28 replies to this topic

[Tarogato]

This is something that comes up again and again, and I'd like to add something to it. First off, the obvious bit - adding a "damage received" stat to the scoreboard would be trivial, it would be nearly identical to the damage dealt stat in terms of how the servers tracks it (and actually it's probably already tracked to an extent)

But this rewards the heaviest mechs foremost. There's two solutions - one is convuluted: score according to percent of health tanked so that you get the same score whether you're in a Locust or an Atlas, but this isn't very helpful. The second solution... is track "damage evaded". So you know that PPC that whiffed past your mech by an inch but didn't connect? Yeah, that counts as damage evaded by you. Arrghh, but how is the server supposed to track near misses like that? Easy, all you need is one ugly hitbox. Imagine it to be like this:




That hitbox surrounds the character. In MWO, mechs have very sophisticated hitboxes that are very accurate to the visual model. So lets say... this new box-shaped hitbox like the red one above... it collects damage for evaded damage. Every shot that passes through this red box is added to your "damage evaded" stat, and then at the end it subtracts that amount of damage that you actually sustained to your mech, so you wind up with two stats: damage sustained, and damage that came near your mech but actually missed you. Then we can start to see how much damage lights are *really* tanking.

Heck, it might even be possible to use the collision detection bounding boxes as the "near miss" hitbox, so it can do double duty and the server doesn't have to track an extra box. How's that sound for an elegant solution?

[mogs01gt]

That would require too much coding and system resources to be effective. What issue does this resolve?

Edited by mogs01gt, 12 April 2016 - 07:45 AM.

[Wintersdark]

Why should you be rewarded for near misses? It's more likely that a near miss is because someone shooting at you just didn't aim well than that you actually avoided it due to your own skill (super silly slow PPC bolts notwithstanding).

Damage tanked via percentage is about the best (and lets face it, most viable to implement) option. Adding a hitbox is fraught with buggy peril and way more complicated as it requires wholly new mechanics as shots need to hit that hitbox AND the actually hit hitbox within it.

In terms of Things PGI Is Likely To Do and Cost Benefit Ratio, your idea fails on both counts even if it's a pretty cool idea.

Ultimately, they'll look at the benefit: Slightly better reward system, vs. the cost: Decent amount of engineering time, potential for game-breaking bugs, and they'll discard the idea out of hand.

mogs01gt, on 12 April 2016 - 07:39 AM, said:

That would be require too much coding and system resources to be effective. What issue does this resolve?

Exactly. It's a lot of work and cost for a minor improvement (though I still contend "near miss" is 99% of the time because the shooting player just missed, not because you dodged) on rewards.

Totally zero chance of happening.

[Tarogato]

Wintersdark, on 12 April 2016 - 07:43 AM, said:

Why should you be rewarded for near misses? It's more likely that a near miss is because someone shooting at you just didn't aim well than that you actually avoided it due to your own skill (super silly slow PPC bolts notwithstanding).

Damage tanked via percentage is about the best (and lets face it, most viable to implement) option. Adding a hitbox is fraught with buggy peril and way more complicated as it requires wholly new mechanics as shots need to hit that hitbox AND the actually hit hitbox within it.

In terms of Things PGI Is Likely To Do and Cost Benefit Ratio, your idea fails on both counts even if it's a pretty cool idea.

Ultimately, they'll look at the benefit: Slightly better reward system, vs. the cost: Decent amount of engineering time, potential for game-breaking bugs, and they'll discard the idea out of hand.


Exactly. It's a lot of work and cost for a minor improvement (though I still contend "near miss" is 99% of the time because the shooting player just missed, not because you dodged) on rewards.

Totally zero chance of happening.

If it really takes that much effort to implement such a simple thing (which already has all the necessary components in place) ... then PGI is a worse developer then I am giving them credit for. And I don't think too highly of them to begin with...

[Ultimax]

I don't care if Heavies and Assaults get rewarded more, they soak more actual damage for the team and often in situations where they can't avoid it (and if they aren't, they need to be and this might help with incentive)


Not every bonus needs to benefit every single class equally - and eating bullets while holding a front line isn't the role of lighter tonned mechs.

Near misses is a decent idea, but its impossible to tell whether you specifically evaded it or whether your enemies just missed.

The damage soaking bonus should also distinguish between how much total armor/structure you have and how much you utilized. If you die with 95% of the damage direct to CT, you should not get a bonus - ever.

