Spider Is Not Broken; Just Well-Designed
#81
Posted 23 October 2013 - 10:22 AM
#82
Posted 23 October 2013 - 10:29 AM
There exists some sort of computational hitbox + hitregistration + ping + HSR + etc. issue that is surrounding the unusual survivability of the Spider in the game as it sits today. Everyone knows it, everyone sees it. Mehlan, the sky really is blue, it cannot be purple as you claim it is. (thats a metaphor by the way, and no I am not quoting you in anyway, just paraphrasing the essence of your posts)
#83
Posted 23 October 2013 - 10:41 AM
Sable Dove, on 20 October 2013 - 05:23 PM, said:
The Spider benefits from convergence issues more than any other mech because it's so slim. There are situations where it is impossible to hit because of convergence. It's also probably a bit smaller than it should be (which causes some of the convergence issues).
It is better than it used to be though.
#84
Posted 23 October 2013 - 11:24 AM
mike29tw, on 22 October 2013 - 11:12 PM, said:
There are more than just "target not registering damage" to this problem. Random instance of damage not registering happens on all mechs, as we saw in the video. Lasers deal damage over 1 sec duration, so chances are even if there are bugged damage not registering, some damage will still register on the target during the whole second.
Then there's another problem where hitbox is not where you see on your screen due to latency or target running at extreme speed. Sometimes if you wanna hit a 150kph+ spider, you have to lead and shoot in front of him, sometimes you have to shoot behind him in order to actually register a hit. You won't notice this type of issue on slower mechs, ie heavy and assault class. Light mechs, and the most notorious spider, exploit this issue the most. That's why i mention medium lasers because if a full duration of lasers deal no damage on the spider, chances are its hitbox is shifted.
In the video being talked about here, what I saw was 20 seconds of damage being ignored, including at least a laser beam, all doing no damage at all. After about 21 seconds into the video, then damage started to stack onto the mech. However, that mech survived 4 other mechs shooting into it for 20 seconds with no damage being taken. This is evidence that, though the Spider may have some other problems, it isn't just the spider this is happening to. Maybe the Spider does have other problems, but we can't go fixing the possible problems with the Spider till we find out what the overall problem is effecting hit registry for every mech in the game.
The Wee Baby Seamus, on 23 October 2013 - 02:00 AM, said:
What you're experiencing here is the age old Server-Client synchronisation "issue".
In a nutshell: What you're seeing on your screen isn't necessarily what's happening in reality. Especially with the very high ping the MWO server location forces onto most players.
Talking about hit registration in this context is just completely misleading and simply wrong
It isn't necicarily high ping that is messing things up, it's inconsistent ping that is messing up HSR's compensation algorithms. It can't guess where a mech is based on it's ping, when that ping jumps from 50-100 in a few seconds time.
I know personally, I run between 30 and 32 ping. I have never (seemed) to have benefits from the lag shield before HSR. Now, when HSR first came in, my Spider (and other fast mechs) died a easier than before, but it didn't chance all that much for me. Even now, even in a Spider, I still can (and often do) die very easily and quickly in a light mech.
There is a problem that PGI itself has acknowledged themselves that hit registration is not functioning correctly. They are working on it. Any other problems that the Spider, or any other mech for that matter, might have can't be found and solved till the hit registration and HSR problems are fixed. If they change it now, it will be like the PPC fix, if you can recall that...
Also, to point out, the videos being presented here is played on the training grounds, as far as I can tell. Training grounds is preformed strictly and completely on your own computer, so there should be no HSR, ping, or "Server-Client Synchronization Issues". They are run on the host computer (your own) with no ping and lacking any ping associated problems. This leads me to personally think it is a hit registry problem, on more than just the Spider (was reproduced on a Jenner as well).
Mehlan, on 23 October 2013 - 01:55 AM, said:
Yea and I ran my testing ground vids after that....
try again....
aniviron, on 23 October 2013 - 01:34 AM, said:
Contrary to what the post above me states (and I have told the poster this several times) you can see that this video is from Oct 17, after the most recent patch; hardly one of the "old couple [of] vids." Even though I've showed it to him at least twice, he has been having a hard time believing it.
As for the lower example, to respond to your video (again), missiles, especially SRMs right now, are very well known to have hit registry problems. Using SRMs to test a hit registry issue only helps to point out the problems with missiles registering their own damage, as much as it could indicate a problem with the Spider as well. ACs still have some of their hit registry problems as well as far as I know.
I would just like to clarify that I am not saying that this might not be a Spider problem, but we can't actually tell till hit registry is fixed better than it currently is, as well as HSR too. Have to fix what is effecting all mechs to see what might be wrong with a single mech. (SSRMs used againt Jenner should help prove the point.)
