Jump to content

Pts 2.1 Doesn't Go Far Enough, Imo.


121 replies to this topic

#1 Sable Dove

    Member

  • PipPipPipPipPipPipPipPip
  • Knight Errant
  • 1,005 posts

Posted 25 August 2018 - 03:30 AM

The point, at least when the idea was suggested 3 or 4 years ago, was to force players to not fire literally all of their weapons at once. With the exception of my absolute hottest build (which was impractically-hot in the live environment to begin with), every one of my builds is perfectly able to use the same alpha strike as in the live server. With a heat cap of 50, almost the only builds affected by the lower cap are mechs that were too hot to be practical already.

I didn't get a chance to play with 40 heat cap, but I doubt it was too low, as some claim. If anything, I suspect 40 was still too high. People are just stuck on the idea that they have to fire all of their weapons at once.

I'll admit that in the limited scope of 4v4, the game does feel a little better to play, but I can't say if it's because of the heat changes, or because 12v12 is overcrowded. TTK is lower simply because there are a third as many mechs shooting at you at any given moment.

I applaud the effort, but the first test didn't go far enough to really test the concept, and the second test backpedaled when it should have gone further.

I can't really speak to the mobility, since most of the affected mechs I don't own or use often.

tl;dr:
Heat cap way too high. Try 35 or so, and buff all heatsinks again, especially internals. Maybe add a tutorial to remind players that weapon groups and chainfire exist.
Reduce max gauss charge to 1 at a time. (maybe cause other weapons fire to disrupt the charge, though that may be going a bit far for now)
Remove ghost heat.

Edited by Sable Dove, 25 August 2018 - 03:46 AM.


#2 Reno Blade

    Member

  • PipPipPipPipPipPipPipPipPip
  • The Blade
  • The Blade
  • 3,461 posts
  • LocationGermany

Posted 25 August 2018 - 08:44 AM

Yea maybe reduce the cap to 30-35 for a test
maybe increase isDHS to 0.24 or reduce cDHS to 0.18 from 0.2
and maybe add 2 heat to Gauss and HGauss on top.

#3 Cypherdrene

    Member

  • PipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 96 posts
  • LocationCabo

Posted 25 August 2018 - 08:49 AM

I actually used your idea as part of my poll, care to vote? Posted Image

#4 Rydiak Randborir

    Member

  • PipPipPipPipPip
  • Kapten
  • Kapten
  • 103 posts
  • LocationJarnfolk Cluster

Posted 25 August 2018 - 10:00 AM

I had typed up a large reply and the forum ate my post. GG

TL;DR: 40 heatcap broke a great many builds that weren't considered OP on Live, and made Lights very problematic to use because they couldn't make use of the increased dissipation rates because they can't boat heatsinks while at the same time gimping their heatcap by a lot. With 45 heatcap the "big scary" (that no one actually used because it was too hot on even Live) 94-point alpha Deathstrike would still be possible, while still gimping smaller mechs by a lot. 50 heatcap at least greatly reduces the "second-strike" capability of the big alpha builds (they have to wait longer to cool before they can fire a second alpha) while still permitting every build that was possible on Live to be possible post-patch.

Additionally, PTS 2.1 values (as expertly shown by Navid) only overtake Live after about 30 seconds of continuous fire. I don't know many situations where I'd fire for 30 seconds straight without any form of movement into cover or cooling off period.

#5 MechaBattler

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Survivor
  • Survivor
  • 5,122 posts

Posted 25 August 2018 - 10:08 AM

40 felt right. It gave IS, with their weak sauce alphas, a better chance. Since it more heavily effected Clan alphas. Though the discrepancy between dubs still put things in favor of the Clans.

But I don't think PGI is willing to really embrace the idea of making people not alpha at every opportunity. Because that's not going to be popular. High alpha damage is way too popular. I dunno maybe things have changed. But the impression I got from more competitive players. They like the higher stakes of pinpoint high alphas.

#6 Quicksilver Aberration

    Member

  • PipPipPipPipPipPipPipPipPipPipPip
  • The Nightmare
  • The Nightmare
  • 11,801 posts
  • LocationKansas City, MO

Posted 25 August 2018 - 10:37 AM

View PostMechaBattler, on 25 August 2018 - 10:08 AM, said:

40 felt right. It gave IS, with their weak sauce alphas, a better chance. Since it more heavily effected Clan alphas.

