Jump to content

Is Pgi Aware Of The Current Catastrophic Ballistic Hit-Registration?


85 replies to this topic

#41 Deathlike

    Member

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

Posted 17 May 2013 - 11:02 AM

View Postl33tworks, on 17 May 2013 - 08:42 AM, said:




You clearly missed the first one.. I'm a little skeptical of the 2nd one, however.. you don't have a paperdoll view to make this easier. It's 50-50 on the second one missing.

Quote

Anyway, the chain firing request. It wasn't a bad idea as the sample size if larger within the same time frame. A mixed bag of everything. This time I actually miss for real, a lot, so you can see the what the difference between a hit and a miss looks like client side.



Oh and I hope you guys aren't watching these videos in these forums windowed? They are 720p if you watch them in youtube
.


I've already seen this behavior before, and it happens and it's annoying. However, it doesn't exactly help that most of your shots seem so frantic that it probably making the issue worse than it should be. Noone has argued that it isn't happening, but to what extent is still debatable.

#42 Pilotasso

    Member

  • PipPipPipPipPipPip
  • The 1 Percent
  • The 1 Percent
  • 365 posts
  • LocationInner Sphere

Posted 17 May 2013 - 01:37 PM

I often get damage on the opposite side where the shot actually came from. Its not ballistic problem or even Streak SRM hitting centre torso problem. The whole hit detection is pretty messed up.

#43 Lugh

    Member

  • PipPipPipPipPipPipPipPipPip
  • The Widow Maker
  • The Widow Maker
  • 3,910 posts

Posted 17 May 2013 - 03:19 PM

View Postl33tworks, on 16 May 2013 - 02:33 AM, said:



Yea but surely its not as bad? The reason i say that is because I seemed to be doing a lot better before the state rewind, relatively speaking. Sure I may have missed more, but other people missed a butt load more. Now thats changed.

Each shot I take I have to concentrate, aim, tell myself "be good", take the risk to fire, Boom it hits, and the enemy is untouched a huge scary amount of the time.

I asked some people of teamspeak when observing me to tell me where a shot I've landed on my screen went on theirs, and they said it went behind the enemy. So I learned to shoot a bit ahead. The next time leading the same ammount is too much.

Its hard enough hitting when moving with ballistics in a 0 ping in this game, having to compensate for an ever changing unknown amount on top of it is just ridiculous.

Imagine if you will knowing you are directly behind your target and there is no way in hell you could miss. Wailing away with an AC20 and 5MLs watching the ac20 not register at ALL, the mlas oranges the armor. You shoot again with the ac20 and again the armor doesn't change at all. The mlas cycle and the center yellows with exposed internals. At two 40 point alphas into the CT of a target moving directly away from you. The mech should be D E D ...D E D. There is no way for 40-50points of internal damage to not have killed this guy. Yet he's still happy to merely be exposed as he turn to face you and you unload another 40 point alpha into his CT it merely yellows the outline not the happy orange of about to be exposed.

And the guy you are doing this to? A fricking Raven or a Cataphract. or a Dragon .. it's rediculous. As I can't stay connected a full match anymore I give them ONE chance each night to keep me connected for a game with my 38ms and 1ms jitter and they can't even do that.

#44 aniviron

    Member

  • PipPipPipPipPipPipPipPipPip
  • 3,752 posts

Posted 17 May 2013 - 05:56 PM

View Postl33tworks, on 16 May 2013 - 02:09 AM, said:

I am would hazard a guess the same shots taken by a Americas player, likely all would all register.

I would love to hear if other people are having similar expriences to me and have video footage.

I also wonder what PGis take on the matter is and if they are aware of it,


I run an 3xPPC AWS-8Q, and have since closed beta. I would guess about 1/3 of my shots 'dud.' I ping average 60ms.

View Postl33tworks, on 16 May 2013 - 03:06 AM, said:

At 00;02. Its not hard to tell. Thats 4 ER ppcs hiting a raven. That should have ripped the hand off the raven if in fact damage did register.


