Jump to content

- - - - -

Hit Box - Plan Of Action - Feedback


330 replies to this topic

#301 AlmightyAeng

    Member

  • PipPipPipPipPipPipPipPipPip
  • Bad Company
  • Bad Company
  • 3,905 posts

Posted 20 December 2013 - 05:32 AM

Artist has to take a leave of absence for personal reasons? "Private personal matter" is a range of things...and every damn one of them is a good excuse.

Thanks for the update Paul...always nice to sweeten a delay with "But we'll put in a couple extra, since it's coming late."

Great update, good news. Thank you.

View PostM0rpHeu5, on 20 December 2013 - 01:01 AM, said:

I don't think that turning the shoulder hitbox from torso to arm hitbox is a good idea. When i shoot something i want to know what i am hitting


From what I recall, a LOT of the hitboxes we have now show significant portions of the arm joint as a side torso. Like on the Jager...those inner 'arm shields' that are supposed to PROTECT the side torsos? Those counted AS side torsos...making them super vulnerable. If they're making THOSE part of the arms, and the big bulky Kintaro shoulders part of the arm hitbox...those are GOOD changes, and ones I've been waiting for.

#302 SpiralFace

    Member

  • PipPipPipPipPipPipPipPip
  • The Devoted
  • The Devoted
  • 1,151 posts
  • LocationAlshain

Posted 20 December 2013 - 06:17 AM

View PostGhost Badger, on 20 December 2013 - 05:32 AM, said:

From what I recall, a LOT of the hitboxes we have now show significant portions of the arm joint as a side torso. Like on the Jager...those inner 'arm shields' that are supposed to PROTECT the side torsos? Those counted AS side torsos...making them super vulnerable. If they're making THOSE part of the arms, and the big bulky Kintaro shoulders part of the arm hitbox...those are GOOD changes, and ones I've been waiting for.


See this is what I'm worried about. Those "armor fins" are physical components of the side torsos that are meant to cover the CT line. But they are PHYSICAL parts of the side torsos. Having things like that cover the arms might seem like a good idea at first, but if both the arms and the side torso are destroyed ant those fins counted as the arms, you all of a sudden get "dead zones" on the model that can absorb 75% of the damage coming into them just because its "counted" as the arm. And unlike the "Physical" arm geometry that falls away once destroyed, Torso geometry that COUNTS as arm geometry will just sit there and absorb shots.

It would make arms nothing more then meat shields for the torso, which is also a terrible idea for the jager since its Arms are a valuable asset to it.

Its one of the reasons that I'm glad the developers haven't decided to fully adopt what was proposed in the "hit box fixes" thread, but we'll see what they do with them.

#303 AlmightyAeng

    Member

  • PipPipPipPipPipPipPipPipPip
  • Bad Company
  • Bad Company
  • 3,905 posts

Posted 20 December 2013 - 06:26 AM

View PostSpiralFace, on 20 December 2013 - 06:17 AM, said:


See this is what I'm worried about. Those "armor fins" are physical components of the side torsos that are meant to cover the CT line.
Supposition. It's what YOU think they're meant to do. I think you're wrong.

View PostSpiralFace, on 20 December 2013 - 06:17 AM, said:

But they are PHYSICAL parts of the side torsos. Having things like that cover the arms might seem like a good idea at first, but if both the arms and the side torso are destroyed ant those fins counted as the arms, you all of a sudden get "dead zones" on the model that can absorb 75% of the damage coming into them just because its "counted" as the arm. And unlike the "Physical" arm geometry that falls away once destroyed, Torso geometry that COUNTS as arm geometry will just sit there and absorb shots.

It would make arms nothing more then meat shields for the torso, which is also a terrible idea for the jager since its Arms are a valuable asset to it.

Its one of the reasons that I'm glad the developers haven't decided to fully adopt what was proposed in the "hit box fixes" thread, but we'll see what they do with them.


I disagree. Looks to me like the 'fins' are armor plates placed directly on the joint between the side torso and the arms, ostensibly as armor to protect the SIDE torsos. Take a look at the fins...thin armor plating, just outside the torso area...engine components aren't bing housed in that...so why should my side torso explode if that's taken away?

And yes, it'd make it easier to use your arms as cover...that's actually USEFUL. Yes, Jager's have weapons in their arms...but what would you rather lose? An arm? Or a side torso (which destroys your arm ANYWAY, and depending on your engine, your mech). So why is that bad again?

Edited by Ghost Badger, 20 December 2013 - 06:36 AM.