Clan alphas were never the problem, it was the fact they were coupled with better DPS. Clan laser vomit has been straight up better than IS at everything for a while now (since before the Mad-IIC dropped) except in the small laser department and that's only because the premiere laser got nerfhammered in the past 2 years and never got buffed to recover.

View PostMechaBattler, on 25 August 2018 - 10:08 AM, said:

They like the higher stakes of pinpoint high alphas.

It's more to do with DPS face-rushing is much less intriguing gameplay if it's the only premiere option in town.

Edited by Quicksilver Kalasa, 25 August 2018 - 10:39 AM.


#7 Sable

    Member

  • PipPipPipPipPipPipPip
  • The Predator
  • The Predator
  • 924 posts

Posted 25 August 2018 - 01:22 PM

View PostQuicksilver Kalasa, on 25 August 2018 - 10:37 AM, said:

Clan alphas were never the problem, it was the fact they were coupled with better DPS. Clan laser vomit has been straight up better than IS at everything for a while now (since before the Mad-IIC dropped) except in the small laser department and that's only because the premiere laser got nerfhammered in the past 2 years and never got buffed to recover.


It's more to do with DPS face-rushing is much less intriguing gameplay if it's the only premiere option in town.

Clan Alphas have ALWAYS been the problem. Where have you been? All the latest balance testing is specifically to address the clans ability to do incredibly high alpha damage.

#8 Sable Dove

    Member

  • PipPipPipPipPipPipPipPip
  • Knight Errant
  • 1,005 posts

Posted 25 August 2018 - 05:12 PM

View PostRydiak, on 25 August 2018 - 10:00 AM, said:

I had typed up a large reply and the forum ate my post. GG

TL;DR: 40 heatcap broke a great many builds that weren't considered OP on Live, and made Lights very problematic to use because they couldn't make use of the increased dissipation rates because they can't boat heatsinks while at the same time gimping their heatcap by a lot. With 45 heatcap the "big scary" (that no one actually used because it was too hot on even Live) 94-point alpha Deathstrike would still be possible, while still gimping smaller mechs by a lot. 50 heatcap at least greatly reduces the "second-strike" capability of the big alpha builds (they have to wait longer to cool before they can fire a second alpha) while still permitting every build that was possible on Live to be possible post-patch.

Additionally, PTS 2.1 values (as expertly shown by Navid) only overtake Live after about 30 seconds of continuous fire. I don't know many situations where I'd fire for 30 seconds straight without any form of movement into cover or cooling off period.


The issue with lights is that they buffed external heatsinks significantly when they should have buffed both internal and externals (arguably, they should have buffed internals more than externals). If they switched it so they increased the effectiveness of internal heatsinks as well, which was the original idea, then at worst, lights would be in no worse position than before.

The obsession with alpha strikes needs to end for the game to ever make progress in this area. The whole point of the reduced heat cap, when it was originally suggested, was that mech would not have 40+ damage alpha strikes without overheating. They'd have to do their 40-50 damage over around 3-5 seconds. Dissipation should have been increased to keep sustained DPS about the same.

50 heat cap plays almost he same as live, with the exception that it's 4v4 in the PTS, which accounts for the vast majority of the difference. If the PTS were 12 v 12, you'd be hard-pressed to see the difference that the 50 heat-cap causes. People would still peek out, alpha, then retreat to cool down.

#9 Quicksilver Aberration

    Member

  • PipPipPipPipPipPipPipPipPipPipPip
  • The Nightmare
  • The Nightmare
  • 11,801 posts
  • LocationKansas City, MO

Posted 25 August 2018 - 05:37 PM

View PostSable, on 25 August 2018 - 01:22 PM, said:

Clan Alphas have ALWAYS been the problem. Where have you been?

Not caring so much about alphas because if they aren't sustainable they aren't a problem, that's where I've been. Even in PUG queue they aren't the strongest thing, dakka, ATM, and MRM spam is. That said, the reason why Clan laser vomit is dominant compared to IS laser vomit is it has the best of both worlds, alpha AND DPS (yes, Clan laser vomit DOES run cooler than IS laser vomit for the damage).

View PostSable, on 25 August 2018 - 01:22 PM, said:

All the latest balance testing is specifically to address the clans ability to do incredibly high alpha damage.

Except it doesn't do that quite well, what it does address is the refire rate on that follow up alpha however.

