I am, on 08 April 2013 - 09:30 PM, said:
You put alot of thought into that, and I thank you for the reply. One thing I noticed was you did not mention reducing the max group size, or increasing the total numbers playing per side. Would you say that is because you think those things would not help, you are just trying to narrow your response to the current available architecture, or because you think that given appropriate designations, a well balanced meta can be achieved through ELO weighting and designations alone?
The goal is to increase group size to allow for player and grouping flexibility (you won't always have 8 people, even if you have more than 4). This produces less arbitrary and less frustrating experiences for the player base. It also decreases exploitative behaviour (you can still play with 5-8 players in the public group queue via sync drops).
Of course you need to recognize the increased effectiveness that group play has on performance; but the system shouldn't provide arbitrary and frustrating restrictions for the sake of a ham-fisted approach to balancing.
Elo multiplers provide for a somewhat more elegant (in that the details and complexity are hidden from the player) balance; a recognition of the various exigent circumstances that can affect the win/loss ratio outside of player skill alone.
At this time, I'd only suggest that; should that prove to be inadequate, I'd explore more fine grained methods for balancing group play, with the goal of making the new player experience enjoyable, while still retaining the fair competitiveness of the overall playerbase.
Quote
I'm just the one who wants to be able to convince his friends to try it again, to try and convince them to like it, in hopes maybe I can get them into playing it with me. By and large, they all have had it with the game for one reason, another, or a handful of reasons. Summarizing all of their gripes into one word is an over simplification, but if I was forced to, I would say that word is "balance".
On that front, although its a feature change, I'd definetly want to include a lobby (which is something PGI are currently... 'exploring' (which is a very nebulous term indeed)), so that friends can have friendly more controlled matches with their friends for whatever reason (competition, introduction, tuition, laughs, etc). It would be one of the simpler things that can be done to make the game significantly more accessible to the player base.
I'd even encourage modding for lobby based play (akin to Starcraft 2) - if only because the player base has an amazing wealth of talent that collectively, no developer can ever hope to match. Amazing things (like the whole E-Sports scene, DOTA2, etc) arise from this sort of encouragement of the playerbase. But I digress.
Quote
So Zap, what do you think. Am I crazy, is there validity in my position, and what are your thoughts on the same?
Feel free to rip me up if you want Zap, I feel alot of venom directed towards me on these forums. What I say, is what I truely believe. Perhaps you could confirm, dispell, or develop my perceptions of waht is preventing MWO from achieving balance. Appreciate your time in responding at all.
I have no secret insight as to the machinations in PGI. I can only assume that they are for the most part a dedicated and passionate lot that truly believes in what they're trying to do.
From what I've gathered, they're not exactly top tier game development material - this is by far their biggest undertaking to date... but what they've achieved to date has despite the balance and the bugginess been extraordinary - the best fufillment of the Mechwarrior vision so far. Despite the numerous flaws, there is an indelible brilliance to the gameplay experience that no other game is able to touch.
But there are higher level goals and decisions that seem to limit their nimbleness in responding to player feedback.
They've purportedly adopted a strategy of listening to the noise and seeing what signal arises: but this isn't entirely evident when you consider how little they've acted upon the volume of feedback on various topics such as ECM and MGs/Flamers.
A large part of that arises from the community's misunderstanding on the actual gameplay mechanics that underpin the entire experience; so they end up making a whole heap of noise that if implemented would actually be detrimental to the game.
However, they've taken that attitude to the nth degree - beyond what is reasonable or effective in my opinion. That is to say, they have an underlying assumption that the player base *doesn't* know what it's talking about... and that once *their* gameplay systems and ideas are fully in, *then* the playerbase can provide adequate feedback.
To an extent, I understand this; but... we already have concrete proof in the form of MGs and Flamers that even when their additional systems are in place that they still underperform (or overperform as in the case of the Raven 3L). While things like PPC affect ECM, PPCs been buffed, Host State Rewind, etc have indeed made a dent on the overpoweredness of things like the Raven 3L - they remain still slightly unbalanced.
Perhaps they'll be right in some instances: once all their proposed systems are in place, the game will be appropriately balanced. But the reality is, as I've stated before, it still means that millions of hours are been played in an unbalanced state. Real players are coming and going based on their impressions of the game in this suboptimal state.
Is it really that difficult to adjust the gameplay values to reflect the current reality of the features? If Host State Rewind isn't in place, how about you give the Raven 3L a big ol' nerf in engine size until it is? If streaks are too effective on ravens, how about reducing the damage for a few patches until other factors arrive to balance them out? If the intent of machine guns is to allow players to disable the internal structure of the mech; including ammo, weapons, engines, actuators; and you haven't gotten the engine and actuator disabling part of things in, why not just buff the machine gun damage until those other factors arrive to show that they're OP?
Is the game really that poorly put together that changing basic variables could create issues with the stability of the game? I would hope not; but then what else other than perhaps obstinacy is stopping them from quickly iterating?
PGI... needs a much more open and transparent form of communication with its player base. Give us an insight into what is actually going on and the thinking behind those decisions. Harness the additional collective cognitive capacity of your playerbase. Fear not the secrets of your developmental process - no one cares about corporate espionage for this kind of product. The important bits of the whole endeavour are 1. the licenses, 2. the work already done.
I know there are interests beyond those at PGI that may determine what can and can't be communicated - but the need to control the message in such a perculiarly rigid way can only emerge from thinking that doesn't completely grasp at the communicative phenomenon of the internet.
Just level with us - we can be adults about it, as long as people don't have to grasp at straws to divine whether or not their actions are indeed reasonable or not. If you're working on too much content at the moment, but look unproductive because that content is slated to launch several months down the track; and you want to hold back... don't. Just let it all out - especially the parts where you double back on an initial decision after re-examining its merits. That's the development process; and it needs to be more transparent.
If we can better understand why something is done the way it is, we can stop arguing in circles, and stop producing counterproductive feedback.
But that would require a degree of humility from PGI to accept that the collective wisdom of the playerbase can have validity beyond their own ideas of what they'd like to see out of the game.