21 replies to this topic

[Skippy The Danger Squirrel]

Ok, so here's a new one..

I just finished playing Crimson Straight. We were defending the saddle and I had a sniper's position over the pass with my Shadow Hawk. I had targets 200-550m away, clear line of fire. I shot a Vapor Eagle 6 times at around 480m and it registered 0 dmg. I finished the match, having fired 66 LB10x and only 191 dmg, most of which was from my medium lasers.

I know the mechanics of this game are wonky, but how in the hell do I HIT a target 6 times and do NO damage???

No. He was not hiding behind a light pole, either. Open shot to the upper torso with his bottom obscured by a building. Or is this the same terrain flaw that I noticed with LOW coming back to bite me in the butt again?

[Appogee]

I've seen that happen when I have my graphics set on low (to improve frame rate) but there is some 'invisible' terrain feature in front of my target, beyond the LOD for my settings.

It can also happen when a target is disconnected. Their hitboxes seem to desync from where we see them.

Edited by Appogee, 03 July 2019 - 07:07 AM.

[Nightbird]

Never had reg issues, only invisible wall ones

[Skippy The Danger Squirrel]

Something else to ponder...

I went back and looked at my combined IS and Clan LB10x stats. I saw an interesting trend:

LB2x (combined) - 1.89 avg dmg ; 5% loss
LB5x (combined) - 4.72 avg dmg ; 5.4% loss
LB10x (combined) - 7.78 avg dmg ; 23% loss
LB20x (combined) - 15.54 avg dmg ; 23% loss

It is interesting that PGI sought to tighten the LB2 and LB5 spread, to make them a better long range weapon, but left the LB10 and LB20 alone, even though they lose about 1/4 of their damage potential due to cluster spread. I don't engage much beyond 500m, which is well within the maximum effective range and should guarantee the greatest percentage of hits.

[Nightbird]

Skrapiron, on 03 July 2019 - 07:55 AM, said:

Something else to ponder...

I went back and looked at my combined IS and Clan LB10x stats. I saw an interesting trend:

LB2x (combined) - 1.89 avg dmg ; 5% loss
LB5x (combined) - 4.72 avg dmg ; 5.4% loss
LB10x (combined) - 7.78 avg dmg ; 23% loss
LB20x (combined) - 15.54 avg dmg ; 23% loss

It is interesting that PGI sought to tighten the LB2 and LB5 spread, to make them a better long range weapon, but left the LB10 and LB20 alone, even though they lose about 1/4 of their damage potential due to cluster spread. I don't engage much beyond 500m, which is well within the maximum effective range and should guarantee the greatest percentage of hits.

If even 1 pellet hits, it counts as a hit, so your math doesn't work.

[Skippy The Danger Squirrel]

No, the math is correct. I divided the total damage done by each combined weapon (IS and Clan) by the number of registered "hits" to get a sense of the slug efficiency.

[Nightbird]

Skrapiron, on 03 July 2019 - 08:27 AM, said:

No, the math is correct. I divided the total damage done by each combined weapon (IS and Clan) by the number of registered "hits" to get a sense of the slug efficiency.

Or your aim accuracy...

[Skippy The Danger Squirrel]

Aim accuracy is 52.35%

34,361 LB10x fired, 17,990 hits. 136,962 dmg. Avg dmg per hit 7.78

Edited by Skrapiron, 03 July 2019 - 09:16 AM.

[Nightbird]

Sigh... a hit is if any one of your pellets hits the target. If you shoot at a target that is partially covered by buildings or terrain, then you'll still get a hit if one of your pellets land, but most are going to be blocked by the obstacle. That's not a problem with weapon efficiency, but a choice to miss most pellets to get a little damage it.

[Skippy The Danger Squirrel]

That is a plausible theory, however, I tend to use lasers on partially obscured targets and LB10 when they are in the open. The spread of shot is greater in the larger caliber guns, meaning as the range extends, the spread of shot expands as well. The "long range" shots (over 500m) result in the fewest "hits", so I tend to use thermal vision and close the range to where I can deliver maximum efficiency.

