Here's the misinformation:
Quote
Childish and stupid argument, whoever came up with it should feel bad.
There is nothing magic or unexpected with ECM. It disrupts radar returns in an area around the ECM mech and within range it blocks tactical data sharing, or it counters enemy ECM within range.
More importantly it can be disabled in many ways:
BAP counterjams ECM within range
TAG negates ECM on any mech under ECM cover
NARC counterjams ECM on the ECM mech
UAV negates all ECM within its range
PPC disrupts ECM on the ECM mech.
ECM can counterjam ECM within range
Also, all of these things are valuable in other ways for LRM/streak mechs. BAP increases radar range and target info gathering while reducing the time to get a lock, TAG increases the effectiveness of the lock, NARC allows a mech to be targeted even when it is no longer in line-of-sight of any friendly mech, UAV can see over hills and cover to reveal all enemies in its range (still line of sight, just higher), PPC does direct damage, and ECM is well what this whole tempest in a teapot is all about. Some, if not all, of these things should be on any mech or with any group of mechs that wants to use LRMs or streaks.
An inability to deal with ECM reflects a failure of tactics, no teamwork, and/or a lack of skill. Period. This is such a non-issue that it isn't funny. And it is in no way 'magic Jesus box' with the number of ways it can be countered.
Quote
...are for tabletop. We aren't rolling dice to see what happens in ten second intervals, this is a real-time sim. It is going to be different from tabletop by it's nature. That is unavoidable.
Quote
Which would be why it doesn't. It negates radar locks within a certain range of the ECM mech. Outside of that bubble, LRMs and streaks work just peachy. Disrupt the ECM in any of the several ways listed above and LRMs and streaks work just peachy. The weapons aren't in any way negated, you just have to have some skill and understand the game to use them effectively. Which is true for every other weapon system in the game, you have to know how to use them in order to use them effectively. Also, it should be noted that the weapons systems that are affected are some of the lowest-skill-required systems available.
Quote
Actually, no I wouldn't. The mechs that can use ECM are typically pretty mediocre in terms of the rest of the variants for those chassis or for other mechs of that weight class. The Spider and Commando ECM mechs are the slowest with hard engine caps, the Raven has bland weapons loadouts (exceeded in all ways by the Jenner D and F) though that's true of the chassis as a whole, and the Cicada has only torso mounted weapons (which are limited in their vertical range). The DDC is the outlier in that it actually has decent hardpoints for its class and chassis (the RS has the same number, with more energy than missile, and the K has less weapons overall), but then that's the only ECM capable mech that I've ever seen in game without ECM more than once. Generally, ECM is something that gives an otherwise unremarkable mech some value. On a mech that is otherwise strong, I'm not going to gimp my build by taking away speed or damage just to get ECM on it. The only mech I have that I would consider it would be one of my Locusts, which is so gimped by its lack of armor and tonnage that trading in a medium pulse laser and two small pulse lasers for an ECM and one medium laster+two small lasers would be a no-brainer. But I can't do that, and I shouldn't be able to do that. It's entirely correct that only a few mechs can carry ECM. Not because ECM is so ungodly overpowered, but because it gives some mechs something to do for their team.
In any case the entire argument is silly, because the same thing can be said about anything. "Would you put endo-steel on anything if you could? Yes, you would. That means it is overpowered." Or DHS. Or energy weapons. Or max engines. Or whatever.
Quote
That's actually exactly how jamming works. To understand jamming you have to understand radar. There's three parts to radar, a transmitter, a receiver, and a calculator. The transmitter shoots out a frequency of invisible light, the receiver receives the echo, and the system calculates the range of the thing that reflected the echo by determining how long it took the light to go there and back. What jamming does is it sends out a higher power signal that drowns out the echo, so the range can't be determined. It's like radar is sweeping a flashlight around a room, and where you can see the light of the flashlight you could calculate the range. Now imagine looking for things that are between you and the sun using a flashlight. Are you going to see the light from the flashlight? No. So you can't figure out the range. Even though you may be able to see the reflection of those things from the sun's light, even things around or behind the sun, you can't figure out the range because you have no idea how long it took that light to get to you because it wasn't the light you sent out. Your light is drowned out by the sun's.