Edited by Ultimax, 12 April 2016 - 09:24 AM.

[Afuldan McKronik]

Ultimax, on 12 April 2016 - 09:23 AM, said:

I don't care if Heavies and Assaults get rewarded more, they soak more actual damage for the team and often in situations where they can't avoid it (and if they aren't, they need to be and this might help with incentive)


Not every bonus needs to benefit every single class equally - and eating bullets while holding a front line isn't the role of lighter tonned mechs.

Near misses is a decent idea, but its impossible to tell whether you specifically evaded it or whether your enemies just missed.

The damage soaking bonus should also distinguish between how much total armor/structure you have and how much you utilized. If you die with 95% of the damage direct to CT, you should not get a bonus - ever.

The system would have to be complicated enough to know when you actively twisted to spread damage to arms compared to when someone took a snap shot and hit an arm.

[Wintersdark]

Tarogato, on 12 April 2016 - 09:12 AM, said:

If it really takes that much effort to implement such a simple thing (which already has all the necessary components in place) ... then PGI is a worse developer then I am giving them credit for. And I don't think too highly of them to begin with...

It's how much effort vs. how much reward, simply cost:benefit.

And the ratio here is poor.

This has nothing to do with PGI's capabilities, and you're a smart dude, you should know this.

Once you start adding hitboxes, things get complicated and the potential for serious bugs is a real thing. Maybe it would go well, maybe not. Maybe it'd work out, maybe not. They'd need to factor in the potential for problems beyond simple design. Maybe your system would allow for exploits: such as deliberately missing your opponent, and him deliberately missing you, so you could farm cbills. Sure, this could be capped (maximum earnings for it? etc) but this is adding design time and complexity. Where should the cap be? It's just not a simple change.

Any developer would be making the same analysis.

Do the changes this idea bring to the table warrant the cost in implementation (see: dev time, QA time, potential conflicts with existing systems, potential for exploiting, extra server load (checking hits vs. another 24 hitboxes in game, thus requiring HSR rewinding) etc).

So, what does your idea bring? Try to look at this objectively.

It allows people to be rewarded because someone took a snap-shot at them and missed.

Put differently: How many weapons can you actively dodge? PPC's from an unquirked mech at a distance, sure. But beyond that, practically nothing. If someone missed you, 99% it's because their aim was bad, not because of what you did.

So, the benefit is terribly minor, the cost isn't.

Whereas: A cbill bonus for surviving a match at lower health is VERY easy to implement. It doesn't impact any systems, it's simply a reward bonus at EOR based on 1-(surviving_health_%) * max_cbill_reward. This is trivial with no danger of damaging other things, no processing load. It's much less exploitable; no more so than simply playing a match and having one side throw the match. The difference, of course, is that only one team can really benefit from it at a time; it'd be much more difficult to arrange and not worth the trouble vs. simply playing normally.

[Tarogato]

Wintersdark, on 12 April 2016 - 10:04 AM, said:

It's how much effort vs. how much reward, simply cost:benefit.

And the ratio here is poor.

This has nothing to do with PGI's capabilities, and you're a smart dude, you should know this.

How much effort (almost none, if PGI isn't completely dufous) vs how much reward (new stats that people have been asking for).


... I think you're overcomplicating it.

Quote

Once you start adding hitboxes, things get complicated and the potential for serious bugs is a real thing. Maybe it would go well, maybe not. Maybe it'd work out, maybe not. They'd need to factor in the potential for problems beyond simple design. Maybe your system would allow for exploits: such as deliberately missing your opponent, and him deliberately missing you, so you could farm cbills. Sure, this could be capped (maximum earnings for it? etc) but this is adding design time and complexity. Where should the cap be? It's just not a simple change.

Deliberately missing the enemy so that the enemy gets evasion rewards? I don't think people are going to be doing this. Besides, it should be a stat with less focus on the reward - actually hitting and dealing damage would give you a better return.

We already have hitboxes, they work fine. It's all stuff that's already in the game.

Quote

Put differently: How many weapons can you actively dodge? PPC's from an unquirked mech at a distance, sure. But beyond that, practically nothing. If someone missed you, 99% it's because their aim was bad, not because of what you did.

So, the benefit is terribly minor, the cost isn't.

Yes, you are being rewarded for the enemy having **** aim. If you are moving, and they can't hit you because you are moving, then it counts as evaded damage. You outplayed them. You moved. But more interesting, I think it would reveal a lot about light mechs - how much damage people are putting toward them without actually connecting. I think the numbers would be staggering and interesting to see, worth the tiny effort.

