475 replies to this topic

[Suko]

100+

I usually play at 60-80, but for a month or so my ISP was being crappy and my connection to MWO was around 110 - 130 for a while. That was noticeably more difficult to play. However, I also think the lag made it harder for my opponents to properly target the "hurt" areas of my mech, so it kind of worked in my favor, too. But I primarily use lasers and was leading spiders with 3x their body size to hit'em.

[boboking]

So who do I talk to about y ping? WEnt from 250ms now to above 300ms. Was fixed once by PGI?

[FallenDevil]

I'm from Indonesia,the usual ping is 230ms-270ms before pgi departed from igp but now it's 280ms-300ms+,hell since few days ago it never reached 280 alway 300+. Tried to use WTFast but the game alway reconnecting or always returning to the mechbay. What do you use for powering up your server? Hamsters? And what ISP do you use? Is it from North pole ISP? The game has become unplayable for me since a week ago. My other MMO game is fine,the ping is relatively normal for people like me,who lives in asia. But for MWO...it's the worst.

Edited by FallenDevil, 19 March 2015 - 01:05 AM.

[Anjian]

This game is full of desyncs at 220ms. It costs me hits, it costs me lives, it costs me kills and it costs me games. I wonder how I can justify spending for your game, Russ, if MWO continues to perform this way. My other games perform so much better than this.

[MayoRe]

I play on a "stable" 237~240ish ping. No issue really. I still able to get nice hits on it.
Problem with delays on zoom/vision only happens when my ping goes unstable with 350ms highs.

[Lily from animove]

honestly above 80 is noticeable.

but a real game impact starts at 140-180, because then using the zoom gets a very noticeable delay. And you start "gaming in the past"

you can not react directly on events happening on your screen. reacting to hits especially of lasers starts to get pointless, because together with human reaction time you have somethin like 300-400ms delay between "server registers hit, until your reaction reaches the server" And by this trying to dodge or spread laserfire or parts of it (especially MPL and SPL) is tehcnically not possible anymore.

[Cola]

In CT, my ping was around 70-90 range. When I transferred to HI my ping was in the 240 range. Major difference in play, my performance dropped significantly, During peek hours my ping would be upwards of 500. After around 6 months I basically stopped playing, no longer able to compete on the level I was accustom with. I would test my connection by zooming in. What happened to the world? I use to play MW4 in the NBT on a dial up connection and I don't remember lag being as bad as it is now days.

[nehebkau]

90-140 ms
At 90 to 140 my miss-rate with ballistics and SRMs starts to drastically go up. Anything above 200 and I go play something else. That also goes for other players I face -- when their pings are above 200 they develop a nice lag-shield due to HSR trying to be a good sport.

[Funkmaster Rick]

My normal ping to the game server is ~25ms. I find this to be excellent, and it makes me feel all warm and fuzzy inside.

An old rule of thumb is that anything above 60ms is not acceptable for competitive play. It's a rule that's held since I first started playing Counter-Strike. It's still valid.

I have not noticed many ping spikes on this game, so I can't say for certain, but in many other games I play, 60 ping is the limit - after that, I notice the lag and my in-game performance drops.

[Rushin Roulette]

Jon Cunningham, on 12 March 2015 - 03:22 PM, said:

Especially for people who have experienced the game at different pings, at what point do you start noticing ping is having an affect on your gameplay?

1-50 ms
50-90 ms
90-140 ms
140-180ms
180-220ms
220-260ms
260-300ms
300-350ms
350-400ms
400+ ms


I have a lot of data from internal tests but I am looking for your experiences. Of course remember that jitter and packet loss fps and many other factors also impact gameplay; I'm just looking for what your subjective experiences are at each range of those pings.

This is actually a very dificult problem to answer. Most players are playing in a fixed location and teh ping stays stable within certain tollerances.

After a while you get used to the range of pings you usually play at. If anything affects that pingrange, you will quite definitely notice, it doesnt matter if the ping goes up or down.

Any lag compensation you would automatically do will substantially throw you off target (either ahead if you are used to higher ping or behind if you are used to lower ping). Its much like the feeling you get after driving a go-cart and then start driving a normal car withpower steering. You have gotten so used to handling that small gocart with no power steering and having to fight it in every corner, that the powersteering will feel like you are in a toy car afterwards.

The only real data you can use is if someone moved from one area/country to another (ie if someone moves from Europe to the American continent or Asia (or the other way around of course). If you really need the data, Ill be happy to provide you with some... I just need a paid return flight to Canada and a PC to play test on .

