43 replies to this topic

[Phaesphoros]

TexAss, on 15 March 2013 - 03:42 AM, said:

Now don't do the test on an Atlas, do it against a commando or spider and then see what will happen.
It should not be just doubled (which would be ok) but trippled or up to 5-7 times as high. That's the bug everyone is talking about.

This is the bug Amaris has investigated. I'm talking about the bug/issue the OP Valore mentions.

Quote

Paul said that damage is doubled in testing grounds so your recorded damage is just slightly higher.

I can confirm this is not true. Stock JR7 has 8 armor on its arms. Shoot 1 SRM2 (5 dmg) + 1 SL (3 dmg) cores it w/o touching the internals, as expected (there's additional splash dmg to side torso and leg, that's what Amaris has observed). Dmg is not just doubled on testing grounds; though many of the stock mechs don't have maxed armor (that's why they die so fast).

[3rdworld]

Here is a 1 second video of a full srm alpha doing nothing to an atlas. you can see my targeting reticle does not even register a hit.

Edited by 3rdworld, 15 March 2013 - 04:18 AM.

[Nightcrept]

Phaesphoros, on 15 March 2013 - 04:06 AM, said:

This is the bug Amaris has investigated. I'm talking about the bug/issue the OP Valore mentions.

I can confirm this is not true. Stock JR7 has 8 armor on its arms. Shoot 1 SRM2 (5 dmg) + 1 SL (3 dmg) cores it w/o touching the internals, as expected (there's additional splash dmg to side torso and leg, that's what Amaris has observed). Dmg is not just doubled on testing grounds; though many of the stock mechs don't have maxed armor (that's why they die so fast).

The bug is supposed to act differently on the four original mechs the system was balanced against. Jenner and atlas are two of them.

3rdworld, on 15 March 2013 - 04:08 AM, said:

Here is a 1 second video of a full srm alpha doing nothing to an atlas.  you can see my targeting ridicule does not even register a hit.

Be kinda poetic if in their attempts to fix their broken mechanics they break the whole game.

[3rdworld]

Nightcrept, on 15 March 2013 - 04:09 AM, said:

Be kinda poetic if in their attempts to fix their broken mechanics they break the whole game.

Here is another one, the atlas is standing nearly still and I am right in front of him. Full alpha, no damage. He is currently being mauled but you can see my reticle doesn't turn red. And yes, this was the same game

Edited by 3rdworld, 15 March 2013 - 04:20 AM.

[Phaesphoros]

3rdworld, on 15 March 2013 - 04:08 AM, said:

Here is a 1 second video of a full srm alpha doing nothing to an atlas. you can see my targeting ridicule does not even register a hit.

What your client shows you (graphics, missile flight paths, explosions etc.) is not accurate until Host State Rewind for missiles is introduced. Meanwhile, your ping is a factor. That is, it is possible you missed the Atlas actually: When the server receives your "trigger SRM launch" message, the Atlas as well as you have already moved somewhat (lag shield). Without paper doll and your ping, this video does not proof anything (note it could as well be a bug of the crosshair hit indicator).

Nightcrept, on 15 March 2013 - 04:09 AM, said:

The bug is supposed to act differently on the four original mechs the system was balanced against. Jenner and atlas are two of them.

So what? There are two different issues 1) excess splash damage (has to do with hitboxes) as described by Amaris and 2) chain fire vs. alpha dmg as described by Valore. I can confirm 1), but my test does not show any effect similar to 2).

Edited by Phaesphoros, 15 March 2013 - 04:21 AM.

[3rdworld]

Phaesphoros, on 15 March 2013 - 04:20 AM, said:

What your client shows you (graphics, missile flight paths, explosions etc.) is not accurate until Host State Rewind for missiles is introduced. Meanwhile, your ping is a factor. That is, it is possible you missed the Atlas actually: When the server receives your "trigger SRM launch" message, the Atlas as well as you have already moved somewhat (lag shield). Without paper doll and your ping, this video does not proof anything (note it could as well be a bug of the crosshair hit indicator).

Thought someone would say that, check the next vid, the atlas is pretty much standing still, and I am directly in-front of him. there is no way none of those missiles did not hit that atlas.

[MaddMaxx]

The Training Grounds damage read-out is borked. Paul mentioned that as well.

Do not use the Training grounds as a DAMAGE meter and then come back here and WHINE! or QQ! Please.

There is an issue with Splash damage application though, way to much, and they are correcting.

[Phaesphoros]

3rdworld, on 15 March 2013 - 04:23 AM, said:

Thought someone would say that, check the next vid, the atlas is pretty much standing still, and I am directly in-front of him. there is no way none of those missiles did not hit that atlas.

