

Thank You Archer: For Highlighting So Magnificently The Inherent Flaws In The Lrm System.
#141
Posted 17 March 2016 - 10:10 AM
LRMs are honestly a bad weapon, PERIOD. I like the idea of them right now and its alot better implementation then some other MMOs with "artillery" or "missile" weapons, but it has too many countetrs and even if said counters were removed, it would still be subpar as a weapon.
Currently there are 3 counters to LRMs:
Cover
ECM
Target Derpivation
The first of these cannot be countered at all. LRMs cannot fly through hard cover, and a competent player can and will be taking advantage of cover. While at lower tiers you do see plenty of people running out onto the open, this is very rare in higher end play (at least at the tier im at), and those that are exposed usually stay close enough to some cover to duck behind it in the even they get targeted by LRMs (or any other weapon system they cannot out DPS or trade with effectively).
The second one is kindof a soft counter, since there are many ways to neutralize ECM, but it is virtually impossible to lock a ECM sniper. ECM on a mech that gets up close is pretty meaningless (BAP, TAG, NARC, or just being close enough to spot them through ECM), but when you look at people running ERLLs sitting at 700+ away, the ECM is actually a counter to LRMs.
As for TD, well, lets just say that 90% of everyone running at high tiers has it, and it makes it virtually impossible to lock anything since smart players dont maintain locks as they rarely face their target the entire time, and TD makes you instantly drop a lock as soon as the player isnt directly loking at you. That, and many maps have plentiful intermittent cover, which makes it near impossible to maintain constant locks nomatter what you or your team is doing.
Im not claiming that LRMs are useless, but for all intents and purposes there is just too many methods to counter them, and really only ECM has enough methods to counter. Target derpivation is probably the most annoying one when playing with LRMs (i dont do it often but occasionally do equip a few lrm launchers when i feel like screwing around). How annoying is it to loose your lock a second before the salvo hits them.
The other issue with LRMs is the spread. The ONLY viable lrm system right now is 3-6 LRM5s, since they are the most tonnage efficient, fire the fastest, and most of all have lower spread. The LRM10 is a good option too, but its heavier and the spread, while still acceptable, is not as good as the 5s. As for the 15 and the 20, I will NEVER use them unless i absolutely insist upon using LRMs, and have only a single hardpoint available. if you have just 2 hardpoints, 2 LRM10s will do you far better then the 15/20, and in actuality is not that bad of a secondary weapon. On a mech like the timber, some of my builds end up with spare tonnage and not enough slots top really use other weapons, so adding 2 LRM10s and 1-2 tons per launcher is not a terrible idea if im not in the mood to use SRMs, and those LRMs can be used to help the team when i dont currently have LOS or am repositioning and out of range of my primary weapons. That said, after i left T4 i never touched any dedicated LRM boat, and even the hunchback which i played on occasion (4J) i still treated as a 50/50 laser/LRM mech, and always tried to be within range and utilize the lasers, while firing LRMs when possible.
All in all, unless they decrease the spread of the larger LRMs liek the 15/20, i dont think there will ever be a reason to use them. With artemis the LRM15 is decent, but you need LOS, and in that even you are far better off with more or larger energy/ballistic weapons, which will do more damage, direct that damage to whatever component you are aiming at, and wont splatter their damage all over the mech. As for the LRM20, its hands down inferior to the LRM15. The 15 is much more likely to land all its missiles on a target (assuming its not a light or very small medium), while the LRM20 will miss at least a portion of its salvo against all but the larger mechs. In essence unless you are targeting assaults/heavies exclusively, the LRM15 will actually do equal damage per salvo in most cases, weighs less, and fires faster. If you must use a LRM20 you have to have artemis, and you cannot afford not to be within LOS (which forces you to face-tank, a very bad idea in higher tiers) to utilize the spread reduction.
#142
Posted 17 March 2016 - 10:11 AM
Bishop Steiner, on 17 March 2016 - 10:03 AM, said:

To date, I don't think I've seen JMan5 in a comp match (or a regular match for that matter) as of late... but that in part has to do with my lack of interest in the pub game at the moment (have been doing more private lobbies/practice).
Edited by Deathlike, 17 March 2016 - 10:11 AM.
#143
Posted 17 March 2016 - 10:14 AM
Bishop Steiner, on 17 March 2016 - 10:07 AM, said:
I just ended up spending a lot more matches having to pewpew leftovers to death than Jman, because if the match lasted more than 6-7 minutes, I sure as heck was gonna be out of ammo, lol.
Sadly, the nerf to the 4J has left it collecting dust for me. Because, ya know, it was sooooo OP before.
I feel like they're cycling quirks for sales. I've been meeting people that spend mad MC on the latest uber-quirked IS mech when PGI does a re-shuffle.
Deathlike, on 17 March 2016 - 10:11 AM, said:
To date, I don't think I've seen JMan5 in a comp match (or a regular match for that matter) as of late... but that in part has to do with my lack of interest in the pub game at the moment (have been doing more private lobbies/practice).
QQ has been CWing mostly. And we've been comp matching some, but we got uber-f*cked on seeding for MRBC.
#144
Posted 17 March 2016 - 10:16 AM
Deathlike, on 17 March 2016 - 10:11 AM, said:
This is partially because most of the former QQ comp team folded into 228 (aka QQ8) after MRBC season 5 and MLMW so they have been trying to rebuild since then (and somehow put Johan in charge XD).
#145
Posted 17 March 2016 - 10:18 AM
Quicksilver Kalasa, on 17 March 2016 - 10:16 AM, said:
I got voluntold. I don't even.
Oh well, I'm trying to make the best of it and make QQ a high level casual team like it was when I first joined it.
#146
Posted 17 March 2016 - 10:19 AM
As for spread, I wouldn't mind seeing the LRM launchers fire in quick salvos of 5 and keep the spread pattern of the LRM5 today. Kind of like how MRMs were shot in salvos of 10 in MW4. The salvos would be shot in quick bursts but basically like someone mentioned already in this thread, keeping the missiles in tight packs. So in a LRM20 rack, it would quickly burst 4 packs of 5 missiles at target. I don't think you would really need to adjust cycle times much with that change, if say your mech had 2 LRM20s, chain fire those should provide the right amount of DPS at a target since the packs of 5 missiles will do the right amount of damage to the target, and possibly allow the target to try to twist off some damage to help it not be too "OP".
#147
Posted 17 March 2016 - 10:22 AM
Saint Scarlett Johan, on 17 March 2016 - 10:14 AM, said:
I feel like they're cycling quirks for sales. I've been meeting people that spend mad MC on the latest uber-quirked IS mech when PGI does a re-shuffle.
Yeah, been seeing that too, and then the QQ when the cycle shifts (you'd think people would learn, eh?)
But you take time to introduce a new shiny... you want it to be Good (not god tier, but good) because you know...more sales, and consumer confidence for the NEXT shiny you offer?
I get it. PGI had a massive patch they were putting together (though since the actual workloads are compartmentalized, and most of the mass was textures..... let's not overdo it, ya know?), and CW3, and the work on the Grand ReScale.
But from the Desktops being quick grabs of Sean Lang's build of the Month, to having that loadout shown as the stock 2R load on the Quick Release (aka just shabby amount of foresight shown, and QA) to the Crappy Quirks/Paul Saying "Oh opps, not final, we'll fix that" /Russ saying: "No! How can you have any pudding if you don't eat your meat!!!!"....
this showed an alarming disconnect/ambivalence from a company whose sole shiny beacon has generally been releasing stompy robots.
#148
Posted 17 March 2016 - 10:28 AM
Saint Scarlett Johan, on 17 March 2016 - 10:14 AM, said:
Sounds like PGI. Working as intended.
Quote
Well.. you didn't have to win vs us in the MSI Tourney...

Seriously though, there were more behind the scenes issues at work, but hey, congrats and gl.
Quicksilver Kalasa, on 17 March 2016 - 10:16 AM, said:
I've noticed that and say whaaaa. Where's Homeless Bill and who let Johan in charge???

#149
Posted 17 March 2016 - 10:35 AM
Deathlike, on 17 March 2016 - 10:28 AM, said:
I've noticed that and say whaaaa. Where's Homeless Bill and who let Johan in charge???

