

A Loadout Guide To The Death's Knell
#41
Posted 30 January 2013 - 06:09 PM
210xl
DHS
endo
Kept the mLas.
Stir and enjoy
#42
Posted 30 January 2013 - 11:49 PM
If i at some point have the money for a 210 "lying around", i will drop the AMS and be happy

#43
Posted 31 January 2013 - 12:04 AM
#44
Posted 31 January 2013 - 12:05 AM
That sounds....ok, i WILL have to try that ^^
#45
Posted 01 February 2013 - 06:19 PM
LLas, MLas SLas x2, Endo, XL195, DHS x12
Not sure why i went with the multiple laser types in the first place but it does a lot better then previous builds i had so i stuck with it.
#46
Posted 04 February 2013 - 11:06 AM
#48
Posted 11 February 2013 - 02:12 PM
http://mwo.smurfy-ne...0dd4805650bf360
I know I can switch out my PPC for a ER PPC, but that seems to make a 6% difference in cooling efficiency, and i can basically chain fire the PPC constantly on any map other than caustic.
The main concern I have when running this mech is running into SSRM lights, as others have said, so I try to stick close to the group.
I do run a COM-2D as well but I feel that build is "easymode" and not challenging enough.
I see this TDK build as basically the anti-heavy sniper light , but it will take a while to get used to its best use.
I have only had it pan out really nicely once, when I was able to get to the outer edge of the map without anyone noticing me, and while everyone was engaged at the front I kept running along the boundaries taking pot shots 2 kills 5 assists and about 340 dmg.
Again SSRM equiped lights will ruin your day.
Also not having to pay rearm costs due to all energy weapons is nice

EDIT: After not getting many good chances to get good shots off with the PPC, I decided so swap it for a LL and an extra ML. seems a LOT more forgiving if you miss your initial shot, although not as satisfying when you get a good solid hit.
Edited by Stanfy, 12 February 2013 - 11:26 AM.
#49
Posted 11 February 2013 - 04:30 PM

I have no idea how to work the Death's Knell. I'll also admit, it's not super-high on my list of Lights to Worry About. That list generally goes RVN-3L, (competent) Jenners, COM-2D. Death's Knells are nastier customers than Spiders, but not really something I feel warrants a lot of attention as a Dragon pilot. That's not a good indicator, really. I sorta feel bad for folks who buy it.
#51
Posted 11 February 2013 - 05:16 PM
Phatel, on 23 January 2013 - 01:44 PM, said:
Let me get this straight...
You are complaining that ECM and netcode make lights so OP that anyone in such a mech becomes a nigh-invincible killing machine ("brawl assaults and win easily" "taking on 3+ mechs at once").... but you also say that any light not taking Streaks and ECM is literally a blight on your team that instantly kills any hope of winning?
You say the netcode is a shield for lights that makes them virtually impossible to hit with direct-fire weapons (which, BTW is no longer true in most cases, ask my Gauss as it cores 3Ls) but then you turn around and say that because of the netcode, any mech with ECM+Streaks is a silver bullet that rapes lights unmercifully?
Before you go being a pessimist and ripping on how "broken" X Y and Z are, at least formulate something that resembles a cohesive appearance. Not trying to be a ****, just saying no one wants to listen to you rant about this and that being broken on a thread about Death's Knell builds...
Carry on.
Edited by RagingOyster, 11 February 2013 - 05:16 PM.
#52
Posted 01 March 2013 - 11:08 PM
http://mwo.smurfy-ne...689e48a0c3dc9bd
It does work surprisingly well!
i even got some nice an satisfying Raven-3L-Kills with it :-D
#53
Posted 02 March 2013 - 04:43 AM
after few matches (when I met MANY 2Ds in opposing team) I was desperate and regreting the buy. I got over my pride and mastered it. It is great mech and with a bit (well, not really such a bit

I use this build:
http://mwo.smurfy-ne...2080f71ae5f9c9d
To be honest I do not understand why are you fitting the XL engine. I tried it short after the beginnings with this mech, but I suddenly found I my deaths rate went very high. I therefore fitted an 200STD engine and it is best for me - good speed and I do not have to bother to loos the torso (which goes down very quick on the Commando:)) ).
One of best games:

Edited by M4jk, 15 June 2013 - 12:12 AM.
#54
Posted 05 March 2013 - 06:51 AM
#55
Posted 05 March 2013 - 06:55 AM
mine runs with a 210XL and 4 Med pulse.
It's a fun mech to play, what deals decent damage and can hold it's own against any other mech.
(maybe not RVN-3Ls.... )
#56
Posted 05 March 2013 - 08:10 AM
DEATH'S KNELL
Edited by omegaorgun, 05 March 2013 - 08:12 AM.
#57
Posted 05 March 2013 - 12:05 PM
#58
Posted 06 March 2013 - 12:51 AM
And XL210 or XL195 (the by far best/most fitting engines for commandos) will take you back by a lot, but they are worth it.
#59
Posted 07 March 2013 - 06:19 PM
4 Med las
xl195
14 dbl heat sinks
shave 1 point armor off each leg. Have just a bit more armor in front than back since they'll be firing at you mostly as you run away
Mix with an average pilot
Here's average result

#60
Posted 07 March 2013 - 06:47 PM
11 user(s) are reading this topic
0 members, 11 guests, 0 anonymous users