#304 Lightfoot

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Bad Company
  • Bad Company
  • 6,612 posts
  • LocationOlympus Mons

Posted 20 December 2013 - 09:07 AM

I see several of the latest mechs to undergo Hitbox re-mapping (Kintaro, Jagermech) are having portions of the side torso sections re-mapped to the arm-shoulder sections and this still needs to be done to the Awesome which still has the largest Center Torso of any mech.

The Awesome should also be handled like the Atlas due to it's even larger Center Torso area, where the CT hitbox is only the front facing plate, then everything else is side torso hit-box, and the enormous shoulders could take up some of the Side Torso section increases. The Awesome model looks great, but it needs normalized hit boxes. I don't think I have ever lost an Arm on the Awesome while in other mechs it's quite common.

#305 Jairn

    Member

  • PipPipPip
  • Ace Of Spades
  • 68 posts

Posted 20 December 2013 - 10:43 AM

We have completely lost sight of why this was an issue.

"Mech hit boxes had bad geometry and were not registering hits correctly" A.K.A. spider invincibility.... and Awesome suckability just to name two. I can see how the changing of the CT on the awesome explained why every hit was registering on their CT. But you could point blank hit a spider multiple times and for some reason the hits were not registering. Do you really think moving the hit boxes or changing the size makes it so that the hits register more reliably? I remember a staff member stating something about hits not registering inbetween components due to geometry and hitbox validation. How do you know a simple artist pass is fixing these issues? What kind of validation are you doing? Do you have a place to shoot at each and every mech from all different kinds of angles to validate hits register correctly? Are you allowing players to test this on a test server? Are you asking for specific testing before and after to find flaws or issues?

IMO, the original (and root cause) issue is being completely ignored. And this has morphed into simply 'changing' where the hitboxes are and splitting damage locations so it wasn't as easy to core out a mech. I am fine with that change and it seems reasonable to do at the same time. But that is not the fix to the original issue. It's a bonus for game play.

So in essence what you are doing is applying a band aid to a symptom and ignoring the actual disease.

To prove my point. A spider is still as frustrating to kill as ever even if you hit it full frontal face on. And to close out an action item you need to validate the end result meets your goals. As your manager I would not allow you to close out this action item even if you did this artist change to each mech. You need to confirm that you fixed the issue. The verdict is out on this one actually 'fixing' the root cause and not just as a band-aide that makes it seem better.

-J

#306 Cimarb

    Member

  • PipPipPipPipPipPipPipPipPip
  • Caladbolg
  • Caladbolg
  • 3,912 posts
  • Twitter: Link
  • Twitch: Link
  • LocationA hop, skip and jump from Terra

Posted 20 December 2013 - 12:38 PM

@Jairn

It's a different issue. HSR is one issue. Hitboxes are another. Yes, they complicate each other and one may affect the other, but so does weapon damage.

Put another way: HSR determines IF a hit is made. Hitboxes determine WHERE the hit is applied if made. Damage determines HOW MUCH the hit does to the area if made.

#307 Jairn

    Member

  • PipPipPip
  • Ace Of Spades
  • 68 posts

Posted 20 December 2013 - 01:08 PM

I am not talking about HSR as the issue. That is another issue entirely. That is what makes this so confusing. There are BROKEN mechs with graphics and hitbox issues. Some are geometry of the mech like the Awesome that can be fixed by shrinking the hitbox of the CT. That makes sense. But there are other issues (unknown) like the Spider that is not related to HSR. I can hit other mechs in a match just fine but hitting a spider is an issue. And since they already did the 'fix' for the hitboxes on the spider I should see an improvement for the point blank hits that should hit SOMETHING but are not even hitting anything. Again, not HSR issues but mech hitbox issues.

This is why PGI is confused and already working on the wrong issue. Changing mech hitboxes is a plus or added bonus but they are missing the real root cause of why people asked for looking into hitbox issues in the first place!

That is why I am calling out PGI on any kind of control and effect type scientific methodology of 'fixing' these issues.

When you 'fixed' the Spider:
What issues did you find with the hitbox?
What did you do to fix them?
What results did you see from testing?
Are these stable fixes that can be reproduced on connections with no HSR issues?
Are you confident of hitbox registration (minus the HSR issues that can be seen with lag and server/client side ping

There is no approach like this being taken by PGI and instead we are glad that there is an 'artistic pass over of each mech'.

Again, ignoring the issue and putting a band-aide on to placate the concerns that people have raised about certain mechs!!