Quote

Whereas: A cbill bonus for surviving a match at lower health is VERY easy to implement. It doesn't impact any systems, it's simply a reward bonus at EOR based on 1-(surviving_health_%) * max_cbill_reward. This is trivial with no danger of damaging other things, no processing load. It's much less exploitable; no more so than simply playing a match and having one side throw the match. The difference, of course, is that only one team can really benefit from it at a time; it'd be much more difficult to arrange and not worth the trouble vs. simply playing normally.

This is actually very exploitable. Everybody just puts an alpha or two into allies at the end of the match--- boom, more tanking rewards. You can easily self-inflict damage with override, fall damage, and MASC. You have to measure the damage that the enemy has dealt to you, not the health you are at.

[Wintersdark]

Tarogato, on 12 April 2016 - 11:04 AM, said:

How much effort (almost none, if PGI isn't completely dufous) vs how much reward (new stats that people have been asking for).

Curious. What is your background, that you're sure that adding a hitbox to every mech in the game, that is transparent (that is, unlike EVERY other hitbox in the game, incoming fire can hit it AND another hitbox inside it, *but* only one other hitbox inside it)? How much testing have you done to determine exactly how big this hitbox should be vs. mech geometry? Is it just a box? Does it fully encompass a mech? If so, does that include the length of barrels (does a Jagermech, then, facing sideways (thus to the firing mech, it would require a very large box due to arm length, despite the rest of the geometry being narrow, so shots that missed by a fairly large margin would still register as "near misses" )

Quote

... I think you're overcomplicating it.

I'm not "overcomplicating it"; it's not my suggestion. That it IS more complicated than you seem to think is your failing, not mine.

I could go on and on, like above, at how really complicated this suggestion is. You don't seem to see that, but your obliviousness to the complexity doesn't remove it. Those (and many, many more) are all things that need to be considered and solved before this could be implemented.

Quote

Deliberately missing the enemy so that the enemy gets evasion rewards? I don't think people are going to be doing this. Besides, it should be a stat with less focus on the reward - actually hitting and dealing damage would give you a better return.

Yeah, that was one potential problem. Whether it's actually a problem or not depends on a lot of factors. But it as well is something that needs to be considered.

After all, group queue, 12v12, you get a match against a fellow 12 man unit. You start the match simply firing at each other and missing, move apart and then fight it out. Yay for potentially huge bonuses! Now, that's unlikely, and solvable by having minimal rewards for doing so, and/or max caps on evasion rewards, or whatever else.

Quote

We already have hitboxes, they work fine. It's all stuff that's already in the game.

Our existing hitboxes are different, and these would need to be assigned to every mech, sized differently for each, behave differently from existing ones, etc. Simply because we have a conceptually similar thing doesn't mean what you're asking for is trivial.

Quote

Yes, you are being rewarded for the enemy having **** aim. If you are moving, and they can't hit you because you are moving, then it counts as evaded damage. You outplayed them. You moved. But more interesting, I think it would reveal a lot about light mechs - how much damage people are putting toward them without actually connecting. I think the numbers would be staggering and interesting to see, worth the tiny effort.

The numbers WOULD be interesting to see, but that's a big can of worms just to get an interesting number. I get your appreciation for data (and I'm a big fan of it too!) but that's all a lot of work to get a small amount of data.

With that said, are you saying that if you were motionless, you wouldn't get the bonus? So the system needs to track your speed vs. shots crossing that "near miss" hitbox? Lots of times, people just miss because they just missed. I see people miss atlases that are walking directly towards them.

Quote

This is actually very exploitable. Everybody just puts an alpha or two into allies at the end of the match--- boom, more tanking rewards. You can easily self-inflict damage with override, fall damage, and MASC. You have to measure the damage that the enemy has dealt to you, not the health you are at.

Friendly damage has a direct cbill penalty. It's exploitable, but only as a very slow cbill transfer mechanic. As well, because this increases team damage, it endangers you for account action should someone report you for FF in the future, as you'd have an unusually high team damage amount.

[Tristan Winter]

I don't see any major reason why it shouldn't be in the game. It's a solid way to reward light mechs and other harassers in a way that is easy to measure and makes sense.

However... the downside is that you'd see Spider 5K's running around all match and thinking they were contributing. They'd die with 50 dmg and say "look at my matchscore, whiners". I think you'd need to think long and hard about how to make the algorithm, because a flat C-bill bonus per missed shot may not have the intended effect.

