Technical Discussion
  >> Gaming Forum


Register (or login) on our website and you will not see this ad.


Pages in this thread: 1 | 2 | 3 | 4 | [5] | 6 | (show all)   Print Thread
Standard User deleted
(deleted) Sun 01-Jan-17 21:41:57
Print Post

Re: Giants shadow


[re: deleted] [link to this post]
 
I tried out the M1903 Experimental. Only two kills with it so far but it seems ok. Had to sit in a tank for a while to get my scout to level one so that I could get the gun
Standard User mrnelster
(eat-sleep-adslguide) Mon 02-Jan-17 01:45:37
Print Post

Re: Giants shadow


[re: deleted] [link to this post]
 
Most of your post can be debunked if you simply accept that matchmaking doesn't work the way you seem to expect. Put simply, if you have 20 marbles in a bowl and twist it as if you were panning for gold, how many marbles do you think might escape being hit by the other marbles?
Standard User deleted
(deleted) Mon 02-Jan-17 19:03:35
Print Post

Re: Giants shadow


[re: mrnelster] [link to this post]
 
I realise matchmaking doesn't work the way I expect, That's why I said it needs to be fixed, Maybe changed is a better word.
Now the BF matchmaking is based on their "skill" ratings for each player (which is taken from KD, SPM and KPM)
I can't remember if it's at the end of the round or the start of the round that it says 'based on your skill you have been placed on' if you have been moved, So if you are on a team that's heavily won and you get moved and that comes up then you've been moved across to the other team because the matchmaking "thinks" that you've got more skill than others to help the weaker team.
So if that's the case (which going by what it says at the beginning or end of the round is how it does it) Then I don't see how on earth it decides that 3 of the top 5 players who overall had higher KD SPM and KPM than other members of the team move over when clearly there are worse players than them which should be moved over to the winning team but were not.
I realise some people will fall through the cracks and it won't be right all the time what I'm trying to point out is that from my experience is that it doesn't work well the majority of the time as a lot of the games are blowouts.
I lost 3 in a row earlier and was 3rd on my team the 2nd match and 1st on my team the 3rd match (in top 10 overall both times) and then I get moved to the winning team that again won quite easily and then strangely enough my new team wins super easy just like the previous rounds.


Register (or login) on our website and you will not see this ad.

Standard User mrnelster
(eat-sleep-adslguide) Mon 02-Jan-17 21:06:36
Print Post

Re: Giants shadow


[re: deleted] [link to this post]
 
I wasn't challenging your observations as to what happens, only your reasoning as to why. They don't really swap sides because they know the other team has better players. They swap about to balance the connections.

You are very lucky to have a connection that allows you to score top 3 so often. But your assumption that those below you are worse players is founded only on your personal experience. And that I'm afraid isn't indicative. Ever wondered why "noobtubers" use the "noobtube"? Maybe they can't kill a damn thing with a rifle because of their connection? Ever wondered why Titanfall added NPC Spectres and grunts? To give players whose connection doen't stand a chance, something to kill. They couldn't admit that, it would have killed sales. As it is, people got bored with it anyway.

There isn't really any point in my pointing out how to analyse the evidence in the in-game lobby's, because you won't get it. If you suffered your worst ever game, every single game, you would be so used to the idiosyncrasy of the connection, you would see and feel every single spike. As digital foundry concluded, you'll never believe it's a problem, until it's your problem...or you won't actually care, more is the point.

I can estimate the score patterns of the winning and losing sides in my sleep. They've been apparent in 90% of games. It's very, very rare that they'll differ drastically.

After years of frustration I finally proved it, and then got hints of "cheating". If you understood how Sony's geo-location software prevented you from deliberately inducing latency to your connection (through distance) you'd see my point. Slow your connection, improve your score; but they don't let you achieve that by design. I fixed the problem to a certain extent, by connecting my PS4 via VPN by proxy. That upsets the geo-location filter and restricts your connection sporadically. So as there is no benefit in attempting to connect to far flung servers, I have ordered a flashed Tomato router. It should arrive this week, and hopefully I'll be able to get back to BF1 without all the crossover cables and a laptop balanced on the chair beside me!

So, imagine spawning into games like this, as regualrly as I have to!

