

[Suggestion] Clan Technology
#121
Posted 07 October 2012 - 09:14 PM
#122
Posted 07 October 2012 - 09:20 PM
Melcyna, on 07 October 2012 - 08:57 PM, said:
when they are fighting at 500m range or less?
i mean seriously now... since NDA is now gone, we can talk of what we know of so far...
now you tell me how on earth would ANYONE not realize their presence with the range of engagement and vision we have?
one guy standing on the caustic edge can see half the map, one guy standing on the snow city ridge can see most of the map baring the tunnel, etc...
oh and i suppose if he packs ER weapon or LRM he can shoot them too..
I mean seriously, the idea that the clan can be NOT noticed with their distinct mech, weapon, and the size of the map we have not to mention our significant view range (much further than the weapon range of most loadout) makes the concept quite ludicrous.
And sure, each IS team will attempt to cripple the other team while still winning the match and do you know what is the BEST method to do so?
Waiting until the clan cleaned one IS team would be STUPID beyond comprehension since that lets them utilize their best strength (ie: their superior individual capability if we follow the rationale that they are simply stronger and balanced by number) and reducing the chance of the IS team of either side from winning, engaging them first is similarly hazardous since they would want to preserve their strength to be the last one standing.
That leaves pretty much only 1 alternative left to maximize their chance, ie: avoiding engagement until light mechs have finished their scout run and waiting until the clan engage the other IS team before engaging the clan.
Once the clan is dead, the other IS team that were engaged would have taken the majority of the damage. This works since ppl in general cannot cope with multiple threat at once (very obvious in the beta when watching the performance of most pilot when they are outnumbered) and thus tend to choose a target (the easier one) and focus on that even if they are being shot to pieces from multiple direction including the back so in all likelihood if the clan engaged one IS team, the other IS team is more or less safe as long as they avoid fighting until the brawl start.
No offense of course, but this is a pretty dumb setting... where the best move is TO NOT FIGHT, and avoid engaging the enemy until the scout finishes their scout run at least to ascertain the participant and waiting until the other teams start engaging each other first.
Same with the clan team if they are present, their best move is also to NOT FIGHT until the other 2 IS team engage each other and to avoid the IS lights before they are ready to strike. Naturally impractical to do on the current known map given there's HARDLY any room to hide ANYWHERE when you have a whole team of mech that needs to be hidden.
In the worst case scenario (assuming the map is much larger than we have now), you have 3 team with lights running around, ALL REFUSING to fight each other since whichever 2 sides fight first automatically loses.
Games that uses 3 sides or more generally are not THIS DUMB, they instead introduce an objective to be achieved that makes the risk of engaging first worth it... instead of simple team deathmatch. ie: for example, King of the Hill objective match etc...
If your game design simply put the 3 team into a team deathmatch, then clan or not, CHAT RESTRICTION OR NOT, then i'm afraid to say but the idea is FUBAR. We already know how such design will end up as...
you seem to be forgetting that the maps we are playing are not the full size maps, neither are the teams the full size, release maps will be MUCH bigger, and the teams will be 12v12, not 8v8. also, your scenario assumes people know they're going to deal with clans in the first place, which they won't initially, and even after the first few matches of that scenario it won't be every drop involving the clan so it could be a while before people would catch on. And yes, that's exactly the way it SHOULD be, the clans would even WANT it to end up that way (canon wise anyway) and it should be more difficult for the clans to win against twice as many adversaries, yet they have the advantage in weaponry and if they work as a team (none of this 'oh let's make the clans pretend to be stupid) there's no reason why they couldn't mow through two greedy IS drop groups.
No I don't expect the IS groups to remain ignorant for long unless the clan group is exceedingly good at staying out of sight, but even a few volleys exchanged before the clans are engaged could make the balance quite a bit more even across the board, and don't forget that the majority of players are going to be pugs, who typically do their own thing even when they should work together. It's all hypothetical but I don't think there's an argument you could make to convince me this couldn't work, do I expect PGI to do it this way? probably not, truthfully the only way I see the clan invasion being handled smoothly is in the form of a PVE storyline with in game salvage being dropped until IS is capable of standing even footing against the clan. Omni mech or not I can't see a equal weight mech equipped with clan gear being significantly outclassed by another who just happens to be able to plop weapons where they want to, tonnage and heat are the primary restrictions and there's still going to be hard points, they'll just be omni hard points.
There's also the contribution reward, which if they applied to salvage would basically make it rewarding to be in the thick of it, though I truthfully doubt that would change the way people play.
Edited by Damion Sparhawk, 07 October 2012 - 09:23 PM.
#123
Posted 07 October 2012 - 11:40 PM
Quote
unless the clan are actually more coherent either through design or process of elimination, THAT SAME PUG constitutes the clan team...
you tell me how exactly we can get pug teams to nevermind hold fire... but to stop breaking ranks and 'doing their own thing'
THIS was part of the point of why non symmetrical team numbers on video games almost NEVER work, in a team based on PUG, ie: RANDOMS you cannot rely on them to do ANYTHING unless you herd them through with either a strict rule (and some way of enforcing it) or a massive reward.
So if the clan is going to be filled with PUG anyway, the whole idea instantly falls apart before we even have to consider it further (of which we have plenty other problems still).
Now what happens if we have NON PUG situation?
We're screwed still...
if the clan is composed of actual team 'clan members etc' ie: coherent players or at least more coherent than PUG, then the game is effectively a 'premade' or 'clan' (clan as in teams and not battletech clan) vs 'randoms' of which there is ZERO chance of it ever being balanced especially without some way of measuring the capability of the participant.
if the IS teams also consist of 'clan' or team members or otherwise coherent players then the entire problem with numerical imbalance that were outlined in the previous posts come straight SMACKING in the face.
Edited by Melcyna, 07 October 2012 - 11:53 PM.
#124
Posted 08 October 2012 - 10:11 AM
#125
Posted 08 October 2012 - 07:39 PM
The whole point why many F2P FPS games concentrate on PvP instead of PvE is PRECISELY because they can dispense away with the AI and constantly creating PvE content (which small developers simply have no resource to do on any regular basis capable of coping with the speed in which players consume them)
if they had to write one, then that defeat part of the goal in the first place with the game, never mind if they actually have the resource to make one.
For human contestant however, if we are to put them like you said (which technically i am keen for, obstacles aside), we have 2 immediate problem....
1. how do you rank the players?
2. how do you pick coherent groups? ie: how to rank the groups?
Open tournament? Outcome determines their capability?
Then what? if X players from that tournament winners are unavailable for the clan event or they simply don't want to participate for the event you plan... then what are we gonna do?
I mean let's face it here, these are still random ppl (everyone on internet is until you get a legally binding contract on them) and yet we cannot bind them with contract or agreement even to do an event or what not so everything is still up in the air as to how reliable they can be for the event we planned.
SO, pretty much the only way it will ever work is to ensure that there is sizable number of them recruited for your event or what not so in the event that some cannot show up or otherwise, you still have enough left to get the event to run through... which comes back to the initial problem again since every time we have a large group of players of various source and makeup we effectively can consider their turnout in terms of response and otherwise as a PUG.
Edited by Melcyna, 08 October 2012 - 08:14 PM.
#126
Posted 08 October 2012 - 07:53 PM
#127
Posted 08 October 2012 - 08:23 PM
that was never the difficult question when ranking players.
But WHAT STAT would be relevant to determine their usefulness in the event? A simple thing like determining how well one can aim for example results in complications like, how do we differentiate those who scores a high dismemberment because they are focusing themselves on gunning on the heavy and assault mechs (where you hardly miss them), vs those who are fighting against the light or medium more often? (who will have harder times accurately placing their shots on specific location)...
For that matter will we take into account as well if someone has a high hit ratio but utilizes a hit scan weapon for the most part like lasers? Or how do we consider ppl who have worse hit ratio but using mostly autocannon weapons? Do we track each specific weapons they use? Do we tabulate that as well into the criteria when we choose the ppl to enter the event?
Alternatively of course we can just say, screw all that, and just pick whichever criteria we like and think is relevant and just adjust on the fly as the event go... shoot first, ask questions later... but good luck with that.
#128
Posted 08 October 2012 - 11:25 PM

