58 replies to this topic

[aniviron]

Mehlan, on 23 October 2013 - 12:20 AM, said:

they're videos, system it just having problems with the links.... the 'hole' your referring to, is based on a video that I believe is well over a month old. In fact, we've already covered this before in another thread...

Yes, and when you said I posted a video that was over a month old, I dropped into the training grounds five minutes after you said that and was able to recreate the bug within 30 seconds, e.g. it has not been patched and is still an issue.

[Mehlan]

aniviron, on 23 October 2013 - 12:32 AM, said:

Yes, and when you said I posted a video that was over a month old, I dropped into the training grounds five minutes after you said that and was able to recreate the bug within 30 seconds, e.g. it has not been patched and is still an issue.

I know you did, after which I went and made Which you've ignored. Went and hit it with lasers and ac/20 again this morning too.

Edited by Mehlan, 23 October 2013 - 03:47 AM.

[aniviron]

Mehlan, on 23 October 2013 - 03:46 AM, said:

Went and hit it with lasers and ac/20 again this morning too.

It's great that you can hit it, but you're also 1: not aiming at the deadzone, and 2: positive hit indications some of the time do not indicate that the mech doesn't have problems hitting most of the time. It'd be as though I said, "I slipped on the icy sidewalk; it's dangerous," to which you replied, "Well I've never slipped there, so because I can walk across it right now just fine it is always fine."

[Mehlan]

aniviron, on 23 October 2013 - 04:33 AM, said:

It's great that you can hit it, but you're also 1: not aiming at the deadzone, and 2: positive hit indications some of the time do not indicate that the mech doesn't have problems hitting most of the time. It'd be as though I said, "I slipped on the icy sidewalk; it's dangerous," to which you replied, "Well I've never slipped there, so because I can walk across it right now just fine it is always fine."

Lets work with your 'example'... My reply would 'be careful, it's icy....' You'd have them tear up the sidewalk to remove the 'danger', I'd get a torch and melt it or coat it with sand. The sidewalk is fine, it's the ice on it that's the hazard.
You own 'vids' are not consistent in dead zone either, nor are they with the original 'hole' video...which should tell you something.

[Mavek]

stjobe, on 22 October 2013 - 07:05 AM, said:

And nobody's asking for them to be able to stand still in front of a heavy or assault and trade blows (and win!). But look at what Bishop said:


He doesn't want a light to EVER be able to win a 1v1 against anything even 20 tons heavier. Not "brawling", not "flanking", not hit-and-run - NEVER.

And that's just wrong.


Actually, it's me that's wrong. Reading comprehension FTW, right? Right. Sorry Bishop.

I think that when Bishop said a light should 'NEVER' be able to solo a 20+ ton larger mech, he mispoke.

A GOOD light pilot should be able to smoke a BAD assault pilot, MOST of the time, etc. But given equal pilots and a loadout in the Med/Heavy/Assault that is not: 2 PPCs, 2 LRM15s, and 2 Small Lasers, the Med/Heavy/Assault should win against a light mech MOST of the time. And to that point I think this is true in the outcomes of most matches, and I do not see a large issue in the game right now with GOOD players with GOOD builds in untouched mechs getting owned by ANY light mech regularly. When it does happen, it is almost assuredly to a VERY GOOD light pilot that knows what he has and how to use it.

But what we do see, a great many of us, is our screen telling us we hit that damned spider, but the spider's paper doll is laughing at us in return, as it blinks away with no change in color or the expected missing parts. Even when the spider is shut down and/or immobile for any reason, AC20 rounds often bounce right off the spider like it's a battleship.

I will say tho, to Steiner's point, there seem to be no shortage of average to bad spider pilots that like to gloat after having a good game against bad pugs, and if the Spider wasn't broken, those Spider pilots would not be having those games and would either be driving something else like a Centurian with a broken hit box, or would not be playing MWO, instead finding some other game to exploit and troll.

Edited by Mavek, 23 October 2013 - 12:02 PM.

[Corwin Vickers]

And I would say it's a connection problem, not a "broken" spider problem.

Also, not landing shots where you want them to land occurs at exactly the same rate and distance on spiders as it does on other mechs. The difference is if you are aiming at the right torso of a highlander and instead hit his arm you don't notice it. If you make that same shot against the spiders it means you totally miss.

Two players with sub 100 ping and good computers will not notice this problem against each other, SRMs not being counted. I have stopped using SRMs because in large numbers (the amounts you generally want to use them for) they do not register their full damage right now. I have had better luck chain firing them but I don't want to chain fire them so instead they just don't get brought.

