81 replies to this topic

[Serapth]

Just logged in, played 12 matches, was top 2 in damage ( on my team ) on all of them, havent won a match.

F'it, this is just simply not fun. Especially when you look around and see three of your teammates running in ****** ( 3PV ) mode.


The problems go soooooo far beyond weight limits. Premade/Pug balance is certainly a big factor. ELO hell is another. Not taking into account customization is another. Trial mechs are another. All combined, they've made a situation that is increasingly less fun.

Frankly I would take that 2/2/2/2 proposal suggested earlier over the complete {Scrap} we have now.

Edited by Serapth, 08 December 2013 - 02:05 PM.

[ShinVector]

Atheus, on 08 December 2013 - 09:27 AM, said:

Was basically a shooting gallery for you there, huh. I'm amazed you weren't getting smoked, though. I assumed you were in a 3L for the first 2/3 of that since nobody seemed to be able to see you, but noticed toward the end that it was a 2X, lol. Nicely played.

Yeah... Long range kiting, hiding, pecking on people while they are distracted and etc... Really only possible in big maps..
Abusing the terrain as well.. Disabling the Atlases before they can do real damage..

The 2X has some special properties.. The TAG torso hardpoint has only 1 point so, your lasers come out the same place. This allows a low chance of your shots getting block by terrain because it is so, close to the cockpit view.

Must have been a horrible for the Atlases.. Believe there were 4 in a pre-made group. The Spider's base cap distraction and assists was perfect as well. I still LOL every time I see the synchronised laser firing against the last Atlas. Just to show what PUG gets you sometimes.

Edited by ShinVector, 08 December 2013 - 05:06 PM.

[Troa Barton]

Not all premades are created equal..
You probably had a very high elo group on your team that all got out their "choice" mechs to grind or mess around. The resulting elo power vaccum has broken many matches.
Also you guys could have split up, used poor judgement or any other method to your destruction.
Weight balance wont solve anything.
Applying a spreadsheet to players wont accomplish anything.
A good jenner pilot will ace an atlas. 100 tonnes VS 35. A 65 ton weight "imbalance" come on man.
FFS people its not the weapons, the weight imbalance, or the meta. You cant perfectly balance players.
Elo kind of works but needs to be calculated to the individual mech that pilot is using. A flat number is just another "spreadsheet" time saver tactic just like wight balancing.

[Jun Watarase]

Merchant, on 07 December 2013 - 03:41 PM, said:

No, not an example of what you say. There are people who can hit me in any Mech they drive, just this Atlas pilot kept shooting late.

First you ignore player skill then later in the same post comment on bad skill claiming bad pilots should not be playing.

Seems you pick and choose whatever supports your belief, a wrongful way of looking at things.

I dont know what you are talking about. Player skill does not determine whether you can hit fast mechs, your ping does. Go try to kill fast mechs with 300 ping and come back to me.

And bad players should be not in high ELO matches, i dont know why you are implying this is wrong. High ELO matches always have newbies running in trial mechs or with stock/bad loadouts and they do nothing but die pointlessly. They should not even be in those matches. This is not difficult to understand.

Everyone here knows perfectly well that the matchmaker is putting high ELO players (Especially in premades) against newbies on a regular basis. I have yet to see any explanation why high ELO players should be getting matched up against newbies....all i hear are excuses for why team A should be stomping team B. Which always seem to boil down into "you are not in a premade, its YOUR fault!".

The lack of lag compensation and wonky hitboxes has also unfortunately resulted in a situation where light mech pilots think they have amazing skill because multiple assaults can unload on them to no effect. A RNG based mechanic where your shots miss because the game decides "noooo it didnt hit! i swear!" is in no way a reflection of skill. Why do assaults cost like 4 times a light mech....because in both the fluff and TT, an assault will completely wipe the floor with a light mech. Light mech piltos should be ******** themselves if they are soloing a assault, not going "lol xd the lag shield makes me invulnerable hahaha look at all his shots miss". There is no reason to pay 10 million + c-bills for an assault when a light mech costs what, 3-4 million fully kitted out, and a light mech can down multiple assaults due to no lag compensation.

