Jump to content

Matchmaker Feedback


396 replies to this topic

#381 Tulip Lee

    Member

  • Pip
  • Bad Company
  • Bad Company
  • 10 posts
  • LocationTennessee

Posted 20 May 2014 - 06:50 PM

My matchfinding seems to be working fine now. Maybe you have to quit and relog?

#382 draiocht

    Member

  • PipPipPipPipPipPipPip
  • 791 posts

Posted 20 May 2014 - 07:18 PM

I've had highly variable wait times from a lucky several seconds wait to an over 15 minute wait...

but the worst of it: these matches, win or lose, have been amongst the shortest, most unsatisfying landslides I've ever played.

Population problem?

[edit: or maybe i've had a string of fluke matches? bbl]

[edit: 3 hours later, and wait times are back down to a couple minutes and my games have returned to vaguely tolerable rolls rather than the vexingly dysfunctional rolls. All is well?]

Edited by draiocht, 20 May 2014 - 10:46 PM.


#383 Kali Rinpoche

    Member

  • PipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 639 posts
  • LocationCrossing, Draconis March

Posted 20 May 2014 - 07:38 PM

Drops happening much faster now it seems.
Looks like 4x3 is out.

Edited by Kali Rinpoche, 20 May 2014 - 07:47 PM.


#384 Caswallon

    Member

  • PipPipPipPipPipPipPip
  • Survivor
  • Survivor
  • 540 posts
  • LocationArboris

Posted 21 May 2014 - 02:16 AM

View PostNikolai Lubkiewicz, on 20 May 2014 - 04:26 PM, said:


As this so far seems to be more related more to the logic of the algorithm than with the hard rules that algorithm places, which are testing of different class limits seems to suggest: The issues encountered today would have theoretically appeared in the use of tonnage limits, BV values, or just about any other conceivable system. We hope to get these logical loops worked out ASAP.

Sooo are you saying whatever you did to matchmaker it would have broke? Not a good way to inspire confidence sir.

OK revolutionary thought here for you to ignore

The Problem:
Cant test accurately due to smaller size of tesy community.
Cant garner best data due to smaller duration of tests.


Solution:
Extend test duration
Offer 1-5 MC to the accounts of all those that test on a per match basis.

100% Guaranteed to increase the test server take take up!

Now played a single match in a Griffen today after reading the addition 15 pages on this "feedback" thread. To see for myself before posting further.
2 Min wait time match had NO evidence of 3/3/3/3 in it. We won 9-3 on Frozen Night. Pretty GG the losers fought hard and did not make any noobish mistakes that I could see they just got a tad unlucky.

Result: well done that unnamed engineer hitting the "PATCH ABORT" Button!



Oh and interestingly as I played on my XP partition for the first time in an age (Same computer as I usually use - dedicated Hard drive with different OS on it) FPS was UP!!!! back to 45 or so whereas on the Win 7 side with newer drivers and so forth I get 15 -20. Go figure that one out. I can't; my mind is offically boggled to think that maybe PGI uses XP as some of their test machines OS'...

#385 DeathlyEyes

    Member

  • PipPipPipPipPipPipPip
  • The Messenger
  • 940 posts
  • Google+: Link
  • Facebook: Link
  • Twitter: Link
  • Twitch: Link
  • LocationMetaphorical Island somewhere in the Pacific

Posted 21 May 2014 - 04:05 AM

View PostSLDF DeathlyEyes, on 21 May 2014 - 03:48 AM, said:

Instead of mucking around with 3/3/3/3, which seemed like a doomed concept from the start, I suggest you focus your attention elsewhere. A dropship mode was discussed in various interviews early on in development. Adding this game mode would alleviate a lot of strain on the matchmaker. Instead of trying to match tonnages a tonnage limit could be instituted in a players dropship. A player could easily be restricted to 200 tons or any other weight. He/she would then be responsible for tradeoffs such as having more mechs or having the ability to take a heavier mech.

This game mode would also have numerous beneficial side effects. One such side effect is reducing the effectiveness of ammo based mechs. Autocannons would no longer be as much of an issue. The reason they were niche weapons in battletech is because engagements were often long drawn out conflicts. With the prospect of 48 mechs attacking from the other team, ammo dependent builds would become less desirable.

This would also reduce the tendency of matches to be one sided. A major reason for the one sided matches is that often times a few mechs are lost early. This reduces the targets and firepower of one team significantly allowing the remaining players to be focused down. With dropship mode, players would be able to drop into combat again with a new mech bringing the fighting strength of their team back to par with the other.

