Go here: http://www.regurge.at/ql/Lorfa/

And then just CTRL-F "m_filter".

for the lazy:

m_filter "0"

Description:

Values >0 turn on mouse interpolation which makes mouse movement smoother. Values range from 0 to 33.

Recommendation:

I recommend m_filter 0. This is because interpolation adds frames in-between counts, and these wouldn't normally be shown as they aren't true to the actual counts that came in. It also adds a tiny bit of latency, which is negligible at values like 1 but becomes significant at values like 8+. It may also add what feels like positive acceleration, albeit this would be extremely tiny.

Some players adore the feeling m_filter 1 gives them and swear by it. Even top players have used it with great success. It's the kind of cvar that technically should be bad, but in practice the issue is not so simple. So you can give 1 a try to see if you like it. However I cannot recommend using >0.

from carmack's plan updates during the q3 dev:

Jun 08, 1998

I spent quite a while investigating the limits of input under windows recently. I foudn out a few interesting things:

Mouse sampling on win95 only happens every 25ms. It doesn't matter if you check the cursor or use DirectInput, the values will only change 40 times a second.

This means that with normal checking, the mouse control will feel slightly stuttery whenever the framerate is over 20 fps, because on some frames you will be getting one input sample, and on other frames you will be getting two. The difference between two samples and three isn't very noticable, so it isn't much of an issue below 20 fps. Above 40 fps it is a HUGE issue, because the frames will be bobbing between one sample and zero samples.

I knew there were some sampling quantization issues early on, so I added the "m_filter 1" variable, but it really wasn't an optimal solution. It averaged together the samples collected at the last two frames, which worked out ok if the framerate stayed consistantly high and you were only averaging together one to three samples, but when the framerate dropped to 10 fps or so, you wound up averaging together a dozen more samples than were really needed, giving the "rubber stick" feel to the mouse control.

http://www.scribd.com/doc/14192/John-Carmack-Archive-plan-1998

So, basically a good thing, right?

no, it's not. you do not want an average of x samples over x frames to be the equivalent of your mouse input :)

ever since "high" usb hz became the norm i see no reason for using it. one should take note that carmack is talking about win95 here, too.

m_filter 33 for drunk mouse!

m_filter 1 and 2 even 3 seem to help, but I noticed a problem they seemed to mess up large mouse movements small movements like when you lg seems smoother but large movements like rocket predictions seem messed up. Rails seem a lot easier mabey I will use m_filter 2 for rails and lg and m_filter 0 for mostly everything else. I need to experiment more.

don't. read up on what m_filter actually does and stop using it :d

http://lmgtfy.com/?q=m_filter+quakelive

Google knows all..

But not all know Google.

m filter is best left on 0 for shaft. It works enabled nicely for rail and rocket though. As far as I understood, it adds latency to the movement of the mouse because it averages something xD carmack wrote an article about it once. Afaik its a cmd for shitty mice.

It was good for when we never overclocked mice...back in 1999!

pilot wheel mouse ftw

I get brain damage every time I try to use m_filter or cg_filter_angles. Best way would be just leaving everything at 0.

anyone with information on why ql filter was reprogrammed? q3 filter seemed like something you could use, or did it still introduce latency?

The code is very likely identical.

m_filter 33 doesn't feel like drunk mouse in cpma :(

Isn't m_filter just 1 or 0 on Q3?

Yep. I've read someone here mention that in Q3 you could use values between 0-1 such as 0.5 but I'm pretty sure that's false. Perhaps it depends on the client your using to run Q3?

Yes, it's possible. TTimo commited the new mouse code to ioq3, so most recent clients were probably already released with it.

Edit: Checked ioquake3 source and only uses last 2 samples to filter, so it must be a feature exclusive for QL.

yo whats that ssre's name again i forgot?
trembolen?

tramadol.

Tramadol is an opioid analgesic.

Apparently an SSRE is a Selective Serotonin Reuptake Enhancer (as opposed to an SSRI which would be a reuptake inhibitor.

Per wiki the only SSRE is tianeptine, brand names Stablon, Coaxil, and Tatinol.

There is a drug called "trenbolone" but that's a steroid.

