StaIker, on 19 March 2012 - 11:12 AM, said:
It seems to me that if the game has to force restrictions in order to create "role based" combat then the roles are not actually performing as required. I don't remember open NR servers being pure assault fests, most players used heavies or big mediums most of the time as they are by far the most flexible platforms for combat with a smattering of other types based on the map and requirements.
MW4 No Respawn servers are not the best thing to judge on.. at all. The reason taking a lighter 'mech could be useful there is that they'd rack up more points per hit, with a large enough modifier that a smaller chassis could actually be competitive in the right hands.. for points. That said, 90% of the 'mechs were either Mad Cats or Daishis anyway: Most of the people piloting smaller 'mechs were generally League players testing out new mechlab designs. Long story short is in these games people only cared about being #1 on the scoreboard and not the victory of their team unless it was an organized League match with persistent results on the line. Quick example: A Wolfhound that has done 25% of the damage of a Novacat could still have the highest score. The fact the team will probably lose from the lack of guns wasn't a major concern in a pub match.
Also, when it comes to role, you have to stop thinking of each unit as a specific "different-but-equal" role like Team Fortress 2, and rather, all different parts of the same entity (the team). That's the reason the tonnage/BV system has worked in the past, because the drop configuration changes the style of the team in question and how they can act as a single unit. Simply put, in actual league games the people who bring the most guns typically will win; there's room for some fast heavies and
definitely room for scouts, but when it comes to choosing 'mechs whose job is to hit things as hard as possible, there's no reason to go lighter.
I'd like to propose a challenge for the next person who wants to say that every 'mech has a role: Name two reasons you would want to chose a Hunchback over an Atlas. The Hunchback is not much faster than an Atlas, carries less weapons than an Atlas and less firepower than an Atlas. This sums up the entire point I've been trying to make, really: You could argue the same about any 'mech with a similar combat role (Catapult vs Mauler, Nova vs Supernova, etc) where there's just no reason to take something lighter in a competitive no respawn match.
StaIker, on 19 March 2012 - 11:12 AM, said:
There is also the problem of trying to coordinate a company that is made up of mechs with vastly different speeds. In order to maintain any kind of combat mass the mediums in the company will be stuck travelling at whatever speeds the assaults can manage for most of the time, I sure wouldn't want to send them out alone when there are 12 enemy mechs unaccounted for.
There's a reason to take faster and more mobile heavies sometimes, for some specific maps where this can pay off but ultimately you still need the most armor and firepower you can bring to a match. That's why heavies fill this role nicely but not mediums. Additionally, the mediums confirmed in the game so far do not move much faster than assaults or heavies for default.
StaIker, on 19 March 2012 - 11:12 AM, said:
I would much rather field 9 heavies than 3 mediums, 3 heavies and 3 assaults. The homogeneous company performs far better than a mixed one in the field, they move at the same speeds and they have similar levels of armor and weapons so they can fight very effectively as a large group. Being forced to take assault mechs when the company really needs to move faster than that is a major tactical liability.
There are definite advantages to taking all heavies and a couple scouts; you are right all heavies or all assaults performs better than a mixed company but that in itself is kind of the problem, however: If there's no limit, all you will likely see are the 2-3 best assaults/heavies and whatever 'mech ends up best for scouting and that's about it, at least in any teams actually playing to win.
Edited by Victor Morson, 19 March 2012 - 12:45 PM.