View PostSable Dove, on 25 August 2018 - 05:12 PM, said:

If the PTS were 12 v 12, you'd be hard-pressed to see the difference that the 50 heat-cap causes. People would still peek out, alpha, then retreat to cool down.

That's because 12 v 12 is just too many people given how maps are design. You can cover a lot more angles with 12 people meaning I'm less likely to close the gap unnoticed meaning you create long range poke wars. Even if alphas weren't a thing, sustained DPS would just take its place and no one would want to step out of cover to get melted by 20 AC2s or something like that.

Edited by Quicksilver Kalasa, 25 August 2018 - 05:44 PM.


#10 Navid A1

    Member

  • PipPipPipPipPipPipPipPipPip
  • CS 2022 Gold Champ
  • CS 2022 Gold Champ
  • 4,938 posts
  • LocationVancouver, BC

Posted 26 August 2018 - 10:14 AM

40 heat cap invalidated every and all builds.

Brawlers,
laser vomit,
lights,
EVERYTHING!

and you want to decrease it to 35?
no thx.

current 50 is almost the perfect point.

#11 Reno Blade

    Member

  • PipPipPipPipPipPipPipPipPip
  • The Blade
  • The Blade
  • 3,461 posts
  • LocationGermany

Posted 26 August 2018 - 11:12 AM

tl;dr: reduce cap to 40, use ~0.2 dissipation but also reduce cooldown of weapons by about 50% or push skills to higher% to give low-hardpoint builds an automatic cooldown benefit (and save yourself the hassle to give every mech with lower hardpoint count some cd quirks).


Some scenarios to compare:
Well, imagine a 30 cap and a dissipation of 0.2 to 0.24 but also longer weapon cooldowns (let's just say doubled for example).
in such a scenario you would have a much better reason to have lots of weapons, even if you could not alpha more than a handful.

why/how?
waiting for 8-10 seconds for a weapon would be longer than waiting 5-6 seconds for your heat (depending on heatsinks).
so if you got yourself 2 or 3 builds worth of weapons (e.g. 3 sets of lasers, or 3 different range brackets, or 3 different weapon types) you could use all of them, just not really all together in the same volley.

-> and i think this is moving more into the right direction than anything we have live or have with 50 cap.

e.g.
boating 2x HLL and 6x ERML now and being able to fire all of it together even on PTS, this would need to split up in such an environment with 30 cap.
But if your weapons have a 10s cooldown, you would want to have SOME weapons to use while waiting for your main gun anyway, right?

The downside of having such huge cooldowns are the builds that only have few weapons, as they would then wait for ages for cooldown without any alternative.
->
So let's scrap the idea of the long cooldown and check the opposite by lowering the cap of 30 AND reduce the cooldown of all weapons (let's say by half)?


What this would do?
first of all, you would not be able to fire all your weapons due to the low cap, but you would be able to fire a few weapons much faster.
low hardpoint builds with a lot of DHS would be able to "spam" few weapons (e.g. the Locust with a single laser is a good example).
But at the same time it would reduce any build that have a lot of hardpoints to either skip some hp or use a bunch of smaller colder weapons (e.g. use a Nova with micro lasers instead of meds/smalls).


Still not good?
So let's try something in between.
let's keep a medium cap of 40-45 with decent dissipation (0.2 for clans, 0.22 for is and 0.15 for SHS) BUT also reduce the cooldowns of all weapons without reducing their heat values.

This would give low hp builds automatic cooldown advantage while the boats can decide if they want to fire an alpha that brings them to 90% or if they want to split their shots and maybe use the cooldown of few main weapons in some more hectic (brawler) fights and keep the big volleys for emergency or when there is enough time to cool off.

but wait? isn't that what we have on live?
somewhat for boats using alphas to ride the red line, but more beneficial for non-boats and mediums with low hardpoint due to faster cd for single weapons vs heat generated/dissipated.


Remaining problems?
Yes the high dissipation will provide some builds a lot more dps than others for similar investment.
e.g. MC2 with 2x UAC10 and 2x UAC5 on current PTS2.1 is running so good, even a average pugger like me is spamming kills with this.
while a KCrab with 4x UAC5 is already overheating after few seconds...

So in the end, there is some remaining balancing needed for heat generated for similar builds on IS vs IS and IS vs Clan and even Clan vs Clan that will require more balancing of all weapon types again in any new environment.

Edited by Reno Blade, 26 August 2018 - 11:18 AM.


