1, use in_mouse 2, thats pretty much bypasses windows.

2, I think the two OS got the exact same sens stuffz, since all the w2k drivers will do on XP as well.

actually that's not true, xp sens is different from 2k. 2k was awesome :(

damn. was hoping someone would tell me it was same :(

yea, perhaps you right, but I remember going xp from w2k and didn't notice anything (I think I had an IMO that time)

i do that, but i want to know what my (same) sens was in windows 2000 compared to windows XP now, in in_mouse 1. (actually now i use different cause am not sure if it was same)

i used default drivers and everything (middle notch), no accel in OS including mouse fix for XP.

dunno sorry, I found in_mouse 1 awfull for me, used -1 before, and I use 2 now

in_mouse doesnt matter its just that q3 had no in_mouse 2 and i can probably use 2 for any sens i used on 1 (not -1, i think)..

and in mouse 1 awful?? I played q3 recently and it felt good...

iirc, "in_mouse 1" is directx (directinput), so...... :)

there isn't

I tried many stuff, I could never achieve it

2k medium accel was nice :/

win2k had accel? i dont think it did.. thought u had to turn it on in drivers.

yup, I was so used to the low one

but not automatic or hidden right?

test

just in the regular mouse settings

It was not the same for sure.

Here is win2k mousesettings from the old days :>
Though I have not used it for a few years I still have it !

Windows Registry Editor Version 5.00

[HKEY_CURRENT_USER\Control Panel\Mouse]
"SmoothMouseXCurve"=hex:00,00,00,00,00,00,00,00,00,a0,00,00,00,00,00,00,00,40,\
01,00,00,00,00,00,00,80,02,00,00,00,00,00,00,00,05,00,00,00,00,00
"SmoothMouseYCurve"=hex:00,00,00,00,00,00,00,00,66,a6,02,00,00,00,00,00,cd,4c,\
05,00,00,00,00,00,a0,99,0a,00,00,00,00,00,38,33,15,00,00,00,00,00

whats that??

Are you saying i can put that in my winXP registry, and it will make the mouse the same as win2k??

Also that is the same as the values in the Registry for my WinXP currently (again i've used the XP mouseaccel regfix - i think i remember it puts those values in)

Yes, thoose settings should make your mousefeel like win2k.

Don't forget to back your own up first if the feel is to weird to go back to :)

hmm my windows XP already had those settings. I think those are applied when you do the standard XP Mouseaccel registry fix/CPL mouse fix etc.. I remember if you wanted to do the fix manually, these were the settings they always said to change them to.

But does this exactly replicate the way the mouse moves in win2k as well, or just gets rid of the lame XP accel? Are you sure these are the values u got from win2k?

In that case, the CPL mousefix is an export of a clean win2k ;)
I never knew that, only had my own export.

that's just the accel part, iirc the sens was different in 2k and it had nothing to do with mouseaccel.

That curve is the old CPL Mouse Fix.
It is not the same as W2K.
The correct Cheese Mouse Fix for your in-game monitor refresh rate would the same as W2K, IF you used pointer speed slider 6/11 on both W2K and now on XP, otherwise a different curve would be needed.

Some parts of W2K mouse sensitivity are the same as XP/Vista/7, some are different.
[Edit, I was talking crap: Using in_mouse 1 or 2 (or 3?) can have a different sensitivity than in_mouse -1, because in_mouse -1 goes thru the Control Panel and applies the pointer speed slider (sensitivity).
(in_mouse 1 or 2 is sort-of like Pointer speed slider = 6/11 and Acceleration = "None".)]

The internal calculations that W2K and XP+ uses are known, so it should be possible to approximate W2K mouse sensitivity on XP+ (in many cases).
(I'm talking about MEASURABLE mouse sensitivity, like inches-per-360, or inches-per-desktop-pointer-pixel.
If the difference between W2K and XP/Vista/7 is described using words like "floaty", "laggy", "I can't put my finger on it", then I can't help!)

W2K and XP+ mouse sensitivity are exactly the same WHEN accel is turned OFF.
(That means the W2K Control Panel Acceleration set to the left-most "None" option, and the XP "Enhance pointer precision" option OFF.)

It gets more complicated if you have accel ON in either W2K or XP+, because the accel calculations are completely different.

Also complicated is what happens for those games that need a mouse fix.
When such a game runs on W2K, no accel is added, but mouse sensitivity is doubled.
When such a game runs on XP+ (with a mouse fix in place), no accel is added, but the mouse sensitivity stays the same.
(My recommendation would be to double the in-game sensitivity to get back to what you had on W2K.)