However, like a broken clock twice each day this broken argument actually has some things that are right:
The area where ECM prevents radar from working isn't a circle. It's a parabola, with the point of the curve nearest the radar with the focal point of the curve being the jammer. You can't see behind jamming. You need a second radar at a distance away to do that. Now, a circle of coverage is simpler to program than a bunch of constantly changing parabolas (parabolae?), so that's a reasonable enough approximation. But understand, an accurate implementation of jamming's area of effect would make things 'worse,' not better.
Also, what is missing is the ability to know a bearing to the jammer and to home-on-jam. The first point is kind of useless in any case, it's just a bearing and you can't use indirect fire on just a bearing. Even triangulation would only provide a rough idea of where the jammer is, and couldn't be used to launch anything effective. The second point is easy to answer (wrongly, though): it's not in table top. But as I said before, that's not entirely applicable here. Arty and air strikes aren't in tabletop either, at least not in any form like what we have here. I would actually be ok if there was some kind of home-on-jam specific air strike that would target a random (or the nearest) ECM mech, and actually be guided to the mech. If they turn off the ECM after the strike is called but before it hits, then the strike would just go to where the ECM was turned off. Call it a HARM strike and it would work just like other strikes, only without smoke and it would home in on an active jammer. Any kind of mech-based home-on-jam system would have to be a direct-fire streak-like missile, and would be of limited use or effectiveness so I wouldn't call for anything like that.
Quote
No, I suggested something above. Also, one thing that needs to get addressed and corrected ASAP is hit registration for Narc pods. I hate watching a Narc I shoot go through an enemy mech. It seems to happen almost half the time and it is infinitely annoying. Maybe the narc needs to be faster or something. Whatever the fix is, it needs to happen yesterday.
[edit to add: it was pointed out in comments that I may be seeing my Narcs getting AMS'ed]
Other things I'd kind of like to see would be a passive radar mode and an IR locking method.
Passive radar mode (anyone here play MW4?):
Tag, narc, ECM, and BAP wouldn't work (those are all explicitly active systems).
Can't have own locks, but can see UAV and other players' locks (so long as they aren't passive, too) -- including other players' tag and narc locks.
Can only be detected by BAP at the normal detection range (ie negates the BAP's extended range).
Can only be detected by normal radar at the range where BAP would detect a shut down mech.
IR locking for SSRM and LRM:
Must be in heat vision, but can work in passive radar mode. Range is limited by the range of heat vision. Normal radar locking would not be possible while in this mode, it's one or the other.
Hover reticle over target for a second or two and get a visual and auditory IR lock notification (this would not 'hook' or select the target in any way).
Once lock is achieved it is held for as long as the reticle is on the target (maybe an IR target decay would allow for a second or two).
Once the missiles are away, they no longer require the lock -- IR missiles are fire and forget. If the lock is lost by the player, they will still home in on the target.
One player's IR lock can't be used by anyone else, and it wouldn't provide any target info. It's not a radar lock.
But these would be nice to have, and lower priority than the Narc hit registration issue. The HARM strike idea might make people think and think again about maybe not bringing ECM, so it's a trade-off. "Do I take ECM and eat a couple of strikes I can't avoid or know that they are coming, or not?" So I would say, in order: absolutely fix narcs, maybe implement HARM strikes, then maybe do the IR locking and lastly maybe allow for passive radar. The last two are more significant programming changes and would take longer anyhow, even if they were something that anyone would want. Which is iffy.
But the bottom line is that ECM isn't as bad a problem as it is being described as being, and certainly not an instant-win gamebreaking magical Jesus box. Take away the hysteria, and there's not much to see. Sure, there's room for improvement or minor changes, but the way it has been set up is actually pretty sensible and a reasonable model despite its inherently unavoidable over-simplification of how jamming works.
As for the 99 actual problems I have, you'd have to get past the top 80 or so to get to where there's anything MWO related. Job, finances, relationships, whirled peas, that sort of thing make up most of them. But so far as MWO goes, I would say that desyncs and hit registration (including narcs, btw) could be in that list. But ECM ain't in it at all. Some people got no priorities is all I can think.
Edited by Ertur, 16 September 2014 - 07:29 AM.





























