Jump to content

Psa This Is Volumetric Scaling


478 replies to this topic

#381 Dimento Graven

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Guillotine
  • Guillotine
  • 6,208 posts

Posted 22 January 2020 - 12:59 PM

View PostNightbird, on 22 January 2020 - 12:21 PM, said:

Not talking about hit rig here or other mechs. Just take the first image in the OP, the commando is 1/4 the area of the Atlas. Per you, the size and hitboxs are right. If you want same armor per square meter what we have is what you want.

The third image shows how large the commando should be if the Atlas is the right size, per volumetric scaling.

The fourth image show how large the atlas should be if the commando is the right size, per volumetric scaling.

What we have now is equal armor per surface area, at least with all the humanoid mechs (I'm not comparing between humanoid and other shapes since that requires more work and wasn't the point of this thread)
I've never said the current situation was fine. What I said is it should be based on surface area.

The current formla if it is in fact surface area based, needs to be adjusted, significantly.

But there are many more 'mechs OTHER THAN Commandos and Atlas's.

Fafnir, and the King Crab come to mind as 'mechs with HUGE amounts of surface area that, regardless of any torsor twisting, the average pilot will be able to drill through a given hit location with ease, where as... say a Flea, Locust, or Piranha, yeah... Good luck...

Just showing a picture of two of the some 700+ 'mechs in the game and say, "See it's good...", nope, that's no argument either.

#382 Dimento Graven

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Guillotine
  • Guillotine
  • 6,208 posts

Posted 22 January 2020 - 01:06 PM

View Postjustcallme A S H, on 22 January 2020 - 12:32 PM, said:

Dimento...

Did you miss another dozen shots on light mechs and need to come post about it?
Participating in the conversation, giving my input.

It's not like I haven't seen you regularly miss the little buggers either.

It's not even so much about missing them, it's frustrating when people hit the little shits and it fails to register, or if they're hitting them with lasers while they're moving, the current implementation prevents them from inflicting any significant damage, thus allowing lights run amongst entire teams of enemy 'mechs, being nigh continuously bathed in laser beams not receiving any significant damage.

#383 Nightbird

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • The God of Death
  • The God of Death
  • 7,518 posts

Posted 22 January 2020 - 01:12 PM

View PostDimento Graven, on 22 January 2020 - 12:59 PM, said:

I've never said the current situation was fine. What I said is it should be based on surface area.

The current formla if it is in fact surface area based, needs to be adjusted, significantly.

But there are many more 'mechs OTHER THAN Commandos and Atlas's.

Fafnir, and the King Crab come to mind as 'mechs with HUGE amounts of surface area that, regardless of any torsor twisting, the average pilot will be able to drill through a given hit location with ease, where as... say a Flea, Locust, or Piranha, yeah... Good luck...

Just showing a picture of two of the some 700+ 'mechs in the game and say, "See it's good...", nope, that's no argument either.


This thread is about volume scaling versus current (surface area) scaling. If you don't think the in game scaling is surface area based, please explain what you think is surface area scaling. If you can't, just admit it and move along.

#384 Dimento Graven

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Guillotine
  • Guillotine
  • 6,208 posts

Posted 22 January 2020 - 01:17 PM

View PostPrototelis, on 22 January 2020 - 12:33 PM, said:

Stunlock is cancer.

Knockdown was removed because it causes desync and this is a server authoritative game.
That was BS. Worst I ever saw was the knocked down 'mech 'warping' one side 10 or 20 meters while trying to get up. Everyone who hyprobolizes the 'desyng' or 'warping all over the place' are lying.

Fact is, several things could have been done to prevent stun locking:

Only allowing a heavier 'mech to knock down a lighter 'mech.

Only allowing knock downs when combined velocities of the 'mech exceeded a certain threshold, so on and so forth.

There were plenty of options, but Russ in his monumental butthurtedness decided to pull it from the game altogether, and lied about 'looking for solutions'.

Seriously if there isn't going to be knock downs, then AT LEAST there should be significant physical damage delivered to a light 'mech humping a larger 'mech's legs. As it is now, the teeny tiny .1 point of damage being delivered now is completely negligible to the point it's mostly ignored.

Want to provide incentive for better light piloting? Increase the damage delivered to legs when enemy 'mechs collide to a full point per pulse and we'll have motivated people from leg humping.