#12 Tarl Cabot

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Tai-sho
  • Tai-sho
  • 7,773 posts
  • LocationImperial City, Luthien - Draconis Combine

Posted 26 August 2018 - 12:12 PM

View PostNavid A1, on 26 August 2018 - 10:14 AM, said:

40 heat cap invalidated every and all builds.

Brawlers,
laser vomit,
lights,
EVERYTHING!

and you want to decrease it to 35?
no thx.

current 50 is almost the perfect point.

I would suggest going to 45 then have HS add a fraction on top of that, and the same amount whether it be SHS or DHS. Every mech comes with 10 HS. If we do 0.50/HS (10HS *0.50=5) that moves it to 50 cap. Or 0.25/HS, 10HS *0.25 = 2.5. I would prefer the 0.25% return and it would be the same whether it was a SHS or DHS. External HS simply add a small amount of gravy. Essentially heat sinks would continue to provide a buffer but at a much smaller percentage, comparing Solaris VII boardgame (2.5sec turns - 120 heatscale) vs BT boardgame (10sec turns 30 heatscale). Solaris for a 10 heatsink mech only had that buffer of 10 on the 120 HS vs BT 10 buffer on a 30 HS scale.

Edited by Tarl Cabot, 26 August 2018 - 12:16 PM.


#13 Sable Dove

    Member

  • PipPipPipPipPipPipPipPip
  • Knight Errant
  • 1,005 posts

Posted 26 August 2018 - 02:26 PM

View PostNavid A1, on 26 August 2018 - 10:14 AM, said:

40 heat cap invalidated every and all builds.

Brawlers,
laser vomit,
lights,
EVERYTHING!

and you want to decrease it to 35?
no thx.

current 50 is almost the perfect point.

Have you considered... trying new builds?

#14 Nammuz

    Member

  • PipPip
  • Ace Of Spades
  • 28 posts

Posted 26 August 2018 - 03:03 PM

I'm not a fan of completely disassociating heat caps from heat sinks, but perhaps lowering it while upping the dissipation a bit would be good.

#15 Navid A1

    Member

  • PipPipPipPipPipPipPipPipPip
  • CS 2022 Gold Champ
  • CS 2022 Gold Champ
  • 4,938 posts
  • LocationVancouver, BC

Posted 26 August 2018 - 03:05 PM

View PostSable Dove, on 26 August 2018 - 02:26 PM, said:

Have you considered... trying new builds?


Show me some new builds.

I just want to know where you are standing on this matter.

#16 Sable Dove

    Member

  • PipPipPipPipPipPipPipPip
  • Knight Errant
  • 1,005 posts

Posted 26 August 2018 - 05:13 PM

View PostNavid A1, on 26 August 2018 - 03:05 PM, said:


Show me some new builds.

I just want to know where you are standing on this matter.

I'd have to play with the new numbers and find new builds naturally. I'm not going to plug numbers into a spreadsheet and pretend I know whether or not the build will be effective.

With 50 heat cap though, the only builds that needed changing from what I normally use was one that's ridiculously hot in the live environment already. And even then, I could still run it if I delayed firing one laser for half a second.

Excluding the mobility changes, PTS 2.1 is barely different from live. Most of the changes in the flow of the game are a result of it being 4v4 instead of 12v12. The heat changes had virtually no impact on alphas, and the tiny change in short-term DPS is far more noticeable because in 4v4 you can actually push into enemies without being annihilated, while in 12v12, it's mostly alpha peeking anyways.

#17 Rydiak Randborir

    Member

  • PipPipPipPipPip
  • Kapten
  • Kapten
  • 103 posts
  • LocationJarnfolk Cluster

Posted 26 August 2018 - 05:41 PM

View PostSable Dove, on 26 August 2018 - 05:13 PM, said:

The heat changes had virtually no impact on alphas, and the tiny change in short-term DPS is far more noticeable because in 4v4 you can actually push into enemies without being annihilated, while in 12v12, it's mostly alpha peeking anyways.


This statement right here shows exactly why you feel the way you do about PTS 2.1, because you have a lack of fundamental understanding on what changed in PTS 2.1 versus Live. As Navid expertly showed in his graphs, the "big alpha" builds all lost between 2 to 5 seconds in their second-alpha re-fire time which DRASTICALLY reduces their ability to effectively receive a push. That alone is enough to boost brawl up in the meta, which has far-reaching effects in terms of the poke meta. Additionally, the improvement in post-30 second sustained DPS further means that brawlers are stronger, which again reduces the effectiveness of the poke meta.