Hint: The winning score is indicative.
Standard User DG834MAN
(member) Tue 03-Jan-17 02:59:14
Print Post

Re: Giants shadow


[re: mrnelster] [link to this post]
 
I find bf1 seems to premote [censored] connections, when my adsl was down i had to resort to a 3g dongle horrendous ping from yoyoing 80-150, seemed to come out on top of most firefights, get adsl back and have noticed players with 200 odd ping you shoot at them and hit nothing, they kill you after line of sight is broken
Standard User deleted
(deleted) Tue 03-Jan-17 10:12:29
Print Post

Re: Giants shadow


[re: DG834MAN] [link to this post]
 
In reply to a post by DG834MAN:
they kill you after line of sight is broken

That really [censored] me off
Standard User deleted
(deleted) Tue 03-Jan-17 18:20:18
Print Post

Re: Giants shadow


[re: mrnelster] [link to this post]
 
Hi Mate,

I think we have crossed wires somewhere.

I don't get top 3 that often I'm usually around top 10 of my team. ( I don't use vehicles I run everywhere so lose out on a fair few flag captures which are worth obscene amounts).

My "reasoning" for the matchmaking is based on what DICE have said happens, They say that it's based on their 'skill' algorithm which I mentioned before.

I've never assumed that the guys below me were worse players, I was trying to point out that by DICEs own matchmaking system in the way THEY say it works (not my opinion) then it SHOULD of moved players below the 3 guys that it did move before them and was pointing out this was not the case and if it was then it wasn't working very well as going by their skill system that it rated these 3 players better than the players below.
Having matches that are won by 300 or so probably 75% of the time is not fun, My point was that the matchmaking was [censored], I don't care how they do it, I'm just pointing out that whatever way they have of matchmaking at the moment doesn't work very well at all from my experience as most matches are very one sided.

I can feel in game when for whatever reason I have a bad connection to someone close in game and have a latency spike to them, I can feel my guy get more clunky and have more jerky movements when a poor connection gets close, They take way more bullets to kill, When it comes to the actual "fight" either they die or I die a second or so later than we should (first person dies and then 0.5sec or 1sec later the other person dies to the dead person after we've apparently traded where as normally when I trade with someone we both die at the same time).

Regarding player skill I'm only going to say from my experience of playing and watching peoples videos/live streams is that the reason MOST of them are down the bottom is because they don't do the fundamental parts of the game correct and just charge in etc, Sure some people will have a horrible connection like you say you do (I've no reason to doubt you so if you say you have that's good enough for me) but from most of the people I see at the bottom of the leader board they are there for a reason.
Standard User mrnelster
(eat-sleep-adslguide) Tue 03-Jan-17 20:34:45
Print Post

Re: Giants shadow


[re: DG834MAN] [link to this post]
 
Much of the stuff I've researched over the last year, seems to concur on a reasonably successful latency hovering somewhere between 90 - 150ms.

Your point about yo-yoing is very important though. Prediction code works quite well at compensating for slow connections. That's why jumping around or manicly zig zagging works a treat on higher latency connections. It stops the server/client from reliably predicting your position. When the latency yo-yo's, all bets are off.

So a hopping player with a line yo-yoing between 100 - 200 ms, is more likely to be struck down and killed by a baby monkey riding backwards on a pig, than by a low ping player.
Standard User mrnelster
(eat-sleep-adslguide) Tue 03-Jan-17 20:56:29
Print Post

Re: Giants shadow


[re: deleted] [link to this post]
 
I'm sure you're right that some players are awful, but your description of what happens when high latency players get near you doesn't match my experience. That sounds more like your connection dropping to me.

My fibre connection plays flawlessly at 56up/18down and 9ms of latency. It doesn't stutter or lag; it just doesn't have aim assist 90% of the time, and it takes two full clips to get a kill. Shotguns don't work sniper rifles neither. Noob tubes are the only one shot kill for me, and even that's rare. I could post you a video every day of grenades going off at a players feet, followed by a full clip from an assault rifle, for them to then simply spin round and one shot me with an Uzi. crazy

That's why I haven't played BF1 since losing my VPN over Christmas....can't be doing with the return to frustration, and I really don't want to grow to hate it. It's too good a game for that! My new flashed router didn't turn up today so I'll have to wait until tomorrow or Wednesday now to set it up.
Standard User deleted
(deleted) Wed 04-Jan-17 11:10:19
Print Post

Re: Giants shadow


[re: mrnelster] [link to this post]
 
Tried getting 20 shotgun kills last night for a medal. I managed 4
Even against a chap with a pistol I lost :/
Pages in this thread: 1 | 2 | 3 | 4 | [5] | 6 | (show all)   Print Thread

Jump to