 |
 |
|
|
Member
462 posts
Registered:
Jan 2006
How does your own location affect to hitting with antilag btw? What exactly is sent to server, the point that your lg would have hit, or your position and angle u shot at? Nothing additional sent to the server right now, server may guess all it need on his own, well, it should be so but unfortunatelly there is things which make this state fail. So let me explain AL yet more time and why own location have issues. Basically, when AL is ON server just keep in mind history(origins) of where players was, so server receive packet from the player and place players at origins where they should be from the client point of view (own position not changed!!!) but that all you know alredy. So what with own location. Let use some weird example so you can get it, say you have ping 300ms (!), packet travel 150ms from server to client and 150ms back to the server. You aim at some point but while you do so you experience rocket hit (for example) and you will gain some knock back which result in you location change but because of high ping you will know about it in 150 ms late and whole 150ms you will naively aim even your own position completely different which ofc result in miss(or at least may result). Well, we all know how RJ looks at high ping, you shoot rocket but you does not see knock back and only after some delay you will notice it, that means what for ping/2 your client show you wrong own location. But luckily, we does not play on such pings so influence of this effect is yet again minimal. 2everyone: All this flame about how AL drastically changing game is EXREMELY exaggerated if we imply sane pings, simply imagine 50ms fraction time and speeds, which is not so common anyway, more like 39ms and 26ms, right? Stop to fight windmills, k? Well, aim getting messed up because of rocket hit is still more an exception to the rule. More often than not your location will NOT be changed by knockback effect. I was mainly interested how your own location affects in hitting the target with hitscan weapons.
Member
357 posts
Registered:
Mar 2006
Here's another interesting summary of wtf we are talking about here: http://www.fabiensanglard.net/quakeSource/quakeSourcePrediction.php
Moderator
383 posts
Registered:
Jan 2006
[SNIP - Qqshka completely unable to understand time]
Slightly ironical, but you can get the point. Qqshka, we already know that you don't understand the concept of time. See here for a thorough explanation. You've tried for how long now? Half a year? And you still don't get it. It's OK, though. I'm sure there are other things you are good at. OK. Bigfoot, I described the situation. Could you tell how extremely lagged player (bigfoot) could see nonlagged (B1aze) before nonlagged? Seems like YOU don't understand the concept of time. The situation is very simple: Non-lagged player sees position of lagged player which he had 10 seconds ago. Non-lagged player could use any weapons he has. Lagged player sees position of non-lagged player which he had 10 seconds ago. Lagged player could use any instant-weapons he has. (sg, ssg, lg) For me it is quite obvious who has upper hand in this situation.
Member
284 posts
Registered:
Oct 2006
I would really like to see a set of screenshots or a side-by-side splitscreen view of the situations discussed here.
What I'm talking about is recording the aforementioned situation with one player at 13ms and other at say 300-500ms, what ever is enough to bring out the effects (and unintended side effects) _clearly_. I think this kind of demonstration would be very useful to people who don't fully understand what principle issues bigfoot and qqshka are discussing and in general, create more awareness of what antilag is actually doing. The demonstration could also show how small the effects are when playing with normal ping differences for comparison.
I can even volunteer to help with recording such demos (as a player), but I don't feel that I have a good enough grasp on AL to actually setup the situations etc.
Medar ^?
I truly feel that Bigfoot in all his splendid candor is on to something, but as a scene I think we need more information and easily understandable presentations to draw conclusions about what the inherent problems actually mean in-game. In this way we could have a much more constructive discussion and point out our wishes concerning fine-tuning etc.
Moderator
383 posts
Registered:
Jan 2006
You can record a demo when hpw will have some advantages and another demos when lpb have upper hand.
What I trying to say that OVERALL even with high ping there are no clear advantage for HPW.
It will be even for instant weapons and LPB will have big advantage for rl, gl and sng.
