i use 0 dont really feel any big diffrence..

why should timenudge be considered a cheat?

ask strenx

It shouldn't, it makes no sense. If people really have a problem with it they should explain why, and try to get it formally banned. Until then, it's as much a cheat as picmip or forcemodels.

mine as well get snaps, rate, and cl_maxpackets removed too

Snaps is disabled in ql now

not surprising

Ye I know. Thank you. I was just wondering if there is anything new besides "strenx said so". Obviously there isn't, besides the usual unqualified "nudge makes you warp" whine.

I think timenudge made sense on OSP, but with CPMA and QL netcode it does nothing but warp the model.

having like 25ms ping and then set TN to -20, brings plasmagun to another level.
For insta weapons i dunno tho.

hey guys i think using a mouse gives you an unfair advantage against the opponent so we should just use keyboard only because it's much more fair

lool good one

I'm also against monitors tbh, some have big ones:(

Monitors? Whats that? My printer prints 125fps

so you still living in the forest?

I'm from Norway.
Enough said I guess.

mine prints at 24 fps; smooth as a film, and i can't understand what's that "tearing" or "vsync" that peoples whines about

human eye cannot see more than 24 printed pictures per second anyway i heard

yes, i've read it too, but there are things that doesn't fits with that: for example, is it related to having a smoother vision with 120 hz instead of 60, or are just different principes? And i've also read that "The Hobbit", the film from Peter Jackson, for the first time in the cinema hystory, has been filmed at the rate of 48 fps, the double of the standard; it is still a pioneristic technique, and opinions on it are discordants, but scenes seems to be smoother and more "natural"... That shouldn't be possible, if eye could see only 24 fps

actually i was sarcastic in that.

ive visited a course in psychologie at the hu berlin once where that issue had been adressed (actually i just went there because of that, im not even a student at that uni :D)
short version of what was told: theres a level at which point the brain stops with single pictures and instead interpret motion. that is however different for each person + their condition and also heavily affected by how much is actually changing on the screen. so you can use less pictures per time for a slow motion than for a fast one.
also it matters how focused the viewer is on one point of the screen or if hes trying to observe the whole thing.

in addition: more pictures per time always leads to a more distinct experience of motion, without a limit but in a nonlinear way. the perception is made for an unlimited amount of pictures out of which a variable amount is processed depending on stresslevel, condition, age, variety of information and so on.

so the conclusion concerning monitors, movies and so on would be: more fps/hz allways increases the possibilitys for your perception without limits. but the benefit deminishes even tho theres neither a softcap nor a hardcap :P

i hope they bring out 24hz LCDs then soon....

SRY BRO THE ONLY UNFAIR ADVANTAGE(TM) I USE IS THIS
http://ap.razerzone.com/razer-gear/get-the-un...titude-tee

if it was a cheat, it wouldve been cheat protected or removed entirely like /kill and so many other cvars from q3.

at one point timenudge was removed from QL but i guess popular demand brought it back.

having no /kill command alone is enough to disqualify this game as n00bish

" is there any weapon which could benefit from # > 0 timenudge ? "

yes

also TN depends a lot on your opponent ping.

layman question: what does actually timenugde do and how can it influence the game?

Hardly a laymans question, noone seems to know what it actually does exactly.

who cares? if it feels good just do it.

Peter Noone? What's he got to do with this?

Well, wartex does. He also had a cfg cleaner in osp times for the max_cvars freeze error:)

Here it goes:
http://forum.drc.su/yet-another-cl-timenudge-...t2301.html

probably nobody knows, but in Q3, it reduced the size of the snapshot-buffer what was used to interpolate the happenings in the "gameworld" for the client. So with negative timenudge, you saw things earlier, but with less interpolation, and that's why enemies looked warpier (but only for you, and did not affect anything on the server or for others).

thats what i remember as the explanation, and its no different in ql really.

key word, CLIENT SIDE / DOES NOT EFFECT OTHERS

Yes, that's why it's cl_, (client_) but QL has unlagged code, which could alter things a lot. Perhaps the difference is not more than the difference between interpolation vs extrapolation, but maybe it's entirely different now.

You never know with QL:) I just had a normal game and after mapchange I had a huge (200+ms) inputlag without touching anything.. I restarted FF after the "game" (more like suffering) and everything was normal again:(

I am speaking from personal experience in game and from what I have read over the years, both in q3 and ql. the timenudge setting of your enemy makes no difference (or is so small that it is not noticeable / doesnt effect outcome) in the way you are required to aim vs him, can be either 0 or -20 the end result is the same.

I agree of course, and our knowledge based on experience is probably right. I'm just saying that we don't have much proof, we are still just guessing here:(

that's proof enough for me really though, not gonna bother with trying to proof it 100% when i am already certain it is so 90%

QL doesnt have unlagged code, its adaptive timenudge built in

As most of the people are used to say; you only have to lead your aim above 80ms, and nudge (only) can't do that for you. Can you link the source of the info please?

i thought so aswell, apparently that is not the case though: http://www.esreality.com/post/1768002/ql-unlagged-netcode/

Hmm well i guess it isnt the same as cpm, which i know for a fact isnt unlagged

It's about a billion times worse than cpm's netcode.

What does this 'unlagged code' term mean, its the 10th time i see this, and i am only guessing:))

I've heard a million conflicting explanations about what cl_timenudge does. What is it, really?

At least some form of explanation. Also, the "built-in lag" in QL is 25ms iirc due to a higher sv_fps setting.

I have tried to use timenudge -20 and -10 on a 50 and 40 ping, and it does help your aiming a whole lot.
It kinda aims for you, you just need to sweep your crosshair around your enemy's general area and it does loads of damage with any hitscan weapon.
That's why I'm not using it now, I love aiming too much to let a command do it for me =P

timenudge is not an aimbot, it doesnt 'kinda aim for you'. gtfo with that strenx logic shit.

Would you prefer if I said that aiming the same way, i get 10% more accuracy?
I'm not hating on anyone using it, but you're hating on me for telling my experience with it. :o

im "hating" on you for thinking that timenudge has anything to do with your aim when infact it doesnt. ofcourse this is what happens when noobs who know nothing about quake blindly quote everything certain "pros" in the game say as if its the fucking bible.

The one you call noob probably rapes u in duel :D.

probably, guess we'll never find out will we eurofag?

anyways, i wasnt calling him a noob in particular, but i was referring to the scum of wannabe pro-duel players in general who have noobish tendencies such as quoting every single thing a guy says as fact as if they cant think for themselves. honestly, you fanboys make me sick.

someone woke up on the wrong side of the bed

Somebody woke up on the wrong side of the Atlantic

What does being good on a game has to do with knowing how the code of that game works?

Yes Strenx would rape arQon, but he has no fucking idea technical-wise about the game compared to him. Nice logic there.

What the actual fuck?
I have no idea what any of your "pros" said about it.
You asked for my opinion and I gave it to you.

