14 replies to this topic

[Navid A1]

I'm sad and disappointed to say that not only LOS LRM fire fails to give you less effective spread, it actually INCREASES spread, in direct contrast to PTS objectives.
I wonder what tests have been conducted to conclude otherwise.

Here is a test on an Atlas, a large target with a large front facing surface:

[The6thMessenger]

Navid A1, on 17 January 2019 - 07:38 PM, said:

I'm sad and disappointed to say that not only LOS LRM fire fails to give you less effective spread, it actually INCREASES spread, in direct contrast to PTS objectives.
I wonder what tests have been conducted to conclude otherwise.

Here is a test on an Atlas, a large target with a large front facing surface:

Nice test.

Maybe it's because they're hitting more legs due to the profile.

[Navid A1]

The6thMessenger, on 17 January 2019 - 10:38 PM, said:

Nice test.

Maybe it's because they're hitting more legs due to the profile.

Which is the exact definition of more spread.

[The6thMessenger]

Navid A1, on 17 January 2019 - 11:15 PM, said:

Which is the exact definition of more spread.

While i agree that PGI should work on this because it's contrary to their center-of-mass approach, I think their definition of spread is basically just how well the missiles are grouped.

Try it with IS mechs though.

Edited by The6thMessenger, 17 January 2019 - 11:27 PM.

[Kaeb Odellas]

The6thMessenger, on 17 January 2019 - 11:25 PM, said:

While i agree that PGI should work on this because it's contrary to their center-of-mass approach, I think their definition of spread is basically just how well the missiles are grouped.

Try it with IS mechs though.

I didn't test LOS vs non LOS firing for IS LRMs, but I did find the LOS mode also hitting legs more often. LRMs seem to target the mech's lower torso as center mass, so more missiles hit the torso in IDF than in direct fire due to the approach angle

[justcallme A S H]

The6thMessenger, on 17 January 2019 - 11:25 PM, said:

While i agree that PGI should work on this because it's contrary to their center-of-mass approach, I think their definition of spread is basically just how well the missiles are grouped.

If it was a reduction of 30% as Chris claimed was the PTS value for LoS. It should take approx 28 LRM15 volleys to kill the Atlas. Give it 30 for benefit of the doubt. Not 8 more volleys.

It's nothing to do with grouping.

[Bad Pun]

I've done some testing too and my results can verify that the spread is absolutely absurd, and indeed in the range of 30%. Though mine was only in determining that the PTS is worse than the Live Client, not so much of Indirect Fire.

[The6thMessenger]

justcallme A S H, on 18 January 2019 - 12:06 AM, said:

If it was a reduction of 30% as Chris claimed was the PTS value for LoS. It should take approx 28 LRM15 volleys to kill the Atlas. Give it 30 for benefit of the doubt. Not 8 more volleys.

It's nothing to do with grouping.

"Spread" has everything to do with grouping, hitting "center-mass" is a different story.




It's like missing with SRM, it's still tightly grouped but hitting center of mass is a matter of aim.

What I am saying is that, the LRMs is probably scripted to hit the CT at a certain-angle, and it's actual point (point-area) of homing isn't dead on the CT but probably something behind, it that may explain the tighter spread but the cluster hitting lower.



That's just an idea though, take it with a grain of salt. I don't know why they would do an off-center homing point, but LRMs do have this wierd trajectory in the end where when you expect them to land on the ground for flying on a really steep angle, they actually pull up for some reason.

Edited by The6thMessenger, 18 January 2019 - 01:53 AM.

[Bad Pun]

That does make sense. So if I see it right, the weapon spread should still be smaller for the direct fire to account for the more flat facing of the low angle? That sounds like it could be the easy fix to me.

[justcallme A S H]

The6thMessenger, on 18 January 2019 - 01:43 AM, said:

Spoiler

"Spread" has everything to do with grouping, hitting "center-mass" is a different story.




It's like missing with SRM, it's still tightly grouped but hitting center of mass is a matter of aim.

What I am saying is that, the LRMs is probably scripted to hit the CT at a certain-angle, and it's actual point (point-area) of homing isn't dead on the CT but probably something behind, it that may explain the tighter spread but the cluster hitting lower.



That's just an idea though, take it with a grain of salt. I don't know why they would do an off-center homing point, but LRMs do have this wierd trajectory in the end where when you expect them to land on the ground for flying on a really steep angle, they actually pull up for some reason.

The grouping has always been randomised in terms of the spread.

This is evident with any missile testing in MWO. SRM is super obvious as no two volleys of missiles will follow the same path, easier to see to.

They all hit within a certain 'area', where they hit within that area however is random. Either way a 30% reduction, as stated, is clearly not occurring. I mean SSRMs were altered to aim torsos more over legs a while back, and they most certainly do. That change could be instantly felt.

[The6thMessenger]

justcallme A S H, on 18 January 2019 - 03:58 AM, said:

The grouping has always been randomised in terms of the spread.

And so it's "group", some are well grouped as in they are tight, others are badly grouped and so the shots are everywhere. I'm not challenging that's not randomized, I'm saying that a group is a cluster of something -- hits, missiles, bullets etc.

justcallme A S H, on 18 January 2019 - 03:58 AM, said:

This is evident with any missile testing in MWO. SRM is super obvious as no two volleys of missiles will follow the same path, easier to see to.