It is almost the year 2014, this is an online shooter, if my crosshair is over an enemy and i fire, i should hit, not have it miss due to lag.

For reference, if a light mech were to try and solo an assault in the TT, the assault would simply stand with his back to something and hit the light mech with a 40% or so chance to hit...including with one shot weapons like the AC20. This forced players to use heavies and assaults to distract the assaults while a light went for rear shots. In MWO that chance is pretty much zero unless you have low ping, lasers or missles, so a light mech can just run at an angle towards a team of assaults and live. Which again, is not skill, but relying on the game's lack of a key core feature.

A light mech can tank more firepower than a lance of Atlases due to no lag compensation. This is extremely wrong. Even my XL engine shadowhawk at 95 kph survives more firepower than my standard engine Atlas.

And seriously when is PGI going to stop allowing bad players to rig matches by stacking teams....there are some people here who think they are the best players in the game but they literally CANNOT do anything by themselves....all they can do is sync drop in optimized loadouts (stacking ECM, AC40s, PPC/Ballistic sniper loadouts, etc) and bash newbies. Ive seen the same people drop sometimes without their premades and they always die with less than 200 damage done. The exact same thing happens when their sync drop fails and they end up in a different match....they die like flies. You know who you are.

Edited by Jun Watarase, 08 December 2013 - 10:27 PM.

[Vanguard319]

Serapth, on 08 December 2013 - 01:48 PM, said:

So basically what you are saying is PUGS could become premades and since they dont, f 'em.

To which I reply, f you.

Dropping premade gives a big advantage. With the recent spawn point change it gives a huge advantage. This is simply called a broken game mechanic, nothing more. That you enjoy exploiting it does nothing to help the long term viability of the game. Add to that ELO hell, were you can play 10 or so matches in ROFLstomps, where you are the highest scoring player on the losing side match after match after match. Game is broken, weight is the least of the problems.


Add lobbys, in lobby give the ability to specify PUG ONLY, even if it increases match wait times, it solves the majority of problems.

Only someone who refuses to adapt would call the random drops a "broken mechanic". As far as lobbies go, it is no different than running Teamspeak, logging into Comstar NA or NGNG, and placing yourself in that little channel labeled "looking for group". What you are proposing is essentially what I said: Pugs using the same kinds of tools as premades, and in essence becoming premades themselves. Where we disagree is your insistance that these features MUST be integrated into the game to be useful.

F me you say? I take no offense, since I have no problem using a VOIP and taking the time to join a group, I'll just get my revenge by crushing you ingame.

[Dirty Old Man]

I PUG a lot most of the time... but I must say... the waiting time does seem to dictate the skillz you might get most of the time of the other puggers.

But all in all, what makes this game so freaky awesome was... one waited and dropped into a game where half the team dies off running to engage the enemy head on, then leaving 6 of random puggers from different factions holding off the onslaught and coming out 12 kills vs 6 dead... winning the game .... with only 1minute to spare on the clock... that is a classic...

We seemed to "clicked" without Pre made, without typed commands... without VOIP.... but somehow.. we knew we had to make the last stand ... and it paid off....

[Atheus]

Here you go: 975 tons vs. 835 tons. A 17% difference in weight, mwehehe.



On second viewing, it was actually 975 vs. 745 when you subtract out the guy who disconnected at the start (a 31% difference in weight), but the match was fairly close all things considered. I wonder how matchmaker calculated ELO for this match... was our team expected to win?

Edited by Atheus, 09 December 2013 - 09:11 AM.

[Wildstreak]

Jun Watarase, on 08 December 2013 - 09:59 PM, said:

I dont know what you are talking about. Player skill does not determine whether you can hit fast mechs, your ping does. Go try to kill fast mechs with 300 ping and come back to me.