It would reward good gameplay at higher levels, perhaps at the cost of leading to undesirable gameplay at lower levels.

[Ultimax]

Tarogato, on 12 April 2016 - 11:04 AM, said:

This is actually very exploitable. Everybody just puts an alpha or two into allies at the end of the match--- boom, more tanking rewards. You can easily self-inflict damage with override, fall damage, and MASC. You have to measure the damage that the enemy has dealt to you, not the health you are at.

The game already recognizes the difference between damage you do to enemies and damage you do to allies.

It shouldn't take much for damage received to have similar flags, including "neutral" for fall damage/masc damage, etc.

[Tarogato]

Meh, I give up. I guess making a simple transparent hitbox that tallies incoming damage is too difficult for Potato Games Inc.

Oh well, out with another decent idea.

[Davegt27]

Sort of off topic but still on stats

I started to look at avg. damage per match for my Mechs so I can come up with the effectiveness of chassis and variants

It would be nice if it was automatically calculated

Also Stats on weapons it would be nice if we could get kills per weapons

Like how often do you actually get a kill with those crap ERPPC's

Edited by Davegt27, 12 April 2016 - 12:43 PM.

[Ex Atlas Overlord]

Ultimax, on 12 April 2016 - 09:23 AM, said:

Not every bonus needs to benefit every single class equally - and eating bullets while holding a front line isn't the role of lighter tonned mechs.

Uh oh, I detect back pedaling and double standards.....

So lights mechs need more firepower and better rewards because "it's a fighting game and everyone should be able to earn the same in combat".....and "stop saying lights have to be scouts, we can be for combat too".

But now "it's an ASSAULTS job to get shot, not every class needs to do the same thing guys."

But yeah OP.

This would actually be an amazing addition to the game.

Maybe PGI would wake the F*CK up and stop making lights so hard to kill (even when you hit then).

[Jetdrag]

Certainly would be nice if something encouraged people to rank damage. Would encourage some more passive teams to engage. Some matches you know you've lost in the first few minutes because your team won't fight.

[MechWarrior5152251]

How can you even know who the intended target was? I often shoot at one mech and hit another mech....

[Khobai]

wait you want to be rewarded because other people have bad aim? that makes no sense at all.


heres how you do a tanking bonus properly.

damage done by you / damage your mech received = ratio of damage done to damage taken.

ratio of damage done to damage taken divided by tonnage of your mech * some amount = your tanking cbill bonus


that rewards you for doing more damage than you take in return. And it divides by the tonnage of the mech so heavier mechs dont have an advantage.

Edited by Khobai, 12 April 2016 - 02:04 PM.

[Troutmonkey]

The Atlas Overlord, on 12 April 2016 - 12:55 PM, said:

Uh oh, I detect back pedaling and double standards.....

So lights mechs need more firepower and better rewards because "it's a fighting game and everyone should be able to earn the same in combat".....and "stop saying lights have to be scouts, we can be for combat too".

But now "it's an ASSAULTS job to get shot, not every class needs to do the same thing guys."

But yeah OP.

This would actually be an amazing addition to the game.

Maybe PGI would wake the F*CK up and stop making lights so hard to kill (even when you hit then).

It's the Assaults/Heavies jobs to tank damage when pushing the line. It's the Lights job to draw fire and attention while getting hit as little as possible. Both are contributing to their team.

Conversely, A light mech tanking dies, and an assault mech hanging back and letting the mediums get shot to pieces isn't helping their team at all.

[Jetdrag]

I think they need to include something in the tutorial. A mechs armour is a team asset. A good team will cover each other and swap out to share damage. It's not a bad thing to take some fire, but I see a lot of folks not wanting to get shot at and they sometimes dgive up desirable positioning even when there is a numerical advantage. That comes with experience but maybe it can be taught.

[Levi Porphyrogenitus]

I'm happy to reward heavier mechs for actually doing one of their jobs and taking fire off of the little guys. Atlases that hang back and poke while the rest of the team engage are only doing half (or less) of their job. Paying them to wade in and draw fire away from their fragile allies only makes sense.

It also pays more if you take damage without dying, since the better you are at spreading damage the more damage you can take and the higher your tanking payout will be. Throw in a small bonus for surviving, a bonus that scales up the lower your health percentage is at, and you should see people be willing to tank hits but not going out of their way to suicide.