The brilliance of PTS 2.1 is it does all this WITHOUT completely negating a single build entirely (you know, how PGI did when they killed GaussPPC), which is the biggest problem with PGI's balancing over the past few years. More builds should be effective, not less (by wholesale removing them).

Additionally, in regards to your last point about 4v4 vs 12v12, it is a lot easier to receive 4 mechs pushing on you than it is 8 or 12 simply by applying your own logic. A reduction in heatcap to sub-50 would further exacerbate the second-alpha re-fire time of laser vomit mechs, and effectively kill that entire archetype from existence.

Edited by Rydiak, 26 August 2018 - 05:52 PM.


#18 Sable Dove

    Member

  • PipPipPipPipPipPipPipPip
  • Knight Errant
  • 1,005 posts

Posted 26 August 2018 - 06:21 PM

View PostRydiak, on 26 August 2018 - 05:41 PM, said:


This statement right here shows exactly why you feel the way you do about PTS 2.1, because you have a lack of fundamental understanding on what changed in PTS 2.1 versus Live. As Navid expertly showed in his graphs, the "big alpha" builds all lost between 2 to 5 seconds in their second-alpha re-fire time which DRASTICALLY reduces their ability to effectively receive a push. That alone is enough to boost brawl up in the meta, which has far-reaching effects in terms of the poke meta. Additionally, the improvement in post-30 second sustained DPS further means that brawlers are stronger, which again reduces the effectiveness of the poke meta.

Yeah, it has a very small effect on the viability of poke gameplay, but it's still very small. The alpha strikes are all the same, so the first two or three mechs over the hump in a push will still get butchered well before there's any significant effect due to the heat. It's also specifically the heavier mechs that get increased sustained DPS, while most mediums and lights get shafted since they can't afford as many external heat sinks.


View PostRydiak, on 26 August 2018 - 05:41 PM, said:


The brilliance of PTS 2.1 is it does all this WITHOUT completely negating a single build entirely (you know, how PGI did when they killed GaussPPC), which is the biggest problem with PGI's balancing over the past few years. More builds should be effective, not less (by wholesale removing them).

Lowering the heat cap also doesn't negate any good build, provided they increase dissipation in proportion to the lowered heat cap. There is the option of not firing literally all of your weapons at once. If they had done as was originally suggested, average DPS would be virtually unaffected on any build, while alphas would have been cut significantly; which would have encouraged sustained engagement and made shielding with arms/STs more viable.

View PostRydiak, on 26 August 2018 - 05:41 PM, said:


Additionally, in regards to your last point about 4v4 vs 12v12, it is a lot easier to receive 4 mechs pushing on you than it is 8 or 12 simply by applying your own logic. A reduction in heatcap to sub-50 would further exacerbate the second-alpha re-fire time of laser vomit mechs, and effectively kill that entire archetype from existence.

It's arguably easier to receive a push from 4 mechs, but it also takes much longer to kill those mechs because you have less firepower focusing them down, which means that mechs are able to disengage and rotate to share damage far better than with 12, because with 12 mechs, you're unlikely to live long enough to disengage, and it's usually too clogged with allies to fall back anyways. Fewer players means higher TTK for the spearhead, and higher TTK gives players more options in a fight. Players are going to be a lot more willing to push, too, because they're far less likely to die in under 10 seconds if they lead the charge. The increased dynamics in skirmishes in PTS2.1 are because of the reduced team sizes far more than a tiny decrease in short-term DPS.

The changes aren't bad; lower burst DPS and higher sustained DPS is an improvement, though the change is barely noticeable in 2.1. But as the topic says, I don't think it goes far enough. Decreasing the cap and increasing the dissipation more would not break many builds, but it would increase TTK, and thus give players more options in the fight. You just have to accept that the game would no longer revolve entirely around alpha strikes.

#19 Rydiak Randborir

    Member

  • PipPipPipPipPip
  • Kapten
  • Kapten
  • 103 posts
  • LocationJarnfolk Cluster

Posted 26 August 2018 - 06:35 PM

Again, the laser vomit archetype would flat out cease to exist in any effective form given the changes you want. The game currently "revolves entirely around alpha strikes" because the massive heatcap allows very hot mechs to double-alpha in a very short amount of time, negating any effective push due to the incoming damage. This is meta-defining. 12 mechs receiving an enemy brawl push are a LOT less effective when an extra 2 to 5 seconds are added onto the time before they can launch their second volley.

