Jump to content

Mwo Is Not Battletech And That's Why It Is Broken


118 replies to this topic

#61 Gas Guzzler

    Member

  • PipPipPipPipPipPipPipPipPipPipPip
  • Big Daddy
  • Big Daddy
  • 14,274 posts
  • LocationCalifornia Central Coast

Posted 10 October 2015 - 09:13 AM

View PostAlistair Winter, on 10 October 2015 - 01:57 AM, said:

How was it a step in the entirely wrong direction?


How was it a step in the RIGHT direction?

#62 Jaeger Gonzo

    Member

  • PipPipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 1,219 posts

Posted 10 October 2015 - 09:23 AM

View PostUltimatum X, on 09 October 2015 - 08:53 AM, said:

This idea has come up so many times, and one thing that I've learned is that some players have a warped perception of how long their mechs get to live because table top games take a really long time.



So, tell me, just how long do the actual simulated battles last?

You have a 10s turn, so Total Average Turns x 10s / 60s = Total Battle Time (in minutes)


So how many turns would an average battletech battle take? (do you need to adjust for number of mechs? probably - I'm sure some BT expert can enlighten).

30 turns, that will give around 5 min, MWO still has it probably less, as in MWO you move in to position 3-4 min and do fighting for 2-3 min, if fight really happen, not speaking about cases when everyone is hiding afraid to pick your head up and be instakilled. but that`s quiet irrelevant. What is relevant is human perception, the amount of tactical choices and movement that TT gives in a given live time of meks and that`s significantly more then MWO.

#63 Hawk_eye

    Member

  • PipPipPipPipPipPip
  • Bad Company
  • Bad Company
  • 325 posts
  • LocationGermany

Posted 10 October 2015 - 09:24 AM

View PostFupDup, on 10 October 2015 - 08:05 AM, said:

This quote right here needs to be repeated and requoted.

The majority of the "muh TTK" people seem to have at least some of a background in the Tabletop game, claiming that mechs were extremely durable in that board game.

But so far nobody has actually provided some anecdotes about how long a TT mech will even survive during combat.

If somebody wants to make a claim that MWO's TTK is too low compared to TT, they need to actually provide some estimates of TT's TTK for comparison. And those comparisons should be based on the number of TURNs, not the total game time, because otherwise calculations and crap would make the game drag out for ten times longer than the "real combat" part of it.


A fun side-note is that some of the same people who claim that our TTK is "abysmal" are the same types who want certain weapons to be "devastating" (e.g. AC/20, Gauss Rifle, PPC). Logically, wouldn't weapons that can kill or cripple mechs in one hit make the TTK actually VERY low by definition? That's what makes them "devastating" after all, if they weren't devastating then they wouldn't kill mechs in a short period of time. And if they didn't kill mechs in a short period of time, then they wouldn't be "devastating" anymore.


In my TT games, lance vs. lance, it took usually between 8 (very short) to 15 (very long) turns.
company vs. company, about 15 to 30 turns, depending on terrain and size of map and stuff.

Note: I mostly played 3025. Games with clans took a bit less time, while IS vs. Clan was roughly equal, but only because we usually played with 4 IS lances (often with Infantry/armor support) vs. one Clan Binary (i.e. 16 vs. 10) so there were more mechs/tanks on the board.

Note2: The thing I´d like to throw in though is, that while yes, the entire battle in TT (10 sec per turn) didn´t last as long as a MWO match, the time to actually _kill_ a mech (from the first shot fired at it until it died, let´s call it engagement time) or the number of shots (not hits) you had to take at a mech to kill it was usually longer/more than what I have seen in my very limited experience in MWO.
I guess this is mainly due to the fact alpha-ing was a _very_ rare thing, only to be done in desperate moments.

Edit:
And also because you missed a whole lot in the TT.
I guess my own performance in MWO (which is _horrible_) would be equal to a regular mechwarrior in the TT, while an average player would be the equivalent of an Elite mechwarrior and the top 10% would be equal to living legends in the TT.
End edit.

Again, keep in mind I am strictly speaking 3025 here.

Edited by Antecursor Venatus, 10 October 2015 - 09:29 AM.


#64 Illya Ghost Bear

    Member

  • PipPipPipPipPipPipPip
  • Survivor
  • 500 posts
  • LocationTaking your planets, eating your cookies.

Posted 10 October 2015 - 09:26 AM

View PostRedMercury, on 09 October 2015 - 08:35 AM, said:

Why are there so many gameplay problems with MWO?