Quote

lol at "unearned advantage" Maybe try not chainfiring AC2s at lights.
Yes, unearned advantage, and what does what weapon I fire have to do with it? Obviously you're trying to make this personal and wrong for doing so, especially since I use lots of other weapons (just not the computer guided ones).

View PostNightbird, on 22 January 2020 - 01:12 PM, said:

This thread is about volume scaling versus current (surface area) scaling. If you don't think the in game scaling is surface area based, please explain what you think is surface area scaling. If you can't, just admit it and move along.
You point to an example of two similarly shaped 'mechs and say, "See, it's working the way it should, no problem here."

When there are more than just the humanoid shaped 'mechs in this game.

If that's the logic that was used, yeah, no wonder it's out of whack.

#385 Appogee

    Member

  • PipPipPipPipPipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 10,966 posts
  • LocationOn planet Tukayyid, celebrating victory

Posted 22 January 2020 - 01:23 PM

For those who didn't pay close attention...

PGI said they did the rescaling using "volumetric scaling and then our judgement".

Can you see now where it all went went wrong?

#386 Nightbird

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • The God of Death
  • The God of Death
  • 7,518 posts

Posted 22 January 2020 - 01:29 PM

View PostDimento Graven, on 22 January 2020 - 01:17 PM, said:

You point to an example of two similarly shaped 'mechs and say, "See, it's working the way it should, no problem here."

When there are more than just the humanoid shaped 'mechs in this game.

If that's the logic that was used, yeah, no wonder it's out of whack.


Look at the last image then, (2x)^2 is 4x, that's surface area scaling since 4x25 tons is 100 tons.

I don't need to explain that area is x^2 right?

View PostAppogee, on 22 January 2020 - 01:23 PM, said:

For those who didn't pay close attention...

PGI said they did the rescaling using "volumetric scaling and then our judgement".

Can you see now where it all went went wrong?


Nan it's surface area scaling then their judgement.

#387 dario03

    Member

  • PipPipPipPipPipPipPipPipPip
  • Galaxy Commander
  • 3,627 posts

Posted 22 January 2020 - 01:47 PM

View PostDimento Graven, on 22 January 2020 - 12:16 PM, said:

No it appeared some people were asking for volume based balancing rather than surface area balancing.

Some 'mechs have SO MUCH surface area that even though they may have 200 armor in that location, it's actually incredibly thin given that the particular location in question can be hit from nearly every direction, and due to that size is even if the target is torso twisting to the maximum capability of the 'mech, even the average pilot has little issue maintaining their aim on some portion of the location, maximizing the damage delivered.

Lights on the other hand are armor dense, PLUS, with the "damage over time" hit registration for lasers, small size of hit boxes, physics breaking movement profiles, imperfect hit registration in general, the 20 armor they have in the same location is effectively at least twice that of the assault.

The result of all the various decisions, nerfs, quirks, et al, is that lights are way to durable in this game, as evidenced by, more often than not, the last surviving 'mechs at the end of the match will almost invariably light 'mechs.


You forgot an important reason for lights often being the last alive. LPT - low priority target, its probably the biggest reason they live as long as they do because they simply don't get shot at as much as larger mechs. In large part due to the fact that larger mechs are a bigger threat since they are more powerful.
At the end of the day even with lights being harder to hit they still have lower performance than larger mechs like assaults because while lights are stronger than they were in other games there is still plenty of things working against them. An assault that for some reason goes LPT can easily do far more damage than a light and even one that is high priority target can.

#388 FupDup

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 26,888 posts
  • LocationThe Keeper of Memes

Posted 22 January 2020 - 01:57 PM

I still don't know why people still take the "last mech alive" metric seriously. I've been the last man standing at least a few times in every single mech I've ever used (including Daishis). A lot of it is based on how aggressive you are and how close to the red team you are (i.e. snipers/pokers tend to live longer than brawlers, cue that annoying Raven saying "How did we lose?").

Usually being the last one alive just means you stayed behind your pug allies and/or managed to slip away when they got overrun (*raises hand).

#389 Dimento Graven

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Guillotine
  • Guillotine
  • 6,208 posts

Posted 22 January 2020 - 02:49 PM

View PostNightbird, on 22 January 2020 - 01:29 PM, said:

Look at the last image then, (2x)^2 is 4x, that's surface area scaling since 4x25 tons is 100 tons. I don't need to explain that area is x^2 right? Nan it's surface area scaling then their judgement.
I get that, but a linear 4x multiplier doesn't work because the speed of the typical Atlas is 1/4th the speed of the typical Commando, and with the current mechanics that speed effectively doubles the armor, or to put it another way the lack of any significant speed and huge *** hit boxes on the Atlas effectively reduces the comparable durability by 75%.

