Jump to content

My 2-4/2-4/2-4/2-4 Idea


79 replies to this topic

Poll: 2-4/2-4/2-4/2-4 Alternative (51 member(s) have cast votes)

Do you like the suggestion proposed here?

  1. Yes (19 votes [37.25%] - View)

    Percentage of vote: 37.25%

  2. Yes, but with comments (7 votes [13.73%] - View)

    Percentage of vote: 13.73%

  3. No (21 votes [41.18%] - View)

    Percentage of vote: 41.18%

  4. No, but with comments (4 votes [7.84%] - View)

    Percentage of vote: 7.84%

Vote Guests cannot vote

#1 Supersmacky

    Member

  • PipPipPipPipPipPip
  • The Messenger
  • The Messenger
  • 239 posts
  • Facebook: Link
  • Twitter: Link
  • LocationGeorgia

Posted 11 March 2014 - 06:41 AM

First, let me state that I am not a fan of the idea of the 3/3/3/3 model that PGI has decided to push out in April. I believe it will result in a mostly negative experience in the end.

That said, I realize that PGI and various players have valid concerns about the way matches are put together now...myself included (had a match last night where the entire opposing team was jagers and cataphracts except for 2 ravens).

So, what I propose is using the class limits, but modifying them to allow more flexibility in team building. Basically, each team would have to have at least two of each class to be able to deploy in a match and no more than 4 in a class.

Benefits:
- This would prevent all assault/heavy teams (thus stomping all over the other team)
- This would prevent all light teams (and thus getting all stomped on)
- Would allow 4 mans to all run the same class of mech
- Would make sure that every team had every class represented (which admittedly 3/3/3/3 does, but in a much more boring and restrictive way)
- Would reduce the potential bottle-neck and wait time for matches that 3/3/3/3 is going to cause. The queue for building a match will have a wider base to choose from and fill out the teams than just 3 of this, 3 of that, 3 of those and 3 of the others, no more, no less.
- Would continue to encourage taking a player's mech of choice in a class rather than just the heaviest in a class).
- Worst case that would happen under this model would be 4 assault/4 heavy/2 medium/2 light vs 2 assault/2heavy/4 medium/4 light. This would likely be a rare occurrence, but even so it would be better than what happens in a lot of cases now.
- Would be new/pug/solo-player "friendly".
- Would give more variety and flexibility than the "oh, here we go again: 3 assaults, 3 heavies, 3 mediums and 3 lights yet again" model. If you always know what you are going to face then the game will become dull and repetitive.
- Would comply with PGIs stated goals and reasons for wanting to use the 3/3/3/3 model (balance, no tonnage limits, etc).
- Would keep teams from knowing the basic make-up of the opposing team before the match even starts. As it stands right now, the 3/3/3/3 model will allow teams to "count cards" and know exactly what they have left to deal with. So, under 3/3/3/3 if they one team has already killed 3 assaults, they know without performing any scouting they will not face any more assaults.

Again, not perfect, but I believe it shows more imagination than the 3/3/3/3 model. Just hoping that PGI will start putting some stock into what their player (payer) base are saying and add a little more flexibility to the model they are planning.

#2 Reitrix

    Member

  • PipPipPipPipPipPipPipPip
  • Survivor
  • Survivor
  • 1,130 posts

Posted 11 March 2014 - 07:04 AM

The thing is, A lot of people currently bring popular assault 'Mechs specifically because the other side is most likely going to be doing the same.

The only thing we can really do is test the waters. When the new rules hit us, pick your favorite 'Mech and drop in. After a week or two, the queue will normalize as people realize they can bring whatever they happen to truly enjoying piloting, because we know for a fact that we wont drop into a Match against 4 Atlas DDC LRM Boats backed up by 3 Highlander poptarts and 2 Stalker LRM boats.

Scouting will still have a purpose because while we know there is going to be 3 Assault 'Mechs, we need to know what they are, preferably before we turn a corner and crash into its CT.

In Conquest, I'll be able to take out my Stalker knowing that my team WILL have Light 'Mechs on it to help secure cap points.