PLEASE STOP USING THE MOST BROKEN WEAPON TO PROVE DAMAGE REGISTRATION PROBLEMS.

[Mehlan]

Tests 9:21 Cst, 10/23/2013


AC/2 Edition

Laser Edition



Lets not forget this little tidbit too...

Quote

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

#13

Edited by Mehlan, 23 October 2013 - 06:42 PM.

[JKohn]

Just came from a game where I hit that little **** multiple times without anything happening to it. Though I noticed that it took damage from medium-long range just fine, but only took damage from SRM's in short range (no streaks), pretty much ignored ERPPC and LBX impacts. I have a lag of 150ms so that shouldnt be much of a problem, and I hit just fine locusts, commandos and jenners running around me.

[Mehlan]

JKohn, on 25 October 2013 - 07:18 AM, said:

Just came from a game where I hit that little **** multiple times without anything happening to it. Though I noticed that it took damage from medium-long range just fine, but only took damage from SRM's in short range (no streaks), pretty much ignored ERPPC and LBX impacts. I have a lag of 150ms so that shouldnt be much of a problem, and I hit just fine locusts, commandos and jenners running around me.

read Corvin's post in regards to latency, then you might check your connection then next time you see it happen.

[Bishop Steiner]

Mehlan, on 25 October 2013 - 07:23 AM, said:

read Corvin's post in regards to latency, then you might check your connection then next time you see it happen.

It's a nice thought, but since I check my latency pretty constantly, and have 0% loss, and an "official" ping considerably under 100, not the issue, at least in my case. I am certain it can be ONE of the contributing factors, but the simple truth is, there IS a hit registration issue, with lights period, to a degree, but for multiple reasons apparently converging in one asset, the Spider is the worst offender of it.

[Mehlan]

Bishop Steiner, on 25 October 2013 - 08:55 AM, said:

It's a nice thought, but since I check my latency pretty constantly, and have 0% loss, and an "official" ping considerably under 100, not the issue, at least in my case. I am certain it can be ONE of the contributing factors, but the simple truth is, there IS a hit registration issue, with lights period, to a degree, but for multiple reasons apparently converging in one asset, the Spider is the worst offender of it.

In this case I was referring to Jkon.. :-) I ask in part, just for data collection Bishop.

[Christof Romulus]

wintersborn, on 21 October 2013 - 07:10 AM, said:

What I find to be moronic is that PGI devs said there is no problem with spiders.
The mechs is so broken that it should be disabled until they fix it, since it is exploiting.

Actually...

I didn't want to admit this, but since I don't believe in giving PGI flack for things that aren't actually true I will - As an ATLAS, I have been struck by shots that have dealt no damage what-so-ever. I have no Spiders, but in my AS7-D, AS7-D-DC, and AS7-RS I have been struck by many weapons that have done no damage to me at ALL. Literally last night I was struck with an Gauss Rifle point blank and I remained at 100%.

Also, recently, there was an issue with the Atlas' head hitbox and "Certain Weapons" - which implies one of two things:
1. That weapons have different hit registration based on the WEAPON and not the HITBOX.
or
2. Every hitbox has a different shape depending on each weapon that strikes it.

What I believe PGI said, without saying it, is "The hitbox of the Spider is not bugged, weapon hit registration is."

[Corwin Vickers]

Christof Romulus, on 25 October 2013 - 11:35 AM, said:

What I believe PGI said, without saying it, is "The hitbox of the Spider is not bugged, weapon hit registration is."

And that's all some of us have been saying and we get called exploiters, liars, blind. The guys I play with tell me I don't even drop in spiders more than 10% of the time.

I post in these threads because I saw what dipshit whiners did to LRMs.

[Greyboots]

First all, let me say that I am NOT denying there's a problem with spiders that needs to be fixed. Identified problems should be fixed at the earliest convenience. I will also admit to owning a spider but that doesn't prevent me from experiencing the exact same situations when I am killing them from a spider or any another mech.

That being said, I kill spiders all the time. I kill them with MGs, LB10x's, LRMs, SRMs and SSRMs. Weapons which circumvent the issues described here by either being self guided, spray an area or give me a nice beam that I can guide manually if needed. Spiders are FAR from unkillable and I believe the OP expressly using ACs as an example is somewhat misleading (but also enlightening, showing us to some extent how ACs are currently ruling MWO But I digress.

Even if this is all fixed? I still will. The weapon convergence issue is unlikely to go away and, although I dare say many will disagree, I don't think it should. Player choice is an important element in any game of this kind and I don't see a problem in it being a choice for a player in having to choose weapons that will overcome the convergence issue and make it a lot easier to kill Spiders.