And since the command "reduces" the delay the game gives you thanks to the ping, it actually helps your aim?

no, it doesnt help your aim at all. timenudge has nothing to do with your aim.

I shouldn't have replied to you in the first place I see. Sorry, my bad.

ur crazy if u think timenudge doesnt help with aim at 50+ pings. the difference between with and without nudge is huge. he wasnt claiming it was an aimbot.

play ql with timenudge at 50 ping and then play cpma with default settings at the same ping and u can easily see what hes talking about. the feeling of just having ur crosshair around ur enemy and being able to hit, is easy to see. it isnt actually that ur hitting where u shouldnt be. its like when the enemy moves u dont for instance have to track the model as well as u had to do in cpma.

sigh timenudge has nothing to do with your aim, period.

first off, its a client connection variable, like snaps, rate, maxpackets, etc, which only affect the connection on your side of things. it can however, help with your hits, but at the same time make you harder to hit, just like using a low maxpackets value.

second, the reason why timenudge allows you to hit more at negative values is because its attempting to compensate for the lag. saying that it helps or has anything to do with your aim is totally misleading.

hell, you said it here, point proven:

the feeling of just having ur crosshair around ur enemy and being able to hit, is easy to see.

are you saying that something that changes hit registration doesnt affect the way u aim? rofl????

also no it doesnt make u warp. zzzzzzz. my fault for assuming you knew anything about it.

ps: it doesnt affect your "connection"

are you saying that something that changes hit registration doesnt affect the way u aim?

thats exactly what im saying. if my aim is terrible, its terrible, with or without timenudge. what makes you think timenudge has anything to do with your aim? lol what a scrub thing to say.

also no it doesnt make u warp.

i never said anything about it necessarily making you warp. i said it could make you harder to hit. timenudge has such a miniscule impact on things though that most people dont really notice the change much anyways unless the person's actual connection is terrible.

it doesnt affect your "connection"

it affects how your connection is perceived by other clients in-game which is why its called cl_timenudge. its obviously not affecting your actual connection outside of the game

it doesnt affect your connection in game. it has nothing to do with what data is sent from ur client. nice try at dodging the fact that u clearly have absolutely no idea what nudge is or what inter/extrapolation means.

I don't think very many people who browse this site really have the slightest as to what commands actually do.

nice try at dodging the fact that u clearly have absolutely no idea what nudge is or what inter/extrapolation means.

Please, oh wise one who's surely studied math and science at an appropriate level to thoroughly explain this, tells us EXACTLY what interpolation and extrapolation is AND how timenudge affects each.

Enlighten us.

try searching esr for nudge or look at the xerp thread.

whatever, my point is that it has more to do with your connection than it does your aim like most of the morons in this thread have said. lol thats like saying the choice of crosshairs, timers, or enemycolors/models will affect your aim (im sure alot of you idiots believe it does). theres alot of commands that affect your aim but commands like snaps, maxpackets, timenudge arent one of them. if i change one of those values my aim will remain the exactly the same as its always been.

get real.

im sure you use default settings with the brown sarge as enemy model.

heh, if i did, it wouldnt change my aim at all, despite whatever your idiotic beliefs are.

i dont have any beliefs about how you aim. i only have beliefs about what affects my own aim.

im sure its possible to aim like kgb playing on default settings with sarge as enemy model. yet to see anyone do it tho. and no decent player i know plays on anything but keel or tankjr enemy model.

it would be ok if u said some settings are mostly placebo. but thats neither here nor there. it doesnt matter if its placebo. if it affects ur aim it affects ur aim. saying u can overcome it doesnt change the fact that if ur comfortable with a certain setting then thats whats best for you.

i think ill stop replying to you now cos u have nothing of value to add to this discussion other than calling people idiots.

im sure its possible to aim like kgb playing on default settings with sarge as enemy model. yet to see anyone do it tho. and no decent player i know plays on anything but keel or tankjr enemy model.

sigh look noob, why would ur aim go from amazing to shit just from not using brightskins? that doesnt even make sense for any good quake player to be that dependant on cvars such as timenudge and brightskins.

obviously those cvars can help you hit more but they dont affect my aim in the slightest. i know this cause ive tried a bunch of other similar q3 spinoff games such as openarena xonotic warsow etc etc where i aim exactly the same as i do in q3/ql. once youve learned how to aim, you dont just forget how. ur aim was either already shit to begin with or something else ie: hacks.

i never said anything about it necessarily making you warp. i said it could make you harder to hit.

I don't see what makes you believe this. It has really got nothing to do with it, other than you may theoretically be able to dodge better with quicker updates, but this is a bit of a stretch.

well like i said it depends on the person's connection. some people use timenudge and they appear just fine in game, while others using it may appear a bit laggy/jittery when they use it. ive come across a few instances of this where people have complained about certain players using timenudge appearing a bit laggy to everybody else.

others using it may appear a bit laggy/jittery when they use it.

This is simply not possible. It is 100 % client side.

If they appear laggy/jittery timenudge is not the culprit.

This is simply not possible. It is 100 % client side.

lol its totally possible, clientside connection commands can affect how your connection appears to others in a server. mess around with timenudge, snaps, maxpackets sometime. just because its 100% clientside dosent mean shit. hell, try hitting a guy using com_maxfps 10 or cl_maxpackets 30 see what happens lol.

maxpackets affects how many packets are sent from ur client to the server. and maxfps affects maxpackets. timenudge does not in any way affect data sent from your client to the server. it only affects how things are displayed on your side.

nudge enough and see what happens

stfu and see what happens, you obviously have no idea what you are talking about.

A 100% client-side feature affects only inside the client, not having any effect on the server-side. If it were to affect the server, it would not be a 100% client-side feature. Timenudge exists only inside the client-side binary (in Q3/QL commonly known as cgame) and it is never sent from the client to the server. It does not exist in the server-side, it is never sent to the server-side and it is never handled server-side. Therefore, it cannot affect server-side. Now because of the client-server networking model, a 100% client-side feature on a certain client [A] can not affect another client [B] because the information is not sent to it [B] (it would also require the server to send such information to the other client [B]).

lol, trolled

lol its totally possible, clientside connection commands can affect how your connection appears to others in a server. mess around with timenudge, snaps, maxpackets sometime.

Snaps does not exist in QL, or rather it is internally just set to the sv_fps which happens to be 40. It is the value of snapshots the client will take from the server. It is a server --> client gig and not totally client side because it involves this network operation.

maxpackets is not a totally clientside deal, it controls how many packets/sec are uploaded to the server from your client. So client --> server.

Timenudge however affects how your client makes use of the information it gets. There is no network involved at this point, it is post-network operation processing.

just because its 100% clientside dosent mean shit.

Yes it does, it means it cannot affect what other players see in any way. It is the same for cvars like cg_drawgun.

hell, try hitting a guy using com_maxfps 10 or cl_maxpackets 30 see what happens lol.