FASA's Battletech was far from perfect, but it was a heck of a lot better than the balance MWO has.

I'm not going to offer solutions in this post, there are plenty suggested by people far more insightful than me, and there are indeed very difficult design choices needed. I just want to point out the balance problems which logically follow because of the design decisions of MWO, because it deviated from Battletech.

The point I want to drive home is that deviation from Battletech led to alternative designs, some of which were bad. These bad design decisions necessitated further deviation from Battletech and further bad design decisions, in order to balance various aspects of the game.

When you make one seemingly small change in one aspect of the game, for example to adapt it to be a 3D shooter, you affect balance all over the place, because balance is an interconnected web. Any seemingly small change can have a global effect, with unintended and unplanned consequences. Fixing these problems often leads to additional problems, problems you didn't even know could happen.


1. Weapon cooldown. A standard Battletech round represents 10 seconds of real time. Movement, heat dissipation, and weapon recycling times are all based on this. By making weapons recycle in 2 to 4 seconds, the effective firing frequency was increased up to 4x, (to make the game more arcade-like, because MWO is a thinking man's shooter). This broke:

A. balance between heat generation and heat dissipation [Result: everyone runs hot]

B. balance between heat intensive (energy) weapons and low heat (ammo) weapons, which are heavier.

C. Ammo counts [Result: ammo runs out sooner]

D. Balance between fast mechs which rely on speed to avoid firepower, and slower mechs, since fast mechs can now be shot at more often [Result: lights are less survivable]


2. Pin point damage when firing many weapons simultaneously. This made mechs very easy to kill. This destroyed the balance of:

A. armor vs mobility vs firepower (which is basically, the core of mech design) [Result: meta where pin point damage is king]

B. pin point weapons (energy and autocannons) versus missiles and LBX [Result: LRMs, SRMs and LBXs need buffs to be relevant, ghost heat, Gauss delay]

C. amount of protection needed for various body parts (e.g. head, CT) [Result: buffed head armor and internals needed, low average amount of damage per kill]


3. Armor increased by 2x. This is required to fix #2. But this, in turn, breaks balance between

A. armor vs mobility vs firepower. Armor and firepower back in balance, but mobility gets the shaft. Speed costs mass, just like guns. If guns fire twice as often (even if effect is the same due to armor buff, they have the potential to hurt, versus 0 potential if a skilled pilot uses speed to go in and out during cooldown), speed should be twice as valuable per ton as well. [Result: light mechs are less useful as skirmishers]

B. ammo counts versus armor. [Result: ammo weapons are hurt because they need more tonnage/critspace for more ammo]


4. Not using most Battletech heat penalties. This is required because of #1. This, in turn, breaks balance between:

A. high heat weapons vs low heat weapons [Result: laser vomit]

B. mechs with high reserve firepower versus mechs with lots of heat dissipation. Though due to customization, who cares about stock mechs? "Screw the lore!" [Result: unless you are a brawler, front load your damage]

C. fighting while remaining in line of sight vs fighting with intermittent line of sight. Since you don't have to roll for a shutdown chance after an alpha, just hide and cool down. [Result: hill humping, poking, and poptarting]


5. Giving internal components health, spreading laser critical checks across the entire hitscan. This breaks:

A. SRM and LBX (and lesser extent, MGs and AC2s ) are niche weapons for finding criticals, now they are useless in that niche. [Result: no one uses LBX, SRMs need buffs to stay relevant]

B. Large damage weapons versus small damage weapons. In Battletech, a PPC was good for punching holes, but not as good for getting criticals. Now it is good at both, e.g. better than a pair of ML.

C. Number of criticals is lower than what one would expect in real Battletech, since components have health and critical hits are spread evenly. This relatively benefits mechs with more internal structure, e.g. heavies.

D. Reduces risk from having armor destroyed with only slight internal damage, e.g. shot in the back. This relatively hurts fast mechs, hurts flanking as a strategy. [Encourages front loading of armor, ignore minor threats to the rear, deathball/nascar]


6. Shooting LRMs without line of sight, without needing a designated spotter who cannot fire othwerise. Increasing LRM range to 1000m. This makes LRMs very powerful weapons. They bring firepower without locality and line of sight. It's like your team's LRM boats are all riding on top of your fastest light mech. To balance this:

A. LRM speed is reduced, which makes them less useful as direct fire weapons. [Result: no LRMs in Tier 1]

B. ECM is made to counter LRM spotting. This ironically makes LRMs useless when ECM is present. [Result: no LRMs in Tier 1, LRMs are feast or famine, depending on whether ECM is present]

C. AMS has to be buffed to counter LRMs


7. Putting Clan mechs on equal footing as IS mechs and expect players to be happy with asymmetrical power without asymmetrical incentives.