Phromethius, on 23 October 2013 - 10:29 AM, said:
There exists some sort of computational hitbox + hitregistration + ping + HSR + etc. issue that is surrounding the unusual survivability of the Spider in the game as it sits today. Everyone knows it, everyone sees it. Mehlan, the sky really is blue, it cannot be purple as you claim it is. (thats a metaphor by the way, and no I am not quoting you in anyway, just paraphrasing the essence of your posts)
To counter your own point, he has provided his own video evidence to counter the point. There is a larger problem that effects all mechs, but seems to settle most on the Spider. This leads me to conclude that this isn't a spider problem, or at least not completely. We can't truly see if something is wrong with the spider till what is effecting all mechs is fixed. Then we can farther test the issues that might still remain with the Spider.
So, yes, the spider might have a problem in itself. On the other hand, it might be something else. Jumping too quickly on "fixing" the spider now might result in having to revert the "fix" later, wasting time and money on "fixes" that might not be needed. We don't have all the data, so we can only speculate what might be the real issue behind all the problems.
Is it just a problem with only the spider? Maybe. Could it be a larger problem that effects every mech in the game, and not a spider problem? Maybe. Is it a combo effect of a problem with the Spider and a larger problem that effects all mechs? I'm inclined to believe this is more likely.
#85
Posted 23 October 2013 - 12:05 PM
Bront, on 23 October 2013 - 10:22 AM, said:
I linked to the source of that thread here already. It's quoted in the post above. /\
#86
Posted 23 October 2013 - 12:06 PM
aniviron, on 23 October 2013 - 04:31 AM, said:
I used SRM2 exclusively because it's the easiest to show it with; all the other damage on the mech was from the SRM2 impacting as I looked for other holes. You will note I am down a number of shots when the video starts. But just because I love making you happy, I went back to the testing grounds to record more. Attached is a video of about half my shots with an AC2 finding the hole in the spider's hitbox. Like I've explained to you before, I tend to use SRM2 as the hits seem to be more precise, and can more reliably be landed in the deadzone. As of this time, I was unable to replicate any 0 damage hits with a medium laser.
68 damage fired.
28 armour (CT+RT)
34 IS (CT+RT)
62 damage taken.
Difference of 6 damage. Which seems like a fair bit (~9%), but considering quite a few shots hit the destroyed RT, dealing half damage, it really doesn't prove much other than sometimes hits do not display damage properly.
Vodrin Thales, on 23 October 2013 - 10:18 AM, said:
All lights were impossible to hit pre-HSR. The Raven was singled out because it had Streaks and ECM (before BAP countered it), and still went as fast as any other light. Only competition for the Raven-3L was the COM-2D, and it was outclassed simply because the Raven was heavier, with more armour and weapons.
Back in those days, it was possible for a Spider to solo two assaults at the same time.
Edited by Sable Dove, 23 October 2013 - 12:10 PM.
#87
Posted 23 October 2013 - 12:18 PM
Quote
There exists some sort of computational hitbox + hitregistration + ping + HSR + etc. issue that is surrounding the unusual survivability of the Spider in the game as it sits today. Everyone knows it, everyone sees it. Mehlan, the sky really is blue, it cannot be purple as you claim it is. (thats a metaphor by the way, and no I am not quoting you in anyway, just paraphrasing the essence of your posts)
Quote
#89
Posted 23 October 2013 - 05:20 PM
People with packet loss and unstable ping were getting disconnected because of HSR. Those are the people who have the most trouble with leading targets and who also warp on your screen.
http://mwomercs.com/...longer-working/
So again, it's not a spider issue, is't a "people have ****** connections" issue.
I think this quote by Mackman is especially relevant.
Quote
And how is it "unfair"? It's been a fact of online gaming since online gaming first came into existence. And there's literally no solution other than to artifically make everybody else's experience worse.
#90
Posted 23 October 2013 - 05:41 PM
Odins Fist, on 22 October 2013 - 10:12 PM, said:
Like I said there is no consistency....
You don't know how many times I have heard similar things from so many different people.
The lack of consistency is because it's not a spider problem, it's a connection problem. Targets you are shooting at that have good stable ping will be just as easy to hit as anything else, no magic involved.
It's the guys with bad connections that make it hard for you to know where they actually are.
mike29tw, on 22 October 2013 - 11:12 PM, said:
Then there's another problem where hitbox is not where you see on your screen due to latency or target running at extreme speed. Sometimes if you wanna hit a 150kph+ spider, you have to lead and shoot in front of him, sometimes you have to shoot behind him in order to actually register a hit. You won't notice this type of issue on slower mechs, ie heavy and assault class. Light mechs, and the most notorious spider, exploit this issue the most. That's why i mention medium lasers because if a full duration of lasers deal no damage on the spider, chances are its hitbox is shifted.
