what is the differens between accelpower and accel?

Mouseaccelpower is the "curve" of the acceleration. When set to its default of 2, the increase in sensitivity with a positive mouseaccel value is linear (picture a diagonal line).

With for example power 3, the increase is exponential (picture the positive side of a parabola).

Mouseaccel is how fast you move up and down this "curve".

Some info here: http://www.quakelive.com/forum/showthread.php...post161401

The accelpower is not written on the link on these guys? isn't that a important fact too?

As far as I know there are no pros using a cl_mouseaccelPower different than the default of 2.

Ok. I found this graph --> http://ambro.insane.si/accel.html there is no Accelpower i can configure this graph with. What accelpower "?" does this graph show?

Rapha 0.48 accel. He can not have accelpower 2? Impossible? Try for your self?

why not? i have mouseaccel 1 and its starting to get slow for me

Huge accel to compensate your retardness

u have low sens and u only shaft on jumppads, whats the point? :D

also in water!

didnt mention that cos i thought its obvious :D shafting someone jumping from ra to water on c7 feels so good <3

That graph is outdated, uses the old TTimo cl_mouseaccelstyle 1 stuff that doesn't exist in QL now.

Rapha uses power 2, cl_mouseaccel 0.48, offset 0, senscap 0.

accel number alone says next to nothing thb.
im using 0.005 and thats not a low value for me since with 0.01 i can do tarded 720turns all of the sudden ^^

type in senscap and see what happens

It should be noted that the greater the cpi the greater the effect of mouse acceleration.

Like lorfa explained...

this is how I imagine power should be described.However I do think that its too fast with power 2 aswell, I play with 0.

http://shrani.si/f/3w/fX/LpPUAnm/accpwr.jpg

Anything to the 0 power is 1, so accelpower 1 is the same as just raising your base sensitivity by 1 (if P is 1 then P-1 will evaluate to 0).

So for example these two are identical:

cl_mouseaccel 0.29 ; cl_mouseaccelPower 0 ; sensitivity 3

cl_mouseaccel 0 ; cl_mouseaccelPower 2 ; sensitivity 4

how did u come to such conclusion?Sorry I havent followed injx and the other sens related debates..would like to understand which formulas did u follow?And is it possible to make an exact q3 copy of sens to ql??

edit:since q3 does not know accelpower command.. does this mean that closest to that sensitivity in ql would be if accelpower is set to -1??

how did u come to such conclusion?

There's no negative power, (P-1) is lower bounded at 0.

Negative power wouldn't work anyways as mouse sens would have to begin at infinity (just look at the graph for x^-1 for example).

Perhaps they could add a feature where if senscap > 0 then negative power would be allowed, result would be that your sens would start at the senscap, and then past a certain velocity would take a nose dive. No idea why you'd want that :-P

would like to understand which formulas did u follow?

x^0 = 1 (why?)

B + (A(v-c))^(P-1)

B = Base sens
A = Mouseaccel
v = velocity
c = Offset
P = power

B + (A(v-c))^(0-1)

The 0 - 1 will evaluate to 0 here since it is lower bounded at 0.

B + (A(v-c))^0 =

B + 1

And is it possible to make an exact q3 copy of sens to ql??

Sure.

edit:since q3 does not know accelpower command.. does this mean that closest to that sensitivity in ql would be if accelpower is set to -1??

q3 will be the same as power 2 in QL (which evaluates to ^1).

There's a reason that QL defaults to this value, because it is the same as q3 mouse accel and is what everybody is used to.

Just leave accelpower at 2 if you wish for mouse accel to behave the same as it does in q3.

q3 will be the same as power 2 in QL (which evaluates to ^1).

There's a reason that QL defaults to this value, because it is the same as q3 mouse accel and is what everybody is used to.

Just leave accelpower at 2 if you wish for mouse accel to behave the same as it does in q3.

But it doesent behave like q3 accel.. its just too fast and I can feel it.I havent read anywhere that accelpower 2 would be a default for q3 accel behave, all I have seen until now was SyncErrors explanation on how to set style 0 again and thats it.

SyncError wrote:
The default settings for power=2 and offset=0 give you the traditional acceleration previously known as style 0.

this I believe only applies for ql.. in q3 we do have offset and its set to 5!!

There's no negative power, (P-1) is lower bounded at 0.

type in accelpower -10 and you will notice that it is different!!
go play for a while with this setting and you will see.

edit: Another thing that I noticed is that cg_zoomsensitivity 0/1 affects my base sensitivity and the way accel behaves.. any comments on that!?

and I shouldnt forget about this too..

http://www.quakelive.com/forum/showthread.php...ensitivity

Theres dozens of such threads, Im not the only one and I sure dont imaginate things.

But it doesent behave like q3 accel.. its just too fast and I can feel it.I havent read anywhere that accelpower 2 would be a default for q3 accel behave, all I have seen until now was SyncErrors explanation on how to set style 0 again and thats it.

The q3 clients that I know of default to in_mouse 1 (direct input) which feels different than in_mouse 2 (WM_INPUT) that QL uses by default.

In q3 1.32b source code I couldn't find any mention of anything that could be an accel power. Which tells me it is simply ^1 (not there being the same as *1 or ^1), the same it would be in QL if you set accelpower to 2.

this I believe only applies for ql.. in q3 we do have offset and its set to 5!!

I don't have a mouse accel offset cvar in cnq3 1.43, or q3 1.32b. Which q3 are you using? Ioquake has it, as does Open Arena I believe.

If you use your configs interchangeably between the two games cvars that don't exist will be created with seta but of course will be ineffectual.

type in accelpower -10 and you will notice that it is different!!
go play for a while with this setting and you will see.