People are killing spiders. This is simple fact. Yes, it's really hard to do with twin AC20 jagers and such but it's far from impossible with a vast array of other weapons. IMO? Fix the hit registration issues and leave everything else alone. If people want to kill spiders so much? Let them put on the right weapons for the job.

[Amsro]

Spiders benefit from perfect convergence, and not being able to lead them properly.

Hit Detection is gimped for ALL mechs, either HSR is flawed or PGI is dropping packets server side.

Either way I don't see this as a spider only issue. Hit detection is flawed all around.

[Bishop Steiner]

Amsro, on 26 October 2013 - 08:34 AM, said:

Spiders benefit from perfect convergence, and not being able to lead them properly.

Hit Detection is gimped for ALL mechs, either HSR is flawed or PGI is dropping packets server side.

Either way I don't see this as a spider only issue. Hit detection is flawed all around.

No, it's not Spider only, the spider just happens to be the flagship for it, as combining the HSR issues with convergence, with known hitbox issues, stupidly tiny size makes it the epitome of HSR headaches. Light Mech piloting to be effective, it ain't easy. And yet we see how many Spiders per match theses days? Considering all the "Pros" in particular seem to flock to whatever is easiest to exploit, I don't think it coincidence.

Let's just say Spider jocks KDr should have a nice giant asterisks next to it.

Edited by Bishop Steiner, 26 October 2013 - 09:13 AM.

[Sephlock]

Bishop Steiner, on 21 October 2013 - 05:47 AM, said:

seems to explain some of the issues.

Problem is, as many can attest, the issues with Spiders registering hits seem to be legion. Similar to the comments at the bottom of his post, I cannot count the number of times I have lined up a short range shot on the torso, or even leg of a stationary (or even shutdown, and yes, an AFK one too) Spider with my JagerBomb, touched off a 40pt alpha and achieved 0 damage.

So it appears we have dead zones on the Spider'***** boxes themselves, convergence issues that are variable and not fixed, various HSR issues, and more.

And yet they decided to buff the Spider with the other Lights this last pass, instead of prepping the fixes, and then holding off on the SDR until they have resolved some of the issues. Grand. Because 1v1 it makes so much sense for Heavy and Assault mechs to regularly get solo'd by a 30 ton scout. (Sorry kiddies, without a pack, or some seriously drawn out hit and run tactics, a Light Mech should NEVER be able to solo a Mech 20+ tons bigger. Otherwise it makes no sense, to spend 4-10x as much building larger mechs. Lights are scouts and flankers, not killers)

And just to make it more fun, and probably harder to diagnose, the issues seem highly inconsistent, as there are days where I can tag Spiders mid run just fine, and then others you can't damage them at all. It's highly not fun for people to have to deal with this, especially since so many of the Spider pilots out there act like it's their skill that is making the difference, instead of the fact they are just exploiting broken game mechanics, much like the Raven-3L pilots of old.

Could it be a combination of factors?

As in: not only are there problems with convergence, but also with how the shots appear to converge on your screen vs how they actually converge on the server side of things, with spiders just being narrow enough for it to make a difference whereas with most mechs the disparity generally isn't big enough for people to notice?

At the very least that would provide a (IMHO) more plausible explanation for the "dead zone" phenomenon than just "there are parts that don't take any damage".

There may be spots on the Spider's narrow body where your side of things renders your shots so that it looks like they are converging on the mech, but a relatively small discrepancy due to latency causes the server to say NOPE YOUR SHOTS DIDN'T CONVERGE THAT WAY!.

[Bishop Steiner]

Sephlock, on 27 October 2013 - 02:39 AM, said:

Could it be a combination of factors?

As in: not only are there problems with convergence, but also with how the shots appear to converge on your screen vs how they actually converge on the server side of things, with spiders just being narrow enough for it to make a difference whereas with most mechs the disparity generally isn't big enough for people to notice?

At the very least that would provide a (IMHO) more plausible explanation for the "dead zone" phenomenon than just "there are parts that don't take any damage".

There may be spots on the Spider's narrow body where your side of things renders your shots so that it looks like they are converging on the mech, but a relatively small discrepancy due to latency causes the server to say NOPE YOUR SHOTS DIDN'T CONVERGE THAT WAY!.

Oh there certainly could be, though the whole point of HSR was to eliminate that. But I don't think HSR has been the cure all PGI hoped for, and there are probably a whole host of small things combining to cause a much larger headache.

[Corwin Vickers]

They were pushing more packets for HSR but the people with bad connections, the ones HSR was supposed to help, were getting disconnected by HSR so they turned HSR down.

Queue today.

http://mwomercs.com/...longer-working/

http://mwomercs.com/...nal-experience/