There are benefits to the 3/3/3/3 system on paper. Just as you are legitimately pointing out some potential flaws. But without putting the system to the entire playerbase and seeing what we do with it, we cannot form a solid conclusion on how it will turn out.

If it turns out that we can't form a match within a reasonable timeframe, then the Devs can remove it and try with a different system.
But we first need to try it. Also, don't say *use the Test Server* because you wont get a large enough pool of players in there to give an accurate representation of queue time effects.

#3 Supersmacky

    Member

  • PipPipPipPipPipPip
  • The Messenger
  • The Messenger
  • 239 posts
  • Facebook: Link
  • Twitter: Link
  • LocationGeorgia

Posted 11 March 2014 - 08:03 AM

With all due respects, I am not from the school of "I know it is a bad idea and has obviously flaws, but let's try it anyway". Trying something out just to see what happens has costs many of businesses dearly and many individuals various body parts.

I never said scouting would be useless, just that there is no reason to scout to find out how many of a given class there are because there will always be the same number of everything in every match...bleh!

Concerning taking out your Stalker, my idea does just as well because you will still have lights, mediums, heavies and assaults on every team.

So, what you are saying is PGIs idea is just as good as any other idea out their (and no better).

And I disagree about using a sample base to draw conclusions. It is done (successfully) all the time. Is it possible beta testing might miss something? Sure. But I would rather have something tested and then rolled out rather than have to something hoisted that had no real world practical testing performed (and to which I suddenly become and unwilling beta-tester).

I never understand the notion of trying something just because. I don't want to spend two weeks, a month, six months testing something flawed that PGI could have fixed before public release. If it is a beta/test environment, sure: I'd be happy to help out. But your statement is basically agreeing the system is flawed, but let's just give it a go anyway. Wow! Really? At the very least you are saying PGIs has no more merit than any other idea out there (including the current iteration). Hardly a motivator to embrace and test it for them while paying for the privileged.

#4 Lukoi Banacek

    Member

  • PipPipPipPipPipPipPipPipPip
  • WC 2018 Top 12 Qualifier
  • WC 2018 Top 12 Qualifier
  • 4,353 posts

Posted 11 March 2014 - 08:07 AM

How does your version encourage folks to take mechs of a class that is not the optimal or the heaviest in class? Or did I misunderstand that comment in your OP.

#5 Gladewolf

    Member

  • PipPipPipPipPipPip
  • Shredder
  • Shredder
  • 464 posts
  • LocationUnited States

Posted 11 March 2014 - 08:39 AM

Adding a little flex to the system doesn't seem like a bad idea, but I think the "flex" should match for both sides of play.

#6 Dramborleg

    Member

  • PipPipPipPipPip
  • 128 posts

Posted 11 March 2014 - 10:08 AM

Ideally all mechs would fulfill equally important roles on the battlefield. Ideally a team of all assaults would be at as much of a disadvantage against a balanced team as a team of all lights because there are strategies they can't counter. Ideally piloting an "inferior" class of mech would be considered enjoyable rather than boring but necessary. Ideally the weight classes would be balanced. Ideally role warfare would be real rather than imaginary. Ideally the matchmaker wouldn't need to force class balancing artificially because there would be equal player populations piloting all kinds of mechs.

Ideally...

#7 oldradagast

    Member

  • PipPipPipPipPipPipPipPipPip
  • Philanthropist
  • 4,833 posts

Posted 11 March 2014 - 11:01 AM

While I would still prefer a system that gives people strong incentive to play all types of mechs, this concept - introducing a tolerance range, basically - is far better than the currently proposed rigid system. Adding tolerances will allow for quicker matching finding while still preventing silliness matches like 8 Atlas vs. random mech pile.