A. To keep them competitive, IS mechs need magic upgrades to weapon systems. But at least these changes are more "local" and spread less badness in terms of balance.


8. Lack of melee or rules preventing collisions. I know this is hard to get right. But it is a core part of Battletech. Without it:

A. IS mechs are further disadvantaged comapred to Clan

B. Heavies mechs are disadvantaged comapred to light

C. Mechs with hand actuators get no benefit for those lost critical spaces

D. balance between brawling vs plinking is broken

E. jump jets are less useful for close quarters combat



There's probably more, but I'm tired.

Once Battletech hit the 3050/Clan Era, all semblance of balance went out the window. Even their game line developers, past and present acknowledge that. Only 3025 TT was remotely balanced, and it still had it's warts.

Guess what? When you play 3025 Stock Mechs Matches? This game becomes eerily well balanced.

Coincidence, or conspiracy? You make the call.

#65 MW Waldorf Statler

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 9,459 posts
  • LocationGermany/Berlin

Posted 10 October 2015 - 09:27 AM

TT=the Pilot can not aiming a single Hitzone by a enemy in 20m, firing of the Center and hit the left Leg ...that good for TT in a Real PC Game is nonsense

With this TT weapons firing to Aerospace Fighters ? seeing MWLL...TT Rules most stupid- Tanks and Aerofighters dominated the game

Only to play of a Table , the Weapons range is like the range of Weapons of the US Civil War

Edited by CSJ Ranger, 10 October 2015 - 09:44 AM.


#66 Jaeger Gonzo

    Member

  • PipPipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 1,219 posts

Posted 10 October 2015 - 09:32 AM

View PostIllya Arkhipova, on 10 October 2015 - 09:26 AM, said:

Once Battletech hit the 3050/Clan Era, all semblance of balance went out the window. Even their game line developers, past and present acknowledge that. Only 3025 TT was remotely balanced, and it still had it's warts.

Guess what? When you play 3025 Stock Mechs Matches? This game becomes eerily well balanced.

Coincidence, or conspiracy? You make the call.

When we played Stock Clan v IS, balanced around 5v8 was not bad either.
But yeah this game just screams for 3025 Stock Mode.

#67 Alistair Winter

    Member

  • PipPipPipPipPipPipPipPipPipPipPip
  • Storm
  • Storm
  • 10,823 posts
  • LocationBergen, Norway, FRR

Posted 10 October 2015 - 11:03 AM

View PostAdiuvo, on 10 October 2015 - 09:03 AM, said:

Homogenization of mechs, which in the end resulted in less choice overall. A complete lack of weapon rebalance, which is the core of the problem, in favor of structure quirks that are meaningless. Somehow buffing the 2 most powerful mechs in the game (Daishi, TBR). Illogical nerfs like Locusts, Mist Lynx. The incorrect assumption that any amount of 'infotech' is as useful to the same degree as armor, speed, and firepower. In general a lack of any attention to the details.

There's more, but that's a short enough summary.

Fair enough. I just view it as unfinished business, trying to experiment with some of the new variables (Infotech) before actually doing a public test where the aim was to make every mech balanced. I don't think PGI is overestimating the usefulness of infotech or ust how OP the Timber Wolf is. But time will tell. It's not long before the next public test.

View PostGas Guzzler, on 10 October 2015 - 09:13 AM, said:

How was it a step in the RIGHT direction?

They moved away from the insane power creep of weapon quirks. The matches with mostly Inner Sphere mechs were absolutely wonderful, because TTK went up drastically. No more Thunderwubs or AC5 Dragons turning everything to pulp in seconds.

This was ruined by the fact that Clan mechs were indirectly buffed (by removing negative quirks), but I hope that they'll find a way to balance the Clans and IS mechs without the use of hyperquirked IS mechs.

And Infotech needed to be in the game, for sure. Using weapon quirks for everything was incredibly dull.

