Matchmaker Hangs And Wait Times
#141
Posted 29 April 2014 - 10:36 PM
yesterday (for me that is GMT+1), it was hard to find a match, but I understand the problem, you can test it for 2 months or even more but the real test is when it goes "live", only then you find real problems, 3/3/3/3 is great feature/idea if you can get it to work, but you really need to have a list of mechs mentioned above, so people could switch to mech that is low numbers.
Other thing is now if you have 4 friends in the group you are doomed since I relay do not think so you will find a match that fast, of course, you really need to make some alert so all the people in group need to know that they can not have 4 or more Mechs of the same type
keep up the good work, for all good things need time to grow,
#142
Posted 29 April 2014 - 10:40 PM
But I think some of the reactions in here, within literally minutes of the patch going live, are a little over the top and unfair.
Give them a day or so to hotfix it. It's not the end of the world.
#143
Posted 29 April 2014 - 10:49 PM
If you thought this would work on patchday you are somewhat delusional. I didn't even download the patch yesterday, because i knew this would happen. My buddies were like "dude download the patch, we need to test 3/3/3/3!", and i was like "guys, you won't test anything today".
I was chilling on the couch when i got the first messages of 5 minutes to find a match and matches played 4vs3. It was bound to happen.
I'm all cool with this, because i never expected it to work on release. Just chill, drink some coffee (i love coffee) and let them fix it.
#144
Posted 29 April 2014 - 10:53 PM
Wants it back. Please? Soon?
#145
Posted 29 April 2014 - 11:02 PM
MischiefSC, on 29 April 2014 - 10:53 PM, said:
Wants it back. Please? Soon?
They said ~10 hours from now... give or take an hour.
#146
Posted 29 April 2014 - 11:06 PM
Nikolai Lubkiewicz, on 29 April 2014 - 10:39 PM, said:
That sounds like a noble goal and if we had inbuilt voice it might actually work out that way, but in the game as it stands there's no real way for a group to impose or suggest order to a group of pugs. If / when we had voice or lance command preset hot keys it could help, but even then adding groups into pugs mostly just means there's a group of four running around with their own coherent plan. Maybe all of them are doing brutal swift focus fire, or coordinated LRM bracketing. Maybe they're all running flamer joke builds. You're still adding that element of unity, good or bad, into the PUG. Effective ELO is qualitatively different for that group. ELO bracketing might help some but I still think it'd cause issues in scrub-tier.
Edited by Osric Lancaster, 29 April 2014 - 11:44 PM.
#147
Posted 29 April 2014 - 11:12 PM
Osric Lancaster, on 29 April 2014 - 11:06 PM, said:
That sounds like a noble goal and if we had inbuilt voice it might actually work out that way, but in the game as it stands there's no real way for a group to impose or suggest order to a group of pugs. If / when we had voice or lance command preset hot keys it could help, but even then adding groups into pugs mostly just means there's a group of four running around with their own coherent plan. Maybe all of them are doing brutal swift focus fire, or coherent LRM bracketing or maybe they're all running flamer joke builds. You're still adding that element of unity, good or bad, into the PUG. Effective ELO is qualitatively different for that group. ELO bracketing might help some but I still think it'd cause issues in scrub-tier.
I honestly think the Devs play a different game to us sometimes with the conclusions they draw. Imposing order unto Pugs. LOLWUT
At F8L We do try as referenced by this thread
http://mwomercs.com/...2-0-thread-f8l/
But sometimes it's just not worth it.
Edited by Tekadept, 29 April 2014 - 11:13 PM.
#149
Posted 29 April 2014 - 11:16 PM
MischiefSC, on 29 April 2014 - 11:12 PM, said:
Well, the MM part doesn't need the patch exactly (they can tinker with that all they like - this was true since the debut of Elo, no patch needed to fix that)... match type selection however does.
Edited by Deathlike, 29 April 2014 - 11:16 PM.
#150
Posted 29 April 2014 - 11:33 PM
JohnnyWayne, on 29 April 2014 - 12:26 PM, said:
Well if the mass doesnt attend its mass test then its only a test.
Which might be closely related to the "mass" test lasting only some hours during one day within half a year ... which hardly can be called a mass test. I remember WoT, where a mass test / Test server was up for a week or so.
#151
Posted 29 April 2014 - 11:46 PM
Quax1102, on 29 April 2014 - 11:33 PM, said:
Which might be closely related to the "mass" test lasting only some hours during one day within half a year ... which hardly can be called a mass test. I remember WoT, where a mass test / Test server was up for a week or so.
War thunder test server is up 7 days a week, with alternating shifts to accomodate different timezones in 6 hour blocks, and a 2 6 hour blocks on weekends. Just saying.
#152
Posted 29 April 2014 - 11:47 PM
YEARS go by - nothing changes.
Game still needs URGENT hotfixes due to ... I got no idea why...
Now seriously.
I am really happy that at last we got the least (clans and voice chat are also the least).
I am not surprised that nothing works as it should have.
I am leaving a bottle to celebrate 2 years of this "open beta".