I like this game, but I'm concerned about the direction it's going with the 3-3-3-3 system and recent game mode changes. The addition of Skirmish and turrets on Assault has swung the game even more in favor of heavier mechs. Skirmish is just a playground for assaults, and the turrets on Assault have badly weakened the "stealth cap" or "base tag" role of light, fast mechs. These changes all favor assault mechs, and simply bludgeoning the community into playing lighter, "lesser" mechs with 3-3-3-3 does not change the fact that the ROLE for those lighter mechs has been reduced. This means they are simply less effective and less fun to play than their heavier cousins. Forcing people to play them won't change that fact - only true role warfare will, but we seem to be heading away from that goal these days.

#8 Supersmacky

    Member

  • PipPipPipPipPipPip
  • The Messenger
  • The Messenger
  • 239 posts
  • Facebook: Link
  • Twitter: Link
  • LocationGeorgia

Posted 11 March 2014 - 11:17 AM

View PostLukoi, on 11 March 2014 - 08:07 AM, said:

How does your version encourage folks to take mechs of a class that is not the optimal or the heaviest in class? Or did I misunderstand that comment in your OP.


You may have missed by intent or I may not have stated it well. With the 3/3/3/3 there will be a tendency to take the heaviest of a class because the heaviest tends to have the best chance of survival (in general, I know there might be exceptions). When you introduce the variable of not knowing how much of any given class might be on the opposing team (such as in my suggestion) you have to think beyond just outlasting the other side or "I am definitely going to see 3 assaults, 3 heavies, 3 mediums and 3 lights).

View PostGladewolf, on 11 March 2014 - 08:39 AM, said:

Adding a little flex to the system doesn't seem like a bad idea, but I think the "flex" should match for both sides of play.


My only issue with this is it is essentially the same thing. In other words: "Hey, we have 2 assaults, 4 heavies, 4 mediums and 2 lights and so does the other team." Seems boring to me if you know before the match starts what you're going to face (within class types).

#9 Bhael Fire

    Banned - Cheating

  • PipPipPipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 4,002 posts
  • Twitter: Link
  • Twitch: Link
  • LocationThe Outback wastes of planet Outreach.

Posted 11 March 2014 - 03:05 PM

I'm looking forward to not seeing 4+ assaults/lights on the other team. Also looking forward to having equal weight classes on each team instead of the off-balanced wonkiness we have now.

So, yeah...3/3/3/3 seems just about right to me.

#10 Moromillas

    Member

  • PipPipPipPipPipPipPip
  • 943 posts
  • LocationSecret **** moon base

Posted 11 March 2014 - 10:13 PM

View PostSupersmacky, on 11 March 2014 - 06:41 AM, said:

First, let me state that I am not a fan of the idea of the 3/3/3/3 model that PGI has decided to push out in April. I believe it will result in a mostly negative experience in the end.

Ok, I'd like to see why you think it might be worse. Something substantive please.

View PostSupersmacky, on 11 March 2014 - 08:03 AM, said:

With all due respects, I am not from the school of "I know it is a bad idea and has obviously flaws, but let's try it anyway". Trying something out just to see what happens has costs many of businesses dearly and many individuals various body parts.

I never understand the notion of trying something just because. I don't want to spend two weeks, a month, six months testing something flawed that PGI could have fixed before public release. If it is a beta/test environment, sure: I'd be happy to help out. But your statement is basically agreeing the system is flawed, but let's just give it a go anyway. Wow! Really? At the very least you are saying PGIs has no more merit than any other idea out there (including the current iteration). Hardly a motivator to embrace and test it for them while paying for the privileged.

Please tell me you're joking. This is actually how the industry works, a lot of the things you see in video games are a direct result of testing of getting a swath of concepts into a build, most of which get dumped when found out they just don't work well.

Why? To find out what works and what doesn't, to find out what has good gameplay and what's shit, to find out what fits the the best, and what just doesn't fly. You seriously think this is a waste of money?

What's going on with the body parts thing?

#11 Supersmacky

    Member

  • PipPipPipPipPipPip
  • The Messenger
  • The Messenger
  • 239 posts
  • Facebook: Link
  • Twitter: Link
  • LocationGeorgia

Posted 12 March 2014 - 06:45 AM

View PostMoromillas, on 11 March 2014 - 10:13 PM, said:

Ok, I'd like to see why you think it might be worse. Something substantive please.