#68 FupDup

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 26,888 posts
  • LocationThe Keeper of Memes

Posted 10 October 2015 - 11:07 AM

View PostAlistair Winter, on 10 October 2015 - 11:03 AM, said:

...Using weapon quirks for everything was incredibly dull.

Using agility and structure on every single mech is equally dull, which is exactly what the PTS did.


The only real positive from the PTS is that we now have the coding ability to mess with sensor properties (although the values were wacky)...every other feature of it was a flop.

#69 Gas Guzzler

    Member

  • PipPipPipPipPipPipPipPipPipPipPip
  • Big Daddy
  • Big Daddy
  • 14,274 posts
  • LocationCalifornia Central Coast

Posted 10 October 2015 - 11:10 AM

View PostAlistair Winter, on 10 October 2015 - 11:03 AM, said:


They moved away from the insane power creep of weapon quirks. The matches with mostly Inner Sphere mechs were absolutely wonderful, because TTK went up drastically. No more Thunderwubs or AC5 Dragons turning everything to pulp in seconds.

This was ruined by the fact that Clan mechs were indirectly buffed (by removing negative quirks), but I hope that they'll find a way to balance the Clans and IS mechs without the use of hyperquirked IS mechs.

And Infotech needed to be in the game, for sure. Using weapon quirks for everything was incredibly dull.


Power creep with weapon quirks isn't a thing. Giga quirks are only present on old mechs. All the new IS mechs have moderately useful weapon quirks at best. Everyone complains about them because their quirks pale in comparison to mechs like the 5SS.

Also, the complete lack of weapon quirks is boring as ****. Variants are balanced only by durability/agility/infotech, and that sucks.

Edited by Gas Guzzler, 10 October 2015 - 11:18 AM.


#70 Davers

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 9,886 posts
  • Facebook: Link
  • LocationCanada

Posted 10 October 2015 - 11:20 AM

View PostAlistair Winter, on 09 October 2015 - 10:40 AM, said:

8- In Battletech, some mechs are inherently more valuable than others and in fair matches you won't see Dire Wolves matched up against an equal number of Awesomes, you won't see upgraded mechs with XL engines, Endo and DHS matched on even terms with cheaper tech variants, etc.

In other games, there's a risk-reward calculation based on things such as repair & rearm, but we no longer have that in MWO. 99% of the MWO population considered R&R a bad thing and 99% of the MWO population were dead wrong. IMHO :ph34r:


What kind of things are we trying to "punish" with R&R? XL engines? Ok, so now light mechs are slower and with less weapons, while heavier mechs are unaffected. Ammo based weapons? Because SRMs, LRMs, and ACs are so good right now? Not to mention that LB-10X ammo costs the same as gauss. How do Omnimechs fit into that whole risk vs reward calculation? Since no one can say what exactly R&R is supposed to fix, it's best we don't have it.

View PostJaeger Gonzo, on 09 October 2015 - 09:46 AM, said:


First example of stupidity and ignorance of TT result in worst outcome you can find with ECM mechanics.



Why is it everyone hates on ECM for being different from TT, but no one complains about the auto hitting LRMs or SSRMs that are certainly not in TT?

View PostDuke ramulots, on 09 October 2015 - 12:02 PM, said:

That's a deflection, not an argument. The balancing feature used can easily be ported over to a PC game.


Main balance in TT came from random hit locations.

View PostUltimatum X, on 09 October 2015 - 02:21 PM, said:


Can you be more specific than this?

All I need is an estimate, a number of turns for 4v4 is a starting point.


If 2 teams are actively engaging the games will be short. 4v4 might last 6-8 turns, 8v8 will be decided (though not all mechs necessarily destroyed) by turn 10-12. Some games can have a LOT of "clean up" if you need to actually destroy every mech due to random hit locations and everyone's favorite minigame, "Catch The Jumping Light Mech".


View PostEchoFreebirth, on 10 October 2015 - 07:47 AM, said:


THIS

Even without changing any of the underlying math or the actual mechanics of MWO, I would make one change inspired by TT that could improve gameplay experience for all: let the matchmaker calculate groups not on the basis of TONNAGE * SKILL but BATTLE VALUE * SKILL.

Example: stock Arctic Cheetah Prime vs Centurion CN9-A
  • MWO: Light vs Medium, therefore the search for balance favors ACH
  • BT: 1334BV vs 945BV, therefore the search for balance favors CN9
Surprised an ACH survives until the end as a 30-ton killing machine? Well in TT you would not be!! Because it isn't just a "light mech", it's a freaking 1334BV mech, just slightly inferior in value to a Catapult C1 that has over twice its tonnage.


