Looking For Video Evidence Of Missed Registration On Light Mechs For The Sake Of Discussing Gameplay Balance.
#61
Posted 23 September 2013 - 05:27 AM
#62
Posted 23 September 2013 - 05:48 AM
There are clearly hit reg problems, and because it is more apparent the smaller you are it's natural that some light pilots might become a little defensive. But to be fair, as the Spider is the most effected there is no way any of us can tell if there are further problems with the chassis until the current issues have been sorted out.
#63
Posted 23 September 2013 - 05:50 AM
One point I believe many folks are overlooking is surface volume...
Fact is, most lights size (thus hit-box volume) is roughly 1/2 to 1/3rd the side of mediums, heavies and assaults. The result is, not only is the target smaller, more difficult to hit... for weapons that are hit-scan / duration weapons, the ability to keep that damage focused on a given hit-box in reduced dramatically and hit-scan damage tends to float all over, thus spreading the damage. Obviously, front-loaded weapons do fair better but the skill-level necessary to consistently apply that damage on lights is a tall order for most that IMHO is further mitigated by HSR...
In short... I don't necessarily think light hit-boxes are borked as much as I they are proportionately difficult to inflict damage upon purely from a percentage basis.
That said, I know that doesn't account for poor damage registration on stationary lights... but it's got to be contributing factor.
Edited by DaZur, 23 September 2013 - 05:51 AM.
#64
Posted 23 September 2013 - 05:53 AM
PEEFsmash, on 23 September 2013 - 04:44 AM, said:
These are from July. I said after the most recent patch, which still nobody has posted.
Last night with all the network issues, getting kicked from game, etc, hit registration went back to being terrible, not on lights in particular but on everything. Multiple shots on standing-still hunchbacks and victors, etc, so it looks like whatever HSR fixes they had done are no longer working perfectly because of the network.
Helen Keller mode engaged... You just can't admit that there is a problem. When someone provides evidence or that's not this patch that was last patch. The game didn't really change! Almost every patch for the past 6 months was a nerf to something, and added colors, mechs, or cockpit items! **** I bet HSR breaks again when they finally (if they ever) get DX11 in
#65
Posted 23 September 2013 - 05:54 AM
Quote
Quote
Edited by Mehlan, 23 September 2013 - 05:58 AM.
#66
Posted 23 September 2013 - 06:03 AM
Mehlan, on 23 September 2013 - 05:54 AM, said:
No you just can't face the music! When evidence is provided you throw it out. Regardless I will continue to make posts without video evidence because honestly why lie? I want the game to succeed not be Light Warrior Online! That is the direction it is going, and that is sad! I WANT ROLE WARFARE! Not March of the cows (assaults), or flood of the rabbits (lights). I want 4 Lights 4 Mediums 4 Assaults. So by all means out in tonnage limits, I don't care if the queue time goes up a bit!
Mehlan, on 23 September 2013 - 05:54 AM, said:
They so called fixed it patched it 9 times already. It still sucks!
Edited by Imperius, 23 September 2013 - 06:15 AM.
#67
Posted 23 September 2013 - 06:20 AM
Every vid I have see of "proof the lights are OP" have a few things going on with them.
1: no one holds their weapons on the light long enough to get the full damage.
2: People seam to think if they are shooting and the light hits a bump going one direction and the shooter hits a bump going the other and the weapon does not stay on target, they think they should get full damage for not holding weapons on target.
It all comes down to a few things.
Learn how your weapons work, and practice you aim.
A light can maneuver faster then a pilot with jumpy, overzealous aim, or wrong mouse config can hold weapons on target. BOO HOO. try different weapons and practice. Lights are FAR FORM OP!
Edited by Funkadelic Mayhem, 23 September 2013 - 06:26 AM.
#68
Posted 23 September 2013 - 06:31 AM
Andrew Porter, on 22 September 2013 - 12:00 PM, said:
As a some-time Jenner pilot, if I'm not dodging I'm losing bodyparts like there is no tomorrow, would love someone to prove me wrong.
But they wont. Predicted amount of evidence that will be submitted = 0, predicted levels of baseless QQ = off the scale.
As someone who flips back and forth between scouts and assaults, I've never had problems hitting Lights and the only time people have problems hitting me when I'm in one is when I'm jumping around like crazy TRYING NOT TO GET HIT.