Yes, your crosshair does not turn red. Other than that, I can't really see what's happening, as the two AWS seem to shoot SRMs at the AS, too (both). The AS CT seems to be orange at the beginning, after some time it's cored orange. Full CT armor might be about 100, internal health is 63. Though I don't know what the colors mean precisely, I think it's reasonable to assume that about 80 dmg has been done to CT: let's assume orange = 50 %, then 50 % armor (50 dmg) + 50 % internal (30 dmg) = 80 dmg.

Edited by Phaesphoros, 15 March 2013 - 04:59 AM.

[3rdworld]

Phaesphoros, on 15 March 2013 - 04:58 AM, said:

Yes, your crosshair does not turn red. Other than that, I can't really see what's happening, as the two AWS seem to shoot SRMs at the AS, too (both). The AS CT seems to be orange at the beginning, after some time it's cored orange. Full CT armor might be about 100, internal health is 63. Though I don't know what the colors mean precisely, I think it's reasonable to assume that about 80 dmg has been done to CT: let's assume orange = 50 %, then 50 % armor (50 dmg) + 50 % internal (30 dmg) = 80 dmg.

You can see an SRM volley from another splat to my lower left shoot just before I do. If it only took 80 dmg to take him to that level, then the 2 splat volleys should have been more than enough to core the atlas.

And I don't really get why you are trying to disprove this. Anyone that has ever used SRMs has noticed this. It isn't isolated and happens a couple of times a match if you use your SRMs consistently.

Heck our 8 man team has thought about not using them because of the terrible hit detection, but figure the dmg/ton is too good to not use. It completely absurd for a shooting game to have this much trouble detecting hits.

I mean it isn't like a atlas backing straight in front of me, out ran the SRMs. I normally have around a 50-70ms ping. That atlas would have needed a half second or more to move out of the way of that volley.

It is bad detection, case and point.

Edited by 3rdworld, 15 March 2013 - 05:07 AM.

[Tremendous Upside]

3rdworld, on 15 March 2013 - 05:02 AM, said:

You can see an SRM volley from another splat to my lower left shoot just before I do. If it only took 80 dmg to take him to that level, then the 2 splat volleys should have been more than enough to core the atlas.

And I don't really get why you are trying to disprove this. Anyone that has ever used SRMs has noticed this. It isn't isolated and happens a couple of times a match if you use your SRMs consistently.

Heck our 8 man team has thought about not using them because of the terrible hit detection, but figure the dmg/ton is too good to not use. It completely absurd for a shooting game to have this much trouble detecting hits.

I mean it isn't like a atlas backing straight in front of me, out ran the SRMs. I normally have around a 50-70ms ping. That atlas would have needed a half second or more to move out of the way of that volley.

It is bad detection, case and point.

I've had the same thing happen any number of times over the past few weeks - and it seems worse since the last patch. Had one game the other night where I nailed a trebuchet at least 6-7 times in a row flush (with 6xSRM6's) and it did next to nothing - and at least a couple of the shots hit while it was standing still/shut down. I had the rest of my team spectating and commenting about it... Happened in the SJR / DV8 matches last night and anyone that was watching the stream saw it. One of the players lined up a clean shot and appeared to have hit the target flush - and no damage registered from the hit.

[Phaesphoros]

3rdworld, on 15 March 2013 - 05:02 AM, said:

You can see an SRM volley from another splat to my lower left shoot just before I do.

Ok, I'll analyze this vid a bit more accurately. Events I can identify:

• Frame 3-6 (0.1 - 0.2 s): first sparks of an explosion at the AS' RA. There also seems to be a PPC flash at the back side of the arm.
  
• Frame 11 (0.37 s): the AS' paperdolls stops flashing.
  
• Frame 27 (0.9 s): AS paperdoll RA begins to flash
  
• Frame 38 (1.3 s): left AWS shoots something, presumably SRMs.
  
• Frame 42 (1.4 s): the SRMs of the left AWS begin to hit the AS' RA.
  
• Frame 42 (1.4 s): Some missiles, presumably streaks, enter at bottom of screen.
  
• Frame 45 (1.5 s): AS' paperdoll's CT, head, LT begin to flash.
  
• Frame 46 (1.5 s): the streaks hit the AS' CT.
  
• Frame 47 (1.6 s): your heat jumps to 77 % and the SRM cooldown counter starts (of all SRMs).
  
• Frame 49 (1.6 s): two volleys of missiles, presumably your SRMs, enter the screen.
  
• Frame 50+51 (1.7 s): two volleys of missiles without exhaust o.O
  
• Frame 51 (1.7 s): the SRM volleys hit the AS
  
• Frame 51 (1.7 s): The AWS to your right seems to be hit by something at its RA (sparks).
  
