There's a program that allows changing how a game gets it's mouse input IIRC but I forget it's name but hopefully someone will come along with a link to it.

Care to explain in detail what you believe is different between direct input and non-direct input? :)

to me, the -1 feel was always superior. it felt more responsive overall.

I used to believe that years ago when I played a lot but I have no idea what it's like these days. It was as if direct input suffered from x number of frames of mouse input lag whereas with non-direct input there was no lag.

I have reasons for why such lag could be occuring but no concrete evidence to prove it one way or the other or much in the way of motivation to research it but I'm always interested in reading what others think about it as I could be motivated to investigate it further.

i could simply never flick as well with in_mouse 1--it felt as though the mouse just didn't go as far as it should have. i always found myself underflicking.

there was also a noticeable difference between mice. i found that, for me, microsoft mice performed better with -1 and logitechs were better with 1.

Well i'm using a Logi G1 and i find that -1 feels much better, I aim better and even strafe jump better! I Have more control over my mouse. I can't explain it... It's just a feeling ;)

Ahh a program, yeah a link would be great. It'd be shit not to have the same feeling in WSW or Q4 1.3, or whatever i play next.

For a test you could try -1 with vsync enabled (r_swapinterval 1 IIRC but need to also make sure the video drivers don't have it set as always off). Doing so might make it feel like direct input as you'll have a frame of visual lag (or possibly two) with vsync enabled.

It won't be a conclusive test but try it if you have the time to mess around and let me know if the non-direct input and vsync on makes it feel like non-vsync and direct input.

Well i didn't test for long as i've only just stopped tweaking sens

My movement was messed up as soon as i tried 1 with vsync.

It definitely felt different from -1 without vsync. Hmmm maybe it was just mouse sensitivty that changed and not actual feeling...

If someone else who's interested at all be willing to try this out then that'd be great. :p

Thanks for your help Strider btw.

The only program I know of, which is able to change the type of input a game gets, is mforce by anir. But it's only able to "feed" non direct input games with direct input.

I've used that before, but it doesn't work with Q3 for me. However, it works with ioQ3.

Hello,

I think Strider is referring to this program:
http://razerblueprints.net/index.php?option=c...pic=5245.0

Unfortunately, it does just the opposite of what you want.
As a matter of principle, it is possible to achieve your desired effect, but I will have to look into hooking of COM objects first (its interprocess communication could be a problem).

I can't promise anything.


- anir

btw is it possible "XP Acceleration Fix" to work without restart? I am asking coz many LAN centers have pointer precision enabled and I cant open control panel to remove that and when I try to apply the patch and restart everything is reverted to the default.

try nomouseaccel2.reg then :)
www.geocities.com/gougougr2000/mousesettings.zip

Actually I've tried that first, but registry fix also needs a restart to work.

The problem is that microsofts 'mouse settings' in control panel makes the system load the values you change. However the regpatch does no such thing. You can of course log out all users and log in again and the new values should be loaded.

in_mouse -1
Captures the position the Windows cursor and passes that info to the game to change the camera angle. Once that is done for a frame, the cursor is returned to the center of the screen. There is nothing else applied to the data until it gets to the sensitivity, acceleration, m_yaw, and m_pitch part of the code.

Pros: -As responsive as you can get
-It's accurate

Cons: -The cursor can clip the edge of the window and not be able to go further, resulting in negative acceleration. The higher your Q3 resolution and framerate, the less you will run into this problem (800x600 is pretty much safe for any sens, 640x480 might have some clipping with low sens). Higher mouse sampling rates or r_finish should also be used to prevent Q3 from getting 2 or more large mouse samples in one frame.
- Must use CPL mouse fix (or alternative) to eliminate Windows enabling Pointer Prescision automatically for games in WinXP.


in_mouse 1
Uses DirectInput to directly acces the mouse, bypassing the windows cursor. It also applied ~12ms worth of mouse smoothing automatically before reaching the sensitivity and acceleration part of the Q3 code.

Pros: -No negative acceleration
- Works independantly of Windows cursor settings (i.e. sensitivity and Pointer Prescision)

Cons: -Uncontrollable mouse smoothing which adds latency to mouse actions.
- Delay when clicking mouse (click -> wait 12ms -> shoot = bad)



Something nice that could be added, if possible, to CPMA now that it's the CPL's game would be a good and clean DI implementation, a la QuakeWorld and Warsow.

so DI also affects clicking a mouse button?

Something that I forgot to add is that all the crap associated with in_mouse 1 is caused by Q3's implementation of DI, not DI itself. Other games have perfectly fine DI, like WSW or QW.

But yea, in_mouse 1 lags your mouse clicks.

nice, didn't know about q3 fucking up the DI. :)

Dunno where you're getting this from, but there's nothing like that in the ioq3 tree. I haven't checked the original engine source though so maybe io fixed it.

I don't know if that smoothing thing is correct, but -1 is definitely not DI, so it is Window cursor Set/GetPos, and 1 is definitely DI, check the console.

1 definitely adds delay to mouse movements and clicks, while -1 doesn't.

I believe the DI input method was introduced in the Team Arena patch (1.17 or 1.27, dunno), before that only -1 existed, but it was called in_mouse 1 (just cause it was default)

Be cool if we could get a clean DI implementation in Q3

I meant the "~12ms worth of mouse smoothing automatically" part: I know the rest of it is true - I'm the guy who bitched at TTimo mercilessly to ADD the DI support back after 1.2x broke it. :P

Haha. Well, this is where I found it: http://finger.planetquake.com/plan.asp?userid=johnc&id=8153

It's a Carmack .plan from before the game was released, and it details 3 input methods he had lined up to use in Quake 3, and it also explains the function of m_filter.


The way I interpert it is that in_mouse 3, which he said would be default at Q3's release, was eshewed in favor of the in_mouse 1 method (which in the .plan is cursor get/setpos) for whatever reason (maybe time costraints?). After TA was released, a new DI input method became the default input handling method of Q3. You'd think that it would be like the in_mouse 2 method in the .plan, but Q3's current DI input has considerable lag to it, something that wouldn't be found in "raw" DI implementations. That leads me to believe that it could be something like the in_mouse 3 method in the .plan.

It's purely an inference though, but it would explain the "loose" feeling on in_mouse 1

What kind of a moron dev would want to smooth a player's mouse inputs anyways. Whats so fucking hard about using a mouse that you need to have what you're telling it to do averaged out over time.

It was a problem when Windows 95 only polled the mouse at 40hz. Games usually don't smooth inputs by default anymore unless it's a shitty console port.

Like UT3? I play cpma now because of that game being bad so lousy. (Don't ask me why I wasn't playing it sooner, UT2k4 for me was a step up from CS, it's a natural progression.)

Even if something was being polled at 40hz, I can't see how adding another delay could possibly help anything playability wise. Graphically I suppose when you're standing still and decide to look with the mouse smoothing schemes could make it appear smoother than 40fps if your framerate was higher than that.

Also what is the smoothing in console shooters for? I wonder what the analog pads get polled at in ps3/xbox 360? Probably something horrible as they put really little care towards the playing performance in consoles. Still don't see a need for smoothing in console shooters, Bond played fine without it.

Anything other than raw mouse inputs, excluding yaw, pitch, sensitivity, and maybe q3 style acceleration, is rather extraneous and hurts the feeling of the game.. I wish devs realized that.