28 replies to this topic

[BlueSanta]

I sent a message to Thomas yesterday and got a reply today:

Thank you and we are aware of the issue.

I do not believe you are right about the SSRMs but certainly the LRMs.

SSRMs pick a random location (1/5) on a mech which they track towards. shoulders, thighs and center. (Unless the target bones are missing which might have been overlooked on some mechs but I have not seen this)

LRMs always home in on center of mass so most mechs who are staring at the missiles as they are incoming with get hit on the center torso.

Either way we have an engineer coming out with a new path for SRMs and LRMs, SSRMs for now will not change as far as I'm aware.

I've encouraged him to reconsider.

[Krzysztof z Bagien]

BlueSanta, on 12 April 2013 - 03:24 PM, said:

I sent a message to Thomas yesterday and got a reply today:

Thank you and we are aware of the issue.

I do not believe you are right about the SSRMs but certainly the LRMs.

SSRMs pick a random location (1/5) on a mech which they track towards. shoulders, thighs and center. (Unless the target bones are missing which might have been overlooked on some mechs but I have not seen this)

LRMs always home in on center of mass so most mechs who are staring at the missiles as they are incoming with get hit on the center torso.

Either way we have an engineer coming out with a new path for SRMs and LRMs, SSRMs for now will not change as far as I'm aware.

I've encouraged him to reconsider.

Well, it seems that Thomas is actually right. Thing is that locations SSRMs lock on look like this:



So - basically wherever they hit they do some damage to CT, especially in case of the light mechs - if they target 3 it's CT, if they target 4 or 5 - it's leg and CT, if they target 1 or 2 it's arm, side torso and CT they damage.
SSRM should lock on center of the component, not on joints. That would mean they hit yellow dots not the teal ones (and have lowered chance to hit the head or even don't aim at it at all).
One more thing: if you launch SSRM at Commando they always hit 3.

[BlueSanta]

You have a valid point and I have asked him to comment on this.

This is kind of the problem with splash damage. We have seen that it is problematic in how it functions. If they can fix the grouping and up the damage a tick, I would at least like to briefly see what it would look like to have splash removed completely.

[John MatriX82]

So, today I've started going around with a STK 5M loaded with 5 Streaks and I was about to make a similar post but Krysztof preceeded me. It looks like that when hitting bigger mechs (let's say heavies and assaults), the spread is similar to the picture poster just above by Krysztof.

By this with reduced splash damage, the CT takes damage, but the wider the mech, the lesser is the "coring" effect, although still quite a good chunk of damage it's delievered in the CT.


When targeting lights, jenners, Cicadas, commandos, spiders, ravens and the mediums (especially HBKs, somewhat to a minor extent Cents and Tbts), the above damage location pattern results in delivering huge amounts of splash damage in the CT only, as I've experienced many times in my cicadas, that barely have the rest of the mech scratched or yellowish but the ct totally gone.

Something must be done, because streaks should literally fly in the center of the leg, in the center of the arm, not trying to stay so close to the ct!!

Streak hitting locations should be like this:


Not only, but splash should be reduced or totally get rid of it (upping dmg from what is now to compensate) and streaks shouldn't be able to have the insane turning ratio capabilities that they have right now.

[Felix Reynolds]

So in response to a lot of what was being posted here, and to what I was finding trying to play my Jenner F again (thinking that 'oh hot damn HSR for lasers + normal Raven hitboxes = SWEEP THE LEG FELIX-SAN!!!) I played around in the Testing Grounds rather extensively.

I went and pulled a bunch of video from those tests in MSI AB, and am uploading to YouTube now, and once it's up will provide the link here.

The tl;dr version- SSRMs aren't hitting anything other than the CT if fired from anywhere inside of a 90 degree arc from the targets front (about 45 degrees on each side, give or take a few depending on how prominent a 'nose' it has). This, as many have pointed out, isn't just a problem on light mechs, but on ever chassis.

Further, from watching flight paths it is obvious that Krzy's post is indeed accurate about where they aim, but I have to disagree and say that the 'main' damage from each missile is being applied to the CT, and the only bit that is bleeding off onto the actually-targeted component is the splash damage.

Which makes SSRM's just a little more than absurd. Oh boy.

Link to come, apologies in advance for the wait but uptick in NZ is horrendous.

-EDIT- Link is finally up, can be found HERE.

Edited by Felix Reynolds, 14 April 2013 - 09:22 PM.

[Varnas]

BlueSanta, on 11 April 2013 - 02:39 PM, said:

It's absurd how some people still refuse to completely acknowledge that this is happening to ALL mechs, not just lights, although they are the ones most affected by it. I would really like to hear a dev like Matt or Thomas acknowledge that this is happening and that it will be fixed.