EDIT: And some players have absolutely no idea what they are babbling about... I mean seriously. Read up about the topics you are posting about before you hit the submit post button (ie. what ping is in this case) and get an Education. There is nothing at all PGI or anyone can do about your Ping apart from spending money on servers in different locations... which would severely split up the playerbase.

Anjian, on 19 March 2015 - 01:12 AM, said:

This game is full of desyncs at 220ms. It costs me hits, it costs me lives, it costs me kills and it costs me games. I wonder how I can justify spending for your game, Russ, if MWO continues to perform this way. My other games perform so much better than this.

Edited by Rushin Roulette, 19 March 2015 - 06:04 AM.

[slide]

Addendum from my earlier post. (page 1)

I am used to a high ping, I am used to HSR being borked. I only play this game on line so I cannot compare it to anything other than what I have experienced in the past from MWO.

Since the last patch (17/3/15) Hitreg has been borked even more than usual. I don't know what you did but something is seriously wrong and I am not he only one to have noticed this.

Example. Yesterday, on Mordor, around 5pm Adelaide time, First match in my Zues 6T. Load out 1LL, 3MPL, 3SRM6. Came up behind a Grasshopper firing across the bridge into the Pugzapper (Cauldron). He was stationary, I was stationary or nearly so. At less than 100m I put a full alpha into the grasshoppers back. 65+ points of damage should have either killed him or had him severely damaged. Nope not even an armour break through. Second alpha. Nothing, armour down to yellow. At which point he turned. I don't care what mech you are driving nothing can stand up to over 100pts of damage in the back. I have killed Jagers in 1 shot with my 4srm4, 2ML Shadowhawk before. That Grasshopper should have been dead (probably twice over). But someone else got the kill after we circled for a while.

FWIW I finished that match alive with 2 kills, 10 assists and 794 damage which just shows how inconsistent this issue is. I really wish I had video of this.

Laser focused builds such as my new Grasshoppers and Zues' are practically useless ATM which is rather disappointing. Just as well the King Crab still kills things although I am sure only half my hits register. But half is better than none I suppose.

Edited by slide, 19 March 2015 - 03:17 PM.

[Zordicron]

I just realized I should post some other things to go along with my earlier one, for analytical purposes.

My ping etc, and play XP can be seen in earlier post.

Other things worth mentioning:
I have an OLD(like 2009 old) BUT potent Core I7 quad core PC. OC'd just a tad, it is one of the early architectures so it gets stupid hot if it is pushed to hard. Withthis, I have a solid amount of RAM, enough(and of good enough speed) to have been future proof for all this time. think cutting edge early 2010.
More recently(last several months) I upgraded my GFX card(s) to an SLI Asus GTX 650 TI Boost set up. OC'd quite a lot with ASUS own software. I have a solid state drive.

What this results in: Despite stupid high ping, that fluctuates because Frontier is bad, the game itself runs at mid 60's FPS, and 97% of the time, smoothly, with no rubberbanding of note or crashes etc. the game itself runs solid, to put it more shortly, on nearly max settings. Loading times are decent EXCEPT when it is trying to get something from the server, like in game store windows, or even when the updater is trying to get the current flash/ad/pictures.

Basically, my PC is potent enough to not break a sweat running MWO, but my ISP (Frontier) is working with hardware from the late 90's and ground lines from 1975, in combination with purposeful routing through their own web hubs all the way to Chicago(I live west of Green Bay a bit, and my pingtrace makes 2-3 jumps, the wrong way- through Frontier hubs- to get south)

the only other thing I noticed on the periodic traceroutes I have done, was whatever the first thing in Canada is called(I forget right now) adds some latency. When I do traceroute directly to some of the hub IP's along the route to MWO, most of them arent really too bad on their own. But the ones in canada return well over 200. I do not know if it is just the Canada net, or if it is related to Frontier and international policy on their end. Further, I 100% feel they throttle my connection after identifying it as a game, as neighbors are able to stream Netflix without much issue on the same lines.

Sorry some of this got off topic, but my frustrations with Frontier Comm. are endless.

[Vlad Striker]

I have ping 160-220 and don't have serious issue believed on match result. But for the BattleField game it would be catastrophic. Sometimes especially with clan lasers I have bad hit reg, something near 60% damage wasted but other gamers have this problem too so it is balanced.
I thought is PGI have hidden dices roll on hit? May be lasers have 1-10 dice roll, SRM 1-6 and so on.

