419 replies to this topic

[kapusta11]

Don't you get it? Only cash whales opinion matter.

[Training Instructor]

People will condescendingly spout off about how you shouldn't be able to exclude a map you don't like, but then insist on excluding a match mode they don't like.

[Dauphni]

For me, the problem isn't so much being forced to play gamemodes I don't like, since I do enjoy them all, but rather not being able to play the ones I enjoy the most. Conquest barely showed up for me despite having it exclusively selected, and that's just bad.

[MischiefSC]

N0MAD, on 09 October 2014 - 09:08 PM, said:

So people who want a choice in what they want are scrubs/kiddies/ play with helmets, what other insults?
While those that dont are Boss/drop gloves/and hunt bears in another word what Elite?.
Dude really?

That's a fair response. I get the desire to choose game modes. Due to the reward system I generally play skirmish when pugging. My humor framed contempt is more towards the folks who went past voicing their opinion and straight to absurdities like suicides, dcs, gimping matches and calling it "civil disobedience" like they are being denied the right to own property he representative government.

It's fine to have a preference in a game, to like or dislike things. Throwing an absolute fit over literally only getting your way sometimes in a computer game is not. It stifles useful debate on the topic and gives the false impression that PGI walked the change back due to outrageous tantrums and not just it providing insufficient payoff overall. There will always be dissenting opinions and there will always be people who support our dissent any change.

The point being this wasn't some disastrous event. PGI said "so you don't like big group size and big Elo mismatch. Doing this will help some, wanna try it?" We said "sure!" Then we found out it only made a significant change on narrow ends of the spectrum so they walked it back.

It was a good thing overall and worth trying. We as a community have consistently failed to support pts stuff so if we're going to make changes to things it's going to mostly be tested in live. We can then give our feedback and move forward. People who go hyper dramatic and lose their **** are going to hurt that overall. The attitude of 'I don't like this so I'm going to pitch a fit and try to ruin it for others so they don't like it either' is where the contempt was pointed.

[MischiefSC]

Vercinaigh, on 09 October 2014 - 11:49 PM, said:

We've honestly not tried that in all fairness. but again, neither do the other units and unless they do too, impact will almost certainly be minimal, we cannot control other units.



As much as I hate to say this, give a lot of people option for easy cbill farming, they're gonna take it, more often than not. Would this maybe help if all the units tried to do this? Sure, but I am sad to say reality won't match the hope.

Ah, but that is what makes it so slick. Other high Elo groups might be running cap fest troll night, going assault only and playing for the lulz of capping in assault. You come on and click BOSS MODE and you start dropping wherever they are. At that point they start having to step their game up because no matter what mode they select.

They may pick the game mode but you are choosing to play them wherever they are.

Make sense? Everyone wins.

[Chrithu]

MischiefSC, on 10 October 2014 - 02:18 AM, said:

That's a fair response. I get the desire to choose game modes. Due to the reward system I generally play skirmish when pugging. My humor framed contempt is more towards the folks who went past voicing their opinion and straight to absurdities like suicides, dcs, gimping matches and calling it "civil disobedience" like they are being denied the right to own property he representative government.

It's fine to have a preference in a game, to like or dislike things. Throwing an absolute fit over literally only getting your way sometimes in a computer game is not. It stifles useful debate on the topic and gives the false impression that PGI walked the change back due to outrageous tantrums and not just it providing insufficient payoff overall. There will always be dissenting opinions and there will always be people who support our dissent any change.

The point being this wasn't some disastrous event. PGI said "so you don't like big group size and big Elo mismatch. Doing this will help some, wanna try it?" We said "sure!" Then we found out it only made a significant change on narrow ends of the spectrum so they walked it back.

It was a good thing overall and worth trying. We as a community have consistently failed to support pts stuff so if we're going to make changes to things it's going to mostly be tested in live. We can then give our feedback and move forward. People who go hyper dramatic and lose their **** are going to hurt that overall. The attitude of 'I don't like this so I'm going to pitch a fit and try to ruin it for others so they don't like it either' is where the contempt was pointed.

Mostly agreed!

But here's what I disagree with: The only data about impact we got in hand was provided by Russ about 8 hours after the patch and that little data already showed it had a positive impact in the group queue elo wise. It didn't mention anything about wait time, which I am sure was already improved as well.

Anything after that is the anecdotal experience of people that dropped between 10 and 40 matches in those two days the feature was enabled.

That is not reliable data. I'd very much like for Russ to present an analysis of all the data they gathered in the full 48 hours and if it shows a further improvement over the data from the first 8 hours. And not only including Elo variance but also the average wait time for a match.