• Frame 51 (1.7 s): Further streaks enter from bottom of screen.
  
• Frame 54 (1.8 s): The AS' paperdolls' RT, LA and both legs begin to flash.
  
• Frame 59/60 (2.0 s): Further streaks enter from bottom of screen.
  
• Frame 73 (2.43 s): The AS' paperdoll's RA stops flashing.
  
• Frame 91 (3.0 s): The AS' paperdolls' CT is cored and orange, head, CT, LT stop flashing.
  
• Frame 99 (3.3 s): The AS' paperdolls' RT, LL, RL stop flashing.
  
• Frame 100 (3.3 s): The AS' paperdolls' LA stops flashing.
  
• Frame 128 (4.27 s): Some missiles, presumably LRMs, enter at the top right of the screen.
  
• Frame 133 (4.44 s): The AS' paperdolls' head, CT, LL, LT, LA begin to flash.
  
• Frame 133-136 (4.5 s): Explosions from LRMs at AS LT and CT.

As far as I see, every part of the paperdoll that flashes up, does exactly 3 flashes.

Maybe I'll add some interpretation later on.

3rdworld, on 15 March 2013 - 05:02 AM, said:

And I don't really get why you are trying to disprove this.

Please don't take this personally. This is just how science works (you cannot prove anything, only falsify). It's just there are so many events, hard to distinguish, in this video.
I don't even deny there is an issue. I just claim you cannot prove or show it clearly with this video.

Edit: Btw. I can barely read the AS' % health unfortunately. This would add a very important indicator to the paperdoll and visual events.

Edited by Phaesphoros, 15 March 2013 - 06:10 AM.

[Sheraf]

Phaesphoros, on 15 March 2013 - 06:03 AM, said:

Ok, I'll analyze this vid a bit more accurately. Events I can identify:

• Frame 3-6 (0.1 - 0.2 s): first sparks of an explosion at the AS' RA. There also seems to be a PPC flash at the back side of the arm.
  
• Frame 11 (0.37 s): the AS' paperdolls stops flashing.
  
• Frame 27 (0.9 s): AS paperdoll RA begins to flash
  
• Frame 38 (1.3 s): left AWS shoots something, presumably SRMs.
  
• Frame 42 (1.4 s): the SRMs of the left AWS begin to hit the AS' RA.
  
• Frame 42 (1.4 s): Some missiles, presumably streaks, enter at bottom of screen.
  
• Frame 45 (1.5 s): AS' paperdoll's CT, head, LT begin to flash.
  
• Frame 46 (1.5 s): the streaks hit the AS' CT.
  
• Frame 47 (1.6 s): your heat jumps to 77 % and the SRM cooldown counter starts (of all SRMs).
  
• Frame 49 (1.6 s): two volleys of missiles, presumably your SRMs, enter the screen.
  
• Frame 50+51 (1.7 s): two volleys of missiles without exhaust o.O
  
• Frame 51 (1.7 s): the SRM volleys hit the AS
  
• Frame 51 (1.7 s): The AWS to your right seems to be hit by something at its RA (sparks).
  
• Frame 51 (1.7 s): Further streaks enter from bottom of screen.
  
• Frame 54 (1.8 s): The AS' paperdolls' RT, LA and both legs begin to flash.
  
• Frame 59/60 (2.0 s): Further streaks enter from bottom of screen.
  
• Frame 73 (2.43 s): The AS' paperdoll's RA stops flashing.
  
• Frame 91 (3.0 s): The AS' paperdolls' CT is cored and orange, head, CT, LT stop flashing.
  
• Frame 99 (3.3 s): The AS' paperdolls' RT, LL, RL stop flashing.
  
• Frame 100 (3.3 s): The AS' paperdolls' LA stops flashing.
  
• Frame 128 (4.27 s): Some missiles, presumably LRMs, enter at the top right of the screen.
  
• Frame 133 (4.44 s): The AS' paperdolls' head, CT, LL, LT, LA begin to flash.
  
• Frame 133-136 (4.5 s): Explosions from LRMs at AS LT and CT.

As far as I see, every part of the paperdoll that flashes up, does exactly 3 flashes.

Maybe I'll add some interpretation later on.



Please don't take this personally. This is just how science works (you cannot prove anything, only falsify). It's just there are so many events, hard to distinguish, in this video.
I don't even deny there is an issue. I just claim you cannot prove or show it clearly with this video.

You should try to do this on your own video where you know the frame rate of the recording. I think they would use 60fps, but it is already compress into a lossy format to be on YouTube. It is better to analyze with raw avi file.

