136 replies to this topic

[Carbon Guardian]

PGI listens way too much to this community. They need to let the community complain without being heard at all, until they make real progress on what they feel are the top priorities in this game. The problem from the start was open beta they let people unfamiliar with the development process of a game see the development process of this game. The community couldn't handle it, and PGI tried to please us. Rewriting code happens all the time in every game especially in F2P games, guys are just shocked cause PGI is being open about it.

I don't see a problem with them knowing that MM had to be rewritten for CW. They needed a match maker, or community would complain that they can't play the game.

[Aym]

Carbon Guardian, on 22 May 2014 - 10:10 PM, said:

PGI listens way too much to this community. They need to let the community complain without being heard at all, until they make real progress on what they feel are the top priorities in this game. The problem from the start was open beta they let people unfamiliar with the development process of a game see the development process of this game. The community couldn't handle it, and PGI tried to please us. Rewriting code happens all the time in every game especially in F2P games, guys are just shocked cause PGI is being open about it.

I don't see a problem with them knowing that MM had to be rewritten for CW. They needed a match maker, or community would complain that they can't play the game.

The problem is the vocal minority isn't surprised they need to rewrite the code, we're surprised they haven't started yet, and are in fact starting EARLY for 3/3/3/3 when they told us many times CW was around the corner, even 90 days away at some point, yet the code to allow CW to start won't be ready for at least a month from now, and that, let me repeat, is only because it is being pushed forward for 3/3/3/3.
Why haven't we seen an updated CW timeline?

[Kmieciu]

Wait. I thought the matchmaker has already been rewritten to allow 3/3/3/3 limit? What were they doing for the last few patches?

[Monsoon]

Aym, on 22 May 2014 - 11:25 PM, said:

The problem is the vocal minority isn't surprised they need to rewrite the code, we're surprised they haven't started yet, and are in fact starting EARLY for 3/3/3/3 when they told us many times CW was around the corner, even 90 days away at some point, yet the code to allow CW to start won't be ready for at least a month from now, and that, let me repeat, is only because it is being pushed forward for 3/3/3/3.
Why haven't we seen an updated CW timeline?

Well it seems like they're confident that they can rewrite the code for MM fairly quickly (~ one month apparently) so that pretty much negates your argument about, "when they told us many times CW was around the corner, even 90 days away at some point". If they don't see rewriting the MM as taking a great deal of time, then they could have been saving this for when they were in the final stages of CW, but since UI had so heavily impacted the release of CW, then the MM kept being put on the back-burner. Now they have a reason to push on ahead with it, that isn't CW related.

Edited by Monsoon, 22 May 2014 - 11:37 PM.

[Appogee]

Carbon Guardian, on 22 May 2014 - 10:10 PM, said:

PGI listens way too much to this community.

Heh.

Carbon Guardian, on 22 May 2014 - 10:10 PM, said:

They need to let the community complain without being heard at all, until they make real progress on what they feel are the top priorities in this game.

Ha Ha Ha!!

Carbon Guardian, on 22 May 2014 - 10:10 PM, said:

The problem from the start was open beta they let people unfamiliar with the development process of a game see the development process of this game.

HA HA HA HA HA HA HA HA HA HA HA!!!

Carbon Guardian, on 22 May 2014 - 10:10 PM, said:

The community couldn't handle it, and PGI tried to please us.

BWA HA HA HA *choke* HA HA HA HA *deep breath* HA HA HA HA HA *splutter* HA HA HA HA HA HA!!!

Edited by Appogee, 22 May 2014 - 11:42 PM.

[slide]

What surprises me about this is that Russ is surprised that when given an effective tool, the community can organise itself.

This community has been trying to organise itself for over 2 years with no tools and in some cases actual road blocks to doing so.

We have been asking for tools to do stuff for ages, now that we are finally getting some (private matches, queue counter) things are going to get moving.

Lesson Russ. Give us more tools, not less.

[PhoenixFire55]

tl;dr version

"Our MM is crap, we always knew it is crap, we were told countless times it is crap, but we still forced you to play with this crap for 1.5 years. People responsible for this crap couldn't do crap, so we now assign another person to sort out the crap."

