831 replies to this topic

[SCCOJake]

Paul Inouye, on 11 June 2018 - 12:30 PM, said:

Gonna throw a wrench in here to see what you guys think of this. Kinda related...

Dropping support for groups of 3,5,6,7,9,10,11. All of the groups of that size COMBINED make up less than 0.01% of games being launched. This would mean the MM would only be dealing with groups of 2,4,8 and 12 which would speed things up drastically. While this is something not really slated for implementation... I'd like to know if anyone has some thoughts on this.

Ugh... No thank you. For all the reasons that have likely already been stated. This might not harm most people, but it will make it less fun for some people. You are adding nothing and taking away something, so please don't limit group drops to even numbers.

[Agent of Change]

Paul Inouye, on 11 June 2018 - 12:30 PM, said:

Dropping support for groups of 3,5,6,7,9,10,11. All of the groups of that size COMBINED make up less than 0.01% of games being launched.

Just a thought on this Paul, and correct me if I'm wrong but would your 0.01% be of ALL games being dropped including each and every solo drop?

What percentage of GROUP drops are odd numbered? or perhaps even better What percentage of ALL drops are groups of any size (using the same metrics you used for the 0.01% number)

I mean it seems to me that if you used total games launched based on hitting the launch button then every single Solo game counts as 12 towards teh count and each group queued game could only count 2-6 towards that count based on how many and how big the groups that fit. it seems that it's obvious using those numbers you could make any given group size look insignificant compared to total drops, it adds up quick. I would suspect you could argue that group drops account for something like 0.04% of all drops and therefore should be removed for the ease of matchmaking. If you want to inform people of the situation using data this seems intentionally misleading even if "true"


Anecdotally, my suspicions of this mathematical weaseling is the fact that 1/3 to 1/2 of the time i'm playing with my group we are in a group of 3 or 5 and the rest 2 or 4. So unless we are SUPER unique as a unit (like only one in a hundred groups routinely plays with an odd numbered group) then your numbers have a whiff of fish.

Edited by Agent of Change, 12 June 2018 - 09:51 AM.

[Vamblery]

8v8 - there is no sense on that change. There is too much another things to balance (tiers for example, skill tree isnt revelation still).
8v8 probably kill balance (isn't good now but is) and mayby fun. DC will have bigger impact. Proposition to make 2lights/2mediums/2heavies/2assault is even worse - biggest advantage of mwo is freedom of choice mech and builds.
I like play 12v12 - You have option to make some tactics. I want play battles not duels. In 8v8 it will be impossible to make some movements.
In my opinion 8v8 will be big move back.

[Vamblery]

Obsidian Hand, on 12 June 2018 - 05:56 AM, said:

I support the idea of going back to 8 vs 8 mainly on the grounds of better feel for each mech class. Honestly, ever since quick play went 12 vs 12, assaults and heavies have had less of an impact in the game, since dropping with 12 fast lights and mediums can pretty much outclass an opposing team composed of a "balanced" mixed team of 12.

Back when the game was still 4 vs 4, each class had much of an impact, like mediums and heavy pilots being more reluctant to face off solo against an assault, or an assault thinking twice about engaging a light, or a light pilot being able to test his or her skill by seeing how well he or she can engage in a frantic brawl with an assault. Each weight class really had an impact back then.

In contrast, in a 12 vs 12, just a single lance of 4 lights can have the combined firepower of a single assault and ALSO be maneuverable enough to avoid damage from the assault and even its surrounding teammates via swarm tactics, then systematically move on to the next single target, making an assault less viable in these circumstances.

Going back to 8 vs 8 quick play would be a good compromise between getting the feel back for each weight class in a single match while still having more players with which to make the game more prolonged and engaging. After all, in the BattleTech universe, mechs are supposed to be rare, powerful machines of war that are deployed sparingly. Save the 12 vs 12 for "epic" battles, such as faction play.

I think that 8v8 fights will be very similar to each other which can kill the pleasure of the game. It will not solve problems and will add new ones.

[Vamblery]

GBxGhostRyder, on 11 June 2018 - 01:22 PM, said:

OMG your trolling us right? Why in gods name would you ever set up a MM to group odd numbers? you had it right long ago when you had check box options players could use.

You should have made it like this long ago a check box system to match up
2v2
4v4
6v6
8v8
12v12
When a player checked what option they wanted to play 2v2-12v12 the MM just matched them up together you could have even gone farther and matched players up using another check box system option system to match up players by classes or to make it real fun you could have made a option to drop 2v2-12v12 using this format.
3 assaults-3 heavy's-3 mediums-3 lights =12v12
2 assaults-2 heavy's-2 mediums-2 lights =8v8
1 assault - 1 heavy's- 2 mediums-1 light =6v6
1 assault - 1 heavy - 1 medium - 1 light 4v4

ETC. There could be a lot of mm drop combinations to consider for maximum fun game play.