Even if you see it differently: The Feature rollback was the direct result of the forum rage. But the important point is that this forum rage would have been preventable. The way I see it most people went to rage mode simply because they learned about this change only from the patch notes, which is indeed far too late and makes the rage very understandable for me.

If the first poll would have included Russ' detailed explanation of the feature, along with making clear the feature will be tested for 2 weeks between two patches and a poll time of at least a week, we would now be testing this until the next patch and get truely useful data. I am pretty convinced of that.

[N0MAD]

What i cant understand is...
If they want a tighter ElO bucket/matchup why not adjust that in the matchmaker code, i mean really they do have the ability to do that no? or tighten the valve on the matchmaker ELO tolerance, isnt this just changing variables in the code or whatever, is that not actually possible?

[Roadbeer]

N0MAD, on 10 October 2014 - 04:47 AM, said:

What i cant understand is...
If they want a tighter ElO bucket/matchup why not adjust that in the matchmaker code, i mean really they do have the ability to do that no? or tighten the valve on the matchmaker ELO tolerance, isnt this just changing variables in the code or whatever, is that not actually possible?

They've done that in the past, all it accomplished was long wait times and the MM ended up throwing top tier and n00bs together after a long wait anyway, or the dreaded "Cannot find match".

[Chrithu]

N0MAD, on 10 October 2014 - 04:47 AM, said:

What i cant understand is...
If they want a tighter ElO bucket/matchup why not adjust that in the matchmaker code, i mean really they do have the ability to do that no? or tighten the valve on the matchmaker ELO tolerance, isnt this just changing variables in the code or whatever, is that not actually possible?

Yeah but that would be a change at the expense of wait time. If they reduice the MMs Elo tolerance it directly lengthens the wait time.

They do that change. Next thing that happens is a forum rage about how it takes ages to find a match.

[spectralthundr]

Destructicus, on 08 October 2014 - 11:51 AM, said:

That's awfully egocentric of you

Some people like conquest, some people like assault.

You don't speak for everyone.

Went over your head, It's not that I dislike conquest, I dislike that half the player base thinks the key to winning is cowering behind a hill for 8 minutes not getting caps, and not shooting at anything. That isn't fun. If it's your cup of tea, that's great. But if I'm going to be stuck playing a mode like that, I expect people to push, take objectives and at least try to shoot the opposing force.

[Kjudoon]

spectralthundr, on 10 October 2014 - 05:21 AM, said:

Went over your head, It's not that I dislike conquest, I dislike that half the player base thinks the key to winning is cowering behind a hill for 8 minutes not getting caps, and not shooting at anything. That isn't fun. If it's your cup of tea, that's great. But if I'm going to be stuck playing a mode like that, I expect people to push, take objectives and at least try to shoot the opposing force.

And those same people you disparaged do not want to play the game style you described.

This is why you can't have a voting system. The two sides will not compromise this position of "I will NOT play with them!!!" and would rather burn it all down than change that.

Not saying I'm not one of them with that position, because I am, but I am at least being honest about it.

[tuffy963]

This whole episode might have been avoided if they had pushed this change to an open public test server and provided time for testers to play on it and provide feedback. I am not sure why PGI continues to test potentially volatile features in production.

[Kjudoon]

tuffy963, on 10 October 2014 - 05:33 AM, said:

This whole episode might have been avoided if they had pushed this change to an open public test server and provided time for testers to play on it and provide feedback. I am not sure why PGI continues to test potentially volatile features in production.

This right here. Emphasis mine.

The test server needs to be up on a far more consistent basis for the invited public to come in and help with testing the changes, helping with map bugs and other things like this. It's a resource that PGI, I'm sorry, you're not using enough. I'm not saying that because I want to be involved necessarily, but it would be good for those who really do have the ability to help you make this a better game and give you far better data before changes are put out in public.

[Chrithu]

tuffy963, on 10 October 2014 - 05:33 AM, said:

This whole episode might have been avoided if they had pushed this change to an open public test server and provided time for testers to play on it and provide feedback. I am not sure why PGI continues to test potentially volatile features in production.

Test server population is abysmal. And testing matchmaker tweaks on the PTS is like testing a formula one car in city traffic. It does not produced the data needed to see if the tweak is good or not.

The problem is with how things in this particular case went down: the second vote wasn't a vote about testing a change that could provide better and faster matchmaking. It was simply a vote whether to rollback a feature that was perceived as forced on by people not in the know. The reason for the change (improving matchmaking) and the fact that it was always meant to be just a limited time test did not play a role anymore in that second poll. The posts made in the thread made that pretty clear. It was all about how people either have all modes on anyways or how someone hates this mode or that mode.