Member
188 posts
Registered:
Jan 2007
[SNIP - Qqshka completely unable to understand time]
Slightly ironical, but you can get the point. Qqshka, we already know that you don't understand the concept of time. See here for a thorough explanation. You've tried for how long now? Half a year? And you still don't get it. It's OK, though. I'm sure there are other things you are good at. OK. Bigfoot, I described the situation. [SNIP] Unfortunately you still didn't learn to follow the discussion, and you didn't take your time to read the posts you are replying to either :/ Here's a list of the more important posts, in chronological order, which you might find interesting: http://www.quakeworld.nu/forum/viewtopic.php?pid=55227#p55227 http://www.quakeworld.nu/forum/viewtopic.php?pid=55239#p55239 http://www.quakeworld.nu/forum/viewtopic.php?pid=55269#p55269 http://www.quakeworld.nu/forum/viewtopic.php?pid=55270#p55270 http://www.quakeworld.nu/forum/viewtopic.php?pid=55274#p55274 http://www.quakeworld.nu/forum/viewtopic.php?pid=55276#p55276 http://www.quakeworld.nu/forum/viewtopic.php?pid=55278#p55278 http://www.quakeworld.nu/forum/viewtopic.php?pid=55279#p55279 http://www.quakeworld.nu/forum/viewtopic.php?pid=55280#p55280 http://www.quakeworld.nu/forum/viewtopic.php?pid=55281#p55281 http://www.quakeworld.nu/forum/viewtopic.php?pid=55284#p55284 http://www.quakeworld.nu/forum/viewtopic.php?pid=55286#p55286 http://www.quakeworld.nu/forum/viewtopic.php?pid=55320#p55320 http://www.quakeworld.nu/forum/viewtopic.php?pid=55324#p55324 http://www.quakeworld.nu/forum/viewtopic.php?pid=55336#p55336 http://www.quakeworld.nu/forum/viewtopic.php?pid=55351#p55351 http://www.quakeworld.nu/forum/viewtopic.php?pid=55366#p55366 Although it may be easier to just read pages 6 and 7. Happy reading!
Member
188 posts
Registered:
Jan 2007
I would really like to see a set of screenshots or a side-by-side splitscreen view of the situations discussed here. Make sure you don't miss prolag.qwd. It's not a side-by-side comparison, but it clearly does show one of the problems.
Member
485 posts
Registered:
Feb 2006
I made some video clips to visualize different levels of lag.
Administrator
334 posts
Registered:
Jan 2006
Would be nice with a description and explanation...
Member
485 posts
Registered:
Feb 2006
What is there to explain? The red image is Xms behind the green one. I uploaded the Avisynth script used to create the clips.
Member
405 posts
Registered:
Jan 2006
[SNIP - Qqshka completely unable to understand time]
Slightly ironical, but you can get the point. Qqshka, we already know that you don't understand the concept of time. See here for a thorough explanation. You've tried for how long now? Half a year? And you still don't get it. It's OK, though. I'm sure there are other things you are good at. Sorry, but its you who failed to understand time.
Member
188 posts
Registered:
Jan 2007
[SNIP - Qqshka completely unable to understand time]
Slightly ironical, but you can get the point. Qqshka, we already know that you don't understand the concept of time. See here for a thorough explanation. You've tried for how long now? Half a year? And you still don't get it. It's OK, though. I'm sure there are other things you are good at. Sorry, but its you who failed to understand time. Nuh-uh.
Member
1435 posts
Registered:
Jan 2006
Member
1435 posts
Registered:
Jan 2006
Ok, this thread was going very out of proportions, out of what the forum rules and standards of civilized debate allow, so a moderator action was taken by me.
There are many other posts in this thread already which also are very problematic, but at least contain something that has some point relevant to the topic, that's why I left them untouched even if many pointed out they contain insults or provocations. From this point on I am not going to use so much tolerance.
Make sure further posts in this thread are to the point. As I've already said elsewhere: feel free to attack other persons' arguments with your arguments. Attacking other persons themselves is forbidden on this forum. If you are unable to make your point without insulting, provoking or being arrogant, do not post to this forum.
Member
119 posts
Registered:
Sep 2007
zzzz if you're a big boy you can take it ;p
Member
357 posts
Registered:
Mar 2006
lets just use bigger player models with normal bbox and call it 100ms supported.. 
Member
35 posts
Registered:
Mar 2006
Quakeworld finally has antilag? Definitely have to test this out.
All I've seen from reading this drama is that people can now hit what they aim at and blAze accuses them of cheating because well... hitting where you aim... preposterous!
Administrator
1025 posts
Registered:
Apr 2006
Quakeworld finally has antilag? Definitely have to test this out.
All I've seen from reading this drama is that people can now hit what they aim at and blAze accuses them of cheating because well... hitting where you aim... preposterous! I think you have to read again, cause it seems you miss the point.
Member
462 posts
Registered:
Jan 2006
blAze accuses them of cheating because well... hitting where you aim... preposterous! Really? Where?
Member
35 posts
Registered:
Mar 2006
I think you have to read again, cause it seems you miss the point. I'd much rather play a game that has antilag than one without. Games without it generally feel awful on anything above LAN pings afterwards. blAze accuses them of cheating because well... hitting where you aim... preposterous! Really? Where? Clearly there is something wrong with the feature if it has to be slipped in secretly like this. I was almost certain some players were using aimbot in povdmm4s, but this would explain it.