Different tech levels might have been unbalanced in TT and clans may be OP in 3050, but Battle Value allows us to craft balance matches just by looking at a simple number.

BV is meaningless in MWO. To begin with, neither of those two mechs would be carrying the weapons, armor amount, or engine size (in the Centurion's case) as the stock versions those numbers are based off of.

#71 FupDup

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 26,888 posts
  • LocationThe Keeper of Memes

Posted 10 October 2015 - 11:26 AM

View PostDavers, on 10 October 2015 - 11:20 AM, said:

Why is it everyone hates on ECM for being different from TT, but no one complains about the auto hitting LRMs or SSRMs that are certainly not in TT?

I actually really, really hate MWO's missile mechanics, with Streaks being the worst offender. I hate PGI's anti-light RNGesus aimbot system. :angry:

#72 Sarlic

    Member

  • PipPipPipPipPipPipPipPipPip
  • Hearing Impaired
  • Hearing Impaired
  • 4,519 posts
  • LocationEurope

Posted 10 October 2015 - 11:30 AM

View PostFupDup, on 10 October 2015 - 11:26 AM, said:

I actually really, really hate MWO's missile mechanics, with Streaks being the worst offender. I hate PGI's anti-light RNGesus aimbot system. :angry:

I have recording someone in my low slung Orions how terrible SSRMs are.

I was absolutely gobsmacked when i almost couldnt kill a cored Awesome. The homing of those missiles sucks ass and it still spreads all over the place instead of what a homing should do and that's one targetting component. It does do damage, it does inflicts but it sucks major butt. On Jenners etc it's much easier because the obvious: giant walking centre torso. Awesome should be do-able, but i had to get closer then i thought.

Edited by Sarlic, 10 October 2015 - 11:31 AM.


#73 Cygnus17

    Member

  • Pip
  • Ace Of Spades
  • Ace Of Spades
  • 16 posts
  • LocationOKC, OK

Posted 10 October 2015 - 11:33 AM

View PostUltimatum X, on 09 October 2015 - 02:21 PM, said:


Can you be more specific than this?

All I need is an estimate, a number of turns for 4v4 is a starting point.


Depends on the skill of the players... we had matches of 4v4 that could last well over 75 turns. Basically you didn't just run every mech into short range and blast away. You used every inch of the map, moved constantly, staying in cover for max protection and weapons rarely even hit (if they didn't have a at least a +4 or more to-hit, you're doing it wrong.) -- and when damage was done it was spread out (unless you were in partial cover but then that was your fault). Once weapons were getting destroyed by internals OR you ran out of ammo, the game ended fairly quickly since mechs got into melee range. One charge or death from above usually did enough damage to end the fight for that mech... kicks could really ruin your day when a mech got legged.

There was a lot of thought into were mechs would be at the end of a turn and where they would move next turn, etc. Long range mechs had to be protected, short range brawlers were used to corral enemies into advantageous positions, etc. That's why mechs had a weapon for every range of combat. You used that LRM-20 on the Atlas -- usually all the ammo before even firing a shot with the AC/20. Rear firing weapons were really important. Jump Jets were key in moving to the best places on the map. Heat was a constant problem, as the more heat you carry over each turn means the harder it was to hit a target which usually meant you spending more heat to fire more weapons to do damage and you NEVER wanted to shutdown. A shutdown mech was basically dead before it could restart. SO there were several turns of moving away and NOT firing just to stay cool) I think we fired a true alpha maybe once a game from a single mech if you were lucky (this from a heavy or larger mech). Firing your TWO weapons from a wasp (med Laser and SRM-2) happened, haha.

If you were dying in 10 turns or less than either you were the unluckiest person on the planet (it happens with great rolls sometimes) OR you didn't understand how to play.

#74 Alistair Winter

    Member

  • PipPipPipPipPipPipPipPipPipPipPip
  • Storm
  • Storm
  • 10,823 posts
  • LocationBergen, Norway, FRR

Posted 10 October 2015 - 11:40 AM

View PostFupDup, on 10 October 2015 - 11:07 AM, said:

Using agility and structure on every single mech is equally dull, which is exactly what the PTS did.
The only real positive from the PTS is that we now have the coding ability to mess with sensor properties (although the values were wacky)...every other feature of it was a flop.