Now you are just being arrogant and selectively ignoring skill again.

Jun Watarase, on 08 December 2013 - 09:59 PM, said:

And bad players should be not in high ELO matches, i dont know why you are implying this is wrong. High ELO matches always have newbies running in trial mechs or with stock/bad loadouts and they do nothing but die pointlessly. They should not even be in those matches. This is not difficult to understand.

Everyone here knows perfectly well that the matchmaker is putting high ELO players (Especially in premades) against newbies on a regular basis. I have yet to see any explanation why high ELO players should be getting matched up against newbies....all i hear are excuses for why team A should be stomping team B. Which always seem to boil down into "you are not in a premade, its YOUR fault!".

Mixed ELO players on same team, I never cease to be amazed at how complaints about this do not take into account 2 factors that MUST be considered.
1 - Amount of players online at any time looking for a match.
2 - Amount of players who admit on these forums they left and / or left, came back, did not like what they saw, left again.
THESE 2 ITEMS ARE NOT DIFFICULT TO UNDERSTAND AND CONSIDER BUT KEEP GETTING IGNORED ESPECIALLY BY 'TOP TIER' PLAYERS AS THEY CALL THEMSELVES.
Those 2 items affect who matchmaker has available to pit against each other, as it takes more time to put someone in a match, it will start, to put it one way, 'lowering standards.' Thus when you consider those 2 items I listed, it is possible they contribute to why you have different ELO in the same matches.

But no, it is SOOOO much eaiser to blame 'the system,' right?

If you want a better solution, it looks like they are consdering another match type soon, perhaps you can ask for Lance vs Lance combat. 4 vs 4 is a typical BT scenario and has less chance of putting different ELO on the same team, else you are stuck with 12 man premades.

Jun Watarase, on 08 December 2013 - 09:59 PM, said:

The lack of lag compensation and wonky hitboxes has also unfortunately resulted in a situation where light mech pilots think they have amazing skill because multiple assaults can unload on them to no effect. A RNG based mechanic where your shots miss because the game decides "noooo it didnt hit! i swear!" is in no way a reflection of skill. Why do assaults cost like 4 times a light mech....because in both the fluff and TT, an assault will completely wipe the floor with a light mech. Light mech piltos should be ******** themselves if they are soloing a assault, not going "lol xd the lag shield makes me invulnerable hahaha look at all his shots miss". There is no reason to pay 10 million + c-bills for an assault when a light mech costs what, 3-4 million fully kitted out, and a light mech can down multiple assaults due to no lag compensation.

It is almost the year 2014, this is an online shooter, if my crosshair is over an enemy and i fire, i should hit, not have it miss due to lag.

You are the first person TMK who has complained about lag in a long time especially since HSR was done. Just what 'wonky hitboxes' are you complaining about? On Lights the only ones I know get mentioned are Locust legs being easier to destroy then they should and Spiders. Stop speaking generally and get specific especially since you are, TMK, the only one complaining about Lights in general.

Jun Watarase, on 08 December 2013 - 09:59 PM, said:

For reference, if a light mech were to try and solo an assault in the TT, the assault would simply stand with his back to something and hit the light mech with a 40% or so chance to hit...including with one shot weapons like the AC20. This forced players to use heavies and assaults to distract the assaults while a light went for rear shots. In MWO that chance is pretty much zero unless you have low ping, lasers or missles, so a light mech can just run at an angle towards a team of assaults and live. Which again, is not skill, but relying on the game's lack of a key core feature.

I already know how Lights work in TT and you got some of that wrong.

Jun Watarase, on 08 December 2013 - 09:59 PM, said:

A light mech can tank more firepower than a lance of Atlases due to no lag compensation. This is extremely wrong. Even my XL engine shadowhawk at 95 kph survives more firepower than my standard engine Atlas.