[Prototelis]

Don't argue with the mathamagician. Srsly.

[Foxwalker]

Anytime PGI makes weapon changes, and you note something weird, (Like this). I would suggest reporting it to support. There may be an issue. You never know.

[LordNothing]

wonky hit detection strikes again. it happens.

i totally understand why it happens it situations where there are too many objects to track. there is only so much you can cram into a udp packet. your only 2 options are compression or reduction in numeric precision (sending float instead of double) and modern net code probably does both.

thats one limit but there are others. invisible or hidden terrain/objects is also a thing. another thing are invalid culls. where the game figures out its impossible to tell if a thing hit or not, usually en an edge case like hitting an edge of a collision hull, you just end up in the divide by zero singularity or some number thats too big or too small to compute correctly (usually resulting in a nan error). in these situations the game can either say 'ok we will just call it a hit', or 'we can ignore it'. i think cryengine does the latter.

[Steel Raven]

Just recently watch my LB-10X spread go through a mech at range. Seen it happen with a PPC bolt so while it pissed me off in the heat of battle while the enemy team was pushing but it was of no surprise. Projectile + Particle Effects seem to screw with the Cry engine's math.

[Polecat]

PPC's don't seem to connect for me roughly half the time. The damage stats at the end of the match look right and I see them pop on the enemy mech's armor, but there's no reticle feedback.

[General Solo]

I think its lost/dropped UPD packets combined with a slow/low Server update tick rate.

With a fast Server update tick rate say 30Hz, when a packets dropped/lost you don't have to wait long for the next update, so the effect is small regarding hit reg. You wait one thirtieth of a second about 33ms for the next update packet.

But if the Server update tick rate is low say 10Hz, you will have to wait longer for the next update packet in this case one tenth of a second or 100ms.

This combined together with high-ish ping say about 150 causes the problems imo.

If your PPC shot was in that lost packet, you get sheet reg. But the damage is counted some how on the end of game screen. But that mech with red CT killed you coz your shot didnt reg when you needed the damage.
I WANT THE DAMAGE NOW NOT END OF ROUND


Do people with sub 100 ping report sheet reg too?

Please post if you do....FOR SCIENCE

Edited by OZHomerOZ, 09 July 2019 - 05:00 AM.

[Prototelis]

Yes, I get **** hit reg occasionally even on fiber on the NA server with a 15 ping.

The nature of online games.

Edited by Prototelis, 09 July 2019 - 05:05 AM.

[Shadowomega1]

OZHomerOZ, on 09 July 2019 - 04:55 AM, said:

I think its lost/dropped UPD packets combined with a slow/low Server update tick rate.

With a fast Server update tick rate say 30Hz, when a packets dropped/lost you don't have to wait long for the next update, so the effect is small regarding hit reg. You wait one thirtieth of a second about 33ms for the next update packet.

But if the Server update tick rate is low say 10Hz, you will have to wait longer for the next update packet in this case one tenth of a second or 100ms.

This combined together with high-ish ping say about 150 causes the problems imo.

If your PPC shot was in that lost packet, you get sheet reg. But the damage is counted some how on the end of game screen. But that mech with red CT killed you coz your shot didnt reg when you needed the damage.
I WANT THE DAMAGE NOW NOT END OF ROUND


Do people with sub 100 ping report sheet reg too?

Please post if you do....FOR SCIENCE

I also have seen issues and I get between 40 and 62 ping. Lately however I have had my AC 20 shells' on my new Charger suddenly disappear milliseconds after leaving the barrel.

[Serial Number]

https://www.reddit.c...ave_projectile/

Yeah it's happening more and more often. Seen it quite a few times aswell.

[Bud Crue]

Did you video it?
I ask because there was a post by Lizzee last week on Reddit with some seriously wonky LBX mechanics. Perhaps your issue is related.

Edit: Just saw the post above.

Edited by Bud Crue, 09 July 2019 - 05:17 AM.