As a hunchback i totally ******* agree with this.

[Spider Monkeylord]

Using an A1 Catapult with 6x SSRM2 in the Frozen City Night testing grounds I was able to run a "time trial" and destroy all of the targets in 3:01 and 3:06. Using 6x SRM4+Artemis it took me 2:43, 2:48 and 2:46. When using SRM4s I was able to kill the assault mechs faster but the light mechs took longer. This should be the case in-game against moving targets (assuming I am not a perfect shot) but not against stationary targets. It suggests that SSRMS are in fact not spreading the damage away from the torso as well as they should.

I have not noticed a significant problem with SSRMs in my Jenner with streaks in-game and have not played any of the ECM+Streak lights. Most of my experience is from the A1 catapult, and from that experience a couple of reports in this thread do not at all match my experience. For one thing, if I shoot while someone is arm-blocking, I hit their arms; the missiles do not go through to the torso. I do, however, time my shots so that I shoot when players are facing me, just as I do with lasers and everything else, so that I don't waste shots on arm blocks. I also get a number of kills by destroying side torsos+XL engines and have legged a number of mechs, especially lights, with streaks, suggesting to me that streaks, at least on an A1, are at least partially functional.

Mech hitboxes and missile exit points are things to consider here too. Spiders, for instance, absorb a large volume of streaks to their CT, although this may be because almost every spider I shoot at just jumps in the air when they see streaks, which is perhaps the worst thing they could do as their transverse velocity drops to nothing and they suddenly have no cover. Still, jenners jumping in the air seem to take more leg shots, and when I am in a cicada I feel like ravens hit me in the center more than commandos (just "playing by feel" but that's what they told us to do with the missile nerf right?). Also on the topic of missile exit points, I want to note that as streaks, in my experience, blow up on the first part they run into, I try my best to fire them at ranges where they do not have much time to spread and will hit the st/ct as much as possible. I assume other streak pilots do the same, and I certainly don't support a change to a "problem" that is caused by good piloting.

Anyway, there are my few cents. I think streaks do in fact track towards the center too well but not so much that they need a significant change, just some balancing. In my experience they do not appear bugged to hit only the CT if there is another body part in the way, at least arm blocking (I haven't really payed much attention to shooting a mech from behind as I am not a very good player and can rarely sneak up on people).

[Felix Reynolds]

So video is up, link can be found HERE.

Few interesting things I made sure to include- I'd really love to get a much more thorough documentation up, but given current working constraints this is about the best I can chuck up so I apologize in advance for everyone who I am sure will point out what I could have done differently/better and where.

1. At about 2:00, I start shooting at the Atlas. This is particularly interesting as I do so from a variety of angles including roughly 75 degrees off his left quarter, and yet the streaks are STILL only doing significant damage to the CT.

2. At roughly 1:15 is where the Catapult test begins, and as you can see, shots from directly off the right side hitting the 'ear' and legs, but moving closer to a 45 degree angle again brings the majority/nearly all of the damage in straight to the CT.

3. At 3:30, I start to shoot a Centurion from a roughly 45 degree angle off his facing, with the AC in his right arm directly covering the straight line between his CT and my reticle. The interesting bit is that the damage does indeed hit the 'blocking' arm, but a step back a few meters (which doesn't greatly change the amount of the blocking arm still facing me) cause the missiles to home straight past it and core him out.

Various other mechs shot at as well, lights especially, which show you just how quickly you can drop one currently with a few good Streak salvos.

Now, I realize this aren't 'combat conditions' and in an actual match everyone would be moving much more erratically, but this should at least put to rest the assertion that streaks actually do 'spread' the way they're supposed to in controlled conditions. The locking positions they home to seem to be the 'joint' between whichever part they've targeted and the CT, and for whatever reason their damage is NOT being spread out the way it is supposed to- from what I observed, unless a component is directly in the way of the CT from the launcher (ie, the Centurion in the last test, the Catapult from broadside on) the only damage spreading to non-CT components is the splash damage, while the actual 'missile' damage gets applied directly to the CT.

If anyone has any comments, questions, or suggestions, please let me know, and hopefully after the work week I'll get some more tests up/see if the new patch has fixed anything.

[BlueSanta]

I messaged Paul about this a couple days ago but he has yet to even view the message. I honestly for the life of me cannot figure out how this issue has escaped detection and why we're the only ones who seem concerned about it. I had hoped that Thomas would check it out, but I have received no indication from him that he's even taking it seriously.