Lowest com_maxfps is 30, lowest maxpackets is 63 in QL. Certainly maxpackets would have the potential to lower the information the server receives from your client and could therefore affect how other players see you as the server tries to make do with the information it has.

Maxfps is used for all sorts of client calculations that probably involve the network as well. So I wouldn't consider it a completely client side gig. At the very least maxpackets operates on a ratio of maxpackets/maxfps and _can_ send less packets with lower fps.

So you see, it depends on what cvar you are talking about. You can't just loop all cvars into one category and say they all can affect how other players see you because this is just not true.

are you saying that something that changes hit registration doesnt affect the way u aim? rofl????

It does not affect hit registration.

It allows you to get updates sooner at the cost of some these updates being incorrect. Sooner updates = less latency which can translate to better aim. At the same time less accurate and less smooth displaying of players can translate to worse aim.

It depends on the connection/person which value is best.

you're right i shouldnt have said hit registration. what i shouldve said is that it changes the position of the enemy models as displayed by your client. that clearly affects where you aim. so to say it doesnt affect the way you aim is a fallacy. it would be like saying osp's netcode had no effect on how people aimed (although in this case the changes are not that obvious).

lol where you aim and how you aim are two different things. obviously timenudge may affect where you aim your crosshair on the screen, but it has nothing to do with how you aim and control your mouse movements whatsoever. if the rest of your settings (fov, accel, yaw, pitch, sens) are poorly configured, timenudge isnt going to make anything better. timenudge is not some miraculous cheat where if i choose to use it im going to all of a sudden become an expert aimer. people saying it "kinda aims for me", etc are idiots who have no clue what theyre talking about.

ok, you owned all who believed it literally aims for you, like moves the crosshair... so nobody.

GJ

europeans do...

i heard it kinda aims for you

remeber Q3 cpma and xerp shit. This CLIENT SIDE command does the trick, it realy helps. Maybe it's same with nudge.
For me , not smooth enemy model (all time, not some warps) are not playable, I don't feel fun at all, thats why I never used nudge in 10+ year of quake, even I play with big pings..

Iirc CPMA client side "tricks" were serverside assisted in a slightly different way how it's implemented in QL, so it's not really fair to compare the two without more creditable info on the QL netcode.

how about smaller ping, say 15-20 ms? Does it affect the game?

Didn't try on that ping, thought it was low enough already!

I use it on this ping and it doesn't seem to help or hinder in any way.

yeah that's pretty much what strenx said "I like to deserve my aim" straight from his mouth :P

It definitely improves hitscan, but I think it's a bit much to say it aims for you - rather it allows your normal aim to function the way it should on a lower ping. Corner shots are another story though.

I haven't noticed any benefit on my lg from timenudge on any ping under 60, it feels way better to me at 0 unless im pinging 70ish or higher. I have heard other people say the same thing you did though with 40 to 50 ping, maybe i'm just the minority regarding that. With rail i've found it to be really random and it feels way more consistent to me at 0 than with any timenudge. The only weapons I find to always benefit are rl and pg

timenudge "it kinda aims for you"

Never liked to crank the timenudge up (enemy movement is smoother without yeah?) but with such convincing arguments as "only a retard would use anything but -20 on all pings BECAUSE" I don't know what to do with myself. Honestly can't swear on that it definitely helps (certainly not to an "aims for you" extent) at all but then again I suffer from the brain damage where I'd rather have 5ms instead of it's-totally-basically-aimbot-trust-me->50ms too.

Negative timenudge supposedly reduces built-in delay which would be a good thing and not remotely "aiming for you" (rather your shots go CLOSER to where you actually aim)... or is that explanation pulled out of someones butt?

and that comes from someone that knows how to aim

man create a cool vid for low ping and timenudge whine, like u did for +back (or wasn't u?)

I've recently considered making a new video (raging about shitty teams in pub CA hahaha) but always get stuck on writing lyrics. :c srs bsns

are the lyrics about your love hate relationship with time nudge?

there are no lyrics, the lyrics are a lie :c u write k

challenge accepted c:

DO IT DO IT DO IT DO IT!!!!!!

also : <3

I tried timenudge. I hit more through rails than before, why? NIGGA IS WARPING. HOW THE FUCK CAN THIS HELP ME? LOL

Just stop. No one believes anything you say about anything.

Usually to test something you have to test it at its extremes to see what it truly does. Go into Q3 and do ./cl_timenudge -500 btw this was common for a while for 56k users. You will notice a difference in projectile speed right away as far as whats going on I have no idea but the rockets are almost teleporting to where you aim the game becomes very unstable though. Hitscan weapons I'm not sure it affects them at all. Doesn't really answer OP but maybe some can form their own opinions from this test.

You will also notice at extreme levels of nudge it looks like the rocket is literally a snap shot being taken over and over thousands of times but its very fast. So it doesn't look like the projectile is flying like you'd normally expect

http://www.esreality.com/post/2229718/nigga-plz/

I think this whole nudge bullshit has been launched by tox on dreamhack summer in an interview when he was complaining about getting shot after a corner :DD

which is the opposite of timenudge.... It's the current 'unlagged' code which does pretty much the opposite of timenudge.

Also I was wondering, is there any weapon which could benefit from # > 0 timenudge ?

Accepted values are between -20 and 0.

There hasn't been positive timenudge available in QL for a some time now.

I have heard people tell me 0 is the best and that they take a 20 % accuracy hit with anything else. I have heard people say the exact opposite.

I set it to -10 and have been using that for a while because it seemed like a happy medium.

oh wow ppl saying it aims for you.

aim for you.

hahaha

it aims for you.

HAHAHAHAHA

something something yakov smirnoff

when I use timenudge I only use the mouse for moving and looking around. When I encounter the opponent and engage into battle, I take my right hand of the mouse and let timenudge do its work (I have keyboard binds for jump and fire). I don’t want to disturb its tracking by my faulty aim. That is the secret to good aim with timenudge, to totally trust it and let the mouse. Just let it, take ur hand of the mouse and put it in ur side.

btw that is a secret that very few know (ins and strenx among them). I am risking a lot by revealing it to the public.

"you just need to sweep your crosshair around your enemy's general area and it does loads of damage with any hitscan weapon" = aimbot for you?

u just mad cause I revealed the secret. Don’t try to cover it up. U don’t sweep like u say. If u sweep, u will cancel timenudge’s aim. You just let it uninterrupted do what it does best, aim for u.

EUROPE FUCK YEAH

EUROPE FUCK YEAH

timenudge has barely an affect for me

I heard it aims for you.

Yeah... in all seriousness though, I don't feel like I gain anything in terms of aim or hits with it, it just makes enemies warp a bit.

you're not supposed to gain anything aim-wise. i wish people stop saying it has anything to do with your aim cause it doesnt. hit-wise yeah, it can help if you use negative values, and usually the guy using the timenudge appears warpy to everyone else, not the other way around.

Timenudge is just a placebo...

timenudge make the make unsmooth. In Beta without timenudge the game was much better.