Well, duh. But it's still a group.

justcallme A S H, on 18 January 2019 - 03:58 AM, said:

They all hit within a certain 'area', where they hit within that area however is random.

And grouping, interchangable with spread, is how well the missiles are packed together, even if they aren't random.

It's still a "group", whether you are shooting rifles, shotguns, pistols, machine-guns -- or in this case missiles, it's still a group.

justcallme A S H, on 18 January 2019 - 03:58 AM, said:

Either way a 30% reduction, as stated, is clearly not occurring. I mean SSRMs were altered to aim torsos more over legs a while back, and they most certainly do. That change could be instantly felt.

Well the grouping part there is just that, they may have made it so that the group is tighter, but it's not properly hitting the center mass, making the change irrelevant. My comment about it is specifically proposing that, they might have assume having a tighter group means more center-of-mass hit, but failed to account the LRMs hitting lower -- PGI failed to connect the 2 variables properly, and now we got a -30% less effectiveness.

If it has nothing to do with spread, imagine if PGI also increased the spread more like from 6.00 to 8.00. Won't you agree that, it would take an even longer amount of time to kill an atlas?

PGI being PGI. Fingers crossed, I hope they fix it, but I wouldn't hold my breath.

Edited by The6thMessenger, 18 January 2019 - 06:31 AM.

[Chris Lowrey]

So a few things I want to touch on here.

First off, lower trajectory within this test, does not produce buffs to missile spread, and we have not claimed it has. What we have claimed is that shots fired within LOS results in higher accuracy of the missiles themselves when compared to indirect fire scenarios on a global level using the testing values. (This is when tested and averaged against all target types, not just particular 'Mechs / classes, which is why we made the note explicitly that this number will vary wildly based on a target's overall geometry.)

The spread is the same regardless of LOS or indirect, but the angle of attack results in higher concentrations of damage and accuracy, especially when compared against volleys fired upon moving targets directly vs. indirectly. While I don't want to discount the findings of this example, I do want to state that first and foremost, that we are looking closely at the values gathered through live matches, and practical results of what happens in a live match will always be more reflective of how the weapon will perform live then it would against stationary targets.

I don't say this to shoe away the testing conducted here, as we are keeping an eye on this and many other impressions gathered here in the Forums, I just want to explicitly point out that we have not claimed that the lower trajectory would result in tighter physical missile spread, but a higher amount of accuracy against targets in most live fire situations.

We are not against potentially exploring some kind of buff to spread for LOS fire, but this is a case where this testing environment is set up to see how far we get with only the reduction in the Angle of Attack. As we do not want to implement a mechanic that overshoots our intended target by "double buffing" a particular aspect that could be covered simply by the Angle of Attack change.

Just want to throw that out there as people continue to test. Again I want to state we are not against examining additional changes, but we do want to make sure we get as much data of the functionality currently placed onto the PTS before moving ahead with anything else. If you feel additional spread tightening in direct fire situations is needed compared to what is being tested now, be sure to express that opinion here on the forums.

[Navid A1]

Chris Lowrey, on 18 January 2019 - 04:54 PM, said:

I don't say this to shoe away the testing conducted here, as we are keeping an eye on this and many other impressions gathered here in the Forums, I just want to explicitly point out that we have not claimed that the lower trajectory would result in tighter physical missile spread, but a higher amount of accuracy against targets in most live fire situations.

Appreciate the response.
Even if you are not claiming that LOS results in tighter physical spread, the end result is basically more spread damage.

You are maybe counting the average applied damage per volley and see that its larger with LOS. average applied damage and spread are two different metrics.

At the end of the day, you want to encourage LOS fights, while this test shows that it is making LRMs less effective in terms of focusing fire.


Here is another test at a longer range with LRM20s

[The6thMessenger]

Navid A1, on 18 January 2019 - 05:04 PM, said:

Appreciate the response.
Even if you are not claiming that LOS results in tighter physical spread, the end result is basically more spread damage.

You are maybe counting the average applied damage per volley and see that its larger with LOS. average applied damage and spread are two different metrics.

At the end of the day, you want to encourage LOS fights, while this test shows that it is making LRMs less effective in terms of focusing fire.


Here is another test at a longer range with LRM20s

Hey, could you test with a friend? Try shooting mechs with equal leg armor, torso-armor, and CT armor. see which goes first?

[Bad Pun]

Well, it really did seem in the outline of the PTS that Direct Fire was supposed to have a boosted effect in some way. Here's the relevant clip:

"The key points of this PTS session will be:
New LRM behavior that will see a shallower angle of attack and boosted performance while firing on targets within direct LOS
Reductions to baseline LRM attributes

While we will be reducing a number of LRM weapon attributes, the net effect of these changes have been explicitly tuned to produce the following results:

Net Buffs to LRM performance in direct fire situations
Nerfs to LRM performance in indirect fire situations"

LRMs are performing more poorly than they did before while in Line of Sight to the target.


Edit to add: I can also confirm that LRMs are currently more effective while fired indirectly in the PTS.

Edited by Bad Pun, 19 January 2019 - 01:55 AM.