#129
Posted 09 October 2012 - 04:21 AM
This have been the case with all the previous stat tracking in various games (FPS or not) and hence why very competitive clan often have trial period.
The current system obviously doesn't exactly track much at all... but even expanded using all the stats that have been known to be tracked by games, chances are spotting the good gems among the player pool is still essentially a gamble until you directly assess them.
That's the problem with an uneven team number...
I mean it will be worthless if you gather a theoretically apt players and they just decide to AFK... you are screwed still...
For the team with the numerical superiority, a few randoms suffering from this is not the end of the world since proportionally they are a small fraction of the team (you lose 1 out of 8 and that's still 7/8 of your force, a loss of 12.5 percent capability theoretically)
For the CLAN team however if they are going to be outnumbered, ONE AFK, or one loose cannon, or one *****, etc.. and you lose 20% of the star's capability.
Overall it's effectively near impossible to balance them since we can't possibly account for all possible factor that can skew the outcome... not that it's easy to do if the team is even, but with it being NOT even the difficulty scales up the wazoo.
#130
Posted 09 October 2012 - 10:45 AM
Keep in mind, the clan invasion is a temporary thing, inevitably the IS are going to gather enough clan technology to start fielding their own, and salvaging clan omni's and whatnot, trying to balance the two is futile you only need to come up with a fun involving way to manage the transition. Sure there will be IS not willing to run clan gear or mechs, but that's a choice and, for all we know they may come up with good ways to mitigate the extraordinary advantage that clan weaponry have over their IS counterparts, we won't know until that comes about. Personally I could care less whether I'm using a IS or clan weapon as long as it does the job