I've never seen the one that people are claiming about a Light holding still (either overheated or whatever) and taking hits like an Atlas. I have seen a Light/Spiders overheat, stop, and then get obliterated in one salvo. This seems to suggest the problem, at least since the latest patch, is less HSR and more...aiming.
#69
Posted 23 September 2013 - 06:34 AM
Quote
Edited by Mehlan, 23 September 2013 - 06:53 AM.
#70
Posted 23 September 2013 - 06:41 AM
HSR is nothing more than a server-side prediction tool invoked to mitigate the issues of poor client connectivity issues.
Key word is "mitigate"... as it's neigh impossible to address all levels of line-quality deficiencies.
Not saying the current iteration is not flawed, nor am I saying PGI can't improve upon the present implementation... I'm saying the expectation for it to solve all connectivity-related issues as it pertains to HSR is grossly unrealistic.
#71
Posted 23 September 2013 - 06:59 AM
DaZur, on 23 September 2013 - 06:41 AM, said:
HSR is nothing more than a server-side prediction tool invoked to mitigate the issues of poor client connectivity issues.
Key word is "mitigate"... as it's neigh impossible to address all levels of line-quality deficiencies.
Not saying the current iteration is not flawed, nor am I saying PGI can't improve upon the present implementation... I'm saying the expectation for it to solve all connectivity-related issues as it pertains to HSR is grossly unrealistic.
Rubbish. HSR simulates client experiences server-side EXACTLY if you know the lag of each player to the ms. Lots of games have done this. The algorithms are well discussed.
There will always be people with problematic connections, but when it works, high frequency lag/ping polling gives the server enough data to detect hits reliably. Otherwise there is no point playing. Maybe you are saying that an internet game can't do server side hit detection reliably - in which case, don't even bother making such an internet game. This kind of game RELIES on perfect hit detection. If you can't provide it, don't bother making the game.
This should have been worked out in Closed Beta as a priority above all other concerns. If you can't do perfect hit detection, there is no Mechwarrior.
Edited by Xajorkith, 23 September 2013 - 07:11 AM.
#73
Posted 23 September 2013 - 07:13 AM
stjobe, on 23 September 2013 - 07:06 AM, said:
Hilarious.
LagWarrior
BugWarrior
Try going through a year and a half of this where every sign of progress is met, a month later, with regression and nonsense.
With 40 staff, I can't believe there is any more than 4 programmers. I have no idea what they are justifying their 8 hours a day on.
Then again, it's free, so I should stop complaining. I'll leave it to the Founder members for whom it wasn't free. I joined in Closed Beta but didn't invest.
Edited by Xajorkith, 23 September 2013 - 07:17 AM.
#74
Posted 23 September 2013 - 07:45 AM
Xajorkith, on 23 September 2013 - 06:59 AM, said:
There will always be people with problematic connections, but when it works, high frequency lag/ping polling gives the server enough data to detect hits reliably. Otherwise there is no point playing. Maybe you are saying that an internet game can't do server side hit detection reliably - in which case, don't even bother making such an internet game. This kind of game RELIES on perfect hit detection. If you can't provide it, don't bother making the game.
This should have been worked out in Closed Beta as a priority above all other concerns. If you can't do perfect hit detection, there is no Mechwarrior.
Rubbish is a bit harsh...
Problem is in order for HSR to work reliably.. i.e consistently, there as to be an acceptable tolerance window. Too narrow... and the prediction state is flooded, too wide and the CPU cycles necessary to maintain the wide disparity in the polar extremes becomes unreasonably intensive.
I'm guess PGI is in one of these two boats right now...
I agree it should have been addressed in closed beta but in fairness, they first attempted to solve it in real-time and didn't bring in HSR until much later.. That said, I do completely agree it needs to be on highest attention list as everything, in particular balancing efforts is dependent upon the accuracy of damage allocation.
Edited by DaZur, 23 September 2013 - 07:46 AM.
#75
Posted 23 September 2013 - 07:51 AM
Andrew Porter, on 22 September 2013 - 12:00 PM, said:
As a some-time Jenner pilot, if I'm not dodging I'm losing bodyparts like there is no tomorrow, would love someone to prove me wrong.
But they wont. Predicted amount of evidence that will be submitted = 0, predicted levels of baseless QQ = off the scale.