Here are some of my personal suggestions for game play rules:

Skirmish: Deathmatch style combat with no focus. 30 minute time limit. If not all mechs are destroyed on one team, the winner will be determined by the amount of tonnage destroyed.

Conquest: A basic 30 minute match of conquest. Double or even triple the resource limit.

Assault: 30 minute match of dropship assault. Both dropships are on the ground, winner destroys the other teams dropship or all their mechs. If time expires winner is the one who destroyed the most tonnage, just the same as skirmish.


Dropship mode would create incentives for players to use a variety of mechs without instituting strict tonnage limits, allowing players to play how they want with various pros and cons. It would remian true to sim style combat without becoming an overly arcadey respawn deathmatch. As it stands the no respawn model in pug queue is failing miserably. I think it's time to try something else.

Finally. If the issue with this mode is a monetary reason, ie PGI can't justify making this game mode because it feels it would not profit from it. Look at it this way. As it stands I only have 1 set of modules that I move from mech to mech. I only have a few engines, most of my mechs share engines and I swap them around. I only have a few mechs because those are the mechs that actually work right now. By adding this game mode you are encouraging players to keep multiple sets of modules and pretty much forcing players to keep multiple mechs stocked. The tonnage limit means mechs that aren't necessary optimal for their weight class could be desirable to fill out those last few left over tons. This means more premium time will be used to earn Cbills. More mechs will be bought for MC and more mech bays will be needed to house those mechs.

http://mwomercs.com/...33-alternative/

Post here share your thoughts!

Edited by SLDF DeathlyEyes, 21 May 2014 - 04:09 AM.


#386 JHackworth

    Member

  • PipPipPipPipPip
  • Bad Company
  • Bad Company
  • 106 posts

Posted 21 May 2014 - 04:52 AM

has PGI even simulated this? It ought to be testable; we don't actually need real live players to simulate matchmaker on 3333. There should be enough match data to do this.

#387 Karl Streiger

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPipPip
  • Wrath
  • Wrath
  • 20,369 posts
  • LocationBlack Dot in a Sea of Blue

Posted 21 May 2014 - 05:28 AM

View PostNikolai Lubkiewicz, on 20 May 2014 - 04:26 PM, said:

As this so far seems to be more related more to the logic of the algorithm than with the hard rules that algorithm places, which are testing of different class limits seems to suggest: The issues encountered today would have theoretically appeared in the use of tonnage limits, BV values, or just about any other conceivable system. We hope to get these logical loops worked out ASAP.


Thats interesting but i really ask myself how this logic can affect complete different systems, that should work in a different manor.
Currently the 3333 system has 3 variables maybe 4 with the inclusion of the premade teams. a tonnage or BV system - may have only 1 - if done right.
But when the logic flaws should also hit those systems it must be a very very basic failure. Something so deep in the code that it hurts to change it - anyhow i wish you luck - and hope you give us an update.
I'm pretty sure hundreds of "comunity" members would help you with the Algorithm - (not the Code) - if you would just take the help

#388 mailin

    Member

  • PipPipPipPipPipPipPipPipPip
  • Big Daddy
  • Big Daddy
  • 2,033 posts
  • LocationWisconsin, USA

Posted 21 May 2014 - 06:00 AM

I had a drop at about 5 am CST on 5-21-2014. A friend and I were in lights and somehow MM decided that we needed to have a total of 6 lights on our team and one Cicada. How does this happen?

I don't know what PGI has done to the Mathmaker system, and honestly I don't care. All I DO know is that between uneven weights on opposing teams and now the dramatically increased wait times, there are a LOT of people who are really frustrated with this game.

I used to be able to say to myself, I have 15 minutes to play, I can get in a drop or two, but now if I only have 15 minutes I'll go do something else.

#389 FatYak

    Member

  • PipPipPipPipPipPipPip
  • Ace Of Spades
  • 585 posts

Posted 21 May 2014 - 06:24 AM

I just don't understand how PGI can continually screw this up. When your customers just expect failure every time your on the verge of losing them for good

Wake up to yourselves PGI

#390 Leroifou

    Member

  • PipPipPipPipPip
  • Giant Helper
  • 126 posts

Posted 21 May 2014 - 06:32 AM

Solo Light Mech. 8pm EST.
10 minutes then Cancelled.
10 minutes then got a match.
1 match/20 minutes...