PTS 2.1 solves this without changing the single-alpha capability of any mech. You take away the "alpha" from "alpha" and you are left with an archetype that does nothing. Additionally, you devolve the game into a state that Energy Draw was guilty of: which build can output the most DPS because alpha is effectively capped. Just because you want to see the game turn into a chain-fire slug-fest does not mean that many fans of MWO, let alone the Mechwarrior series, would be happy to fundamentally see the game's combat change. Killing alpha mechs outright does not improve build diversity. It pigeon-holes everyone into DPS builds, which is what Energy Draw did. PTS 2.1 solves that problem.

Want a DPS build? Great. Want a poke alpha build? Great. More options for combat, not less.

Edited by Rydiak, 26 August 2018 - 06:37 PM.


#20 Quicksilver Aberration

    Member

  • PipPipPipPipPipPipPipPipPipPipPip
  • The Nightmare
  • The Nightmare
  • 11,801 posts
  • LocationKansas City, MO

Posted 26 August 2018 - 06:36 PM

View PostSable Dove, on 26 August 2018 - 06:21 PM, said:

Yeah, it has a very small effect on the viability of poke gameplay, but it's still very small. The alpha strikes are all the same, so the first two or three mechs over the hump in a push will still get butchered well before there's any significant effect due to the heat.

I don't think you've ever played against a coordinated push. The difference between 6 mechs breathing down your neck and 7 is pretty large. Whether this is enough to stop that? Probably not since the key to the best laser vomit builds is actually dual Gauss these days so until that is addressed the difference of 2-3 seconds may not be enough to make brawling as viable as it was during cSPL's heydey. That said it does matter and it isn't an insignificant change.

View PostSable Dove, on 26 August 2018 - 06:21 PM, said:

Lowering the heat cap also doesn't negate any good build, provided they increase dissipation in proportion to the lowered heat cap. There is the option of not firing literally all of your weapons at once. If they had done as was originally suggested, average DPS would be virtually unaffected on any build, while alphas would have been cut significantly; which would have encouraged sustained engagement and made shielding with arms/STs more viable.

Curtailing alphas significantly isn't needed as facerushing with sustained DPS is already pretty potent (see dakka MCII or Anni as an example). There is a subtle balance that must be kept intact. That said, higher agility should help make shielding more impactful against laser vomit as well so the idea that we need to do something drastic is way premature.

View PostSable Dove, on 26 August 2018 - 06:21 PM, said:

It's arguably easier to receive a push from 4 mechs,

Let's set something straight. The important thing about facing pushes is having eyes that can catch the pushing team in time to allow you to setup (because very few places are unassailable on maps). The lack of eyes means you are potentially more vulnerable and less able to punish a push and have less good surface area to spread without making yourself light bait. You also don't necessarily have the firepower to instantly burn a mech making you even more susceptible. 4v4s are the easiest to push in hands down. As with 1v1s, you'll also see a lot more riskier play in 4v4s as dire star-esque builds become more viable due to the fact there is less numbers to punish that mech. 25% of the opfor being destroyed instantly is much more beneficial than 12.5% or 8.333%

View PostSable Dove, on 26 August 2018 - 06:21 PM, said:

Decreasing the cap and increasing the dissipation more would not break many builds, but it would increase TTK, and thus give players more options in the fight. You just have to accept that the game would no longer revolve entirely around alpha strikes.

Decreasing the cap and increasing dissipation just pushes the game more and more into DPS facerushing. If you wanted tactics or a thinking man's shooter then your goal should not be to make cover a non-essential thing and to not further encourage deathballs in pub queue.

The game has not always centered around alpha-strikes, hell even during the heydey of the HGN-733C that wasn't true (otherwise why bother with UAC5s). DPS has always mattered and even with laser vomit it does (that is the reason Clan laser vomit is currently so dominant over IS laser vomit) so let's drop the stupid pretense that all that has ever mattered in this game is alpha strikes, otherwise the boogieman builds like the 6 PPC Stalker, Dire Star, or Gigadrill Dire would be a LOT more common in the game.

Edited by Quicksilver Kalasa, 26 August 2018 - 06:44 PM.






5 user(s) are reading this topic

0 members, 5 guests, 0 anonymous users