Did it ever occur to you that one of the details for this is based on the Mech you are driving? Lately I have seen in many matches an Atlas shows up, the other side gets some players who gang up on it. It is called target priority and not everyone's is the same as you. Probably the fact that it is an Atlas with a toughness rep so people try to eliminate it either for its threat value or just to prove to themselves they have a big e-peen. I'm not one of those but it does not mean I limit my perspective to only MY viewpoint, beliefs and playstyle. Why not try asking people to pick target priority out of some groups of Mechs and see what they say? Discussion provides interesting revelations.

Jun Watarase, on 08 December 2013 - 09:59 PM, said:

And seriously when is PGI going to stop allowing bad players to rig matches by stacking teams....there are some people here who think they are the best players in the game but they literally CANNOT do anything by themselves....all they can do is sync drop in optimized loadouts (stacking ECM, AC40s, PPC/Ballistic sniper loadouts, etc) and bash newbies. Ive seen the same people drop sometimes without their premades and they always die with less than 200 damage done. The exact same thing happens when their sync drop fails and they end up in a different match....they die like flies. You know who you are.

When is PGI going to stop allowing some people to post as if they were experts at everything under the sun? The amount of wrongful things I have seen posted is crazy such as those who claim to know every detail about what affects this game yet obviously and selectively ignore certain details.

[Thorqemada]

The MM is pretty bad often - i dont see any advantage over the old weightclass MM - the amount of Roflstomps is equal if not increasing as Players play the ELO and the matches with a Tonnage/Player mismatch seem to increase.

Much work to be done here...

[hercules1981]

Jun Watarase, on 08 December 2013 - 09:59 PM, said:

I dont know what you are talking about. Player skill does not determine whether you can hit fast mechs, your ping does. Go try to kill fast mechs with 300 ping and come back to me.

And bad players should be not in high ELO matches, i dont know why you are implying this is wrong. High ELO matches always have newbies running in trial mechs or with stock/bad loadouts and they do nothing but die pointlessly. They should not even be in those matches. This is not difficult to understand.

Everyone here knows perfectly well that the matchmaker is putting high ELO players (Especially in premades) against newbies on a regular basis. I have yet to see any explanation why high ELO players should be getting matched up against newbies....all i hear are excuses for why team A should be stomping team B. Which always seem to boil down into "you are not in a premade, its YOUR fault!".

The lack of lag compensation and wonky hitboxes has also unfortunately resulted in a situation where light mech pilots think they have amazing skill because multiple assaults can unload on them to no effect. A RNG based mechanic where your shots miss because the game decides "noooo it didnt hit! i swear!" is in no way a reflection of skill. Why do assaults cost like 4 times a light mech....because in both the fluff and TT, an assault will completely wipe the floor with a light mech. Light mech piltos should be ******** themselves if they are soloing a assault, not going "lol xd the lag shield makes me invulnerable hahaha look at all his shots miss". There is no reason to pay 10 million + c-bills for an assault when a light mech costs what, 3-4 million fully kitted out, and a light mech can down multiple assaults due to no lag compensation.

It is almost the year 2014, this is an online shooter, if my crosshair is over an enemy and i fire, i should hit, not have it miss due to lag.

For reference, if a light mech were to try and solo an assault in the TT, the assault would simply stand with his back to something and hit the light mech with a 40% or so chance to hit...including with one shot weapons like the AC20. This forced players to use heavies and assaults to distract the assaults while a light went for rear shots. In MWO that chance is pretty much zero unless you have low ping, lasers or missles, so a light mech can just run at an angle towards a team of assaults and live. Which again, is not skill, but relying on the game's lack of a key core feature.

A light mech can tank more firepower than a lance of Atlases due to no lag compensation. This is extremely wrong. Even my XL engine shadowhawk at 95 kph survives more firepower than my standard engine Atlas.