#390 Dimento Graven

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Guillotine
  • Guillotine
  • 6,208 posts

Posted 22 January 2020 - 02:56 PM

View Postdario03, on 22 January 2020 - 01:47 PM, said:

You forgot an important reason for lights often being the last alive. LPT - low priority target, its probably the biggest reason they live as long as they do because they simply don't get shot at as much as larger mechs. In large part due to the fact that larger mechs are a bigger threat since they are more powerful. At the end of the day even with lights being harder to hit they still have lower performance than larger mechs like assaults because while lights are stronger than they were in other games there is still plenty of things working against them. An assault that for some reason goes LPT can easily do far more damage than a light and even one that is high priority target can.
While I agree that quite a few people don't shoot at lights, I do disagree with labeling them as "low priority targets".

I do see part of what you're saying in my matches. It's frustrating as heck watching some potato chase after some weaponless medium/large, all the while they're being shot in the *** by a light 'mech.

I do not count light 'mechs as 'low priority' though. If there's an assault that's not firing at me, and light that is, or is attempting to move around me, I will prioritize on the light, since he's allowed to leg hump with impunity I want to encourage him to stay as far out from me as possible.

As far as lights being "low performers" that's only on the alpha side of the scale. With boatable MG's and microlasers, there's a few builds that have higher sustained DPS's than any assault.

Even then, I'd wager you take an Atlas, give it the exact same speed, movement, and weapons profile as an MG/micro laser boating light, and the Atlas would STILL be significantly easier to kill just because of the extreme difference in hit box sizes.

#391 Nightbird

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • The God of Death
  • The God of Death
  • 7,518 posts

Posted 22 January 2020 - 02:56 PM

View PostDimento Graven, on 22 January 2020 - 02:49 PM, said:

I get that, but a linear 4x multiplier doesn't work because the speed of the typical Atlas is 1/4th the speed of the typical Commando, and with the current mechanics that speed effectively doubles the armor, or to put it another way the lack of any significant speed and huge *** hit boxes on the Atlas effectively reduces the comparable durability by 75%.


Bottom line, you don't want surface area scaling. Speed is not in this discussion

#392 Dimento Graven

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Guillotine
  • Guillotine
  • 6,208 posts

Posted 22 January 2020 - 03:00 PM

View PostFupDup, on 22 January 2020 - 01:57 PM, said:

I still don't know why people still take the "last mech alive" metric seriously. I've been the last man standing at least a few times in every single mech I've ever used (including Daishis). A lot of it is based on how aggressive you are and how close to the red team you are (i.e. snipers/pokers tend to live longer than brawlers, cue that annoying Raven saying "How did we lose?"). Usually being the last one alive just means you stayed behind your pug allies and/or managed to slip away when they got overrun (*raises hand).
As I typically play assaults 'mechs and typically brawl with assaults, I can tell you those few times I was last 'mech standing was notable because the drops were excessively heavy (we're talking 6 or more assaults) with either no lights at all (or only one per side), or it was because I'd killed the little f'ers off early on in the match.

#393 justcallme A S H

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • CS 2020 Referee
  • CS 2020 Referee
  • 8,987 posts
  • LocationMelbourne, AU

Posted 22 January 2020 - 03:05 PM

View PostDimento Graven, on 22 January 2020 - 01:06 PM, said:

It's not like I haven't seen you regularly miss the little buggers either.


Difference being I don't blame hit-reg. I just know I missed.

#394 Dimento Graven

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Guillotine
  • Guillotine
  • 6,208 posts

Posted 22 January 2020 - 03:05 PM

View PostNightbird, on 22 January 2020 - 02:56 PM, said:

Bottom line, you don't want surface area scaling. Speed is not in this discussion
I want surface area scaling, factoring in all the various 'mech body types where the surface area of a Commando might be 1/4th that of an Atlas, due to the shapes of other 'mechs that surface flat formula stops working. The KGC, Fafnir, Corsair, Awesome, probably even the Stalker have hit boxes.

Heck the KGC can have a light 'mech standing underneath it shooting at as large a hit box area as 'mech standing above it. Yet the KGC with even more surface area than an Atlas still has the same amount of armor as the Atlas, making its armor effectively even more 'thin'.