I did! I experience no difference between power 1 and power -10 or even -200.

edit: Another thing that I noticed is that cg_zoomsensitivity 0/1 affects my base sensitivity and the way accel behaves.. any comments on that!?

I don't experience this effect, zoomsens only affects my sens when I am zooming as it should.

I tested it with a placebo test. I had mouseaccel set to what I use (not 0), and I bound pgup to zoomsens 1 and pgdn to zoomsens 0.

First I spent some time with each setting to see what difference could be felt if any.

Then I pressed both keys at the same time many times such that I didn't know which key took effect when I stopped. Then I ran around for a bit and tried to guess which one was set.

Out of 15 trials I was wrong 11 times, which tells me that I definitely cannot tell the difference.

I use this client http://www.esreality.com/post/1565939/quake3-...t-support/

with raw mouse input.. when I compare the sens with this client and go to ql.Sensitivities are too different, cant exactly explain why, but they are!!

http://www.quakelive.com/forum/showthread.php...ensitivity

Im not the only one that noticed this.. 1 is slower, like it would have negative accel.

with raw mouse input.. when I compare the sens with this client and go to ql.Sensitivities are too different, cant exactly explain why, but they are!!

Well, that is rather odd! Someone will have to look into the source code to try and come up with a hypothesis.

Have you tested it with mouseaccel 0 and just trying to see if your cm/360 is the same? You can use viewpos to tell you exactly when you've made a 360 turn.

http://www.quakelive.com/forum/showthread.php...ensitivity

Im not the only one that noticed this.. 1 is slower, like it would have negative accel.

This one I just don't believe though. DCR was saying this on the forum and he is quite adamant about it. I think he made some believers.

It really doesn't make any sense that it would do this. I don't think the algo would even exist in memory until you zoom.

Its something wrong with ql accel.As soon as I turn it off it feels better, more q3 alike, but when I turn it on its just odd???It suddenly becomes concave/convex feel, like its totally different angle im shooting from when playing ql and compare to q3!?Its not as smooth, too fast, not really shaky but like im shooting through enemy models?And no its not the netcode its accel!!

and for zoomsens Im not a believer, I found this on my own and stumbled across this thread in the way.Its more noticable with low sensitivity.

edit:
Just try that client.. copy the settings in your q3 cfg and type in in_mouse 3.Im certain you will see the difference!

So last question would be:

a)If I have these following settings in q3

sens 1.6
accel 0.3

would this translate into ql like this?

b)
sensitivity 1.5
cl_mousesenscap 1.5
cl_mouseaccel 0.3
cl_mouseacceloffset 0
cl_mouseaccelpower 2

true or false?

also what does velocity stand for?speed/mouse dpi??

also what does velocity stand for?speed/mouse dpi??

Counts/millisecond.

Unless you have m_cpi set to your cpi and yaw is 0.022, then it is cm/second.

a)If I have these following settings in q3

sens 1.6
accel 0.3

would this translate into ql like this?

b)
sensitivity 1.5
cl_mousesenscap 1.5
cl_mouseaccel 0.3
cl_mouseacceloffset 0
cl_mouseaccelpower 2

true or false?

I would say false.

Offset 0 (haven't had a chance to check your modified q3 1.32b client, but assuming offset doesn't exist in it either) and power 2 are fine.

Sens 1.5 is 0.1 lower than the sens 1.6, I assume because you are experiencing some feel difference between the games. Is your m_yaw and m_pitch the same in both games?

The senscap 1.5 makes no sense to me though. This effectively renders your accel 0, as the sensitivity can never go higher than 1.5

what would be the correct settings then?I dont really understand whats the equation from that formula you pasted??What am I supposed to derive out from that formula sens,accel or what?

you didnt write what I am supposed to calculate out from that formula like sync wrote for example:

new_sens = old_sens * (old_yaw * m_cpi / 2.54)

yes pitch/yaw is the same

It should be exactly as you wrote except sens 1.6 and senscap to 0.

I tested that client out and it feels extremely close to what I have in QL. The thing is that I use senscap in QL and it doesn't exist in the modified 1.32b so I had to endure at least that much difference in mouse feel.

The other thing I was thinking of is that in the "changelog" he writes:

"- in_mouse 3 has now even less lag due to delayed mouse movement registration (with this type of input mouse move registered right before frame is rendered)"

Perhaps QL waits for each frame instead which could make it feel a tiny bit different and might explain what you are experiencing. Without the QL source code for mouse input it is not possible to be sure though :-/

Other than that, with senscap 0, accelpower 2, m_cpi 0, everything should be identical.

man difference is enormous.. especially on crt. :D

anyways, thx for the help guess I will have to figure something out on my own.

what was the command for /cl_mouseaccelpower in quake 3?

I don't think there was any.

wmo 1.1a @ sens 4.25

accel = 0

in_mouse = -1


= I RAPE FACES OFF SERVERS



THE END

high sens rapist? lmao

Low tbh.

so... what mouseaccelloffset does?

cl_MouseAccelOffset

The velocity at which mouse acceleration starts.

Measured in counts/millisecond when m_cpi = 0 which it is by default.

When m_cpi = the cpi of your mouse, it is measured in cm/second.

Negative values are permitted, in which case the accel behaves as if the velocity started at a higher speed, this is silly when cl_MouseAccelPower is 2 though because it is identical to raising the base sensitivity.

To convert an offset velocity to m/s in the case of m_cpi = 0 run the value through this:

[(v * 1000) / (CPI / 2.54)] / 100 = m/s

So for example cl_MouseAccelOffset 5, at 400 CPI is 0.3175 m/s.

If m_cpi is set to your cpi, then offset is already in cm/second, so you can simply divide it by 100 to get m/s.

Using positive values has the effect of lowering the sensitivity at a given velocity when mouseaccel is > 0.