Our resident box dweller is on vacation. Me and my Pretty Baby eagerly await his return.
I don't mind being in charge, since when they put me in charge and I started to get some things together we started seeing people logging in to the TS server again and playing with us. Hell, we had a record turnout one night of like 21 people in channel...
I just want people logging in to play with.
Also, so I don't seem like I'm derailing Bish's thread I'm currently typing up a book in Notepad to copy over.
#150
Posted 17 March 2016 - 10:39 AM
Aresye, on 16 March 2016 - 10:56 PM, said:
1. It contributes to poor gameplay, as players hide behind cover and rely on teammates for locks.
2. The TT mechanic is designed for a dedicated spotter, and has movement/engagement penalties (i.e. You can't run around in a light mech holding perfect lock for a teammate whilst simultaneously engaging 3-4 mechs like you can in MWO.)
3. TAG, UAV, and NARC offer more than enough indirect fire options.
just give 1/2 the points to the spotter(less points for lrm user)? or something like this
i like to spot, not sure how much is reflected in score though
#151
Posted 17 March 2016 - 10:45 AM
nar9000, on 17 March 2016 - 10:39 AM, said:
just give 1/2 the points to the spotter(less points for lrm user)? or something like this
i like to spot, not sure how much is reflected in score though
Indirect fire from LRMs pays out absolute BUTT for XP and Cbills. The only reason to use it is when it' helping out the team (like on the way, no LOS, killing a light chasing a teammate, etc), otherwise you are best off in all regards getting LoS locks for yourself.
Also, a good NARCer can make BANK.
#152
Posted 17 March 2016 - 10:46 AM
Moomtazz, on 17 March 2016 - 09:43 AM, said:
What I am saying is treat LRM launchers of any size as multiple LRM5 launchers. So an LRM 20 would take up 1 hardpoint but could be toggled to fire in 4 groups of 5 missile, or one group of 20. Have different heat profiles depending on 5 or 5x groups. That should fix the problem you are all complaining about.
So even if 2 LRM15 weigh more than 6 LRM5, they only take up two hardponts.
No need for a toggle which would add pointless complexity, and new game mechanics that may not even be possible (note how we can't have mode switching weapons for the solid/cluster LBX).
Just equalize all lrm launcher cooldown so each launcher fires at the same speed. Currently, 4x5, whether group or chain fired, put missiles out MUCH faster than an lrm20. Hence the 50% DPS increase for the 5's. Then adjust all lrm launchers to have the same spread irrespective of how many missiles are fired.
But even with that change, the lrm5's are still somewhat better, and lighter.
#153
Posted 17 March 2016 - 10:56 AM
Bishop Steiner, on 17 March 2016 - 10:45 AM, said:
Also, a good NARCer can make BANK.
Without a narcer, though, I generally won't fire a lot of indirect LRM's (depending on the situation) as you tend to just waste a lot of missiles and damage armor. To make good money via LRM damage, you need to be wrecking mechs, not sandblasting them.
I tend to make 350k-500k per match with my mad dog and a narcer (premium time but no cbill bonus on the mech). It's silly. And the narcer makes off like a bandit as well.
#154
Posted 17 March 2016 - 11:04 AM
So maybe we could buff the larger launchers in a different way by adding a secondary effect, like screen shake, that gets more extreme when large launchers are involved?
e.g.
LRM5 - no effect
LRM10 - 0.25 sec HUD freeze effect
LRM15 - 0.5 sec HUD freeze effect
LRM20 - 1.0 sec freeze
#155
Posted 17 March 2016 - 11:07 AM
Dogstar, on 17 March 2016 - 11:04 AM, said:
So maybe we could buff the larger launchers in a different way by adding a secondary effect, like screen shake, that gets more extreme when large launchers are involved?
e.g.
LRM5 - no effect
LRM10 - 0.25 sec HUD freeze effect
LRM15 - 0.5 sec HUD freeze effect
LRM20 - 1.0 sec freeze
honestly not a big fan of using peripherals to try to make the basic combat mechanics "good".
Russ thought that "ECM defeat" was a good enough peripheral to keep PPCs crap for over 2 years.
Let's get the weapon basically dialed in in basic combat stats, then see what peripherals work?
Wintersdark, on 17 March 2016 - 10:56 AM, said:
Without a narcer, though, I generally won't fire a lot of indirect LRM's (depending on the situation) as you tend to just waste a lot of missiles and damage armor. To make good money via LRM damage, you need to be wrecking mechs, not sandblasting them.
I tend to make 350k-500k per match with my mad dog and a narcer (premium time but no cbill bonus on the mech). It's silly. And the narcer makes off like a bandit as well.
Still luff mah LURMDawg.
#156
Posted 17 March 2016 - 11:09 AM
Bishop Steiner, on 17 March 2016 - 09:11 AM, said:
I'd like to see you go solo against him.

He's a good pilot, but personally I feel most of his praise comes more from the fact he was willing to run LRMs in comp play, rather than having some special, uncontested ability.
Give Proton a 4J and a guarantee you he would be just as effective (if not moreso) than Jman.
#157
Posted 17 March 2016 - 11:11 AM
Bishop Steiner, on 17 March 2016 - 11:07 AM, said:
Russ thought that "ECM defeat" was a good enough peripheral to keep PPCs crap for over 2 years.
Let's get the weapon basically dialed in in basic combat stats, then see what peripherals work?
Still luff mah LURMDawg.
My brawling lurmdog has the highest stats of any mech in my stable, in every way. More of everything per match, and by a large margin. It's my only mech that gets consistently more kills, damage, (and kmdd, not just "stolen" kills), xp, etc per match than my Direwolf (and that's comparing to pre-agility nerf Direwolf stats.)
#158
Posted 17 March 2016 - 11:13 AM
and HUD freeze effects belong with PPCs, not LRMs
i'd be much happier if we got stagger effects for aim disruption from impacts - ballistics and missiles both
Edited by Thunderbird Anthares, 17 March 2016 - 11:14 AM.
#159
Posted 17 March 2016 - 11:17 AM
Aresye, on 17 March 2016 - 11:09 AM, said:
Give Proton a 4J and a guarantee you he would be just as effective (if not moreso) than Jman.
Possibly so, though not the point. Point was made that no good pilots would touch LRMs, I countered that.
#160
Posted 17 March 2016 - 11:19 AM
Bishop Steiner, on 17 March 2016 - 11:17 AM, said:
I specifically edited that statement for a specific reason. I've actually seen comp players run them before, but not in any serious manner (they did it for fun, and not to prove any point).
3 user(s) are reading this topic
0 members, 3 guests, 0 anonymous users