#395 Dimento Graven

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Guillotine
  • Guillotine
  • 6,208 posts

Posted 22 January 2020 - 03:09 PM

View Postjustcallme A S H, on 22 January 2020 - 03:05 PM, said:

Difference being I don't blame hit-reg. I just know I missed.
Maybe you just don't know the difference.

I know when I miss and I ***** at myself, on stream, when I do it. I'll explain what I did wrong and nag myself to not do that.

Then there are the misses that are debatable, and I will typically say something along the lines of "maybe that was me".

However there are the glaringly hugely obvious hit/damage registration issues where you can see the shots just pass through without registering, or worse yet, registering the hit, even giving the flashy shape and color changing indication, but paper doll and percentage registering no change.

Or are you trying to tell me that out everything PGI has F'd up over the years, the one thing that's 100% perfect, 100% of the time, is hit and damage registration?

#396 Prototelis

    Member

  • PipPipPipPipPipPipPipPipPip
  • Shredder
  • Shredder
  • 4,789 posts

Posted 22 January 2020 - 03:13 PM

View PostDimento Graven, on 22 January 2020 - 01:17 PM, said:

That was BS. Worst I ever saw was the knocked down 'mech 'warping' one side 10 or 20 meters while trying to get up. Everyone who hyprobolizes the 'desyng' or 'warping all over the place' are lying.



A mech teleporting in a game which revolves around focused precise damage to specific components isn't important. Okay. lol.

Edited by Prototelis, 22 January 2020 - 03:13 PM.


#397 Nightbird

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • The God of Death
  • The God of Death
  • 7,518 posts

Posted 22 January 2020 - 03:32 PM

View PostDimento Graven, on 22 January 2020 - 03:05 PM, said:

I want surface area scaling, factoring in all the various 'mech body types where the surface area of a Commando might be 1/4th that of an Atlas, due to the shapes of other 'mechs that surface flat formula stops working. The KGC, Fafnir, Corsair, Awesome, probably even the Stalker have hit boxes.

Heck the KGC can have a light 'mech standing underneath it shooting at as large a hit box area as 'mech standing above it. Yet the KGC with even more surface area than an Atlas still has the same amount of armor as the Atlas, making its armor effectively even more 'thin'.


Then you're happy with the scaling between the Atlas and the Commando, stop complaining about lights then

#398 Dimento Graven

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Guillotine
  • Guillotine
  • 6,208 posts

Posted 22 January 2020 - 03:34 PM

View PostPrototelis, on 22 January 2020 - 03:13 PM, said:

A mech teleporting in a game which revolves around focused precise damage to specific components isn't important. Okay. lol.
So it broke the game SO much that they completely removed knockdowns, but yet when the same thing regularly happens with fast moving 'mechs with large latency or due to HSR having to reconcile significantly disparate latencies, it's "ok" and "no big deal"...

#399 Prototelis

    Member

  • PipPipPipPipPipPipPipPipPip
  • Shredder
  • Shredder
  • 4,789 posts

Posted 22 January 2020 - 03:36 PM

View PostDimento Graven, on 22 January 2020 - 03:34 PM, said:

same thing regularly happens with fast moving 'mechs with large latency


This is false.

A little warping happens in all games; it doesn't happen frequently in this game.

#400 Dimento Graven

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Guillotine
  • Guillotine
  • 6,208 posts

Posted 22 January 2020 - 03:40 PM

View PostNightbird, on 22 January 2020 - 03:32 PM, said:

Then you're happy with the scaling between the Atlas and the Commando, stop complaining about lights then
I think you're intentionally missing the points I've made.

However, I'll break it down:

1. Not every 'mech in this game is a Commando, or an Atlas.
2. There are 'mechs with more surface area than the Atlas.
3. There are 'mechs with smaller surface area than the Commando.
4. A flat 4 to 1 ratio doesn't work.
5. Even at the current 4 to 1 armor ratio, assaults are effectively less armored than lights due to lack of speed and the hit box size differences and how damage is registered and applied in game.

View PostPrototelis, on 22 January 2020 - 03:36 PM, said:

This is false.

A little warping happens in all games; it doesn't happen frequently in this game.
I notice it at least 4 or 5 times a night, say, once an hour, at least.

Set up some actual rules and logic around knockdowns, what 'little' warping there is (and again even at its worst it was minimal, with minimal shifts) and in the end there probably wouldn't be that much more of a difference.





8 user(s) are reading this topic

0 members, 8 guests, 0 anonymous users