To be honest after 5 years of watching this game go down hill on development because of bad ideas like the way Solaris 7 was conceived I'm surprised the game is still alive at all just my point of view.

And why games should be identical each time (for example - 2/2/2/2). Let us multiply, thanks to which gameplay gains.

[NaturalBornGriller]

Everyone complains about the match maker.

Lowering the number of players in any given drop is going to exaggerate any problem the matchmaker has by distilling all of the self leveling functionality of a population. This means that any one player will have more of an impact on the outcome of a game the fewer players are in said game.

Average player sodium content will go up as you boil off the water if you catch my drift. This means you need to dilute the peaks and valleys (the LRM assaults/discos being the valley, and the peaks being high level, dedicated players)

This is all basic level statistics guys, you try to do a scientific test and the fewer data points you have the pointier the graph is. That translates into very random outcomes in games, invariably.

I have a feeling that 'in beta where everything was flowers and butterflies with the old maps etc etc' is a combination of rose colored nostalgia glasses, the airs of superiority because 'they have played this game so long they must be right' and the sheer undeniable reality that everyone who plays a game in beta/early phase is a straight up boiled potato because they don't even know you can open their weapon doors yet (or whatever).

SO all that is an argument against having low numbers of players in a match. The argument against having high numbers of players is almost identical, however.

Everyone wants to make a difference in a match.. If your skill and actions are so diluted by the sheer number of enemies then the game stops being fun..

If you have so many players in a game that it makes everyones FPS drop into the toilet, or even half or a 3rd of the players FPS drop into the toilet it starts to stray into the realm of 'you can't play this game unless you spend a thousand dollars on a video card or enjoy losing'. Is that pay to win? Not unless PGI starts selling video cards or is sponsored by Nvidia anyway (because we know if they were sponsored by ATI then they would make the game run super well on even potato ATI products )


So it's a balancing act. I think 12v12 is a good level playing ground. I think 8v8 is too little and 16x16 might tax the average players machines too much. I think 24x24 would probably dilute player skill too much.

Edited by NaturalBornGriller, 12 June 2018 - 10:59 AM.

[Jasrion Blastlehoffer]

Revenant Kitsune, on 06 June 2018 - 04:06 PM, said:

My main concern with this would be with the current issue of the disconnects. Losing someone to a d/c in 8v8 would hit a lot harder than it already does in 12v12. Almost every match I go in to there is at least one disconnect.

That is a very VERY good point! BUMP!

[Agent of Change]

NaturalBornGriller, on 12 June 2018 - 10:46 AM, said:

*Snip*

Awesome post full of Logic

*Snip*

Thank you! this kind of logic is what we need to be applying to these problems!

And Re: The beta nostalgia... as someone who was there I can assure you that is exactly the case IMO. It was NOT all that great compared to now but soem folks are die hard for 8v8 because they feel they'd be at the top of rolls and they'd rather selfishly make the change and tell the spuds to git good rather than understand that an environment that is consistently frustrating to the people losing will result in them leaving not improving thus further increasing teh issues with matchmaking times due to population.

[WarHippy]

Paul Inouye, on 11 June 2018 - 12:30 PM, said:

Gonna throw a wrench in here to see what you guys think of this. Kinda related...

Dropping support for groups of 3,5,6,7,9,10,11. All of the groups of that size COMBINED make up less than 0.01% of games being launched. This would mean the MM would only be dealing with groups of 2,4,8 and 12 which would speed things up drastically. While this is something not really slated for implementation... I'd like to know if anyone has some thoughts on this.

Hell No!
When I do drop in groups it is almost always in those categories that would be dropped. Stop screwing over groups you guys did enough damage to them already way back when.

[Lukoi Banacek]

Ed Steele, on 11 June 2018 - 11:53 PM, said:

Get out and vote for 8v8! The game will be better and will run better on crap computers! Most of the people voting against it were not around when it was 8v8!

You have zero idea how many were around for 8v8 and are still voting no.

I've been around since CB and I voted no, among others.

[DAEDALOS513]

Revenant Kitsune, on 07 June 2018 - 01:06 AM, said:

My main concern with this would be with the current issue of the disconnects. Losing someone to a d/c in 8v8 would hit a lot harder than it already does in 12v12. Almost every match I go in to there is at least one disconnect.

Jasrion Blastlehoffer, on 12 June 2018 - 11:11 AM, said:

That is a very VERY good point! BUMP!

Meh I'd say 85-90% of the time they reconnect pretty quick

Edited by DAEDALOS513, 12 June 2018 - 01:25 PM.

[Vamblery]

CapperDeluxe, on 10 June 2018 - 06:32 PM, said:

8v8 means less likely to have more guns in your face at once, thus increasing TTK, which is a good thing for this game.

not really. How often You see 12 mechs in fireline?

[NocturnalBeast]

Lukoi Banacek, on 12 June 2018 - 12:08 PM, said:

You have zero idea how many were around for 8v8 and are still voting no.

I've been around since CB and I voted no, among others.