When talking about birds, we call that a wing. ;]

#45 SmokinDave73

    Member

  • PipPipPipPipPipPip
  • Elite Founder
  • Elite Founder
  • 355 posts
  • LocationAlpheratz, Outer Sphere Periphery

Posted 17 May 2013 - 11:07 PM

I wonder when a DEV will make a comment on this post. I appriciate egomane's feedback but I would like to have some feedback from PGI on the current hit registration situation. The game is pretty much RNG for hitting people at times and I am tired of playing a game that makes you guess/second guess whether or not you hit the guy standing right in front of you. This is the first major discrepancy in the game that has made me stop playing all together.

#46 Kitane

    Member

  • PipPipPipPipPipPipPipPip
  • Bridesmaid
  • 1,009 posts
  • LocationPrague, Czech Republic

Posted 18 May 2013 - 02:12 AM

The following isn't about hit registration, it's about the other problem mentioned in the thread.

One side effect of HSR is that incoming damage will always look a bit wonky (aka the enemy fire will appear to go wide and miss your mech and it will still hit). That is intended and how HSR is supposed to work.

T=0: The mech A shoots at mech B
T=1: The server receives the info. Before HSR it would evaluate the shot against position of mech B at time T1. With HSR it will rewind the simulation and calculate the shot against previous known position of mech B at T0.
T=2: The mech B receives the info about the shot and displays it to the client, however the shot is still rendered how it happened at T0.

Incoming fire is visually delayed by combined latency of both mech A and mech B connections. HSR fixes accuracy at cost of reduced visual fidelity on other clients.

Edited by Kitane, 18 May 2013 - 02:13 AM.


#47 Chavette

    Member

  • PipPipPipPipPipPipPipPipPip
  • Little Helper
  • 2,864 posts

Posted 18 May 2013 - 03:54 AM

View PostKitane, on 18 May 2013 - 02:12 AM, said:

Incoming fire is visually delayed by combined latency of both mech A and mech B connections. HSR fixes accuracy at cost of reduced visual fidelity on other clients.


That is correct. The point of this thread is to get an answer if they plan to do anything about the current de-synced mess. Its hard to get better when you have no idea what factors the chance of you hitting the enemy.

#48 Hayashi

    Snowflake

  • PipPipPipPipPipPipPipPipPip
  • Bridesmaid
  • Bridesmaid
  • 3,395 posts
  • Location輝針城

Posted 18 May 2013 - 12:35 PM

I've experienced both having shots that clearly miss me actually deal damage to my side torsos after I avoid them completely, as well as shots that hit me (and actually jolt my aim) do zero damage to the paper doll. And I also lead targets when using lasers since aiming directly doesn't actually hit. Can't comment too much on ballistics since they've never been reliable for players with 280 ping like me... but I tried twin PPCs for a short while and did 100-380 damage a match, where 50% of the shots that are supposed to hit don't flash the crosshairs red - but I can't tell if it's a near miss that hits the ground or if it actually hit the target due to the range. On the quad MPLAS I'm more used to, my average damage goes from 400 to 800 damage. It's the same person aiming the weapons... and coming from Asia, we (together with Aussie players) have multiple years more experience in lagshooting than most. So while it's hardly conclusive it does seem to me that ballistics is still somewhat off.

There still seems to be a hit registration issue in my experience. It's already improved massively since prior to ballistics HSR I couldn't even do 100 damage reliably with the PPCs and when I saw a 4 AC/2 Phract I did a mental whoopee. But that doesn't mean the issue is completely fixed yet.

While my ping is high, it's extremely stable and I have zero lost packets though.

The devs probably have better tools to see what's going on than we have as players... just hope they use it. Right now it seems a fraction of players get nearly full accuracy on their ballistics, while the remainder play a RNG game.

Edited by Hayashi, 18 May 2013 - 01:54 PM.


#49 Chavette

    Member

  • PipPipPipPipPipPipPipPipPip
  • Little Helper
  • 2,864 posts

Posted 18 May 2013 - 01:11 PM

Kitane pretty much nailed the problem. Your position of shooting is rolled back to the position you were minus your ping. So if my ping is 150ms, I pop from behind a building, and shoot, I can miss If the enemy hasnt been in sight for 150ms+. However, even after I went back in cover, I can shoot and it will connect with the guy if I havent been in cover for 150ms plus when I press the button.

#50 5th Fedcom Rat

    Member

  • PipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 893 posts

Posted 18 May 2013 - 04:50 PM

All I know is, Ac20s and PPCs are definitely LESS accurate for me ever since ballistic state rewind was put in. I have a ping of between 20-50 (which is usually fixed during the match, and doesn't fluctuate much if at all).

