48 replies to this topic

[YueFei]

Training Instructor, on 05 July 2013 - 03:52 PM, said:

And he gets to the heart of the current meta issues.

A mech running half armor and an xl engine so it can boat some insane weapon loadout should be a glass cannon, like the rare 2xGauss 2xPPC K2 Catapults during closed and early open beta. Any mech with full torso armor shouldn't be a glass cannon at all.

Let's all reflect on the fact that even if an AC40 Jager runs with 50/10 armor distribution on the front/back of both side torsos (and that's the max).... he's still fully capable of two-shotting a copy of himself.

When we signed up to play Mechwarrior Online, did we all envision that mech duels could end in 4 seconds without lucky headshots?

[WildeKarde]

I think one of the issues is it pushes everyone to look at taking a similar build.

Just finished a game where the enemy team had:
Catapult with twin AC20's
CTF 4X with twin gauss
CTF 3D with twin PPC's and gauss
Jager with twin AC20's

Now newer players are going to get hammered by these mechs hitting them and the lesson to learn is they should buy & build the same mechs to play. Therefore current players will use them and new players will learn they need to use them.

How much fun will we have when all the mechs just have the same builds and it's attrition to who drops first?

Edited by Jake Hendricks, 05 July 2013 - 04:05 PM.

[Liberator]

Pinpoint damage is the problem, pgi should just test out some wot like aiming reticles that get bigger if you swing your aim about or move to fast.

To further reduce alphas make stagger fire the only option, so if you fire 3 ppc and a gauss, you would fire them 1, 2, 3, 4, and unless you wait a bit after each shot, your aiming circle will have gotten quite big. (The targeting computer idea would also help, firing weapons need cpu, and you only have so much cpu per 0,5 sec, so firing lots of small weapons is ok, but to fire big guns you have to wait a bit.)

This is supposed to be an beta, so try out new crazy ****.

[Rebas Kradd]

The convergence issue presents a difficult razor's-edge problem for the devs to solve. Allow convergence and you get pinpoint high alphas, which is destructive to unaware lights, most larger mechs, and noobs. Disallow it, however, and nobody will ever hit lights or any targets at long range. I realize nobody would mourn over losing the long-range game right now, given its dominance for several months, but it belongs in the game as much as brawling.

The idea of removing convergence terrifies me. It would be reshaping the entire game with less than two months to go before launch. Alpha strikes are supposed to hurt, to the point of possible death. Not sure where anyone got a different idea. Removing convergence in this environment of double armor (which isn't going away) would not simply make alpha-strikes more sensible. It would nerf them completely. Alpha strikes spread all over a mech wouldn't stand a chance of killing any mech that wasn't already a breath of wind away from death. Not unless you're at point blank range. There's not as much wiggle room for convergence as people are implicitly thinking.

It's much simpler and less disruptive to simply nerf the 6PPC boats and let pilot skill and communication handle the other cheese builds. Plenty of people here seem capable of that. Then bring back SRM's, rebalance the other weapons, and watch the cheesebuild problem sink back to manageable levels without any need of a nerfbat. I prefer that to reinventing the wheel.

Edited by Rebas Kradd, 05 July 2013 - 04:29 PM.

[Gaan Cathal]

Rebas Kradd, on 05 July 2013 - 04:22 PM, said:

The convergence issue presents a difficult razor's-edge problem for the devs to solve. Allow convergence and you get pinpoint high alphas, which is destructive to unaware lights, most larger mechs, and noobs. Disallow it, however, and nobody will ever hit lights or any targets at long range. I realize nobody would mourn over losing the long-range game right now, given its dominance for several months, but it belongs in the game as much as brawling.

The idea of removing convergence terrifies me. It would be reshaping the entire game with less than two months to go before launch. Alpha strikes are supposed to hurt, to the point of possible death. Not sure where anyone got a different idea. Removing convergence in this environment of double armor (which isn't going away) would not simply make alpha-strikes more sensible. It would nerf them completely. Alpha strikes spread all over a mech wouldn't stand a chance of killing any mech that wasn't already a breath of wind away from death. Not unless you're at point blank range. There's not as much wiggle room for convergence as people are implicitly thinking.

Except that the game was designed from the early stages to have time delayed convergence. That code is in the game, it's even integral to the XP system with an Elite efficiency specifically for speeding it up. At some point the time delay was reduced to ~0, rendering both that XP expenditure moot and a powerful balancing tool invisible. It wouldn't be hard (within reasonable parameters) to revert convergence such that it is no longer instant, and indeed convergence speed was defined per weapon. Thus one can give beam weapons instant convergence (IIRC they've always had it) and Gauss/PPC etc the slowest convergence speeds. (And at the same time maybe hook convergence distance to targeted mech, that'd solve a lot.)

As for the AC/40 specifically, it's a good example of why I think damage drop-off needs to be exponential, not linear. AC/20s shouldn't be doing 10 damage at AC/10 optimal. Not unless they weigh twice as much as an AC/10.

[Nutlink]

AC40's used to be a much larger problem before XL engines and other weapon improvements came around. Now they're a PITA, but not nearly as powerful as they used to be.

[Takony]

Convergence nerf would chase away (not just new) players faster than trial mechs and pugstomps combined ... weapons not firing where you aim, well I understand the intent behind it but the outcome would be unhittable light mech sh1tstorms murdering everything else, and/or more hill humping and stationary sniping to have optimum convergence.

[MustrumRidcully]

SixBottles, on 05 July 2013 - 10:19 AM, said:

oh c'mon thats complete BS...

4 ac5's WILL DO only 20 pinpoint damage but will U MORE THAN DOUBLE your rate of fire, so u will do A LOT more damage...
same goes for the uac's and ac2's...

ill agree that the ac20 is the best build of your examples but simply writing false statements is going to make this discussion more pointless then it already is...

In 5 seconds, the Dual AC/20 will have fired twice, for 80 damage.
In 5 seconds, the Quad AC/5 will be fired 4 times, for 80 damage.
In 10 seconds, the Dual AC/20 will have fired 3 times, for 120 damage
In 10 seconds, the Quad AC/5 will have fired 7 times, for 140 damage.

It takes a while for the AC/5 to work out a decisive damage output advantage. This is probably one of the main problem with the current balance, I think. You can deal lethal amounts of damage (120 damage can core a Cataphract and side core an Atlas) within 8 to 12 seconds, so the high heat cap allows even heat intensive builds like those Quad PPC builds to reach this damage. It also means that the "first-strike" advantage of high single shot attacks is critical, and that pinpoint accuracy is absolut crucial.

That doesn't have to mean we actually need to "slow down" the game much, or need much more armor, but we have a much tighter room in which we need to balance weapons and mechs.

I suppose one would need to balance by "DPS10seconds" or "DPS160damage" or something like that...
"how much damage can this weapon deal in 10 seconds". Or "How long does it take for this weapon to deal 160 damage"

Edited by MustrumRidcully, 06 July 2013 - 01:23 AM.

[Gaan Cathal]

Takony, on 05 July 2013 - 10:50 PM, said:

Convergence nerf would chase away (not just new) players faster than trial mechs and pugstomps combined ... weapons not firing where you aim, well I understand the intent behind it but the outcome would be unhittable light mech sh1tstorms murdering everything else, and/or more hill humping and stationary sniping to have optimum convergence.

This is a valid concern, and why I'm convinced that convergence should never be tied to the speed of the mech. A clear convergence progress indicator on the HUD and delayed convergence done right would simply require locking the target, or leading on close terrain (the latter of which you need to do now anyway).