I have edited my post for your reading pleasure, please go back and re-read it. I realize that we will all have our own opinions, but I am hoping to see more people agree with me than not.

[50 50]

Paul Inouye, on 11 June 2018 - 12:30 PM, said:

Gonna throw a wrench in here to see what you guys think of this. Kinda related...

Dropping support for groups of 3,5,6,7,9,10,11. All of the groups of that size COMBINED make up less than 0.01% of games being launched. This would mean the MM would only be dealing with groups of 2,4,8 and 12 which would speed things up drastically. While this is something not really slated for implementation... I'd like to know if anyone has some thoughts on this.

Can certainly understand how having even numbered groups would certainly help the tetris game of match making.

Making use of friends/unit lists and the looking for group feature should make building an even numbered group easier and players that do form groups tend to use this anyway so I don't see it being a big drawback.

I can understand there could be some difficulties in making up the numbers when you lose a player, particularly at the higher group numbers as if it can't be done easily, then there are more players out of action and the group would need to split up leaving a player behind.
For example:

We get a group of 8 together, have a few games and then lose 1 player bringing us down to 7.
Let's say that for what ever reason it's hard to get that extra 1 player to get the group back to 8.
The group would need to split into a 4 player and a 2 player and leave 1 behind to get back into the matches.
There is the similar scenario for the 12 player group that loses 1 and then could struggle to make up the extra numbers.

It isn't a massive issue but there are a couple of points to take into consideration.
First, having some better functions to be able to split or manage the group by lances by allowing individual lances to launch and better player to lance allocation.
There is some functionality in the Private Lobby and the Battlegrid that allows us to assign players to lances that would be really good to standardize throughout the game and improve the UI.
If the group screen was better organised into the lance structure it would then be beneficial to have the ability to launch individual lances that have enough players (2 or 4) and are readied up.
Add to that a linked launch option when there is more than one lance ready so as the group waxes and wanes it is easy to keep the majority of the group dropping into matches while the odd numbers are resolved.
It should only ever be a case of needing 1 more player in the group and it gives us true modular grouping functionality based around the lances as the building blocks for a full company drop.
Determining who can launch a group would also need to be looked at in this scenario so maybe it would be beneficial to assign lance command and company command roles with these options.
I would also suggest that the when we combine groups it encompasses this functionality and visibility. This may be particularly beneficial within Faction Play.
Not sure I'm explaining that concept too well so I might try a picture or short vid when I get the chance.

There is also the suggestion, as mentioned earlier, about allowing solo players to opt into the group queue.
The problem with this is that it is a function that would apply once searching for a match and therefore does not work with the change to groups which would prevent the group from launching if they do not meet the number requirements.
Perhaps it could be more of a flag in the friend/unit list that marks the player as looking for group just so we have a bit more visibility there. Improving the looking for group UI a little as a screen that we need to be utilizing more would be beneficial.

For Faction Play, perhaps the call to arms could be changed to a feature that groups could use to call in additional players to add to their group?

While I still believe that the game would benefit from having flexibility in the team sizes, adjusting the limits on groups is a positive step and can be combined with some good group management functions to really bring it all together.

[Lukoi Banacek]

Ed Steele, on 12 June 2018 - 02:39 PM, said:

I have edited my post for your reading pleasure, please go back and re-read it. I realize that we will all have our own opinions, but I am hoping to see more people agree with me than not.

PROBABLY still not an even remotely accurate assessment given feedback on Outreach among others. Why not stick to your opinion (all fair there) rather than making baseless claims to shade your point? Just think it's disingenuous to pretend you've got access to metrics none of us have.

[NocturnalBeast]

Lukoi Banacek, on 12 June 2018 - 03:33 PM, said:

PROBABLY still not an even remotely accurate assessment given feedback on Outreach among others. Why not stick to your opinion (all fair there) rather than making baseless claims to shade your point? Just think it's disingenuous to pretend you've got access to metrics none of us have.

I never gave an exact number, nor do I claim to know exactly how many people feel a certain way, my statement should not be interpreted as anything other than an opinion.

[Lukoi Banacek]

Considering that the forum users are supposed to only make up a very small percentage of the playerbase, 3200ish votes is fairly promising.

[Tarl Cabot]

As noted, I voted maybe and what I would like to see, but I would not vote for 8vs8. What I remember is getting players who did not know how to setup or play an assault and have it be the critical piece of the puzzle. When I wanted to win than lose I had to run assaults, primarily Highlander!!!

The difference in gameplay from 8vs8 to 12vs12 was extreme (solo queue) lances did not rush in, took more time to engage and having a crappy assault player did not matter AS much as it did in the 8v8. It made me feel as if I had more freedom to pilot non-assaults. More nascaring with 12vs12 than with 8v8. Dang if we do, dang if we dont...

[Throe]

[deleted by user]

Edited by Throe, 08 November 2018 - 04:29 PM.

[NocturnalBeast]

Vote 8v8! It is almost there!