Did Quake 3 need a fix on XP? (I think so, and W2K sensitivity would be doubled).
Does QL need a fix? (No, in_mouse 1 or 2 fixes that).

When Accel is enabled in the W2K Control Panel, the accel curves look like so:
http://urawa.cool.ne.jp/tb-works/Win98SE_G_TS.gif
(1="Off", 2="Low", 3="Medium", 4="High")
The pointer speed slider works the same way it did with accel off.

When accel is enabled in the XP+ control panel, the accel curve looks like the pictures here:
http://msdn.microsoft.com/en-us/windows/hardware/gg463319.aspx
The pointer speed slider works completely different than when accel is OFF. IF you have the correct Cheese Mouse Fix for your in-game monitor refresh rate, and use pointer speed slider position 6/11, then that is the same as 6/11 with accel OFF, but if you don't use 6/11, or don't use the correct Cheese fix (CPL for example), then the scaling/sensitivity is different.

What Pointer speed slider setting did you use on W2L/Quake 3?
What Pointer speed slider setting do you use on XP/QL? (and what in-game monitor refresh rate do you have?)...

(Sorry, more questions than answers...)

hey, this is great, very interesting information, u seem like u might know this. But a few things:

Using in_mouse 1 or 2 (or 3?) can have a different sensitivity than in_mouse -1, because in_mouse -1 goes thru the Control Panel and applies the pointer speed slider (sensitivity).
(in_mouse 1 or 2 is sort-of like Pointer speed slider = 6/11 and Acceleration = "None".)

in_mouse 1 in q3, at least in XP, was always affected by what the Pointer speed slider was in control panel (ie. not automatic 6/11 in-game).. if i had the slider @ 8/11 my sensi in quake3 was wayy higher.

W2K and XP+ mouse sensitivity are exactly the same WHEN accel is turned OFF.

Wow, i don't believe it. I used 8 then, and i dont think it seems to feel the same in XP, and zealot also attests to it being different than XP. Ofc it was a while ago, and like i said i cant be sure..

in W2K i used 6/11, no accel, in_mouse 1 and refreshrate@ 120. SEnsi 8 in q3.

XP, always use 6/11, no accel, in_mouse 1 and refresh rate 120...

Basically same mouse.

Also complicated is what happens for those games that need a mouse fix.
When such a game runs on W2K, no accel is added, but mouse sensitivity is doubled.
When such a game runs on XP+ (with a mouse fix in place), no accel is added, but the mouse sensitivity stays the same.

I dont understand how does this correlate to what you wrote in the beginning of your post? But i can say that for sure, it is completely impossible that the sensitivity i used in Win2k is double in XP.


Do you know if the mouse ever changed in Win2k depending on the update for the OS? Ofc i cant remember what version of Win2k i was on, but im thinking of my settings in summer 2002, was when i used it.

"in_mouse 1 in q3, at least in XP, was always affected by what the Pointer speed slider was in control panel"

Yes, you are right.
Sorry, I assumed that since it was DirectInput, it bypassed the control panel scaling. It DOES bypass the control panel scaling and accel, but Q3 must re-apply the pointer speed slider scaling itself, which is a good idea, otherwise your sens would change between in_mouse -1 and in_mouse 1 if you didn't use 6/11...

"...i dont think it seems to feel the same in XP, and zealot also attests to it being different than XP"

This I am sure of. I've measured W2K an XP with MouseMovementRecorder, and there is no difference in sensitivity. I've even peeked inside the executable that scales mouse input, and the code (x86) does the same calculations.

LAG (between mouse input and a game responding) might be different, and accel is different, but sensitivity is the same WHEN accel is off ("None" in W2K).

"...what happens for those games that need a mouse fix."

The doubling on W2K is because of how those old games try and turn accel off. They call a Windows function that turns the W2K control panel Acceleration to "Low" ("Low" normally=Double the pointer speed when the mouse moves faster than 7), but also munge the acceleration threshold from 7 to 0 so that "Low" accel kicks in for every movement.
(See listing 4 at the end of this link: http://drdobbs.com/database/184410376#l4
Where it says : "...newmouseparms[3] = {0, 0, 1};", that sets the accel to Low and munges the thresholds to 0.
Unfortunately, W2K "Low" means "Enhance pointer precision" in XP, and so the accel bug was born when people started gaming on XP.