[C91]

My ping's usually at ~135 but I some massive hitreg issues with Gauss since a week or so. I definately see the trace but it doesn't recognize any hits. Happens to around 25-40% of all my shots, even at "close" range (150 - 400 metres). Not sure why that's happening now, especially to Gauss only as I don't have any issues with other weapons.

[Gozer]

Back in the US I had pings fro 80-135. Any time I hit pings higher than that it was because someone was using a torrent in the house and I instantly knew it.

Now here in Australia I'm averaging in the 250-275 range. Problem is it varies, a lot, and any time I see something touch 300 I see game errors, move glitches, stutters, etc. Since the 17th I have yet to have a game where I didn't see some sort of hit detection error either people hitting me (lag shield) or me hitting others. :|

[David Sumner]

Kalteh, on 12 March 2015 - 03:26 PM, said:

Usually I am at 110-130 ms. At least I have been for the last 2,5 years. Lately my ping suddenly jumps up to 180ms for some games. At that point hitreg becomes just crazy and I stop playing for a few hours...

Really? Wow.
I LIVE in the 200-300 range.

Actually, worse for me is when my GPU heats up and FPS drops below 20.
At 5FPS, a fast moving light going across my screen can cover 30 degrees between screen redraws.


What rages me is all that wonderful unsolicited advice from your team mates telling you what a useless piece of s**t you are because LRMs are worthless.

Guess what. 5,000+ matches. I've got a pretty good idea what works for me. And Yes, LRMS are bad, given that I usually get about 1/2 of what other mechs in the same tonnage claim. But it's BETTER THAN WHAT I GET OTHERWISE.

Here's the main issue I have with NOT using LRMs.

You: 100ms
Me: 260ms
Diff: 160ms

Time taken for a "normal" human to see and react to an image change (like a mech appearing)
~300ms

So, by the time I see you, you have already seen me and reacted.
End result? You shoot first, I end up dead.

So, It would be back to LRMs, but I've had enough. I might try again in a year, but for now the game has enraged me enough that I'm uninstalling.

PGI, my wife thanks you.
My bank balance will thank you for the extra income I earn
It won't miss the money I never spent on you because every time I was willing to, you did things that P***SSED me off.

Having said that, I'm a programmer, and I work with stuff that involves process optimisation of millions of tons of shipping.
I have some understanding of just how hard it is to balance this game.

For those who don't, here's an example.

1. Assume you have 2 teams of 12 clones, all exactly the same skills.
2. Assume team A all have exactly the same Stock TDR-9S
3. Assume that team B all have exactly the same Stock TDR-9S, except they get clan tech instead of IS.

Now assume that doing that gives a team (Let's admit it will be team a 1% advantage per mech of killing an opponent first.

That's 49.5% vs 50.5%

Well, that's pretty close right?
Nope.

Firstly, first kill gives an advantage to the large team. And that advantage increases exponentially the bigger the gap.
Secondly, even without that, just assuming that each mech fights one opponent solo, then the survivors repeat until there is only one person left, you get this.

101% x 101% ... to 101% times 12
Which is
112%
Meaning that in 12 a side, even a 1% variance means that one team will win 9:8 (9 of 17 matches)

If you get to CW, it's even worse.
101% times 48 becomes 161%
That means one team wins 8:5 (8 of 13 matches)

Of course, the top players like to win, so they switch to the team that wins more because of its tech.
Which makes it even more one sided.

Sum Ting Wong, on 12 March 2015 - 06:53 PM, said:

My ping is always 260-300 (usually in the 270s) because I connect from Australia with a stable connection.

Upfront TL;DR: I can never trust what I actually see in front of me and spectating other players makes everyone look like they have terrible aim but still get hits.

Subjective full opinion:

This latency produces an effect similar to old school online mechs games in the days of dialup 56k and the first generation of cable connections. I usually have to aim 0.5 times the width/length ahead of a moving assault and up to 2 times the width/length ahead of a fast light to hit the part I want (depending on movement speed of course). Changes in speed and direction of a target requires quick mental calculations to adjust the length of lead required. If my connection was unstable and my ping fluctuating, I would have written this game off as impossible to play. Since this latency is stable, it's merely extremely aggravating and a little unfair.

Trying to use a laser vomit mech on a moving light without spraying and praying at those levels of latency is almost impossible, even at close range. Combine that with pre existing light hit reg problems and unless they're standing still for me, I prefer to just run away and leave them for my American team mates with better pings.

And yes, there have been quite a few times where in close duels I KNOW I should have survived and killed my opponent but the latency pulled a Kanye West on me. "I'MA LET YOU FINISH YO KILL BUT HE HAS A BETTER CONNECTION THAN YOU!" which leads me to die rather unfairly.