PGI has clearly explained that the PTS was just a part of the process and that it was never intended to be a complete attempt to rebalance the game. They just needed to look at whether Infotech worked as a balancing tool, and then get a sense of what kind of impact it made. It's dumb that they didn't communicate this before, and we all have our doubts about whether the next PTS will actually be anywhere near balanced. But it's still intellectually dishonest to talk about the last PTS as if it was a real attempt to completely rebalance the game, when people know that it wasn't supposed to be that at all.

From what I understand, it seems like their plan is to use 4 different variables to balance the quirks. Infotech, mobility (both agility and speed), durability and weapon quirks. The weapon quirks will be very rare, they'll probably have more to do with range and projectile speed than DPS-related stats. Previously, we lacked infotech quirks and both the mobility quirks and durability quirks were almost only there to compensate for some specific vulnerability (such as the Hunchback's hunch) instead of actually being a significant strength (such as the massive durability bonuses for the AS7-D in the public test).

So I'd say it's a step in the right direction, from that point of view. I fear PGI will disappoint me when we get to the next step. They have a history of saying "This is only the beginning" and then it turns out the beginning was 95% of the end result. But never the less, I figure it a step in the right direction.

Put it this way: if you gave someone like Mcgral carte blanche to rebalance the quirks for every mech in the game as he saw fit, using the 4 new variables (infotech, mobility, durability and weapons), I think he could easily come up with a more interesting game than the status quo, with something that more closely resembles role warfare (while not really living up to the name, untill they fix the skill tree)

View PostGas Guzzler, on 10 October 2015 - 11:10 AM, said:

Power creep with weapon quirks isn't a thing. Giga quirks are only present on old mechs. All the new IS mechs have moderately useful weapon quirks at best. Everyone complains about them because their quirks pale in comparison to mechs like the 5SS.
Also, the complete lack of weapon quirks is boring as ****. Variants are balanced only by durability/agility/infotech, and that sucks.

I definitely disagree with you there. I would much rather have the quirks balanced with durability, mobility and infotech (with weapons being a minor variable) instead of the status quo.

All the new IS mechs do have relatively modest weapon quirks (although it was a noticeable pleasure to play the game with no weapon quirks at all), but many of them have enough power creep in-built anyway. Compare the hardpoint inflation for the King Crab vs the Atlas, for example.

#75 Davers

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 9,886 posts
  • Facebook: Link
  • LocationCanada

Posted 10 October 2015 - 11:42 AM

View PostCygnus17, on 10 October 2015 - 11:33 AM, said:


Depends on the skill of the players... we had matches of 4v4 that could last well over 75 turns. Basically you didn't just run every mech into short range and blast away. You used every inch of the map, moved constantly, staying in cover for max protection and weapons rarely even hit (if they didn't have a at least a +4 or more to-hit, you're doing it wrong.) -- and when damage was done it was spread out (unless you were in partial cover but then that was your fault). Once weapons were getting destroyed by internals OR you ran out of ammo, the game ended fairly quickly since mechs got into melee range. One charge or death from above usually did enough damage to end the fight for that mech... kicks could really ruin your day when a mech got legged.

There was a lot of thought into were mechs would be at the end of a turn and where they would move next turn, etc. Long range mechs had to be protected, short range brawlers were used to corral enemies into advantageous positions, etc. That's why mechs had a weapon for every range of combat. You used that LRM-20 on the Atlas -- usually all the ammo before even firing a shot with the AC/20. Rear firing weapons were really important. Jump Jets were key in moving to the best places on the map. Heat was a constant problem, as the more heat you carry over each turn means the harder it was to hit a target which usually meant you spending more heat to fire more weapons to do damage and you NEVER wanted to shutdown. A shutdown mech was basically dead before it could restart. SO there were several turns of moving away and NOT firing just to stay cool) I think we fired a true alpha maybe once a game from a single mech if you were lucky (this from a heavy or larger mech). Firing your TWO weapons from a wasp (med Laser and SRM-2) happened, haha.

If you were dying in 10 turns or less than either you were the unluckiest person on the planet (it happens with great rolls sometimes) OR you didn't understand how to play.