I vote for remove timenudge

doesnt timenudge remove built in lag so u will be able to see ur enemy faster? like if u both have 10 ping and other guy use -20 timenudge, he will have huge advantage ? ( timenudge is not allowed on lan also)

yes you can see into the future with -20 timenudge!

More complicated than that Kaide. Packets arrive from the server at 40 pps, so you get a packet every 25 ms. Since your game runs at 125 fps, you need to interpolate the position of the player in between these packets arriving.

The netcode releases the packet to the client game at a certain rate and when the packet becomes "available" to the client game code it can use it to interpolate the position of the enemy. Timenudge releases the packet earlier. This means you have the position earlier, but it also means that the interpolation is less smooth - which is why it mostly makes sense on connections with high latency.

Even with 0 ping, you'd still have up to 25ms lag due to the 40 pps (i.e., the snaps). So -20 in timenudge would still account for something.

What happens is that it is a trade-off. For a connection with considerable latency, you can trade accuracy of the opponents movement for knowing the position earlier. But we are not taking unlagged code into consideration here, and it may alter the situation.

Personally, you shouldn't really call out timenudge I think.

I get braincancer from threads like this.

no, because cheater threads cancel it out somehow, so I recommend you have a look at those as well;)

They only add to the effect. I'm sending you my medical bills now.

I do brain surgery for free to fellow Quakers:)

I use it as a replacement for projectilenudge, since that one isn't working at the moment. Obviously, I use timenudge on servers where ping >40 to compensate for rocket / plasma projectile lag.

where is the difference between timenudge and xerp actually? there was already enough whine because of xerp, no need to repeat that

I think the effects of timenudge on your aim are arguable, while the effects of xerp were obvious. "Everyone" hit like 45% shaft. Other than that, timenudge has been there since the beginning of Q3, which is probably why many quakers are okay with it.

Just my 82374 cents.

That's a lot of cents !

A whole lot non-sense

the funny thing is that xerp is built in to quake live and people still can't aim with lg

i set timenudge -20 and i see an accuracy increase instantly without doing a shot!

As a medical professional i always prescribe -20 timenudge to all my patients that are afflicted with everything from a torn anal cavity to having earwigs burrowed in their frontal lobe.

With the recent breakthroughs in refinement of rare elements we are now able to distill it from Strenx's tears, so supply should be plentiful. Praise Allah.

I can confirm this!
As a brain surgeon specialized to remove earwigs from the frontal lobes of e-sportmen, your prescriptions are taking out the food from the mouths of my family.
Now - in order to get new patients- I'm forced to go out at nights and shoot random people into the head with a rusty nailgun, which would be a violation of my hippocratic oath if i would care.

amzilling! gravity is high when antelopes run

holy fuck that made me laugh wtf

Every advantage is unfair if noone else has the advantage or the incentive.

With that said, I don't even know what the fuck timenudge is, but it rhymes with limefudge, and I love fudge.

ahahah you dumbfuck euro idiots try and rag on us Americans for making cheater threads as if european community is any better, rofl most Europeans think timenudge is an aimbot ahahah this makes think europe is worse off than the american scene.


EUROPE FUCK YEAH

EUROPE FUCK YEAH

So you can't count on top of lacking any reading comprehension skills, swell.

Tomorrow, Strenx is going to use nudge ... so we can see if it works or not.

Stay tuned.

I'd lol if he actually installed an aimbot instead and said he was using timenudge, and then he hits like 60% LG and is all 'I le told you so!'

Seriously though, from the things he's said about timenudge it's clear that he doesn't understand it at all. It'll be interesting to see how much difference it makes -- maybe placebo effect, increased confidence from thinking he has this magical 'cheat' command turned on?

DABT DAT. im a constant user of timenudge when playing my 70 ping and it does help me a lot, but when playing on swedish servers with ping 40, then its no different really.

i think strenx is gonna use timenudge -20 all tourney, do no different than he usually does and say, "omg timenudge superhack i win zotac with it ha-ha-ha!!"

I wouldn't be suprised if he used aimbot. At his level is impossible to detect and therefore, safe to use.

hahaha impussyble to detect muhahaha.

95 percent of the time wrong aimbotfov combined with 0 position preaim combined with 95 percent onklick flickshots

impussyble to detect (legit)

exec stermy.cfg
cl_timenudge -20
50% LG

Just stumbled across this discussion ....

I dont know q3 code itself but am a programmer with a fairly good understanding of the mechanics of network games.

The "problem" that the client net code has to solve somehow is that it gets periodic rather than continuous updates on object locations from the server (e.g. players).

In all q3 engine games the update period is determined by sv_fps which is 50 ms @ sv_fps 20, 33 ms @ 30 sv_fps etc basically 1000/sv_fps milliseconds (although you will get less frequent updates if your client "snaps" setting is lower than sv_fps).

Now the thing is that even on a "fast" sv_fps the updates are over 30 ms apart and this is an relative eternity in a fast paced action game with the client rendering well over a hundred fps usually.

So from a client coding point of view you have three option:

A) just update object positions on screen as and when when you receive a snapshot. Although very simple to code this the very nasty side effect that nothing would appear to move smoothly - all object/player movement that you see would be a succession of small warps.

b) INTERPOLATE. here when a snapshot is the client does not immediately move game objects to the new position but instead it generates a set of smaller movement (ideally one per physically rendered client frame) to smoothly move the objects/players between the locations in the current snapshot and the locations in the previous one. These intermediate small steps are generated using the game physics model which are the same on the client and server thus interpolation gives the most accurate possible correspondence of object movement between server and clients because its using real data (unlike extrapolation .. see next point).
The price for this though is that the client must operate AT LEAST ONE SNAPSHOT BEHIND the server which means that the actual delay between something happening on the server and you seeing it is network latency + snapshot period.

c) EXTRAPOLATION. this is an attempt to get smooth movement but without the snapshot period delay. Here when the client receives a snapshot it "guesses" what the position of the object will be in the next snapshot so that it can generate the smoothing movement increments without having the actually wait for the next snapshot. Great you might think, we have smooth movement with 0 delay except for raw ping. Hey, now your thinking. so what if we guess 2 or even 3 snaps ahead so that were also compensating for ping as well as snapshot period ! wouldnt that be cool, I wouldnt have to ever "lead" my targets just aim straight at them (for hitscan weps anyway) ? well yes and no ...

-ve timenudge settings forces more extrapolation (as someone mentioned it seems like QL does *some* by default ...) possibly to the point where it completely counteracts even ping latency HOWEVER Given what we said about how extrapolation works there is a price to be paid: the client can only use its physics model to do the predictions and thus cannot include any other player control inputs in its calculations. So yes, you might see someone come around a corner a split second before they see you if you are using -ve timenudge BUT if your opponent changed his mind as the last instant you would be shooting at a client generated phantom hitting nothing and giving away your position.