Have you been asleep lately? Just check any number of posts on the forum regarding PGIs 3/3/3/3 plan. If you want to see my thoughts, just search for my posts. With all due respects, I am not going to repeat myself here just because you are to lazy to look it up.

View PostMoromillas, on 11 March 2014 - 10:13 PM, said:

Please tell me you're joking. This is actually how the industry works, a lot of the things you see in video games are a direct result of testing of getting a swath of concepts into a build, most of which get dumped when found out they just don't work well.

Why? To find out what works and what doesn't, to find out what has good gameplay and what's shit, to find out what fits the the best, and what just doesn't fly. You seriously think this is a waste of money?

What's going on with the body parts thing?


I don't know where you are coming from and what you think I wrote, but I suggest you re-read. I never said you don't test things. My comment was in response to the idea of just putting something into full on production without any sort of testing. And, no, being employed "in the industry" that is not how good companies do it. We have test cycles, code reviews, smoke test, beta tests, customer focus groups, etc. We never just "stick something in and see what happens." Companies that do turn out crap. You don't push something out on the public and in production without a good idea of how it will work. There are always surprises and bugs that didn't get found during the development cycle, but you have a development cycle and beta testing specifically to limit the exposure to those sort of problems. You don't take a "beta" product and make it full on public full of crap just to find out what works and what doesn't. Games that do (and there are a long list of them) end up on the discount shelf (or disappearing completely) fairly quickly.

But, to give you the benefit of a doubt, I am going to assume you either misread or misunderstood what I wrote.

#12 Moromillas

    Member

  • PipPipPipPipPipPipPip
  • 943 posts
  • LocationSecret **** moon base

Posted 12 March 2014 - 07:52 AM

View PostSupersmacky, on 12 March 2014 - 06:45 AM, said:

Have you been asleep lately? Just check any number of posts on the forum regarding PGIs 3/3/3/3 plan. If you want to see my thoughts, just search for my posts. With all due respects, I am not going to repeat myself here just because you are to lazy to look it up.

I don't know where you are coming from and what you think I wrote, but I suggest you re-read. I never said you don't test things. My comment was in response to the idea of just putting something into full on production without any sort of testing. And, no, being employed "in the industry" that is not how good companies do it. We have test cycles, code reviews, smoke test, beta tests, customer focus groups, etc. We never just "stick something in and see what happens." Companies that do turn out crap. You don't push something out on the public and in production without a good idea of how it will work. There are always surprises and bugs that didn't get found during the development cycle, but you have a development cycle and beta testing specifically to limit the exposure to those sort of problems. You don't take a "beta" product and make it full on public full of crap just to find out what works and what doesn't. Games that do (and there are a long list of them) end up on the discount shelf (or disappearing completely) fairly quickly.

But, to give you the benefit of a doubt, I am going to assume you either misread or misunderstood what I wrote.

No no, this is the first and second line in the OP. I'm not going to wade through crap to find some remark on it, I'll just have assume your reasoning as to why 3/3/3/3 would be detrimental is either nonsense, or non existent.

You are remarkably proficient at straw man. If you double check, you'll see that I never said half of those things. Pushing junk onto the public indeed. Holy hell.

No, the ability to "stick something in and see what happens," as you say, gives developers an incredible amount of creative freedom, the results from it are not a waste of time or cost. Just to be clear, we're talking about dev builds, and NOT releases. For example, you have multiple concepts for the ONE weapon. Testing all of these concepts reveals which is fun, which is not, what doesn't work very well, or just doesn't mesh with other systems. This is something fundamental, it's not a waste of time or cost, it certainly doesn't mean something crap is going to end up in the release. I really hope this doesn't get warped into another straw man.

#13 C E Dwyer

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 9,274 posts
  • LocationHiding in the periphery, from Bounty Hunters

Posted 12 March 2014 - 07:56 AM

Any system that lets a team bring a four man organised group all piloting assaults is a bad idea, its a major factor in what makes teams unbalanced now.

