Shotgun spread and "hit reg problems"
rantology
Join Date: 2012-02-05 Member: 143750Members, NS2 Developer, NS2 Playtester, Squad Five Gold, NS2 Map Tester, Reinforced - Shadow, WC 2013 - Gold
Join Date: 2012-02-05 Member: 143750Members, NS2 Developer, NS2 Playtester, Squad Five Gold, NS2 Map Tester, Reinforced - Shadow, WC 2013 - Gold
<div class="IPBDescription">just an observation</div>So myself and GORGEous were just testing out some of the randomness of the shotty spread, thought I would share and see if anyone else has opinions on said topic.
<div align='center'>video:</div>
<center><object width="450" height="356"><param name="movie" value="http://www.youtube.com/v/HPcXC1Ml1rc"></param><embed src="http://www.youtube.com/v/HPcXC1Ml1rc" type="application/x-shockwave-flash" width="450" height="356"></embed></object></center>
some things to note: The shotgun fires 10 bullets total, the hits are displayed as a single red line. Missed bullets get their own individual green line, so 4 green lines and a red line = 4 missed bullets, 6 bullets hit.
# of Weapons 0 shotgun hits to kill a cara fade: 27 pellets (or about 3 shots assuming 100% pellet accuracy)
# of Weapons 0 shotgun hits to kill a cara skulk: 10 pellets or one shot, assuming 100% of the pellets hit
Anyway.. I think when people complain about hitreg issues they may in part also be unknowingly complaining about the randomness of the shotgun spread as it's not uncommon to hit a skulk dead on (or seemingly) 2-3+ times and still it lives. Some of that may be hit registration, but I think another thing to point out is that the shotty really seems to be a glorified melee weapon ... even after the recent "buff". The models in the video are pretty damn close to me and still you see 50%+ of the pellets just straight up miss due to the spread.
no real point to this thread I guess, just an observation of how random the shotgun can be.
edit:
eh? made this picture- this is the actual shotgun spread, so this exact spread rotates randomly and that's what you get. You can see 40% of the pellets are pretty much in the top-leftish of the spread (in this picture) so if you happen to roll the dice and that corner lands favorably on your target you will get a significant amount of extra damage on that shot...
<img src="http://i.imgur.com/LLtpx.jpg" border="0" class="linked-image" />
.
<div align='center'>video:</div>
<center><object width="450" height="356"><param name="movie" value="http://www.youtube.com/v/HPcXC1Ml1rc"></param><embed src="http://www.youtube.com/v/HPcXC1Ml1rc" type="application/x-shockwave-flash" width="450" height="356"></embed></object></center>
some things to note: The shotgun fires 10 bullets total, the hits are displayed as a single red line. Missed bullets get their own individual green line, so 4 green lines and a red line = 4 missed bullets, 6 bullets hit.
# of Weapons 0 shotgun hits to kill a cara fade: 27 pellets (or about 3 shots assuming 100% pellet accuracy)
# of Weapons 0 shotgun hits to kill a cara skulk: 10 pellets or one shot, assuming 100% of the pellets hit
Anyway.. I think when people complain about hitreg issues they may in part also be unknowingly complaining about the randomness of the shotgun spread as it's not uncommon to hit a skulk dead on (or seemingly) 2-3+ times and still it lives. Some of that may be hit registration, but I think another thing to point out is that the shotty really seems to be a glorified melee weapon ... even after the recent "buff". The models in the video are pretty damn close to me and still you see 50%+ of the pellets just straight up miss due to the spread.
no real point to this thread I guess, just an observation of how random the shotgun can be.
edit:
eh? made this picture- this is the actual shotgun spread, so this exact spread rotates randomly and that's what you get. You can see 40% of the pellets are pretty much in the top-leftish of the spread (in this picture) so if you happen to roll the dice and that corner lands favorably on your target you will get a significant amount of extra damage on that shot...
<img src="http://i.imgur.com/LLtpx.jpg" border="0" class="linked-image" />
.
Comments
At least when i reimagine how a lost shotgun fight went its usually because i made too many partial hits and not deadcenter hits... (if you make deadcenter hits, and got the feeling where the "deadzone" range starts - its very reliable)
Tho depending on distance you can also work with partial hits(you need to get a feeling for this), so i usually know if i dont get the perfect hit - ill need 3 shots now or whatever.
edit: ok now with that pattern picture and explanation... guess its not only bad aim...
Tho i never felt like its the hitreg - its randomness of bullets and/or just bad aim. (and ofc bad performance that leads to bad aim, i hate it when performance drops and i suddenly get/feel a resistance while aiming, or ppl warp around randomly.)
Reread rantology's post. The one red trace you see represents <b>all</b> the shells that hit. Only the green misses show individually.
Perhaps the people who have problems have issues that go deeper than the games mechanics..........
So in short up the chance of hitting, and then nerf damage to compensate.
Perhaps the people who have problems have issues that go deeper than the games mechanics..........<!--QuoteEnd--></div><!--QuoteEEnd-->
Yeah.