So over 2 and a half hours for a 4v4 game? LOL Maybe if you used 10 map sheets or something crazy like that. Standard game of 4v4 is 2 map sheets with players being about 1-2 turns out from being in long weapon ranges. Add in Headshots, Through Armor Criticals, Falling Damage, Ammo Explosions, the power of physical attacks, and few mechs running around in max armor, and I really have to wonder what game you were playing. :P

edit: My "2 and a half hour game" reference is based on the idea that both players each take their turns entirely in one minute each, and no breaks are taken. A more 'realistic' time might be 3 minutes bringing the game time up to 7.5 hours for a 4v4, which is just utterly insane, and more than 5x the time allotted to official BT Tournaments.

Edited by Davers, 10 October 2015 - 11:52 AM.


#76 MischiefSC

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPip
  • The Benefactor
  • The Benefactor
  • 16,697 posts

Posted 10 October 2015 - 11:53 AM

View PostGas Guzzler, on 10 October 2015 - 09:13 AM, said:


How was it a step in the RIGHT direction?


I actually sorta like the sensor changes. Also some structure buffs are not bad; I like longer TTK as a given rule. It favors precision and punishes bad aim. It promotes skill gaps and that's good.

The weapon buff changes were garbage but if you assume (dream/hope/pray/whatever) that an actual weapon rebalance is coming, it's not a bad start.

#77 Cygnus17

    Member

  • Pip
  • Ace Of Spades
  • Ace Of Spades
  • 16 posts
  • LocationOKC, OK

Posted 10 October 2015 - 11:58 AM

View PostDavers, on 10 October 2015 - 11:42 AM, said:

So over 2 and a half hours for a 4v4 game? LOL Maybe if you used 10 map sheets or something crazy like that. Standard game of 4v4 is 2 map sheets with players being about 1-2 turns out from being in long weapon ranges. Add in Headshots, Through Armor Criticals, Falling Damage, Ammo Explosions, the power of physical attacks, and few mechs running around in max armor, and I really have to wonder what game you were playing. :P


Yeah, there is a reason our Star Fleet Battles games could last more than a day too... you can do more than just overload your weapons and try to ram the other ship. Revolutionary war tactics are not used any more for a reason, lol.

Generally we played on a pool table, we played with tanks and infantry as well, had lots of things on the map like buildings, forest, water, etc. I don't remember the name of the company that made the styrofoam hex terrain we used to play with but it was great. We created a lot of stuff with model train scale things. And most of us had at least a company or two worth of miniatures -- otherwise we had to use the little card board uglies from the game. I'm sorry if you used those crappy maps from the box set to play on... like I said, some of us actually used movement instead of just running right at the enemy mechs from the start. Half the game was spent NOT shooting -- getting into or out of a position, cooling off because you did fire several weapons and none of them hit, so you need to not shoot and run and hide for a couple turns, etc. If you got into melee range before you used all your ammo or were down several weapons then that was your fault and some poor planning.

#78 C E Dwyer

    Member

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

Posted 10 October 2015 - 11:58 AM

In table top the clans were even more broken, stupid, and unbalanced than here.

There are a lot of things wrong with M.W.O, but they almost got the clan IS balance to be fun then screwed the pooch with the artic cheater ebon jag hell bringer.

#79 FupDup

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 26,888 posts
  • LocationThe Keeper of Memes

Posted 10 October 2015 - 12:01 PM

View PostCathy, on 10 October 2015 - 11:58 AM, said:

In table top the clans were even more broken, stupid, and unbalanced than here.

There are a lot of things wrong with M.W.O, but they almost got the clan IS balance to be fun then screwed the pooch with the artic cheater ebon jag hell bringer.

The first set of Dominant Clan mechs added to MWO was actually the Timber Wolf, Dire Wolf, and Stormcrow.

I also hesitate to list the Cheeto in there, because while it's the overall best light it's still a light mech at the end of the day. It still has to fight against the much bigger robots, including the other Clan mechs listed in these posts.

#80 MischiefSC

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPip
  • The Benefactor
  • The Benefactor
  • 16,697 posts

Posted 10 October 2015 - 12:10 PM

View PostFupDup, on 10 October 2015 - 12:01 PM, said:

The first set of Dominant Clan mechs added to MWO was actually the Timber Wolf, Dire Wolf, and Stormcrow.

I also hesitate to list the Cheeto in there, because while it's the overall best light it's still a light mech at the end of the day. It still has to fight against the much bigger robots, including the other Clan mechs listed in these posts.


I dunno. If you watch a match and one side has two ACH and the other side a pair of ravens or whatever.... the ACH will generally carry it. If nothing else they can tank better than any other mech and will soak a ton of fire.





1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users