In the end I think the MM should just be left alone from now on in. We deserve the state it is in.

[Bilbo]

spectralthundr, on 10 October 2014 - 05:21 AM, said:

Went over your head, It's not that I dislike conquest, I dislike that half the player base thinks the key to winning is cowering behind a hill for 8 minutes not getting caps, and not shooting at anything. That isn't fun. If it's your cup of tea, that's great. But if I'm going to be stuck playing a mode like that, I expect people to push, take objectives and at least try to shoot the opposing force.

Funny, that's the same reason I like the other modes less.

[Mechwarrior Buddah]

spectralthundr, on 10 October 2014 - 05:21 AM, said:

Went over your head, It's not that I dislike conquest, I dislike that half the player base thinks the key to winning is cowering behind a hill for 8 minutes not getting caps, and not shooting at anything. That isn't fun. If it's your cup of tea, that's great. But if I'm going to be stuck playing a mode like that, I expect people to push, take objectives and at least try to shoot the opposing force.

I dislike conquest because it always boils down to killing the other team's fast mechs and lights. The team thats heavier/slower always seems to lose

[Vercinaigh]

Roadbeer, on 10 October 2014 - 04:55 AM, said:

They've done that in the past, all it accomplished was long wait times and the MM ended up throwing top tier and n00bs together after a long wait anyway, or the dreaded "Cannot find match".

Yeah pretty much what happens now, only back then it was even worse across the board. I remember that crap...all we want is decent games like anyone else. :/

Edited by Vercinaigh, 10 October 2014 - 11:23 AM.

[HUBA]

ok, thats my last attempt to show you that the MM barely improve with this change. It only might get better matches for the top 20% of the player. Lets build a very simple MM It sorts the groups by size and take the biggest group to begin with. Then it tries the 2. group and look how big the elo difference is. If the difference is not to big the open places on both sides will be filled in the same way (looking on the biggest group to fill up with). The only thing that is different without voting, If the group you compared with have no game mote with yours common it will be rejected. So if you select all modes your changes to find a match (in 12 vs 12) is the same than with voting. the only slide advantage can be made by the small fill groups because they have not to have a common game mode.

Some statistic: 3 game modes with free choice gives you 7 possible selection, All (ACS), AC, AS, CS, A, C and S (obviously for Assault, Conquest and Skirmish) evenly distribute all groups have about 15%. If you Select all you can play 100% of these groups. If you select 2 you have only one group that not fit your selection meaning 85% possible groups and with one selected there are still 60%. It narrows down very quickly when you have to group 4 or even 6 groups that have to have a common game mode (I guess 20% with one GM selected and 4 groups and 7% on 6 groups. someone can do the math if he want)

So this simplified MM shows there is a potential to get better games if you ignore the player wish for a specific GM. Thats why I suggested to select all GM and also talk to your mates. If you manage to communicate a weight limit that all follow there might a good chance to get good games.

Elo for me: I won the last 6 games and probable increased my elo by 50 points. Am I better know so that the MM have to find a better player for me? tomorrow I might loose some games in a row. Do I need then other player again? I'm pretty sure in the elo range I'm at, there is no differences between player even if there 100 points apart. Lets make a example: Both sides have the same mechs and the same elo but on one side all player have 100 points higher elo then the other side. thats 1200 Points in total an I guess it would be more even than many games I played. Same example but lets even out the elo by one player. Give the "better" side a noob so they are practically a mech down. Or give the other team an elite player. I guess that would even wore then being one down

Now the downsides of ignoring the player wish. I have to play maps that I don't like and thats ok if this maps come not to often. But because I have less fun on this maps I also bring less effort into my play. This might be different for player who want to win every match but I think for casual player it might be common that they are less effective on maps they don't like. So the tighter elo you gain by forcing player into a GM they don't like will be negated by player only bringing 80-90% effort info play.

[MischiefSC]

Tightening the MM tolerance for Elo doesn't fix the issue - if the best matches for your Elo are in Skirmish and you're playing Conquest you'll never see them with a hard limit.

That's a bit that gets missed a lot. Elo, weight, whatever factor for balancing you want, it can only be applied to people who can drop in the same queue as you. Group/solo and then by gamemode selection. Those things divide people in the matchmaker every much as if they were not all playing MW:O at the same time. They are not available, regardless of Elo.

I think having it as an elective might work - that way the people who absolutely positively must control the gamemode they play or else they are victims of a human rights tragedy can do so while those who care more about a close, challenging match more than game mode can do so - which will make changing gamemode no longer a viable means of trying to avoid competitive matches.