Well ... good luck Karl, you seem like a person who actually wants smth to get done.

[Alik Kerensky]

Bartholomew bartholomew, on 22 May 2014 - 07:12 PM, said:

Solo puggers are not going to do CW anyways.

You can stick this bullocked opinion where the sun don't shine.

[OneEyed Jack]

Yeah... forcing people to not play the mechs they logged on wanting to play is "fun" and "viable."

The Clan releases are going to be very, very ugly affairs.

[Chemie]

If Karl is on it, I feel like it will get done and get done right.

[Chemie]

A few months ago, MM code was able ot handle balancing off number of chassis (it did not force 3/3/3/3 but did force each side to be the same). Your explanation makes no sense in that context as the code was there for balance.

Having said that, we all know the MM was broken, and even though it was supposed to balance weight it never did and often had very mismatched ELO to boot. Of course, I had assumed this was a function of the low queue count but maybe it really is the code (which is kind of a good thing).

[SgtKinCaiD]

First, thanks for trying to fix the matchmaker, it's a real burden in game.

But the 3/3/3/3 limit is not the answer. You don't get a better score because you're using a heavier mech.
The tonnage mismatch between 2 teams is an aggravating factor not the root of the problem.
I must admit though that 12 assaults vs 12 mediums will not end well so maybe class limit are needed but not to this extend.
The real problem as already stated is the difference of skills between players of the same team.

Actually, with the current matchmaker, this 2 match should be even :
1st match :
Team A = 12 players with an average ELO of 1400, ELO team sum up = 16800
Team B = 12 players with an average ELO of 1400, ELO team sum up = 16800

2nd match
Team A = 12 players with an average ELO of 1400, ELO team sum up = 16800
Team B = 4 players with an average ELO of 2500, 4 players with an average ELO of 1400 and 4 players with an average ELO of 300, , ELO team sum up = 16800

Guess what ?
Yeah maybe the first match will end with a 12-4 score, but it will be a landslide victory, one team slowly taking the advantage until the other can't fight back.
The second match ? A lame roflstomp ending in less than 6 minutes because it's not a 12v12 match but rather a 8vs12 : the 4 300 ELO players will get crushed in the first minutes of the match leaving their team with a missing lance.

Now why the new MM is better ?
If you read carefully the post about the launch module (http://mwomercs.com/...93-feb-27-2014/) and especially the paragraph "What the Solo Public Match Queue Looks Like", you will understand that for every match there is a ELO bucket associated and so the 24 players in the match are in the same ELO bucket.
And while the ELO sum up of each team should be different, the match is balanced because the 24 player have roughly the same skill.
Thanks to that PGI ditch the rule of mixing high ELO player with low ELO payer in order to have an average ELO score which is easier to match.
The match were better not only because of the 3/3/3/3 but because both team had *REALLY* the same skill across the board.

TL;DR
The current matchmaker only consider the ELO sump up of each team to balance a match. <= BAD
The new matchmaker only takes player with roughly the same skill to balance a match. <= GOOD

[Gorgo7]

SgtKinCaiD, on 23 May 2014 - 03:45 AM, said:

First, thanks for trying to fix the matchmaker, it's a real burden in game.

But the 3/3/3/3 limit is not the answer. You don't get a better score because you're using a heavier mech.
The tonnage mismatch between 2 teams is an aggravating factor not the root of the problem.
I must admit though that 12 assaults vs 12 mediums will not end well so maybe class limit are needed but not to this extend.
The real problem as already stated is the difference of skills between players of the same team.

Actually, with the current matchmaker, this 2 match should be even :
1st match :
Team A = 12 players with an average ELO of 1400, ELO team sum up = 16800
Team B = 12 players with an average ELO of 1400, ELO team sum up = 16800

2nd match
Team A = 12 players with an average ELO of 1400, ELO team sum up = 16800
Team B = 4 players with an average ELO of 2500, 4 players with an average ELO of 1400 and 4 players with an average ELO of 300, , ELO team sum up = 16800

Guess what ?
Yeah maybe the first match will end with a 12-4 score, but it will be a landslide victory, one team slowly taking the advantage until the other can't fight back.
The second match ? A lame roflstomp ending in less than 6 minutes because it's not a 12v12 match but rather a 8vs12 : the 4 300 ELO players will get crushed in the first minutes of the match leaving their team with a missing lance.