In a wierd way a high -ve timenudge is actually doing the same basicphysics based predictive calculations as an aimbot but it has the same flaw - the more unpredictable the enemy player movement the more likely the extrapotion based predictions are to be wrong. So in practice -ve timenudge will probably aid hitting unskilled opponents especially when they are airborn but actually make things worse against very skilled players with unpredictable dodging patterns.

so you basicaly saying that negative timenudge = extrapolation, while positive timenudge = interpolation?

well -ve timenudge pushes object movement forwards in time but that is relative to some unknown amount of snapshot sampling "lag" (i.e. how many data points the client needs to collect before it starts to move something on your screen).

so .. I dont know for sure if the -20 limit on QL is enough to force the client to start "guessing" before it receives actual data (i.e. interpolating).

The teltale sign of extrapolation is players routinely warping all the time as the clients corrects its extrapolation errors to align with real server data. (note that such warp is purely on your client and nothing to do with other players having laggy connections)

>so .. I dont know for sure if the -20 limit on QL is enough to force the client to start "guessing" before it receives actual data (i.e. interpolating).

oops - I meant "extrapolating" ...

"In a wierd way a high -ve timenudge is actually doing the same basicphysics based predictive calculations as an aimbot but it has the same flaw - the more unpredictable the enemy player movement the more likely the extrapotion based predictions are to be wrong. So in practice -ve timenudge will probably aid hitting unskilled opponents especially when they are airborn but actually make things worse against very skilled players with unpredictable dodging patterns."

Sorry, but I just don't understand this (I understand that it's harder to predict and to aim at skilled strafing enemies, (regardless if a piece of code or a human is doing that), but could you please explain how is that related to TN.
When it comes to predicting the movement of a very skilled player who changes direction very rapidly, there is not much difference between extrapolating from 3-4 or only from 2 snapshots, is there?

"When it comes to predicting the movement of a very skilled player who changes direction very rapidly, there is not much difference between extrapolating from 3-4 or only from 2 snapshots, is there?"

As long as we are interpolating then there is no "prediction" as such by the client renderer - merely a smoothing of real movement coordinates HOWEVER the fewer snapshots that are used to interpolate the movement, the faster that movement gets translated to on screen position (but at the expense of overall smoothness of movement).

I think the QL client is predicting, and it does not wait for the snapshots to arrive from server, so it's extrapolating based on the already received snapshots.
Timenudge "tells" the engine how large the snapshot-buffer should be, and how many snapshots will be used in the extrapolation. Less timenudge means less waiting for snapshots until extrapolation and visualization of the given "frame" (moment) can begin. So basically every client is guessing what is happening on the server.

Someone please correct me if I'm wrong.

In QL sv_fps is 40 and snaps is locked at this internally by the client.

sv_fps 40 is one frame every 25 ms.

So why then does timenudge max out at -20 and not -25?

Say I have 0 latency and I am getting a snapshot every 25 ms.

Am I correct in saying that with timenudge 0 I am one snapshot behind (one is buffered) and this snapshot is used to interpolate the two positions smoothly? My effective latency is 50 ms yes? For purposes of discussion because there is latency from the screen/renderer/possibly other sources.

In the same scenario above if I set timenudge -20, would I then use the previous snapshot for interpolation, is there interpolation? Do I save an extra snapshot every other frame since 20/25 is 0.8?

At what point does extrapolation start? Is it completely client side? What information is used to generate an extrapolation?

In the case of 0 interpolation, the enemy player would be drawn at 40 fps (with sv_fps=snaps=40) yes? With timenudge -20 the enemy player position appears to be drawn at a lower framerate I assume due to less interpolation, is there any way to calculate this value precisely?

Like, could I do something like com_maxfps 120,maxpackets 120, timenudge x where x is the value that makes the enemy player update at exactly every other frame like 60 fps?

q3 code works according to B and theres no extrapolation possible. So on negatve >1000/sv_fps things would warp accordingly (aka effectively your fps would drop to 30 if sv_fps = 30 and timenudge is worse than -33.

Up till that value however the intrapolation just gets reduced. (So on -9 for example you'd have at least 3 fluent (intrapolated) frames per 5. This does mean however that you can see where your opponent is at more accurately.

QL however is able to extrapolate due to some code SyncError wrote.

BUT in quakelive things are a lot more f*ckedup. The worst part is that there is unlagged code running on the server. (Which is prolly the 'build in correction' which you believe to be due to awesome extrapolation... while I can't figure out why nobody else other than me is whining his ass of because of the retardedness of the solution).
What QL does is that on the server they 'rewind the shot' in time in order to check if it should've hit or not. This also completely obliderates all use for timenudge of instant hit weapons (other than sound issues which I've also pointed out to ID ages ago which is a different subject imo) because it will just make sure that your shot gets rewound less + the nmy will have ghosting on as described in the post above so more shots will feel to go through. This rewinding leads to the 'being shot behind walls' and the enemies lg not drawing anywhere near you even though you get hit.
Also it makes warping extremely advantageous and fluctuating pings (as long as they stay below 70) almost equal to automatically winning the game.

What they should, imo, do is write a few lines of text that make timenudge float accordingly to ping and remove the unlagged code.

I've actually given them the code for the small paragraph above... which I'm further unable to comment on but let's say they found 2 parts where the current netcode is better (or better put where the current netcode doesn't cause for glitches in the engine) and thus they don't want it in. Allthough all ppl I talked to said it pwned the crap out of the current netcode. The glitches are pretty bad for spectators (but that should be easy to fix) and the sound system sometimes cracks a bit (not sure why that is although if I would be allowed to fix it I probably could...)

Atm each time that I re-iterate my wish to let people re-test my few lines of code they tell me I should write it myself in q3 and make people test it there... but q3 doesn't have syncerrors code and I'm not sure where / how he fixed it :(.

btw. q3 servers usually ran at 30fps (was depending on server ofcourse but 30 was default) ql servers always run at 40fps.

does packetup affect timenudge ?

I think packetdup is only for outgoing while timenudge is about incoming packets.

sry i mean smoothclients , does smoothclients affect timenudge ?

it would, but it's not working in QL.

I just tried timenudge -20 and it kinda aimed for me

Ladies and gentlemans, the winner of the "Whine Of The Month" award iiiiiss..............
............
............
............
...Tiiiiiimeeeeenuuuuuudge!!!!

...Ta-dum-tschhhhhhhhh....

just tried timenudge and i didnt even have to move the mouse

lol

I tried timenudge, turned my comp off, and when I came back later I had 10 wins on my account.

...me only 5 wins, but 2 of them against cypher

Why is cg_nudge enabled with -8 by default?
and i dont see any difference

Your cfg probably inherited it from CPMA. Delete it from the cfg and do a /clearcvar cg_nudge inside the game.

just tried quakelive and it kinda aimed for me.

you just need to sweep your crosshair around your enemy's general area

I just tried it and it kinda aimed for me

It's disabled on localhost..

Wtf was with that demo though.