-J

#308 Cimarb

    Member

  • PipPipPipPipPipPipPipPipPip
  • Caladbolg
  • Caladbolg
  • 3,912 posts
  • Twitter: Link
  • Twitch: Link
  • LocationA hop, skip and jump from Terra

Posted 20 December 2013 - 01:45 PM

I legged three spiders in the first two minutes of a match last night (a well spent 82 damage if ever there was one) before getting XL-sniped by their friends, so I think they are making improvements.

That being said, HSR and hitboxes are the two issues. There isn't a third regarding what you are talking about. HSR determines whether you hit the mech "somewhere" or not. Hitboxes determines where you hit "if" it connected. Somewhere between these two systems is where it is messing up. Since it is happening even with mechs standing still, I would bet it is in the hitboxes - likely a gap or computer-confusing overlap - which is what they are working on fixing in this series of passes (and they mentioned both of those situations in one of the recent posts). Will it necessarily fix the entire issue? Probably not on the first pass, but it will make it better. They need to do a lot more testing with these changes than they have been, though, because we are going to find issues way faster than their people can, just by sheer numbers.

Edited by Cimarb, 20 December 2013 - 01:46 PM.


#309 Jairn

    Member

  • PipPipPip
  • Ace Of Spades
  • 68 posts

Posted 20 December 2013 - 01:50 PM

Cimarb there is a third issue.

What you see on screen is a graphical representation of the mech. But behind that the server sees multiple boxes and shapes that make up the mech. Those imaginary shapes are what the HSR and hitboxes actually are.

The issue is that geometry is overlapping and when you destroy a part (Centuriion for example) your hits are not hitting the left over CT but their LT and RT which are gone. This causes you to have a shield (which should not be there) for your CT and you can run around with near impunity because the damage that is being registered is on the RT and LT and it's being transferred (albiet with a dmg transfer rate that is lower than a clean hit).

That is the third issue. IF this artistic pass is fixing that great. But I fear it is not.

#310 Cimarb

    Member

  • PipPipPipPipPipPipPipPipPip
  • Caladbolg
  • Caladbolg
  • 3,912 posts
  • Twitter: Link
  • Twitch: Link
  • LocationA hop, skip and jump from Terra

Posted 20 December 2013 - 02:12 PM

That is the hitbox overlap I was talking about. It's not a different issue. The ghost section damage transfer you are talking about has nothing to do with spiders not registering a hit, as the hitbox doesn't change shape - it is there or not.

#311 Sable Dove

    Member

  • PipPipPipPipPipPipPipPip
  • Knight Errant
  • 1,005 posts

Posted 20 December 2013 - 05:20 PM

Blackjack: Minor changes. Arguably a buff, since more CT armour can safely be moved to the front.
Kintaro: Buff. More XL friendly, smaller CT.
Jagermech: Buff. More XL-friendly, no change to CT.
Cataphract: Buff. More XL-friendly, smaller CT.
Catapult: Smaller head, bigger CT. Will get fewer annoying deaths by headshot, but will die faster overall.

The people in charge of these changes needs to be reassigned.


To be honest, you don the 'fireproof suit' because you fear backlash over the delays, but really the delays would be fine if you weren't buffing two of the most powerful mechs available, while the poor Catapult gets changes that are likely going to make it even easier to kill.

If you were going for consistent, well-done hitboxes, like the Cent used to have (bugs notwithstanding), then okay. The Jagermech and Cataphract would be understandable. But you're not going for consistency. The Catapult gets an even bigger CT, the Centurion gets a bigger CT, Cicada gets a bigger rear-CT, while already-powerful mechs get smaller CTs, and/or smaller LT/RTs, and larger arms in return. Wow, I sure am glad I can take off that Jager's AC20 more easily now! Too bad he's still got another one.


Cicada and Centurion get nerfs, but Jagermech and Cataphract get buffs? What's the thought process that goes into these changes?

Edited by Sable Dove, 20 December 2013 - 05:22 PM.


#312 Bront

    Member

  • PipPipPipPipPipPipPipPipPip
  • Urban Commando
  • Urban Commando
  • 4,212 posts
  • LocationInternet

Posted 22 December 2013 - 07:34 AM

View PostJairn, on 20 December 2013 - 01:50 PM, said:

Cimarb there is a third issue.

What you see on screen is a graphical representation of the mech. But behind that the server sees multiple boxes and shapes that make up the mech. Those imaginary shapes are what the HSR and hitboxes actually are.