#131
Posted 10 October 2012 - 12:00 AM
still, given the track record of previous past attempt with uneven teams? I am not exactly holding my breath...
#132
Posted 10 October 2012 - 12:15 AM
Its also likely that Clan mechs will have heavier weights in match making and unable to mount IS weps and IS mechs unable to mount Clan tech.
#133
Posted 10 October 2012 - 09:30 AM
Laserkid, on 10 October 2012 - 12:15 AM, said:
Its also likely that Clan mechs will have heavier weights in match making and unable to mount IS weps and IS mechs unable to mount Clan tech.





Melcyna, on 10 October 2012 - 12:00 AM, said:
still, given the track record of previous past attempt with uneven teams? I am not exactly holding my breath...
considering we won't know for a few months yet at the earliest, I wouldn't suggest holding your breath either

#134
Posted 17 October 2012 - 08:48 AM
it dose more damage weighs wore and genertes more heat thats why even experiecned is pilots prefer them
#135
Posted 17 October 2012 - 12:30 PM
thejunsk, on 17 October 2012 - 08:48 AM, said:
it dose more damage weighs wore and genertes more heat thats why even experiecned is pilots prefer them
if it was true (in spite of your bad spelling) that clan gear did more damage weighed more and produced more heat linearly then there wouldn't be an argument, however in some cases they do more damage weigh less and generate less heat, or equal heat to their IS comparable (as in the case of the ERPPC, which does 15 damage for 15 heat and weighs a ton less, compared to IS 10/15) and that is the reason for the argument, it's completely unbalanced in any consideration.
#136
Posted 17 October 2012 - 01:00 PM
Grendel408, on 12 September 2012 - 08:41 AM, said:
Dude, not sure if this is the best idea. If someone is winning a lot, he'd make more money, then he'd be able to buy clan tech "easier" and would increase his advantage.. The opposite is also true.
They'll have to figure it out some other way to balance this.
I've read, as some people posted here too, that there's a chance that they're kept outnumbered..
As everyone else, I'm curious about it. =]
[ ]'s
#137
Posted 17 October 2012 - 01:06 PM
Damion Sparhawk, on 17 October 2012 - 12:30 PM, said:
You forgot to mention that beam weapons have greater range ! (at least in MW4M)
[ ]'s
#138
Posted 17 October 2012 - 02:07 PM
JokerPW, on 17 October 2012 - 01:06 PM, said:
You forgot to mention that beam weapons have greater range ! (at least in MW4M)
[ ]'s
and yes, clan weapons typically have greater range than their IS counterparts, which is a much bigger consideration in TT than in this game, though certainly not irrelevent either.
#139
Posted 17 October 2012 - 02:14 PM
thejunsk, on 17 October 2012 - 08:48 AM, said:
it dose more damage weighs wore and genertes more heat thats why even experiecned is pilots prefer them
I recommend having a look at the weapons and equipment lists.
Classically, Clan-built weapons and equipment often weighed the same or less than the IS counterpart(s), required the same number of critical spaces or fewer than the IS counterpart(s), had equal or longer effective ranges as the IS counterpart(s), delivered equal or greater damage per salvo than the IS counterpart(s), and generally produced the same amount of heat per salvo as the IS version of the same item... often, all at once.
Altogether, this combination of characteristics meant that a given Clan-built item outclassed its IS-built counterpart in nearly every meaningful metric - it was simply equal or better in almost every possible way.
Hence, the concern that
1.) much of the playerbase will jump to the Clans the moment they become available, so that they can get easy access to the "superior" Clan tech,
2.) the result of (1) would be that the IS factions are drained to the point that the Clanners outnumber IS players by substantial margins (the opposite of "how it should be"),
3.) what IS players are left will have to generally, if not totally, abandon IS tech for Clan tech just to remain competitive with same-skill (or even somewhat lesser-skill) Clan players (who would then have overall population advantages in addition to the tech advantage... and who generally cannot be expected to adhere to Clan RoE, thus negating the the other primary offsetting point to the Clans' tech-based advantages), and
4.) the game on the IS-vs-IS side devolves into a race to see who can get the most "superior" Clan tech first, with IS tech largely going abandoned and unused.
#140
Posted 17 October 2012 - 02:24 PM
Remember this is real time, 5 years from now, I don't think clan tech will be very accessible at all yet.
I expect it will be a rare treat to get a small clan laser even one year after the initial invasion.
2 user(s) are reading this topic
0 members, 2 guests, 0 anonymous users