Does/did Quake 3 do that and double W2K sensitivity? I don't know. If Quake 3 needs a fix on XP to stop accel, then it did and does.
Does/did QL do that and double W2K sensitivity? I don't know. If QL needs a fix on XP to stop accel, then it did and does.

I don't know for sure, but I don't think W2K mouse changed from SP to SP or update.

"otherwise your sens would change between in_mouse -1 and in_mouse 1 if you didn't use 6/11..."

Just curious what exactly you meant by this? In q3, sensitivity in in_mouse 1 and -1 are different with 6/11



thx for the response and research. The stuff on the sensi in Win2K and WinXP being the same with no accel is a little amazing to me.

"LAG (between mouse input and a game responding) might be different"

What differences are there between the two OS about this? If there is any difference, is there any possible way to maybe replicate what it is in Win2K in XP? Like in the registry?


I'm a little lost at your WIn2k accel explanation, and how that transferred to XP mouse being horrible, but again i know both QL (maybe not in_mouse 2, but still think so) and Q3 need an accel fix on XP. Maybe the sens values are the same, or i can also see 8 being slightly faster than 8 is in XP, but i cant see it being DOUBLE in XP.. positive

"In q3, sensitivity in in_mouse 1 and -1 are different with 6/11."

That is strange.
6/11 passes mouse movement straight thru, sensitivity = × 1.0
1 mouse movement moves the pointer 1 desktop-pixel.
Direct Input passes mouse straight thru, sensitivity = × 1.0
In what way are they different in q3? What is the scaling factor difference?

Lag might be different

I haven't seen anthing that cause lag in either OS, but I've only studied the Control Panel sensitivity and accel for each OS, and how some games (original Quake and Half-Life engined games) handle mouse input once they get it.
Other parts of the OSs somewhere in the path between USB hardware mouse input and finally sending movement to a game may cause lag, or may not, I haven't looked at that.

If you still have Q3 to run, there is a test to 100% find out how it handles mouse input... It is a little involved, but I can tell you how if you have Q3 and are keen to try.
Part 1 is follow the instructions starting with "Does my game need a mouse fix?" from my fix link:
http://www.esreality.com/?a=post&id=1846538
If the EnPtPr column footer goes red, then the game PROBABLY needs a fix because it uses the "Low"/munge W2K fix. Further testing is needed to be 100%
If the EnPtPr column footer stays black, the game doesn't use the "Low"/munge W2K fix.

yes am doing it right now

You wanted me to do this test with in_mouse -1? How can i do it when i already have the CPL mousefix? Don't we already know that quake 3 needs a fix?

CPL mouse fix changes WHAT happens when EnPtPr (EPP)goes red.
CPL doesn't stop EnPrPt from going red.
If it goes red then probably a fix is needed. At that point you can have CPL loaded or not, it doesn't matter for what the test shows.

Test both in_mouse -1 and 1 if you can. (kill off MouseMovementRecorder.exe and run a new one for each test.)

yeah, i got caught up in reading the whole post and stuff. Lotsa info i wasnt read up on. Glad people like you are around to help everyone out with things like this that are way more complicated than they should be (fu m$).

i hope im doing it right, but never did Mouse Recorder go red, it stayed black and always read off. I ran the Recorder and then ran quake 3, went to OSP mod in the Menu, and opened up a map (dm6). My cfg there has in_mouse 1, i window and look at Recorder and it was black. Shut everything down and ran Recorder and started quake 3 same way, put in_mouse -1 + in_restart and moved around then windowed again to check. Was still black and off. To double check, i shut down Recorder while q3 was windowed and in_restarted in_mouse 1/-1 each time before running Recorder, then fullscreening before going back to check again.

So what do you think and whats next?

'EnPtPr' staying black means 100% that Q3 does not need a mouse fix.

And that means it does not use the W2K Accel = "Low"/munge method of turning accel "off" (which actually turns it ON for XP).
(At least the current version you are running now doesn't. Could an earlier version you were using on W2K have used Accel ="Low"/munge? Can't be sure.)

And that means Q3 in-game W2K and XP+ mouse sensitivity are exactly the same (with EPP off), because they are only different when accel is on.

Which might not be what you wanted to hear.

Here (if you have time) is the 110% test that Q3 does not need a fix: Open my Fix ZIP file, and double-click and apply "Test_Flatline.reg", then reboot.
Run Q3 and move around a map.
If your mouse/aim has gone extremely speed limited, then I have it wrong somehow and accel is being enabled, otherwise that is more proof that Q3 does not need a fix.