Ok.
Keep us posted how great the shotgun is at pub stomping 3 day old players.
(it's important to note that lerks actually can't be 1-shot with carapace)
<a href="http://www.wolframalpha.com/input/?i=plot+%5B%5B-2.8%2C+4%5D%2C%5B-1%2C3.5%5D%2C%5B3%2C3.5%5D%2C%5B-3.5%2C0%5D%2C%5B0%2C1.5%5D%2C%5B0.4%2C0.5%5D%2C%5B-2%2C-2%5D%2C%5B1%2C-0.5%5D%2C%5B2%2C-2.5%5D%2C%5B-1%2C0%5D%2C%5B-.39%2C.8%5D%5D" target="_blank">http://www.wolframalpha.com/input/?i=plot+...-.39%2C.8%5D%5D</a> (plot of all 10 pellets + center of distribution)
current distribution of pellets is skewed towards one side slightly, in this example and in the picture, top left. so the center of the distribution is actually [-.39,.8] and not [0,0] where it should be (where your crosshair is)
simple fix: change all the GetNormalizedVectors in shotgun.lua so that the center of the distribution is [0,0]
The fade issues are more just bad collisions making blocking a fade impossible, and it so easily to get out when stuck as a fade, and the blink effect being pretty ambiguous as to where the model actually is.
Ok.
Keep us posted how great the shotgun is at pub stomping 3 day old players.
(it's important to note that lerks actually can't be 1-shot with carapace)<!--QuoteEnd--></div><!--QuoteEEnd-->
Words, out, of, my, mouth.
I think if the shotgun is buffed anymore though, it will become way too good? I think once performance is much better it would be a better time to evaluate maybe..
The fade issues are more just bad collisions making blocking a fade impossible, and it so easily to get out when stuck as a fade, and the blink effect being pretty ambiguous as to where the model actually is.<!--QuoteEnd--></div><!--QuoteEEnd-->
qft
<img src="http://i.imgur.com/8UMZM.png" border="0" class="linked-image" />
The problem is that the current spread is not weighted properly to the center of your screen, so it's average damage is actually higher slightly off center, which is really misleading, and can still lead to particularly lucky or unlucky shots.
luck shouldn't be a factor at all. yes. unfortunately/realistically that's how a shotgun works in most games, so working within the confines of the game system the best solution is to normalize the distribution to [0,0]. it's just not a priority for the devs right now. the trick is to getting something that's consistent (for shotguns) and balanced. i mean, you could put all the points on 0,0 but then you'd have a sniper rifle (which i did last night, it was pretty funny actually). i thought eh was working on something so i didn't bother working out specific points. and yeah there are better ways to do it but you've got to fit performance and design constraints. a realistic shotgun in ns2 would be way overpowered because they're accurate up to 40m usually which is bigger than most rooms currently in the game.
What I do notice very often, however, is a lack of perceived machine gun hitreg. I have no idea what causes it -- whether it be actual networking problems, low fps, low server performance, alien animations, alien hitboxes or something entirely different -- but I know that hitting skulks, lerks and fades with the machine gun is significantly more difficult than it was in NS1.
holy ###### it's someone from tf2 who knows what they're talking about
please pay attention, guys!
What I do notice very often, however, is a lack of perceived machine gun hitreg. I have no idea what causes it -- whether it be actual networking problems, low fps, low server performance, alien animations, alien hitboxes or something entirely different -- but I know that hitting skulks, lerks and fades with the machine gun is significantly more difficult than it was in NS1.<!--QuoteEnd--></div><!--QuoteEEnd-->
+1 to the last point. Although the feel of the shotty is that it's not very consistent in dealing damage at close range either. Just like how for random reasons I magically can't land 5 bites in a row.
I know that in NS1 I could time the blast so all the bullets hit (yes, effectively melee weapon, but a 1-shot one on Skulks baby), but that seems impossible in NS2.
Perhaps the people who have problems have issues that go deeper than the games mechanics..........<!--QuoteEnd--></div><!--QuoteEEnd-->
Personally I agree with you here antacid,
Those saying its impossible to 1 shot a cara lerk with what level weapons upgrade are you referring 0 or 3?
Either I have been really unlucky and copped random bullets or you can indeed 1 shot a lerk.
I do find it interesting that no one has an issue with skulks being 1 shotted with carapace when marines have not upgraded weapons.
The upgraded skulks should not die to 1 shot of w0 guns.
w3 fair enough, w2 possibly but no way w0 should 1 shot a carapace skulk after all it is an upgraded alien lifeform not a default one.
Untill the performance issues are addressed people who dont have the best machines will find hit reg issues tweaking now will leave us with very OP weapons (the SG is already OP'd..1 shot a cara skulk with W0?!!)
It is literally a dice roll whether it works or not that depends on latency. There is no person that plays this game on a regular basis who thinks shotguns are going around on weapons level zero killing skulks in one shot.
You do exactly 170 damage for the perfect shot and thats exactly how much health the skulk has. The majority of the time the skulk still gets away with less than 1 hp.