I won't be critical of PGI's new idea until I see it working or not

Edited by Cathy, 12 March 2014 - 07:57 AM.


#14 Supersmacky

    Member

  • PipPipPipPipPipPip
  • The Messenger
  • The Messenger
  • 239 posts
  • Facebook: Link
  • Twitter: Link
  • LocationGeorgia

Posted 12 March 2014 - 09:07 AM

View PostMoromillas, on 12 March 2014 - 07:52 AM, said:

No no, this is the first and second line in the OP. I'm not going to wade through crap to find some remark on it, I'll just have assume your reasoning as to why 3/3/3/3 would be detrimental is either nonsense, or non existent.


Well, at least you are comfortable in your ignorance and laziness. I am not obligated to provide information you are unwilling to get for yourself.

View PostMoromillas, on 12 March 2014 - 07:52 AM, said:

You are remarkably proficient at straw man. If you double check, you'll see that I never said half of those things. Pushing junk onto the public indeed. Holy hell.


This coming from the guy who won't even take the time to look up the tons of easily found comments regarding the pitfalls of the 3/3/3/3 system. LOL

View PostMoromillas, on 12 March 2014 - 07:52 AM, said:

No, the ability to "stick something in and see what happens," as you say, gives developers an incredible amount of creative freedom, the results from it are not a waste of time or cost. Just to be clear, we're talking about dev builds, and NOT releases. For example, you have multiple concepts for the ONE weapon. Testing all of these concepts reveals which is fun, which is not, what doesn't work very well, or just doesn't mesh with other systems. This is something fundamental, it's not a waste of time or cost, it certainly doesn't mean something crap is going to end up in the release. I really hope this doesn't get warped into another straw man.


If you had actually read a comprehended the whole thread of this topic you would have seen that I was responding to some one's post stating that the 3/3/3/3 should be stuck in to public/production without any testing done. You then replied to my post countering that point which means you defacto support the idea of making public untested functionality. Are you really having that much trouble comprehending the flow of this topic or are you purposely being obtuse? To help you our, here is the quote from the original poster:

View PostReitrix, on 11 March 2014 - 07:04 AM, said:

There are benefits to the 3/3/3/3 system on paper. Just as you are legitimately pointing out some potential flaws. But without putting the system to the entire playerbase and seeing what we do with it, we cannot form a solid conclusion on how it will turn out.

If it turns out that we can't form a match within a reasonable timeframe, then the Devs can remove it and try with a different system.
But we first need to try it. Also, don't say *use the Test Server* because you wont get a large enough pool of players in there to give an accurate representation of queue time effects.


To translate: let's just do it without any testing and see what happens.

Thus, my reply to that comment which let to your comment. So, it is really you that is not keeping up and creating straw men. My replies are not in regards to a reasonably tested product change, but to the "let's just see what happens" mindset of the original poster. I actually agree with you about testing and it was never my position otherwise. So, I would appreciate it if you would stop creating straw man arguments in order to try and win this one. Unless, of course, you believe untested functionality should be released to a public production environment in which case I thoroughly disagree with you.

#15 Supersmacky

    Member

  • PipPipPipPipPipPip
  • The Messenger
  • The Messenger
  • 239 posts
  • Facebook: Link
  • Twitter: Link
  • LocationGeorgia

Posted 12 March 2014 - 09:29 AM

View PostCathy, on 12 March 2014 - 07:56 AM, said:

Any system that lets a team bring a four man organised group all piloting assaults is a bad idea, its a major factor in what makes teams unbalanced now.

I won't be critical of PGI's new idea until I see it working or not


I am not saying you are incorrect, but is there any support to your statement that 4 mans are so horribly unbalancing? According to the (questionable) metrics of PGI they state that 4 mans are an extreme minority. If that is the case, how then are they having such an balancing effect on the game? And 3 mans bringing 3 assaults is good, but 4 is bad and unbalancing? And if 4 mans are so incredibly rare as PGI indicated, then how much rarer would a 4 man running all assaults be? So, there is no other factor that is really causing the match to be skewed? It couldn't be lack of tactical cohesion, lack of tools in game to foster better team work, lack of skill, lack of experience or any other "lack" in the game right now?