Tramadol is a very unique drug that affects a variety or different receptors.
According to wikipedia, tramadol acts as a mu-opioid receptor agonist,serotonin releasing agent,norepinephrine reuptake inhibitor, NMDA receptor antagonist, 5-HT2C receptor antagonist, alpha5 nicotinic acetylcholine receptor antagonist, TRPV1 receptor agonist, and M1 and M3 muscarinic acetylcholine receptor antagonist.

In a medical context is almost always prescribed to treat mild to severe pain as it's main metabolite O-desmythladol (or something like that) is a mu-opiate agonist.
Nevertheless Tramadol is (rarely) prescribed off-label to treat OCD, in cases when patience don't respond to typical SSRI's. Tramadol has a number of advantages over typical SSRi's. First off since it is an SNRE and not an SSRI its seratonegic affects are noticed immediately and it doesn't need to taken for for 4-8 weeks before its seratonegic affects become noticeable. Additionally in rare cases where patients with severe OCD dont respond to SSRI's mu-opiate agonist have been proven to be a very affective alternative treatment. Though they are obviously rarely used for that porpoise considering their addictive nature.
Taking all that into consideration, tramadol is one of the more effective medications for treating severe and unresponsive OCD. Weirdly enough it is also often easier to get a hold of then traditional anti-depressants since despite its opioid properties itis often available OTC in many European countries (I think it's a schedule 5 drug in the US).
SSRE's are very rarely used, the only SSRE medication that is used for its anti-depressant effects that I know of is Tianeptine, which is ony available france IIRC. SSRE's have a number of advantages over SSRI's such as an overall more tolerable sideeffect profile and the fact that they do not need to be taken for 4-8 weeks till they start working. I imagine that the next generation of anti-depressants would consist of more SSRE type and melatonin receptor agonists type drugs.

Wow, NMDA antagonism too. What a weird drug!

I wish I had some to play with now :-P

Heh, feels nothing like ketamine/MXE/PCP or other recreational NMDA antagonists if that's your referring too (no dissociative effects). IME it is also not very useful for reducing amphetamine and/or opiate tolerance, though it is a fun mild sertonegic opiate in its own right (nothing compared to tapentadol or levorphanol).

You seem like my kind of guy Lorfa, if your ever in Israel perhaps we can meet up, smoke some DMT, organize QL configs and geek out over gaming mice while watching ana kasparian host some TYT
=). Just bring some desoxyn with ya please.

where did this conversation suddenly come from?

Heh, It's a long story. I was talking to daddy on some QL server one day about random shit and I guess since he's banned from QL and on prohibition here on ESR it was the only way he could contact me regarding his question. Since it was a private discussion that we had I can't tell you the story with you disrespecting his right to privacy, but as you can notice it was to do with drugs.

I'm a bit of a pharmaceutical nerd so when people refer to tramadol as "simply" an opiate antagonist I start to geek out and feel the need to correct em'.

and I have no idea why daddy_ is so damn hostile to strangers, I guess because he's just so used to people giving him shit for being a cheater that he feels constantly attacked, even when it doesn't seem to be the case.

P.S.
M_filter 0 ftw. If I had to make a choice between the two, I'd go with a low value of cg_filer_angles and bind it only to LG/RG like stermy does, but I personally wouldn't mess with it (despite being one those rare individuals that actually prefer mice with a decent ammount of angle snapping/mouse prediction.

Yeah!! Sounds great! haha

<3 Ana

well that was close, good thing i didnt get trembolen

u could if u wanna turn into zyzz within 2-3 years :P

just have to hope u dont have a heart defect.

u aint funny. i get that you're bored and try to enter a discussion you have no place in or clue about, but still.. a lame attempt

err...

ok dude.

I know that in Q3 a quite a few of the top players used m_filter 1. Among them was cypher and jibo who had the best LG aim during the cpma/vq3 era. I might be wrong, but iirc then m_filter 1 'smooth-ens' out the last 8 pixel of mouse movement.

Looks like vanilla Quake 3 m_filter is just on/off
https://github.com/id-Software/Quake-III-Aren...put.c#L426

iirc it was very useful if your mouse polling rate was lower than your framerate. RIP 30hz serial mice

They were 60 Hz I swear!

Yeah though, 60 Hz without m_filter was difficult to tolerate.

fells like more air-acceleration with filter 1

m_filter 1337 kinda aims for you

(heard that)optional "_____kinda aims for you" never gets old :)