Abso fricking lutely. It would be fine if EVERYONE had the same ping, but the difference makes it pointless for me to try anything but those "useless" LRMs my teammates keep slagging me for.

[David Sumner]

Bulletsponge0, on 13 March 2015 - 05:23 PM, said:

that isn't working for me..its not writing any file at all

"Access Denied" - don't write to the C drive root folder.

try tracert www.mwomercs.com > %USER%\results.txt

[David Sumner]

Tracing route to www.mwomercs.com [192.99.109.192]
over a maximum of 30 hops:

1 1 ms 1 ms 1 ms 192.168.0.1
2 165 ms 166 ms 129 ms llu.bng1.tvc.orcon.net.nz [60.234.8.4]
3 141 ms 148 ms 149 ms xe-3-3-0.cre1.sky.orcon.net.nz [121.98.9.72]
4 60 ms 67 ms 118 ms 121.98.9.137
5 167 ms 148 ms 147 ms ae1-0.cre2.nct.odyssey.net.nz [121.99.12.3]
6 162 ms 160 ms 117 ms ten-1-3-0.bdr01.alb01.akl.VOCUS.net.nz [175.45.93.37]
7 268 ms 274 ms 277 ms ten-0-7-0-1.cor01.akl05.akl.VOCUS.net.nz [114.31.202.121]
8 288 ms 207 ms 243 ms ten-0-0-1-3.cor01.lax01.ca.VOCUS.net [114.31.199.31]
9 258 ms 257 ms 263 ms bundle-100.cor02.lax01.ca.VOCUS.net [114.31.199.49]
10 262 ms 264 ms 265 ms bundle-100.cor01.lax01.ca.VOCUS.net [114.31.199.48]
11 264 ms 264 ms 264 ms bundle-101.bdr01.lax01.ca.VOCUS.net [114.31.199.61]
12 402 ms 279 ms 282 ms any2ix.coresite.com [206.72.210.214]
13 * * * Request timed out.
14 230 ms 231 ms 230 ms 192.99.146.86
15 234 ms 233 ms 233 ms 198.27.73.233
16 232 ms 230 ms 230 ms mwomercs.com [192.99.109.192]

Trace complete.


You can see the effect of my local exchange at the top.
C'mon, you gave the other half the street fibre this year, why do I have to wait another 18 months?

Then a big jump at VOCUS (international step?)

[9thDeathscream]

Honestly do you think they care. I don't. Unfortunate we high pingers are in the minority so were ******. Regional servers is the only answer, but that aint going to happen so why bother. Between. unresponsive mechs, broken HSR and woefull hit detection. Not to mention a ******* Match Maker that punishes you for winning. Its not worth it. My Kdr has slowly been going backwards. I am sick of only getting 100,000 cbills and 750Xp per match if lucky. As I said high pingers in the minority.

Its not worth it for them to worry about us 200+ people who suffer. STFU buy our mechs and play in CW.

PGI you have had years to try to fix these issues and there has been little to no progress. So i am out. Such a pitty great game ruined.

Edited by Akulla1980, 21 March 2015 - 09:05 PM.

[David Sumner]

Actually. Asking us is pretty much irrelevant.
If we have a ping, that ping is pretty much all we know.

You should have the stats to run some numbers.
That should tell you about likely "enjoyment" levels.
If ping is affecting weapon choice, it's limiting options,
if it's decreasing Elo, W/L, Avg Dmg, time in match, etc, then it's reducing the "fun factor" by making you spend more time "waiting" than "playing"
If it's increasing Matches to level, time to get First Win of the day, etc, it's increasing the "grind" factor.

Avg Ping vs Elo - assuming Elo not broken (see below)
Avg Ping vs W/L - Use Total Matchs vs Losses, not W/L because your web site mech stats prove that Losses are not always captured if you leave the match early, so the ratio is wrong. And this is why I suspect Elo is also wrong. And yes, this was still happening as of the start of March 2015.
Avg Ping vs weapon types
Avg Ping vs KDR
Avg Ping vs Avg Damage
Avg Ping vs Avg time in match
Avg Ping vs Avg Ping of enemy vs W/L
Avg Ping vs # of matches played to get "First Win Of The Day"
Avg Ping vs # of matches per day
Avg Ping in Matches vs Avg Ping in CW
Avg Ping vs # of matches to get a mech to Elite or Mastered
Avg Ping vs Avg XP per match

Edited by David Sumner, 22 March 2015 - 12:10 AM.