Again, I am not just trying to pick apart what you are saying, but I believe there are a lot more factors at work than a 4 man assault group unbalancing the game. Last night I was in a match with two premades. One was 4 assaults and the other 4 heavies. They were a sync drop and all on TS together. Our entire team got rolled. So, this one experience is runs contrary to your statement. Beyond that I have seen premades get bashed just as easily regardless of being groups on TS or whatever. Yet, weight classes and premades get the blamed.

#16 Lukoi Banacek

    Member

  • PipPipPipPipPipPipPipPipPip
  • WC 2018 Top 12 Qualifier
  • WC 2018 Top 12 Qualifier
  • 4,353 posts

Posted 12 March 2014 - 09:38 AM

Smacky......I still do not see how this system encourages people to play anything other than the optimal/heaviest Mech in a given class. So I don't know if I'll face 2, 3 or 4 Assaults....I'm still only going to play the best assaults.

You wrote:
"- Would continue to encourage taking a player's mech of choice in a class rather than just the heaviest in a class)."

How so? Frankly, how does either version of the MM influence player Mech choice outside of simply limiting the # of them available?

Don't get me wrong, I think your idea is interesting and has some merit, but ultimately I do not think it would encourage all of the behaviors you think it will, nor is it all that much better than 3x4 imo.

So far the best system I've seen offered was Roland's Dynamic-BV system for matchmaking, and even that lacked a couple of key caveats (tied specifically to CW imo) that would make it the best option.

D-BV > Tonnage MM > Flex 2-4> 3x4 as far as I can see.

If Multitalented's old league can manage a somewhat functioning D-BV system with the handful of metrics he had available, I've no idea why PGI couldn't do the same with their awesome array of available metrics, all the way down to a very granular level, that would truly influence game-play the most accurately.

#17 Supersmacky

    Member

  • PipPipPipPipPipPip
  • The Messenger
  • The Messenger
  • 239 posts
  • Facebook: Link
  • Twitter: Link
  • LocationGeorgia

Posted 12 March 2014 - 09:48 AM

Hey, Lukoi:
Admittedly that bullet item is just my view of it and I have nothing other than my opinion and the opinion of others I have listened to on the subject to support it. So, I will concede there is no hard data to back that up.

I also agree with you concerning D-BV > Tonnage MM > Flex 2-4> 3x4

My point in the poll is just to introduce idea that is "better' that the mundane, arbitrary and restricting 3/3/3/3

#18 Lukoi Banacek

    Member

  • PipPipPipPipPipPipPipPipPip
  • WC 2018 Top 12 Qualifier
  • WC 2018 Top 12 Qualifier
  • 4,353 posts

Posted 12 March 2014 - 09:49 AM

Well I cannot argue that 3x4 is just boring as heck, so your model wins in that regard. At least there will be some unknowns!

#19 Monsoon

    Member

  • PipPipPipPipPipPipPipPip
  • Elite Founder
  • Elite Founder
  • 1,631 posts
  • LocationToronto, On aka Kathil

Posted 12 March 2014 - 09:54 AM

Sorry, but your idea sounds worse then the proposed 3/3/3/3 idea.

I'm all for testing 3/3/3/3 and seeing how it actually is, not just how it seems on paper.

Granted none of the proposed systems I've seen really work for me. Weight limits normally would be my preferred choice, but that wouldn't have any impact on a all Highlander/ShadowHawk/Scout Meta group.

Edited by Monsoon, 12 March 2014 - 10:04 AM.


#20 Supersmacky

    Member

  • PipPipPipPipPipPip
  • The Messenger
  • The Messenger
  • 239 posts
  • Facebook: Link
  • Twitter: Link
  • LocationGeorgia

Posted 12 March 2014 - 10:14 AM

Hey, Monsoon. Is there something specific about my idea that appears worse? I'd really like to read it and I am sure there are others that would too.





1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users