Now why the new MM is better ?
If you read carefully the post about the launch module (http://mwomercs.com/...93-feb-27-2014/) and especially the paragraph "What the Solo Public Match Queue Looks Like", you will understand that for every match there is a ELO bucket associated and so the 24 players in the match are in the same ELO bucket.
And while the ELO sum up of each team should be different, the match is balanced because the 24 player have roughly the same skill.
Thanks to that PGI ditch the rule of mixing high ELO player with low ELO payer in order to have an average ELO score which is easier to match.
The match were better not only because of the 3/3/3/3 but because both team had *REALLY* the same skill across the board.

TL;DR
The current matchmaker only consider the ELO sump up of each team to balance a match. <= BAD
The new matchmaker only takes player with roughly the same skill to balance a match. <= GOOD

You SO smart!

[SgtKinCaiD]

Gorgo7, on 23 May 2014 - 04:01 AM, said:

You SO smart!

Is it ironic because i tried to be constructive ? Or maybe i'm not clear enough (english is not my mother tongue) ?

[EyesBurn]

SgtKinCaiD, on 23 May 2014 - 04:22 AM, said:

Is it ironic because i tried to be constructive ? Or maybe i'm not clear enough (english is not my mother tongue) ?

He is just a 10 year old troll,dont bother with him

[SmurfOff]

I gotta say, I respect PGI for the ability to double-down on what seems to be a completely lost cause. Every time they have introduced a system that the community feels is wrong, they push through the fog and do what ever it is they feel is right. It takes a thick skin to ignore veterans, public test server data, feedback, and logic in a quest to develop their vision.

I look forward to the upcoming clan invasion, and can't wait to see how well thought out it will be.

[Innocent]

Hopefully with the admission that the legacy matchmaker code is bad they can move forward. It seems all they have been trying to do is polish a **** and there is just so much polish you can put on feces.

[Kekrebos]

I remember back in closed beta when it would just match weight classes. It wasn't perfect, but it was far better than the last half year of matchmaker 'improvements' we've had.

I play several games that are very similar to this. I'm not a big stat watcher, however I am aware of my stats so I know generally how I'm doing. I tend to do mediocre to decent. This is the only game that uses an ELO style matchmaking, which you would think would average everyone out to about a 50% win/loss rate given enough time. However I've noticed that progressively with each wipe of the stats I end up getting significantly worse stat wise. Maybe I've just been unlucky, but I honestly haven't been logging in much as I end up getting frustrated with constant lopsided losses.

I had high hopes for 3/3/3/3 fixing this, but that broke, and it seams to have gotten even worse. Now I have hope again that a complete overhaul of the matchmaker may make the game fun for me to play again. Thank you PGI for being open about this, it made me not walk away. Just please keep it simple. The more complicated you make it, the more chances to break or have outliers it will have.

Edit: quote fail.

Edited by Rallog, 23 May 2014 - 04:41 AM.

[Moku]

3/3/3/3 isn't a solution. Go back and tweak your matchmaking (ELO) formula includes some experience or battlevalues so newbies don't drop with pros.

[C E Dwyer]

Bartholomew bartholomew, on 22 May 2014 - 07:12 PM, said:

You are gonna need group support for CW anyways. So Roadbeer is right that you might as well get it wrote in and ironed out.

Solo puggers are not going to do CW anyways.

Define CW before you make sweeping statements like that

If you mean private matches with no effect on the world map your probably correct.

If you mean solo puggers don't want to have any effect on the interactive galactic map then I have to say that your talking crap as the only reason I have stayed and tolerated dull arena matches, brainless skirmish death mode, and continued to fund this game is in the hope this autumn there is something worth fighting over