Why? This is the absolutely normal routine for me (and for many other ppl), exactly how QL looks like when I play online. Have to guess where is enemy and could be off by 3-4 keels. It's like that without TN also, but this "hit everywhere" effect is something new ofc.
Weird indeed, it's like a bigger hitbox.

it's just u :p

:(((

Imo it just feels like a bigger hitbox. Its just that what you see on your screen is not necessarily what the server thinks you have seen with backward reconciliation.

I assume that under certain ping conditions (your ping AND enemy ping, maybe with timenudge) the client image is more precise compared to what the server thinks is correct. Therefore people just naturally hit more, because the shots register on the model. Otoh, I think its totally possible to mess up the client extrapolation with the given commands, which results in shots you never hit on your screen, but hit on the server.

/me keeps dreaming of an antilag-free world

whatever the server thinks is the one and only "correct" state of the gameworld. The client just guessing and correcting if the guess was wrong/off.

Also you can't really strafe with 600-700 ups, so it shouldn't be possible to be that off in such a short time period.... but anyway - as InDepther pointed it out already - it's just me, and I don't really give a fuck about ping and netcode anymore tbh, I just play and try to have some fun:p

actually the server isn't consistent either which is one of my major anoyances in ql.

Btw sharpe is just trolling he fucked up his settings here to just make a bs demo

elaborate please

there is no "state in which things happen real (chronological) time" anymore... I made a much more widely post above about how ql works / what needs to change etc.

absolutely not imo! I'm pretty sure that the QL client is predicting based on the already received snapshots, and you actually see things happening "earlier" than how you receive info about them from the server. That's the very reason why the client is "wrong" sometimes and why you can see things "off" compared how they really are on the server. There is a universal time "running" on the server, and that's the base of all happenings in the game-world.
Sorry if I'm wrong, but I can't even imagine how it would be possible to implement your version tbh.

I can't imagine how it is on the server either... which is why I suck beyond believe online.

Fact is the server places shots (from instant hit weapons) back in time. So when they reach the server (for an example) the server time is xx:xx:60, player 1 shot player 2 while p1 has ping 50 -> server will look at where p2 was at xx:xx:35 (1/2 ping earlier) to see if p1's shot hit.
To make the server run the hit will be registered at xx:xx:60 (or 61 if the alg. takes 1ms) but it will be based upon p2's previous possition.

The server receives and timestamps snapshots from each client containing inputs only, and simulates the game-world based on those inputs. Yes, it checks back in time, but it's important that it's checking where p2 was on the server at that exact time, and not on your client.
Or am I wrong?

shouldn't it be like, that the client decides if hit or not, and then send the info to the server?

well, it could be, but the server still has to decide if the client was right or was wrong, don't you think?

given that you don't use timenudge it is the same possition (no TN && ping < 60 --> where p2 was on the server = where p2 was on your client when you pressed fire).
This because the timestamp of both clients are behind that of the server.

So server tells you what the gamestate @ xx:xx:50 is this is sent to both clients at the same time (being servertime xx:xx:50) (they could recieve it at different times according to the server) but the instant they recieve it the local time at the recieving client would be xx:xx:25 as they are always 1sv_fps behind (give or take a few ms and assuming you don't use tn).

The timestamp depends on your local time so... If you press fire while it's interpolating (aka your client is sure that what is drawn on your screen has already happened on the server) you can just hit him.... So you could effectively hit someone on your local gametime of xx:xx:26 which would register on his pc at xx:xx:50+1/2*your ping + 1/2* his ping + (on average) 12.5ms for the next sv_fps to happen.


Ooops btw before anyone else notices it... I missed a few 0's there.... 1ms = a 1/1000th of a second not 1/100st ;)
so plz just correct it for yourself as I gtg :<
But in general it's well possible that you'll get the hit registered to you 100ms after it I hit you. (if we're both on ping 50).
I didn't do the math there btw... just from experience :>

Someone having fun with sv_cylinderscale. :p

haahahashahahah you liar

:D

You made the hitbox biggers, timenudge -20 doesnt do that. I guess thats the joke.

Btw how do u get that weird rocket effect?

no shit sherlock.

wolfcam, fx scripting

Could you upload this cfg? Really nice looking.

http://www.sendspace.com/file/onqtgs

here you go. A few things might be different because it was in wolfcam, but this is a ql cfg

Without reading most of the comments here I just want to say that I personally use timenudge 0 and think this is how Quake should be played.

It's not a cheat since it's doable with a console command, but I would be a slightly happier Quaker if it was forced / disabled to 0.

My main reason is basically that I don't trust it and that you can hit people around corners. This scares me as I usually want to have a logical explination to things :)

you obviously didn't read the threads about tn here

you're complaining about the same thing tox did

and it's not thanks to tn, if your opponent uses -20 tn, he'll hit LESS (or the same amount, dunno about how hit detecition is finally done in ql) exit-rails around a corner!

it because ql has a kind of unlagged netcode, you ARE STILL IN THE ROOM while he hits you (effect gets worse with higher enemy ping)

Fair enough.

I personally don't put any thought into this normally. Just want everyone to be on on same and in general I prefer human prediction over antilag no matter what pings you have. I used to play Quake with +150ms for a few years and still have the same opinion about it.

The problem is the unlagged code. Always has been.

Agreed. I found it was easier to adapt to 100+ ping with 0 timenudge and just compensate as best you can, rather than have -20+ timenudge and be frustrated. haha

There's no reason to use anything other than 0. The difference it makes is negligible. It makes your targets appear less smooth for you though, thus more difficult to track them. It obviously does not 'kinda aim for you' nor does it affect how others perceive you.

It's a console command, yes, which shouldn't be there at all, in this overcomplicated, obsolete piece of mess, which I don't even know why I bother pointing out.

You know, I picked up this game 2 years ago, I can't compete nor do I wish to compete with those who've been at it for a decade.
Yet, it amazes me and I cannot comprehend why people are so uninformed and clueless. Just take a look at the recent netcode threads. People spouting all kinds of bullshit. People who think they can analyze, as soon as they use setting X one game and setting Y the other. Then immediately jump to conclusions, even though the maps were different, the opponents and their pings were different, and so were the situations.

Then there's the pro argument. Oh, the pros must surely know what they're talking about. Not necessarily. You can be talented and become successful at this game, without ever giving a fuck about netcode or TN. Just don't come here then, claiming your stupid BS.

Hitting people around corners has nothing to do with TN.
I do it all the time, and others do it to me all the time, even tho it appears as if I wasn't in line of sight anymore for the shot.
Did it never occur to you, that this happens in pretty much every online game(not just shooters), that has a netcode with a buffer or delay? It has been explained before.
This is unavoidable unless:
-you want to penalize higher ping; the further from 0 the more disadvantaged you are and vice versa
-we had an infrastructure that allowed the usual ping to be single digit, and get rid of the buffer.
In QL, you can play perfectly fine 0-80ms, I'm sure you have noticed that, and there is a good reason for that.(Keep in mind, it's always at least 2 factors, your ping and your opponent's, not just yours.)
It may not be perfect, but for a game with such a pace, I'd say it's pretty fucking good.