And seriously when is PGI going to stop allowing bad players to rig matches by stacking teams....there are some people here who think they are the best players in the game but they literally CANNOT do anything by themselves....all they can do is sync drop in optimized loadouts (stacking ECM, AC40s, PPC/Ballistic sniper loadouts, etc) and bash newbies. Ive seen the same people drop sometimes without their premades and they always die with less than 200 damage done. The exact same thing happens when their sync drop fails and they end up in a different match....they die like flies. You know who you are.

If Ping is causing such a problem for u because it's around 300 or so, don't play the game period. Every one knows there r hit detection problems with hsr wait till u read up on it getting better or play a few matches after a new patch comes out. Complaining about it is just a waste just like your LRMS r useless post!

[aniviron]

I know it doesn't happen much, just venting my frustration at being outtonned 930 to 735, a nearly 200 ton difference. That's all. It was a brutal five-minute *** whooping (with 11 of us dead for three of that) that seemed to be unbalanced by more than just them having the equivalent of 2 extra atlases of weapons, as well.

[hercules1981]

aniviron, on 12 December 2013 - 01:20 AM, said:

I know it doesn't happen much, just venting my frustration at being outtonned 930 to 735, a nearly 200 ton difference. That's all. It was a brutal five-minute *** whooping (with 11 of us dead for three of that) that seemed to be unbalanced by more than just them having the equivalent of 2 extra atlases of weapons, as well.

I honestly think the 11, 7 , 7 , and zero damage where to blame more then tonnage bro but good try.

[aniviron]

hercules1981, on 12 December 2013 - 07:05 PM, said:

I honestly think the 11, 7 , 7 , and zero damage where to blame more then tonnage bro but good try.

Like I said, a lot went wrong with the match; 11 players died in under three minutes. Weight definitely played factor, but as I said, it "seemed to be unbalanced by more than just them having the equivalent of 2 extra atlases of weapons."

[Albert Meyburgh]

Atheus, on 04 December 2013 - 01:49 PM, said:

I'll take a stab at why "ELO" and matchmaker are fairly helpless against this game. There are a CRAZY number of variables to determine how well a given player will perform. Just to name a few:

• Pilot efficiencies - an experienced player in a brand new mech will perform at about 50% of their ability, simply because their mech is too slow, and too hot without efficiencies.
• Loadout - even if you're generally good at building mechs, players are forced to play in variants they wouldn't personally want to pilot in order to build out elite on a chassis.
• Upgrades - Some players won't bother with the 1.5-3 million in basically mandatory upgrades on chassis variants that they don't particularly like, hoping their team will carry them through basic efficiencies.
• Consumables - these powerful modules can have a huge effect on a battle when used at the right time. ELO doesn't evaluate separately how you perform when using air/artillery strikes vs. not, then evaluate your loadout, etc.
• Chassis - some variants are just plain strong, while others are not. Matchmaker isn't going to give you an ELO boost just because you're in your Highlander 733C, then drop it down super low because you're in an Awesome. They use the same ELO figure.
• Environment compatibility - We still don't get to pick what map we play on, but there's no secret that ballistics are particularly strong on hot maps, or that LRM's are particularly strong on alpine and caustic. How exactly do you compensate for that, though, when each mech might be carrying all 3 of those types of weapons in different mixtures? You can try, but as the meta shifts the parameters for weapon affinity will shift with them.

In a game where people all use basically the same equipment, ELO is pretty effective. MWO has a massive array of variables to account for, though, and quantifying them is pretty hopeless. Players will have an affinity for one system over another, while another player will be just the opposite, and the game is never going to be able to compensate for player P trying out the gauss rifle when he's actually really bad at using the gauss rifle, or player H loading up a mech that runs a little hot, then discovering he's landed on Terra Therma. All PGI can hope for is that players of a certain ELO will randomly land both strong and weak loadouts on both teams. That said, perhaps using "big data" you can evaluate all matches and come up with some sort of ultra-advanced ELO that takes every little detail into consideration and comes up with a loadout compensated ELO customized for each player's weapon affinities. It's certainly not impossible for something like that to happen, but for that to happen PGI would need to have Gandalf and Spock and also probably a good fortune teller on their staff. If I were Spock, though, I would probably apply to another company before PGI, though.