The issue is that geometry is overlapping and when you destroy a part (Centuriion for example) your hits are not hitting the left over CT but their LT and RT which are gone. This causes you to have a shield (which should not be there) for your CT and you can run around with near impunity because the damage that is being registered is on the RT and LT and it's being transferred (albiet with a dmg transfer rate that is lower than a clean hit).

That is the third issue. IF this artistic pass is fixing that great. But I fear it is not.
Not only is what you're describing a hit box issue, the Cent's "Phantom Arm" issue was actually a hitbox they had to fix that was specifically called out in the patch update (There were spots on the CT that were registering as "arm" instead.

#313 AlmightyAeng

    Member

  • PipPipPipPipPipPipPipPipPip
  • Bad Company
  • Bad Company
  • 3,905 posts

Posted 26 December 2013 - 12:15 PM

  • Jagermech (Pelvis area split for leg/CT distribution, more of the shoulder added to the arms)
Makes sense to me...why doesn't the Awesome get the same treatment, with those ENORMOUS shoulders?

Edited by Ghost Badger, 26 December 2013 - 12:15 PM.


#314 Ashvins

    Member

  • PipPipPipPipPip
  • Bad Company
  • Bad Company
  • 174 posts

Posted 08 January 2014 - 06:55 AM

Something I have noticed of late (since the DEC 17TH patch) is frequent deaths from hit's from behind hitting FRONT CT. This is not in just one mech type, I've seen it across the board in over 12 different mech chassis. I don't know if this is a hit detection issue hit box issue or netcode but it's getting really annoying.

#315 DeathlyEyes

    Member

  • PipPipPipPipPipPipPip
  • The Messenger
  • 940 posts
  • Google+: Link
  • Facebook: Link
  • Twitter: Link
  • Twitch: Link
  • LocationMetaphorical Island somewhere in the Pacific

Posted 16 January 2014 - 11:06 PM

Still no mention of terrain, the worst offender in this game. Can we please get proper terrain hit boxes on maps with jagged terrain. Most particularly horrible is Tourmaline Desert.

#316 Necromantion

    Member

  • PipPipPipPipPipPipPipPip
  • The 1 Percent
  • The 1 Percent
  • 1,193 posts
  • LocationBC, Canada

Posted 18 January 2014 - 01:59 PM

Spider still borked and registration not working.
Awesome "fix" was not needed
Cataphract "fix" broke them and theres no point in running a standard anymore because 9/10 times your ct gets nuked even when trying to twist to avoid dmg.

#317 EoRaptor

    Member

  • PipPip
  • Ace Of Spades
  • 37 posts

Posted 19 January 2014 - 05:25 PM

View PostNecromantion, on 18 January 2014 - 01:59 PM, said:

Spider still borked and registration not working.
Awesome "fix" was not needed
Cataphract "fix" broke them and theres no point in running a standard anymore because 9/10 times your ct gets nuked even when trying to twist to avoid dmg.


The cataphract fixes aren't in the game yet, and won't be for a few days?

#318 Deathlike

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPipPipPip
  • Littlest Helper
  • Littlest Helper
  • 29,240 posts
  • Location#NOToTaterBalance #BadBalanceOverlordIsBad

Posted 19 January 2014 - 05:39 PM

View PostSable Dove, on 20 December 2013 - 05:20 PM, said:

Cicada and Centurion get nerfs, but Jagermech and Cataphract get buffs? What's the thought process that goes into these changes?


Quick hint: Which is better for PGI... the X5 and YLW, or the FB and Ilya?

#319 zudukai

    Member

  • PipPipPipPipPipPipPipPip
  • Trinary Star Captain
  • Trinary Star Captain
  • 1,707 posts

Posted 19 January 2014 - 07:57 PM

View PostDeathlike, on 19 January 2014 - 05:39 PM, said:


Quick hint: Which is better for PGI... the X5 and YLW, or the FB and Ilya?

neither because the user already has payed for MC.

#320 Deathlike

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPipPipPip
  • Littlest Helper
  • Littlest Helper
  • 29,240 posts
  • Location#NOToTaterBalance #BadBalanceOverlordIsBad

Posted 19 January 2014 - 07:57 PM

View Postzudukai, on 19 January 2014 - 07:57 PM, said:

neither because the user already has payed for MC.


Who said this was for the ones that bought them?

Edited by Deathlike, 19 January 2014 - 07:58 PM.






4 user(s) are reading this topic

0 members, 4 guests, 0 anonymous users