PEEFsmash, on 22 September 2013 - 11:42 AM, said:
For the sake of discussing gameplay balance, hit registration is a top priority. Many threads have suggested that light mechs are failing to register damage even when hit cleanly. I do not think this is a real issue as of the latest patch, but is rather the result of self-deception and misperception.
I am looking for video from you or someone else playing during the most recent patch where two or more shots fail to register on a light mech or fail to damage a light mech in the same match. This is said to be a serious issue that makes lights "unhittable", so if it is happening, providing a single match where this happens 2+times should be no problem. To simplify: 1 video of 1 match where there are 2 failed registrations on light mechs.
So, please post video evidence of this gameplay balance issue occurring. It would further our ability to properly discuss this important gameplay balance issue, and possibly substantiate or disarm the claims being made about light mechs.
I will fill this spoiler tag with all of the Video Evidence of this gameplay balance issue as it is posted:
Khobai, on 22 September 2013 - 12:19 PM, said:
SgtKinCaiD, on 23 September 2013 - 12:37 AM, said:
lol @ people like this. Its like reading somebody telling you apples don't exist even though I am eating one right now. Simply hilarious.
Yes there are hitbox issues on some mechs like the spider. It is a completely SEPARATE issue to damage not registering upon impact, that has to do with netcode and thus varies from time to time and person to person, the former is static design fault that affects everyone and the latter doesn't. and P.S netcode registration issues affect all mechs, its just the lights move so fast and are smaller you are more likely to miss the mech entirely as far as the server is concerned instead of a slightly different location on a slower, bigger assault mech.
I am not going to bother posting videos because last time I did that when we last we had this thread and people simply said those shots were missing even thought you could clearly see in the videos the shots hitting the mech (projectile stops, sparks etc).
I will say at the moment, registration is at its WORST. Almost 90% of shots on a moving target 80kph+ will simply not register. Sometimes leading works, sometimes it doesn't. Before HSR, leading ALWAYS worked. Now its completely inconsistent so little can be done besides building frustration.
The sad part is when PGI implement laser HSR it worked perfectly even for me on 250-300 ping. Ever since they did that bandwidth optimization about out 3-4 patches about it all is just getting worse and worse.
Edited by l33tworks, 23 September 2013 - 07:56 AM.
#76
Posted 23 September 2013 - 08:10 AM
Quote
Nope. Its called the Dunning-Kruger effect; players think theyre more skilled than they actually are. You're missing 90% of your shots because youre actually missing, not because you're hitting and the game isn't magically registering damage.
Because I have absolutely no problems hitting light mechs with most weapons, particularly lasers and ballistics. The only weapons that arnt registering hits properly are SRMs, but thats a known issue, and its already been mentioned.
And yes the Spider also has some weird hitbox issues going on right now, but its got nothing to do with hit registration, since its unique to the Spider. The Spider's hitboxes need to be redone, like how they redid the Raven's hitboxes; nobody has had trouble killing Ravens since then.
Edited by Khobai, 23 September 2013 - 08:14 AM.
#77
Posted 23 September 2013 - 08:10 AM
Video footage from today 2PM UK time.
I backed up my claim about hit detection issues.
inb4 excuses and hand waving away of unmistakeably clear video evidence.
/thread
#78
Posted 23 September 2013 - 08:15 AM
Funkadelic Mayhem, on 23 September 2013 - 06:20 AM, said:
Just watched this, wow... Don't know if any of them normally pilot lights, but none of them should.... That was probably the worst display of light mech piloting I have ever seen... And trust me, I've watched video of myself playing in light mechs, so I have seen some pretty bad piloting...
#79
Posted 23 September 2013 - 08:27 AM
There are several camps of thought to this issue, and most know that there's an issue.
1) People who see something NOT happening when they fire a shot , but can't do anything about it.
2) People who know they should have taken damage but didn't, and purposely continue to exploit that system.
3) People who continue to claim that nothing is wrong whom hope to keep it as is
4) People who don't know what's supposed to happen
#80
Posted 23 September 2013 - 08:30 AM
Khobai, on 23 September 2013 - 08:10 AM, said:
The Raven's hitboxes were never "redone." The only change is that their leg hitboxes now extend larger than the physical model. The side torsos still extend along the beak like they always have.
6 user(s) are reading this topic
0 members, 6 guests, 0 anonymous users