Nice analysis! I had pitched using a bloom filter/histogram approach to make the ELO decoupled from JUST weight class in order to increase ELO indexing granularity to address the plethora of relevant parameters (some you mentioned, but there are more of course) but improvements to this (as well as other fun stuff that I couldn't possibly comment on) are on hold while UI changes in the lobby system are released as we only have so many hours in the day :/

Thanks for all those providing critical thinking in their feedback, we don't always respond, but we do 'lurk' the threads!

[Jun Watarase]

Albert Meyburgh, on 13 December 2013 - 12:56 PM, said:

Nice analysis! I had pitched using a bloom filter/histogram approach to make the ELO decoupled from JUST weight class in order to increase ELO indexing granularity to address the plethora of relevant parameters (some you mentioned, but there are more of course) but improvements to this (as well as other fun stuff that I couldn't possibly comment on) are on hold while UI changes in the lobby system are released as we only have so many hours in the day :/

Thanks for all those providing critical thinking in their feedback, we don't always respond, but we do 'lurk' the threads!

I dont know what that means. Can you explain why high ELO players in premades keep getting matched up against random newbiess in trial mechs please, in simple english? The 12-0 stomps are not fun for anyone except the players rigging matches for free wins.

Edited by Jun Watarase, 13 December 2013 - 08:38 PM.

[Tabrias07]

Albert Meyburgh, on 13 December 2013 - 12:56 PM, said:

...
improvements to this (as well as other fun stuff that I couldn't possibly comment on) are on hold while UI changes in the lobby system are released as we only have so many hours in the day :/

It worries me that literally everyone is too busy with the new store to tweak the broken matchmaker.

[Albert Meyburgh]

Tabrias07, on 14 December 2013 - 01:11 AM, said:

It worries me that literally everyone is too busy with the new store to tweak the broken matchmaker.

I didn't mean to imply that everyone is busy working on the new store. I, for example, don't work on that at all. If a feature requires UI, and the guys that work on UI are busy already, then we have to add the work to their queue and so the feature can't be released until their other work finishes. While we wait for that, we of course work on other things, whether they be features, or concept prototypes, or fixes, or optimizations or what have you!

Pardon my clumsy communication as I am an engineer not a poet.

[Kamenjar]

hercules1981, on 12 December 2013 - 07:05 PM, said:

I honestly think the 11, 7 , 7 , and zero damage where to blame more then tonnage bro but good try.

Total BS. I just had a match with the winning team having 1/3 tonnes more. The 4-atlas lance pushed immediately and their team followed, completely roflstomping us. There was absolutely nothing we could've done even with pro players on our side.

If one guy did zero damage, he soaked up some of it, if nothing and team could've compensated. With 1/3 of tonnage more you need more than 3 players to be total idiots to balance.

Here's the match:


You can see how there are more than a couple of ways that you can rearrange lances in order to balance these tonnages.

I'm not sure why this is even up for debate. The matchmaker is broken - PERIOD and there's no amount of gold mech's that players can buy to compensate for the lack of programming skills in PGI.

Edited by Kamenjar, 16 December 2013 - 07:40 PM.

[Jun Watarase]

Albert Meyburgh, on 16 December 2013 - 10:24 AM, said:

I didn't mean to imply that everyone is busy working on the new store. I, for example, don't work on that at all. If a feature requires UI, and the guys that work on UI are busy already, then we have to add the work to their queue and so the feature can't be released until their other work finishes. While we wait for that, we of course work on other things, whether they be features, or concept prototypes, or fixes, or optimizations or what have you!

Pardon my clumsy communication as I am an engineer not a poet.

Why are you dodging my question?

[ShinVector]

Jun Watarase, on 16 December 2013 - 08:33 PM, said:

Why are you dodging my question?

Let's not scare away people who are willing to communicate shall we ?