cursor crossed the 10cm window while my mouse traveled 1.65 on my pad.

If it means anything, 1024x768 res, 85hz, and in CS:S I use raw input.

..but that.. isn't what I asked..

I need # of pixels per distance on pad with 6/11. That's why I suggested a paint program, as some of them tell you what pixel coordinates you're on in the image. You could then move an exact distance on the mousepad, 10 cm just being an example of a reasonable length to measure.

Sorry about that.

5cm on my pad was 790px in photoshop. I dragged a selection and cropped the image.

This is with 6/11 in windows right?

790 / 5 is 158 counts/cm, which would be 401.32 cpi.

Probably actually 400. Interesting that the ms ball mouse could be 400 cpi and their optical would be too.

Assuming it is 400 cpi, what is exactly that you were trying to do now? It's actually not clear from your post.

"Now that I'm craving to go back and play some Quake/TFC, I'd like to match my new sensitivity in XP itself. "

I'm not sure how to interpret that.

that's true for logitech ball mice too, i didn't have to change my sens going from ball mouse to optical. at the time i figured they did this on purpose to help people migrating to a new tech :)

Yes very cool.

I just never trust mouse companies to do something sensible, so when they do I'm like woah :-P

Acceleration is disgustingly inconsistent and so is a high sens of 11 cm for 360. 20-30 with no accel is bestest. And yeah, I am late to the party.

11cm is extremely low compared to what I used for years.

Yes, 6/11 in XP.

In CS:S I can alter the mouse accel using a slider or simply typing in a number.

I want to take my CS:S sensitivity settings and convert it to my XP settings so that I can use this sensitivity that I've gotten used to on older games that don't have this feature.

What values do you use here:

http://www.esreality.com/index.php?a=post&...pid1903738

Should be able to like, extrapolate back to XP so a windows accel custom could be recreated if I knew what they would look like in QL.

m_yaw 0.022
m_pitch 0.022
m_customaccel 3
m_customaccel_exponent 1.1
m_customaccel_max 0
m_customaccel_scale 0.04

What fps do you use in CS:S?

Unfortunately I've had it capped at 300, but my computer rarely gets a steady 300. I haven't necessarily noticed a difference when i get better or worse FPS, but it would be safe to assume that I'm usually around 200.

Does m_customaccel_scale have any effect for you?

Seems like it shouldn't have any effect when m_customaccel is 3

It appears to have no effect. Set at 1, .5, .01 all feels the same as .04.

I appreciate the help thus-far, but I'd really like to play those classic games :)

I found another potentially very useful thread here:

http://85.17.208.103/post/2173972/re-sensitiv...pid2173972

Still I'm stuck trying to figure out what this means (what is pow?):

pow(raw_mouse_delta, m_customaccel_exponent - 1) * sensitivity

Supposedly this is the explanation for m_customaccel 3, as you can see m_customaccel_scale is not considered.

It might mean:

mousesensitivity = ( rawmousedelta^m_customaccel_exponent - 1) * sensitivity.

I'm not sure if it is fps dependent, since I don't have CS is there a chance you could test say half the FPS and see if it affects the sensitivity dramatically?

If not then it might be something like this in QL:

sensitivity 0
m_yaw 0.022
m_pitch 0.022
cl_mouseAccelPower 1.1
cl_mouseAccel 46 // Where does he get this number?
cl_mouseAccelOffset 0
cl_mouseSensCap 0
m_cpi 0

If that's the case it would look like this in QL:

0 + (46(x-0))^(1.1-1) =
0 + (46(x))^0.1

Where x is counts/ms.

This results in a rather flat curve though, sens would go from say 1 to 2 in 0.1 m/s (at 450 cpi), then 1 m/s only makes sens ~2.5.

Perhaps accel was an arbitrarily high value here to generate a reasonable sens value. Increasing it only slowly shifts the curve upwards in sens.

If this can be made to be correct, the values could be plugged into markC to generate a comparable windows curve.

i can confirm that m_customaccel 3 is fps dependent.
you are also right about pow, which refers to the c math function where the parameters are base and exponent respectively.

i can confirm that m_customaccel 3 is fps dependent.

Great :-(

I wonder where to figure it into the equations, and also why injx didn't mention it in that thread.

you are also right about pow, which refers to the c math function where the parameters are base and exponent respectively.

Ohhhhhhhhhhh *smacks head*

That seems very obvious now that I look at it. Sigh I am so clueless sometimes.

Thanks for the clarification.

This isn't making a lot of sense to me, so I spent the last good bit of time doing some trial and error and I've got something I'm alright with.

The only annoying part is that when I go into steam overlay or the choose class menus in TFC, my cursor moves VERY slow :P