Anyway, I'm just tired of the stupidity and retardation that ensues from people deciding to talk about shit they don't know jack about.

Complaining about timenudge usage in QL is ridiculous. It doesn't make a big difference in QL's netcode. I personally use 0, and i've tried -20 and hit worse.

Timenudge has a very bad history, partly factual, partly based on wrong assumptions, rumors, and missing information, most due to the fact that, after all, the real info are buried into the code.

This is what I know about it, which is what I recall understanding over the years, but may be as inaccurate or obsolete.

How it is supposed to work

Many entities in quake are moved around by mean of interpolation. Interpolation is a generic term, and the actual meaning depends on the situation.

Let's start with a simple example: a rocket. Your client receives a start position, a direction, the time the rocket was shot, and a flag that tells this has to be interpolated linearly. Given the current server time (also sent in the packet), your client computes where the rocket is and displays it to you.

This is a simple case because - if we disregard the moment when the rocket is fired - the only latency that counts is your ping. One can guess that by simply cheating on the interpolation time, the rocket could be moved ahead the right amount to give you a perception of where it actually is according to the server.

Opponents are much trickier for a few reasons. One is your client needs more data to move them around. IIRC, to move a player model, your client needs two snapshots: one will contain the start position, one will contain the end position, and as time progresses a linear interpolation of the two is done.

Now, why is this different from a rocket? Because latency-wise, there is no longer only the ping, but also the rate at which the server sends you packets (snaps). You cannot interpolate a player with one packet (as you can with a rocket), so a player is 1/snaps (tipically 25ms = 1/40) delayed with respect to simpler entities.

Granted, this, inserting a "cheated interpolation time" into the interpolation creates a weird extrapolation. The player no longer moves from A to B, but from somewhere between A and B to somewhere outside A and B along the AB direction. Once a new snapshot containing C arrives, the player is immediately moved to the position it should have according to B and C at that time, which often does not match its previous position, and hence we see the opponent warping slightly. This happens about 40 times per second, the extent of the warp depending on many factors, such as how the opponent moves, how ping fluctuates, whether packets are lost, etc.

What some people believe it does

The first comment is that whatever timenudge was in the past, it may now be a completely different beast as the code is different (IIRC I saw a log of Syncerror saying tn was rewritten).

Talking of bad history, I can find many 2003 posts from the italian community where people would swear the timenudge would not just affect the client that uses it, but also the opponents playing on the same server. The point is, I never saw a proof backing those claims.

Some people also believed that you had to use timenudge to play online competitively. I remember playing ra3, and seeing enormous differences with LG. Again, I never saw a good study on the subject, although the trend was so widely accepted that it may be safe to say it was true.

In general, what many people experience is that by using timenudge, they hit more. Placebo? I don't think so. Routine? More likely. Timenudge alters the perception of where things are, ultimately influencing the timing of shots and aim corrections. A simple guess would be that the timenudge-altered-game suits certain people better, and they are naturally inclined to prefer it.
Other players may instead prefer smoother, more stable visual clues, so that they can adapt more efficiently as the medium is more consistent.

Who is better, in the long run, is also not clear. Eventually good players meet in lan, where the game is again different from what it is online, with or without timenudge.

sorry kids, but arguing about a cvar which is more than ten years old to be a cheat is just...insane.

As it has been said for many times now, its reduces the TIME of the CLIENTSIDE Buffer for incoming packets, and therefore gives more extrapolation, which can, but doesnt have to be wrong.
so you can find the value, where your lagometer shows the optimum amount of extrapolation for you, and the weapon you have in use at that time.

for me its -15 for hitscan -20 for everything else, because other weapons are based much more on accurate prediction.extrapolation works better, in some how learnable as prediction, better than human prediction.

now do with that whatever you want, but stop calling it a cheat, as it is a CVAR

god damn this french kid is the most underdeveloped fool on the quakelive cheatcommunity.

fucking euroidiots keep saying timenudge is a cheat what a bunch of clueless twats

This thread needs some xerp.

Why would anyone want to play with a unsmooth-as-hell model and how is supossed to make you hit easier if it's so damn unsmooth it hurts the eyes.

In qlive I use 0, but haven't played on anything greater than about 110 ping yet, so I can't really comment. 100+ ping is def. laggy but still playable.

The antilag combined with timenudge on "ok" ping, about 40-50, def. feels...weird. Not good or bad, just very odd to me, so I keep it at 0.

Still prefer low ping (~10) or med. ping (40-50) with timenudge 0 than anything else.

And that's my opinion that no one cares about. :D

INS is right. With timenudge you deal so much damage that it is unbelievable. When enemy uses TN, i always take enormous amount of lg damage from noobs who cant do one strafe or circlejump ... its cheat definitely. I also eat rail while I take cover behind the wall. The whole concept of quake dies with this nonsense because quake is positional game. Everything depends on position so whats use of dodging and running behind the wall if enemy clearly see you on his screen ... You hit rail and dodge behind wall and when you dont see his model, he hits you. When I use it I feel like Im dealing 2X more damage with lg and rail. People think that something is wrong with server or lags so they use it and think that they are good, but they dont realize that quake is very fast game and that they simply aim very bad. I play with timenudge 0 and I realized that I hit every time my crosshair is over my opponent. Without timenudge you will truly learn to aim and to play. I strongly recommend not using it. Its a cheat and it makes you think you play better. I hope it gets banned from quake.

Agree, but since there are a lot of players who use it, also a lot of them who warp because of it, or even when you get a feeling you got hit after you saw a projectile that you dodged/missed (did not miss).
Because of the said reasons I prefer timenudge on -10 which is somehow a golden middle, plasma works easier and it doesn't give you that much of a false "aim" as much as -20.
Another thing when playing 40-60ping server TD is -10 for me, but when playing 15-30 I rather go with TD 0 sometimes -5(for 30-40).
And yes it's a cheat for 20ping with -20TD....I kept saying that in q3 osp and I keep saying it now.

the rockets being inaccurate can be fixed with projectile nudge btw... although I dunno how they implemented this so I'm not entirely sure where to set it to :(
But projectiles are only affected by your own ping.

I prefer my plasma with -20, everything else is 0.

I can confirm i get much better shaft with -20 timenudge, its like my shaft is thicker and stronger.

i'm sure your gf likes tn then.

a lot of dudes who use timenudge are unhittable for me ;s they think they dodge like a pro, but rails just go trough ))

Players with Virgin ISP are worse, MUCH worse.

hasnt it been confirmed countless times that cl_timenudge is completely clientside

From what I understand of Weird's posts using a negative timenudge draws you more accurately on the enemy screen. If he means extrapolation then I suppose your model will indeed appear "warpy". The strangest thing about all of this is I thought syncerror fixed the extrapolation code making it smoother. Sort of drunk posting so feel free to correct me if I'm remembering this wrong.