People keep saying "hitbox" but in reality the mech and hitbox are together on the server, they are BOTH not where you see them on your screen because your screen is off by your ping times 2.
Also, I think you always want to lead in front of your target. I think the only case where you would have to lead behind them was if they slowed down quickly and is probably the issue people have with hitting shutdown mechs. The mech is behind where you are shooting because your client hasn't been updated on where it actually stopped moving, your client is just guessing. That is why when the mech starts back up it warps forward.
People with bad connections would have to test this though since with my connection this effect is very mild.
#91
Posted 23 October 2013 - 06:03 PM
AC/2 Edition
Laser Edition
The spiders CT seems to replicate that of the ct displayed in the paper doll... it's a narrow line along the center flares a tad below the cockpit up to the head. If you want to hit the CT the best area is to aim is the crotch or just below the cockpit. If you look at Ani's or my video...you can trace what is CT vs LT. You'll note that each shots hits and registers.
Edited by Mehlan, 23 October 2013 - 06:41 PM.
#92
Posted 23 October 2013 - 06:12 PM
No other mech in the game would get away with that," the spider is fine" pffffft.
#93
Posted 23 October 2013 - 06:30 PM
KHETTI, on 23 October 2013 - 06:12 PM, said:
No other mech in the game would get away with that," the spider is fine" pffffft.
I hate to point this out... but someone beat you to the punch...
stjobe, on 22 October 2013 - 06:01 AM, said:
That's no Spider.
That's a Cataphrat, ignoring 20 seconds of damage. From several people. From all weapon types as far as I can see.
#94
Posted 23 October 2013 - 06:32 PM
KHETTI, on 23 October 2013 - 06:12 PM, said:
No other mech in the game would get away with that," the spider is fine" pffffft.
Really, gee what does PGI have to say about this?
Quote
CERES METALS AAlcadis Revised Underground Complex, B5
St.Ives
Friday October 4th 3050 22:47
BUG: We have finally reproduced and idenified an issue that has to do with our projectile hit detection. This causes some bullets to pass through targets on the server or in some cases pass a single component and hit front/rear instead. I estimate it happens about 1% overall and depends on several factors.
Ya'll get that?...
Quote
p.s. Sorry Khet, not targetting you specific with that...just the quote was convenient.
Edited by Mehlan, 23 October 2013 - 06:33 PM.
#95
Posted 23 October 2013 - 06:57 PM
Roughneck45, on 21 October 2013 - 08:12 AM, said:
People miss their shots against them more than anything else, then whine about hit reg. Sure, the hit reg needs a lot of work, but the spider is just a difficult target to land shots on. It's really the only advantage it has.
thats absolutely true, trying to hit something going 200 plus and add the crappy video clipping and lag.. you dang right, its impossible to hit the damn things. Which begs the question, why did PGI put something into the game that the graphics engine can not even handle right?
what where they thinking..???
oh right, they whernt..
Edited by Crazy Billy Joe Bob, 23 October 2013 - 06:59 PM.
#96
Posted 23 October 2013 - 07:06 PM
Every time I see a Spider now this is all I can think of ^
#97
Posted 23 October 2013 - 07:11 PM
Quote
CERES METALS AAlcadis Revised Underground Complex, B5
St.Ives
Friday October 4th 3050 22:47
BUG: We have finally reproduced and idenified an issue that has to do with our projectile hit detection. This causes some bullets to pass through targets on the server or in some cases pass a single component and hit front/rear instead. I estimate it happens about 1% overall and depends on several factors.
Combine this known problem, with the thinness and leanness of the Spider, well. Can we now see why they are being effected more than other mechs? They are the thinnest (in all ways possible) mechs in the game.
Crazy Billy Joe Bob, on 23 October 2013 - 06:57 PM, said:
thats absolutely true, trying to hit something going 200 plus and add the crappy video clipping and lag.. you dang right, its impossible to hit the damn things. Which begs the question, why did PGI put something into the game that the graphics engine can not even handle right?
what where they thinking..???
oh right, they whernt..
This isn't a "graphics" problem, but a "lag", "instable pin", "hit registry" issue. It's across all mechs, including the spider.
#98
Posted 23 October 2013 - 07:33 PM
#99
Posted 24 October 2013 - 10:22 PM
The problem is that they think they hit because their client told them they hit.
They still haven't learned that this is a server authoritative game.
End of story.
#100
Posted 24 October 2013 - 10:50 PM
Green Mamba, on 21 October 2013 - 08:38 PM, said:
Most of the light pilots here (and in general) pilot Jenners. Jenners are still considered superior to Spiders by quite a margin.
17 user(s) are reading this topic
0 members, 17 guests, 0 anonymous users