Member
569 posts
Registered:
Feb 2006
obviously the cheat was not that they hit where they aim. But because they utilized a feature, they knew the opponent didnt knew about.
After some playtesting i noticed i did get better results with 50ms and antilag compared to 13ms. Because opponent would start dodging later when geting hit, most of the time.
Only 3 povs tho.
Member
462 posts
Registered:
Jan 2006
Lol, how is that accusing anyone of anything. On the contrary. I just mentioned that I had noticed hpws hitting too well for their ping under normal qw physics. Only one I ever accused of anything was the person who secretly turned antilag on, and it wasn't about cheating but being lame. Learn to understand what you read.
Member
35 posts
Registered:
Mar 2006
After some playtesting i noticed i did get better results with 50ms and antilag compared to 13ms. Intentional.
News Writer
309 posts
Registered:
Sep 2006
On second thought and giving the antilag some time and some games as well, I must say that I'm converting my opinion. My game has benefited from it: now I'm able to compete pretty well at WG whereas it was pretty burdensome without it (ping +50). All the problems with finding a server for an official game will be diminished now for antilag makes it preety even for almost everyone. All in all, I'm a convert now and give my thumbs uo for the concept 
Member
231 posts
Registered:
Jan 2006
Can someone plz enlighten me what the drawbacks are? I don't really understand.
/jOn
Member
133 posts
Registered:
Dec 2008
I see (as much as I understand it) only one drawback which is just the other side of its advantage:
cl_fakeshaft 0 (lagged shaft) is no more a real shaft.
With high and medium pings cl_fakeshaft 1 and cl_fakeshaft 2 are more close to reality.
cl_fakeshaft 2 is available in ezquake nightly bulds and it tries to imitate 13ms lagged shaft witch is roghly equal to antilagged shaft.
Member
231 posts
Registered:
Jan 2006
I see (as much as I understand it) only one drawback which is just the other side of its advantage:
cl_fakeshaft 0 (lagged shaft) is no more a real shaft.
With high and medium pings cl_fakeshaft 1 and cl_fakeshaft 2 are more close to reality.
cl_fakeshaft 2 is available in ezquake nightly bulds and it tries to imitate 13ms lagged shaft witch is roghly equal to antilagged shaft. So its recomened to use cl_fakeshaft 1 with ping 13-25? Right now im using 0,8 dont know why, but it feels ok /jOn
Member
133 posts
Registered:
Dec 2008
:-) With low ping everithing is OK. At least difference is not too big.
Probably number 0,8 makes sense because real shaft now is somewhere between cl_fakeshaft 0 and 1 and it is quite close to 1
Member
38 posts
Registered:
Nov 2006
Loooong thread, have to admit I didn't read all of the posts Played some with this earlier tonight and I have to say it feels ..... weird, using lan-like shaft on wargamez weird in a good way though! But I dont know if this is a good change. And not just because it favors more LG-based players over RL-based players . It completely breaks the game physics. I don't think anyone can deny that (otherwise, why not implement it for RL as well? but that would only show the flaw more I think?). In the end it probably comes down to what we feel is worth breaking the underlying physics for To me personally, even if the flaw is only slightly noticable in actual games, it doesn't outweigh the advantages. reppie
Member
252 posts
Registered:
Dec 2006
We all know that qw players play because there are no compromises, (compared to other games). As I said when this was news, you can't get something for nothing, privileging erroneous viewpoints is a drastically perverse compromise to high ping. As we all know clientside prediction of self-movement is very nearly perfect except when perturbed by weapons (including your own), but your clientside's prediction of other players gets worse and more filtered/simplified as your ping increases, because all it has to work with is current velocity, orientation (viewing direction, which of the 4,294,967,296 different points you can be looking at from a given position), and last known movement commands, (and i learned thanks to sputnikutah's link it doesn't attempt to use velocity of changes in orientation in its prediction). The only time a latent client will represent other players' movement accurately is when he is moving in a straight line or not moving, otherwise it will be more or less distorted, this can be overcome by retrospective hit calculation (using server's memory of motion that really occured plus hits that would have occured) rather than using the clients pov, but it's still a drastic perversity, because it places the onus of mentally correcting (or accounting for) erroneous motion on the nonlatent player. 'on 120 ping i have beaten mortuary dirtbox and reload' (tm) mz adrenalin
'i watched sting once very boring and not good at all' (tm) mz adrenalin
[i]'i shoulda won all
|
 |
 |
|
News