Well I'm honored to be mentioned but I do have some small alterations.

1) it depends on how you interpret 'more accurately' in the first line.
Ofcourse for your personal side (aka when your nmy uses tn) the offset to where the player on the server is at is still the same (your ping) regardless of his tn. However his possition will be closer to the possition he has on the server (his ping + tn). (note that tn itself is negative (so ping + tn < ping)).
So you could say that his possition (on his own screen) becomes closer to where you see him (on your screen)...

2) if you don't use tn yourself you always interpolate the nmy possition -> he won't warp (unless! the server lacks information due to connection errors (ping spike or packetloss) or other reasons (of which some could be linked to netcode / timenudge)).

3) the fix that syncerror made is with extrapolation so it will only affect the client who uses tn. Before the fix QL could not extrapolate opponent movement. Now it can.

Say you have tn = -20 --> 5ms of interpolation between the last 2 snapshots and then 20ms which gets extrapolated while you're waiting for a new snapshot.
Before the fix all nmy models on your screen would freeze during this time, as they couldn't be extrapolated. Now they mainly follow the path they were on to start with. So they will go continue walking to the side they were moving towards at the last recieved snapshot.



The thing that mainly affects the game though is that his shots will become 'less unlagged' due to point 1. Basically everything he does has a different timestamp. Now in Q3 this didn't matter at all unless the timestamp became > server time (=use TN on lan). However in QL it matters sooo much because the server uses the timestamp to do it's unlagged hit detection (which I heavily disaprove). So his shots will be 20ms closer to your own model when your opponent uses tn = -20.

I'm still not 100% happy with how I formulated 1.... but I can't really find the right words for it :(

Because the player possition is somewhat more complex as each client 'controls' their own possition as long as it's not massively different from the server. Which is slightly different from projectiles which are completely controled by the server.

why is it banned from lan then?

no idea why it would, seeing as it has 0 effect on localhost

You'd think this is true, but you can go and test it against some bots. It makes a difference.

because the server can't deal with it.
It ignores packets which are "in the future".
If your ping < 40 and you use tn=-20 the server 'could' start dropping your information --> warp / shots not always registering and many more things. As your ping approaches 0 this will become increasingly bad, where on a true ping 0 (aka 0ms of rendering / 0ms input lag / 0ms network lag / 0ms processing lag on the server.... etc.) you can only do actions for 1/5th of the time while all, non movement related, actions will be dropped for the remaining 4/5th.

(I'm also not really awake atm so I could make some minor mistakes....)

shouldn't be around, imo.
just another exploit low pingers will use to get an advantage
i remember people were using -50 until it got capped back in q3, lol

if everyone would use autotimenudge ql would be a lot better though.

it's worse than usual tn cause it fluctuates with your ping and feels kinda inconstant. I only use it when ping is over 60ms when the delay compensation is way more noticeable than with tn -20. Also many highpingers actually hate tn cause it's uncomfortably warpy for them, especially LCD users

Now you know how I feel playing against those people who's pings fluctuate or have high ping.
Unlagged puts their lag on my side (which is unfair from every moral stance) while autotn puts the lag on their own side.

p.s. the form in which lag is presented changes a bit but the above is true. Now I could explain this... but I already did... multiple times.

autotn doesnt put the lag on their own side, moreover it's gonna be worse for you than for the highping player, you'll get fluctuating enemy hitrate on your screen and more warpy enemy position not only because of ping fluctuations but also their client sending position with fluctuating +/- few ms timestamps which do not rly compensate the ping fluctuations since that's impossible

I'm sorry but fluctuating enemy hitrate is due to unlagged.
Next to that they don't warp because of using timenuge (they can't unless their ping < -tn*2).

small correction, given that tn is on the client side, it could lag a few ms behind your real ping, so it could increase warp on extremely laggy connections. But these connections are completely unhittable anyway.

hitrate does not fluctuate with stable high ping
current autotn formula is -ping/2

I'm talking about pure autotn fluctuations influence. For example the ping fluctuates from 40ms to 60ms, autotn would fluctuate -20 to -30, but -20 does not happen in the same 40ms packet and influences the next packet(i believe) which can have 60ms or w/e so in summary the fluctuations are up to 1.5 times more noticeable. And that also influences enemy position unlike usual timenudge which is constant and does not make your warp stronger for opponent.

and 40-60ms example is not that extreme, i've seen worse

Maybe they should get autotn value from average ping for last 15 seconds, or for even more precise value using average from 80% of those ping readings, cutting off lowest 10% and highest 10% or smth. But to be honest every implementation can have ways to abuse them. In most cases ping fluctuations are worse for the player rather than his enemy anyway

I also proposed that at some time....
This actually has more downsides though due to how the ql server accepts packets. As it could start ignoring packets when your ping is on a warpy low.
Ofcourse they could alter how the server accepts packets... which I also said how to fix.... damn I think I put in too much effort while not getting paid for it :(

btw I'm pretty impressed with your knowledge on timenudge and synchronisation.
I do think that you haven't thought the unlagged code through well enough though. Or perhaps you feel that the "I get hit while I'm already behind the wall" (and the problems related to that) is in some way fair.
Personally I can't deal with that part of the 'netcode' (hitdetection) because it's due to my opponents connection and not my own.

hitrate also doesn't fluctuate on stable tn.
I know what the formula is (or well I know what it should be)... as I wrote it on their (focus) forum back in 2010.

you're somewhat right on this. On the parts where the ping jumps the fluctuation could be more noticable. Which is what I wrote in the 'small correction'.
However for the players whereabouts this doesn't really matter as the server only looks at the last recieved packet from the player (last one he accepted) and puts the possition which is in there and pulls it forward towards 'now' (on the server).
Aka the only way in which you will warp more is when -tn*2 > ping, which could occur when ping drops significantly and tn will be 'too slow to adjust'.

I tried both autonudge and various single values on different server locations to see what all the fuzz is about tn. Whenever I use tn be it auto or single value I just aim worse. Game feels less smooth to me and with autonudge the whole shit even starts to lag (on 40-50ms):O

the problem is that, because your opponent won't use it, you're getting a negative experience as you're solving the whole lag issue on your side.
You get the warp from solving your own ping and you get hit behind your model because of his ping. So the general gaming experience will be worse.
When both players use it it's just a matter of getting used to it and it will suddenly make sense again.

So the whole "hi@hit me behind the corner" would actually vanish if everyone did or rather get forced to use autonudge? Or is antilag still the major factor for it? It is just sad how stuff hits you even tho you are already behind the corner. The way it is right now just annoys the hell out of me in duels ...

Well it's not possible to erradicate it completely but by far most of it will be fixed.
Basically it'll be up to your own connection.

actually it is possible to fix it completely but in order to do so both players will experience massive warpy opponents... which I think is not a nice gaming experience ;)