On the other hand, AC2s and Ultra AC5s seem to register better now (or perhaps, are just firing at their proper rate so more hits actually land on the enemy).

.

Edited by 5th Fedcom Rat, 18 May 2013 - 06:34 PM.


#51 aniviron

    Member

  • PipPipPipPipPipPipPipPipPip
  • 3,752 posts

Posted 18 May 2013 - 06:04 PM

View PostKitane, on 18 May 2013 - 02:12 AM, said:

The following isn't about hit registration, it's about the other problem mentioned in the thread.

One side effect of HSR is that incoming damage will always look a bit wonky (aka the enemy fire will appear to go wide and miss your mech and it will still hit). That is intended and how HSR is supposed to work.

T=0: The mech A shoots at mech B
T=1: The server receives the info. Before HSR it would evaluate the shot against position of mech B at time T1. With HSR it will rewind the simulation and calculate the shot against previous known position of mech B at T0.
T=2: The mech B receives the info about the shot and displays it to the client, however the shot is still rendered how it happened at T0.

Incoming fire is visually delayed by combined latency of both mech A and mech B connections. HSR fixes accuracy at cost of reduced visual fidelity on other clients.


While this is all true, it's not really what OP is talking about. His (our) problem is that outgoing fire is not hitting, rather than that incoming fire appears to miss and hits or vice versa. Becuase HSR is set up to prioritize the shooter, outgoing fire that appears to hit should always hit, but it does not.

#52 Chavette

    Member

  • PipPipPipPipPipPipPipPipPip
  • Little Helper
  • 2,864 posts

Posted 18 May 2013 - 06:07 PM

View Postaniviron, on 18 May 2013 - 06:04 PM, said:


While this is all true, it's not really what OP is talking about. His (our) problem is that outgoing fire is not hitting, rather than that incoming fire appears to miss and hits or vice versa. Becuase HSR is set up to prioritize the shooter, outgoing fire that appears to hit should always hit, but it does not.

Its just the matter of viewing it from the guy getting shot or the shooter.

#53 xXBagheeraXx

    Member

  • PipPipPipPipPipPipPipPip
  • Elite Founder
  • Elite Founder
  • 1,707 posts

Posted 18 May 2013 - 06:34 PM

just glazed over the post and im going to assume you are a European player with 150+ ping? Ive played games with lousier netcode than this....before the hit-box fixes and improvements i had to lead lights legs by about half a length to a full mechs length ahead to actually get hits to register...this seems to have been improved somewhat, as ive clobbered quite a few raven and jenner legs lately with gauss and large laser salvoes....however I have a much better ping. You might still have to keep using Gun-fu and pretty much hazard a guess and "lag lead" the target a bit until you get hits to register. If you begin to get damage on fleeting targets again, then its probably just some shoddy netcode on thier part that still needs to be worked on.

#54 Koniving

    Welcoming Committee

  • PipPipPipPipPipPipPipPipPipPipPipPipPip
  • The Guide
  • The Guide
  • 23,384 posts

Posted 18 May 2013 - 06:55 PM

Try shutting down less. Run standard PPCs and fire at range, then see if you fair better. Myself I have no issues with the exception of Tourmaline, where invisible walls floating above the rocks tend to absorb my shots. One way to tell if you can or can't hit the target if you fire is to carry a Tag with you.

Tag the target. If it flashes red on the target, it's okay to shoot the target. If it does NOT turn red, you can't hit the target no matter what -- start leading ahead until it does turn red, and then fire the PPC.

#55 Jay Z

    Member

  • PipPipPipPipPipPip
  • Deadset Legend
  • Deadset Legend
  • 436 posts
  • Twitch: Link
  • LocationAustralia

Posted 18 May 2013 - 10:28 PM

I have noticed and tested extensively that some shots do not register damage when they should. Specifically with PPCs even against slow or stopped targets. I fire, I see the burst of energy coming off their torso and the reticle turns red but the ragdoll remains undamaged. This was confirmed by a teammate on teamspeak who witnessed my shot landing and not passing through. I know when I miss, and I can with certainty conlcude that a small percentage ~10-20% of PPC shots do not register (clear line of sight, reticle turned red, enemy ragdoll/ damage percentage unchanged). My ping is 270.