#391 Sjorpha

    Member

  • PipPipPipPipPipPipPipPipPip
  • Philanthropist
  • Philanthropist
  • 4,478 posts
  • LocationSweden

Posted 21 May 2014 - 06:45 AM

Tried about 10 drops today, seems to be the same as before. No 3333 or anything, I must've missed the test and now it's reverted? Load times were fine.

#392 Shae Starfyre

    Member

  • PipPipPipPipPipPipPipPip
  • The Widow Maker
  • The Widow Maker
  • 1,429 posts
  • LocationThe Fringe

Posted 21 May 2014 - 07:41 AM

Less and Less people are playing until this is fixed (wait times extending); drop ELO, erase it completely; complete random drops, 1 (up-to 4 Mechs) team per side.... GO!

Any other system will have adverse effects on wait times and will never account for the skill of a pilot, the hardware he runs on, or communication; balance weapons, mechs, roles, maps ... there is no need for limitations to who drops and in what...people will learn from inuitive design, meta will be abolished, and balance will be achieved.

It is this simple (besides, until Clans hit the scene, everything you do referencing balance is moot); we need all aspects of the game in play before you can assess balance.

Listen to no one else; get back to basics; pump out the clan warfare; drop the balancing act on drops (we know it is temporary); get back to a random mix of everyone playing, and then balance the game starting not with who drops with who, but as I stated above....Mechs, Weapons, Maps, Scaling of Mechs, etc..

ELO can never appreciate who we are! 3/3/3/3 can never anticipate the depth of our passion... we want to play, we want mechs, weapons, maps, and diversity...challenge.

Work out the details later..when you have access to all the details.

Edited by Aphoticus, 21 May 2014 - 07:54 AM.


#393 Bigbacon

    Member

  • PipPipPipPipPipPipPipPipPip
  • The 1 Percent
  • The 1 Percent
  • 3,096 posts

Posted 21 May 2014 - 07:44 AM

View Postmailin, on 21 May 2014 - 06:00 AM, said:

I had a drop at about 5 am CST on 5-21-2014. A friend and I were in lights and somehow MM decided that we needed to have a total of 6 lights on our team and one Cicada. How does this happen?

I don't know what PGI has done to the Mathmaker system, and honestly I don't care. All I DO know is that between uneven weights on opposing teams and now the dramatically increased wait times, there are a LOT of people who are really frustrated with this game.

I used to be able to say to myself, I have 15 minutes to play, I can get in a drop or two, but now if I only have 15 minutes I'll go do something else.


one of the two rounds I had yesterday we had 7 lights (1 spider, 3 firestarters, 1 jenner, and 2 ravens I think), 1 cicada, and the rest were heavy/assault.

AND!!!!

it was a fantastic round...very close and edged out an win in the last 30 seconds (time did run out) but the ECM cicada was able to get into their base range and take out a badly damaged mech and then run like hell with the other guys who was still alive. edged out a 10-9 win.

Edited by Bigbacon, 21 May 2014 - 07:46 AM.


#394 Zolaz

    Member

  • PipPipPipPipPipPipPipPipPip
  • Elite Founder
  • Elite Founder
  • 3,510 posts
  • LocationHouston, Tx

Posted 21 May 2014 - 07:53 AM

Posted Image

#395 Hellen Wheels

    Member

  • PipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 1,326 posts
  • LocationDraconis March

Posted 21 May 2014 - 08:19 AM

View PostAntharPrime, on 20 May 2014 - 12:59 PM, said:


Really hard on those of us that have been hoping against hope for it to turn around.

Isn't it though? To their credit, something has been done to alleviate the wait times, I'm down to 2 minutes or less this morning to find matches with our European, Russian, and Asiatic friends. Five matches found within an hour today, all within 2 minutes or less. Hooray for "progress" of two steps forward, three steps back.

#396 Bigbacon

    Member

  • PipPipPipPipPipPipPipPipPip
  • The 1 Percent
  • The 1 Percent
  • 3,096 posts

Posted 21 May 2014 - 10:25 AM

wait times are back to normal for me today. that is a plus.

all rounds were losses by extreme stompings but least I am having fun again and being able to play.

Still think AC/2s have some wonky stuff going on that I am not seeing with the AC5.

#397 VXJaeger

    Member

  • PipPipPipPipPipPipPipPip
  • The Wrath
  • The Wrath
  • 1,582 posts
  • LocationFinland

Posted 21 May 2014 - 10:43 AM

Wait times are back to normal, good enough so far ;)





3 user(s) are reading this topic

0 members, 3 guests, 0 anonymous users