[sycocys]

I do notice a lot of times that the alpha shots don't register all or sometimes any of the hits. Also sometimes the system chokes on the data and doesn't register the damage until after my next volley.

[Phaesphoros]

Sheraf, on 15 March 2013 - 06:10 AM, said:

You should try to do this on your own video where you know the frame rate of the recording. I think they would use 60fps, but it is already compress into a lossy format to be on YouTube. It is better to analyze with raw avi file.

Of course it is. Analyzation would be simplified and less error prone with higher frame rate, uncompressed video, higher resolution, stable frame rate - and, most important - better distinguishable events (less events not necessarily, this might affect the bug).

This is why I said

Phaesphoros, on 15 March 2013 - 06:03 AM, said:

I just claim you cannot prove or show it clearly with this video.

Edited by Phaesphoros, 15 March 2013 - 06:23 AM.

[Bilbo]

Bluten, on 15 March 2013 - 03:26 AM, said:

I knew the training ground was messed up because Mechs die at least 2x as fast as normal. It's as if they don't have any armor at all. It feels like you're just hitting internals right from the start; although the HUD figure colors say otherwise.

They are stock mechs with their, usually, craptastic stock armor and ammo in places no sane man would put it. Also they are standing quite still.

[3rdworld]

Phaesphoros, on 15 March 2013 - 06:03 AM, said:

Please don't take this personally. This is just how science works (you cannot prove anything, only falsify). It's just there are so many events, hard to distinguish, in this video.
I don't even deny there is an issue. I just claim you cannot prove or show it clearly with this video.

I am shooting an 80 damage alpha. The atlas takes negligible damage to any location other than its CT (the left torso goes from orange to red, but that is it).

The 2 awesomes were 9Ms using a total of 6SRM4s which is 60 dmg. Given the lack of much spread damage taken by the Atlas these missiles would have been sufficient to take the CT from Orange armor to Orange Internal. My 80 damage alpha did not register as a hit per hit indicator, nor did it land on the CT of the atlas ( the volleys just before mine would have to hit somewhere but the atlas paperdoll certainly did not indicate 60 damage from the Awesome volleys anywhere except the CT), nor did it land anywhere else, as it would have surely done more damage to the atlas' armor than what was indicated.

I most certainly did not miss, and the atlas backing away from me did not have the required time to move. it would have taken several seconds for it to get up the hill far enough for the missiles to not in the very least strike its legs (which the paperdoll does not indicate that as happening).

Therefore the only conclusion is the Awesomes missed the entire mech (unlikely given explosion placement and range) and my hit indicator failed to register a hit which spread very little of the damage to any location, or my volley did no damage.

Occam's razor clearly points to my volley doing no damage as the logical culprit.

Before the volleys paperdoll


After volley paperdoll



You can see the atlas only took slight damage to the head, CT, and Left torso. Any other damage was negligible.

Edited by 3rdworld, 15 March 2013 - 06:54 AM.

[CarpetShark]

John MatriX82, on 15 March 2013 - 02:49 AM, said:

I guess the OP is mentioning huge problems in SRM hit-detection. I've myself experienced this, either you blow up mechs one shot or you land a solid alpha with 3 ASRM6 from point blank to a static, maybe even shut down target and there's no hit registration at ALL.

This explains it!

I was in a match a few days ago and cornered a legged Spider in River City. He had machine guns. I pumped volley after volley of SRMs into him and _nothing_ damaged him. He, meanwhile, killed me with the machine guns.

Yes, I should have run off and done something else. I just couldn't believe I was doing nothing at all to him.

Quite the bug.

[Mechwarrior Buddah]

wow people can follow that video? Its so short

[Arkadash]

Banky, on 15 March 2013 - 05:16 AM, said:

I've had the same thing happen any number of times over the past few weeks - and it seems worse since the last patch. Had one game the other night where I nailed a trebuchet at least 6-7 times in a row flush (with 6xSRM6's) and it did next to nothing - and at least a couple of the shots hit while it was standing still/shut down. I had the rest of my team spectating and commenting about it... Happened in the SJR / DV8 matches last night and anyone that was watching the stream saw it. One of the players lined up a clean shot and appeared to have hit the target flush - and no damage registered from the hit.

^^^me too.

[LordBraxton]

GODzillaGSPB, on 15 March 2013 - 01:58 AM, said:

Yeah, I totally agree. If a single volley still one-shots me through the fully intact 50-ish armor of my medium mech that could not outrun the faster splatcat, it was probably me. Me, beeing blown up. That's what you meant. Right? Mhm...

sorry man but thats one slooooowww medium