However, I have been hit with PPCs, gotten cockpit shake and seen the enemy mech (in range) and not received damage so I know that others have experienced this bug.

Although it can be frustrating, I still love this game and will keep playing, it adds a fun little completely random "dodged the bullet" chance which is actually not too far off the tabletop.

#56 SmokinDave73

    Member

  • PipPipPipPipPipPip
  • Elite Founder
  • Elite Founder
  • 355 posts
  • LocationAlpheratz, Outer Sphere Periphery

Posted 19 May 2013 - 06:10 AM

View PostHayashi, on 18 May 2013 - 12:35 PM, said:

I've experienced both having shots that clearly miss me actually deal damage to my side torsos after I avoid them completely, as well as shots that hit me (and actually jolt my aim) do zero damage to the paper doll. And I also lead targets when using lasers since aiming directly doesn't actually hit. Can't comment too much on ballistics since they've never been reliable for players with 280 ping like me... but I tried twin PPCs for a short while and did 100-380 damage a match, where 50% of the shots that are supposed to hit don't flash the crosshairs red - but I can't tell if it's a near miss that hits the ground or if it actually hit the target due to the range. On the quad MPLAS I'm more used to, my average damage goes from 400 to 800 damage. It's the same person aiming the weapons... and coming from Asia, we (together with Aussie players) have multiple years more experience in lagshooting than most. So while it's hardly conclusive it does seem to me that ballistics is still somewhat off.

There still seems to be a hit registration issue in my experience. It's already improved massively since prior to ballistics HSR I couldn't even do 100 damage reliably with the PPCs and when I saw a 4 AC/2 Phract I did a mental whoopee. But that doesn't mean the issue is completely fixed yet.

While my ping is high, it's extremely stable and I have zero lost packets though.

The devs probably have better tools to see what's going on than we have as players... just hope they use it. Right now it seems a fraction of players get nearly full accuracy on their ballistics, while the remainder play a RNG game.


Could not of explained it better myself. I have 290 ping because I am from Australia and have a very silmilar experience zero packet lost stable connection ect. The HSR have been great but since the may 7th patch I have noticed so much more ballistic and laser shot's just not registering any damage at all.

#57 Daneel Hazen

    Member

  • PipPipPipPipPip
  • Nova Commander
  • Nova Commander
  • 173 posts
  • LocationFlorida

Posted 19 May 2013 - 06:29 AM

Maybe some of it is parallax error... although the targeting computer should compensate maybe that's part of weapon convergence.

#58 Oderint dum Metuant

    Member

  • PipPipPipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 4,758 posts
  • LocationUnited Kingdom

Posted 19 May 2013 - 06:41 AM

Maybe it's intentional to make any non NA players play on regional servers....

Then again maybe not!

#59 Waking One

    Member

  • PipPipPipPipPipPip
  • Overlord
  • Overlord
  • 427 posts

Posted 19 May 2013 - 07:11 AM

View Postl33tworks, on 17 May 2013 - 08:42 AM, said:

The First alpha of PPCS misses and hits the ground behind the enemy mech. Note the balls of snow being kicked up from the miss

The second alpha hits the enemy mech. Notice the sparks and flames from the hit. The first video was full only this.
Damage is registered, but interestingly even at 884m distance, 4 ERPPcs should have done more damage. That commando was smoking before that shot. I beleive he was at 64%





You never targeted it before hitting it so we have no idea what happened there, but my guess is not all of your ppcs actually hit him.

Stop expecting the game to correct for your bad aiming.

#60 jay35

    Member

  • PipPipPipPipPipPipPipPip
  • The Hammer
  • The Hammer
  • 1,597 posts

Posted 19 May 2013 - 08:38 AM

View Postl33tworks, on 16 May 2013 - 02:09 AM, said:

Out of a total sample of 9 Alpha shots, ALL direct hits, 4 of them registered and 5 did not.

I am would hazard a guess the same shots taken by a Americas player, likely all would all register.

Nope, it's just as bad here.

Try a Jager with AC5s or AC20s, it's just as hit or miss as PPCs and often easier to see with ACs that your shots were indeed unquestionably on target yet did not register.

Edited by jay35, 19 May 2013 - 08:43 AM.






9 user(s) are reading this topic

0 members, 9 guests, 0 anonymous users