"(At least the current version you are running now doesn't. Could an earlier version you were using on W2K have used Accel ="Low"/munge? Can't be sure.)"

you lost me here. what does the current version of this OS have to do with Win2k? Oh if u mean Q3, I am pretty sure if not absolutely positive quake 3 needed the accel fix in XP, I would think with the CPL fix applied it could affect the results.

I can not remember anything about this win2k low/munge thing. This is weird because as far as i remember and even now as far as i know you need a mouseaccel fix on WinXP for quake 3 AND QL. Again, i don't remember ever needing anything for Win2k -_______- @ m$.

Clarification: "(At least the current version of Q3you are running now doesn't. Could an earlier version of Q3 you were using on W2K have used Accel ="Low"/munge and needed a fix? Can't be sure.)"
Our testing so far shows Q3 does not need a fix in XP.

"I can not remember anything about this win2k low/munge thing. This is weird because as far as i remember and even now as far as i know you need a mouseaccel fix on WinXP for quake 3 AND QL. Again, i don't remember ever needing anything for Win2k -_______- @ m$."

Maybe an earlier version of Q3 did need a fix, but our testing so far says Q3 does not need one now.
W2K never needed a an external fix, only XP+ sometimes needs an external fix, for some games.
For W2K, some games had an internal in-game fix, which is what I'm calling the "Low"/munge code discussed in that Dr Dobbs article: http://drdobbs.com/database/184410376#l4
The developer describes it as :

"... use the SPI_SETMOUSE function to force the mouse speed to medium and to turn mouse acceleration off,...

By "mouse speed to medium", he means MouseSpeed set to 1 which means Acceleration="Low", and by "turn off mouse acceleration", he means set the threshold to zero.
I just made up the "Low"/munge label to describe it, but whatever it's called, it is the root cause of the XP accel bug some games have.

"if i run that how do i set it back to exactly what i have now?"

Download the CPL fix and re-apply it afterwards:
http://www.techpowerup.com/downloads/763/CPL_Mouse_Fix.html

(BTW, When using XP, the Cheese Mouse Fix is better than the CPL Fix.)

"Which might not be what you wanted to hear."

im not sure what i want to exactly hear anymore. I am now thinking of just finding win2k and installing it and seeing it for myself. So i was wondering since it seems like you might know, what dates were each of the Service Packs released or where i could find out such information? Since we are talking about summer 2002 when i used it. i tried doing a search and it looked harder than it should be.

i will go ahead with the test later on, and maybe we share aim?? I am busy right now so i cant do it, but check your pm.

(Then i can update in comments here for others to know)

Wikipedia has some info on W2K service packs:
http://en.wikipedia.org/wiki/Windows_2000#Service_packs

SP1 on 15 August 2000, SP2 on 16 May 2001, SP3 on 29 August 2002 and SP4 on 26 June 2003.

if i run that how do i set it back to exactly what i have now?

Aha! (maybe) it's making more sense now...

"In q3, sensitivity in in_mouse 1 and -1 are different with 6/11."

Is the in_mouse -1 senstivity DOUBLE that of in_mouse 1?
If so, that would make sense: in_mouse -1 is affected by "Low"/munge doubling sens.

i am pretty sure quake 3 didnt need a "mouse fix" in XP, only the CPL mouse fix for registry which is what i used (still do in QL, guess i should look into other fixes, like yours). But do you remember if q3 needed a mouse fix in Win2K? I do not remember using a mouse fix to play quake 3 on Win2k, but again this was long time ago..

If it didnt, then it looks like apparently the sens settings back then will be the same now with the setups i used? And the sensi 8 i had on Win2k is the same as a sens 8 now?

CPL is a mouse fix, so if Quake 3 needed CPL to remove accel then it used the buggy "Low"/munged method of removing W2K accel, which would have doubled sensitivity.

well it needed this in XP, but im pretty sure not in W2K as far as i can remember.

i just cant imagine that my 8 then would be 16 now (in XP).

Well in_mouse -1 would have doubled sensitivity... If you had in_mouse 1 I'm not sure what would have happened.

yes i used in_mouse 1 with the 8. Do you think it would have still stayed the same, or you simply dont know at all what it would have been then?

It would need to be tested to be sure (see the last part of comment # 30 above).

I should have thought about in_mouse -1 versus 1 before, because only in_mouse -1 is quaranteed to double sensitivity IF Q3 does use the "Low"/munge fix.

With in_mouse 1 I don't know, it would have to be tested to see what actually happens.
[Edit: VERY PROBABLY in_mouse 1 would NOT double sensitivity on W2K, even if Q3 used the "Low"/munge fix.]