Really - this is yet an open beta folks
Oh lord, let the clan mechs be balanced from the very start, let the inner sphere live
#153
Posted 30 April 2014 - 12:26 AM
#154
Posted 30 April 2014 - 12:31 AM
Raggedyman, on 29 April 2014 - 12:37 PM, said:
Great
#155
Posted 30 April 2014 - 12:38 AM
Nikolai Lubkiewicz, on 29 April 2014 - 08:37 PM, said:
3/3/3/3 is not currently applying to the private lobbies, as it was disabled until it can be fixed. As a premium time user, you can modify the rules of private matches which you have started and have all the lobby tools needed to help keep both sides in order while lobby leader, so there's no reason to be disappointed.
Sorry this is utterly bulls*. As a premium player, I actually (for repeated times regarding to the last patches) cannot play public games that make me earn C-Bills, XP, GXP. Was the same with you DX-"patch" that ruined my installation for over a week. And has been the same with nearly every patch before. With every bug update (you guys call it "patch") we loose time to play the game because something is not working as intended.
And this makes me loose premium time.
#156
Posted 30 April 2014 - 12:40 AM
Jak Darren, on 29 April 2014 - 02:36 PM, said:
Battlevalue. Use Battlevalue.
I don't see how you can buckle on so many other things, but are afraid to touch the matchmaking system already designed for this.
A blind and mentaly challenged person who looked at battletech would know that BV would be a key ingreedient to any multiplayer game based on the franchise. It would take any rational person 10 seconds, maybe to a minute to recognize this. Your talking to PGI, they are not rational. BV is how TT fixed the problem of imballanced matches what 30 years ago. Its had two major revisions to get it right, and a solid 20 years of testing. Who would want to take advanatage of all that?
#157
Posted 30 April 2014 - 12:42 AM
A match can last up to 15 minutes. Even if there only were 25 players on the PTS, after 15+ minutes I should have been the first in line for a match.
There were lots of reports like mine. PGI chose to ignore them and went on with the plan. No wonder they failed.
And 3/3/3/3 matchmaking won't work without some in-game information to the players (wait times for each class) for the players. Without feedback from the system, people won't know which class of mech to take, therefore the queues will never decrease.
Xenok, on 30 April 2014 - 12:40 AM, said:
A blind and mentaly challenged person who looked at battletech would know that BV would be a key ingreedient to any multiplayer game based on the franchise. It would take any rational person 10 seconds, maybe to a minute to recognize this. Your talking to PGI, they are not rational. BV is how TT fixed the problem of imballanced matches what 30 years ago. Its had two major revisions to get it right, and a solid 20 years of testing. Who would want to take advanatage of all that?
And how exactly are you going to measure BV in MWO,huh ? Do you even know what makes a mech or a weapon useful in this game? Are you going to multiply the mech tonnage by max engine rating and divide by hitbox surface area?
As for the weapons: will you use the DPS/HPS metric? What about pinpoint vs spread damage? Are you going to factor in the hit detection issues and HSR errors? Projectile speed and range? Weapon durability?
Battletech BG was such a simple game, that you could calculate everything going on in your mind. MWO is such a complex system that you would take a year to calculate what happened in a single match.
Edited by Kmieciu, 30 April 2014 - 12:59 AM.
#158
Posted 30 April 2014 - 12:43 AM
Tekadept, on 29 April 2014 - 11:46 PM, said:
Just my saying .. but why should PGI adopt useful behaviour from other game developers. They clearly are not a friend of "best practice" or anything related. Otherwise they would have studied why WoT and WarThunder happen to have a functional MM that also includes "mech types" (aka heavy/medium/light tanks in WoT for example).
#159
Posted 30 April 2014 - 12:54 AM
Xenok, on 30 April 2014 - 12:40 AM, said:
A blind and mentaly challenged person who looked at battletech would know that BV would be a key ingreedient to any multiplayer game based on the franchise. It would take any rational person 10 seconds, maybe to a minute to recognize this. Your talking to PGI, they are not rational. BV is how TT fixed the problem of imballanced matches what 30 years ago. Its had two major revisions to get it right, and a solid 20 years of testing. Who would want to take advanatage of all that?
Problem is that weapons in mwo can be better when paired up in MWO because of the lack of convergence. It leads to tricky balancing. Eg. Two PPCs and an AC5 on one highlander > a PPC on a thud, an AC/5 on a raven and another PPC on a hunchy. You'd need to measure the average performance of each chassis somehow. Roland suggested doing it by usage or popularity, and others have suggested using average match scores or something but the point is you can't make a straight, flexible BV system without convergence or any other number of factors fixed.
Edit:
Kmieciu, on 30 April 2014 - 12:42 AM, said:
Yer, like those. I keep thinking they should have granulated armor into 'plates' per section numbering based on hit-box surface area. Oh well.
Faith McCarron, on 30 April 2014 - 12:26 AM, said:
I do appreciate the improved communication, I just wish it was more of them listening and responding. We might not have 50 threads with different fixes for the same thing in feature suggestion if there was a little dialogue there. If 'managment' gave us a tenth the consideration Karl Berg and Niko have the game could only improve.
Edited by Osric Lancaster, 30 April 2014 - 12:58 AM.
#160
Posted 30 April 2014 - 01:24 AM
PGI...
2 user(s) are reading this topic
0 members, 2 guests, 0 anonymous users