Odd Uac Jamming?
#1
Posted 13 August 2018 - 11:59 AM
Onto the actual test that i ran. I took a variety of mechs that had a given combo of UACs on them and took em into testing grounds. I moved around a bit and shot in various directions to simulate being in combat, all while shooting non-stop till it ran out of ammo for one of the weapon calibers or my heat rose to the point that i needed to stop and let it cool to 0. Ofc this may not be a 100% accurate representation of real games as its testing ground, but from my real game experiences with dakka, i have to say its pretty much spot on what i experience in terms of the unreliability of certain combos.
Essentially what ive found out is that the most jam prone combo to have is 2 UAC5s, or 2 UAC10s, with the UAC5s being notoriously easy to get double jams on. My record during testing is 5 double jams in 2 minutes, which is basically a death sentence during a game, as it means you are screwed. Dual UAC10s come in at a close second behind that, with double jams a bit less common but very uncommon to have more then 1 UAC10 firing in a sustained shootout. They are also prone to having both guns jam at the exact same trigger push when linked together. UAC2s are a similar story, although most mechs running them will have 4+ guns so the odds of a quad jam are relatively low but do actually happen which makes no sense as the odds of all 4 guns jamming on a single trigger is like under 0.1% which means it should statistically NEVER occur during a game, while i managed to get a quad jam (once on the 1st doubletap) twice before running a hunch2c out of its ammo.
Ive also extensively tested combinations of said weapons, with the biggest one for me being 1 UAC10 and 1 UAC5 (mostly because its the ONLY 2 UACs that sync well in both range, velocity, and general tactics). Out of the 10 trials of that combo i ran, only 2 times did i get a double jam on the 1st click, which is statistically plausible since the chance of a double jam is ~3%, or 1 in 33 shots. Ive also found it extremely rare for both guns to be jammed simultaneously in general, so of all the combos, this one provides the best sustained firepower in situations where you really need to keep the guns shooting (like a brawl). In general, weapons seem to jam at the expected rates though, so using said combos isnt equivalent to a magical no-jam exploit, it just results in far less jamming overall in comparison to other choices.
Also one other thing ive tested is firing same type UACs on different triggers (specifically mouse 1 and 2), which results in a drastically decreased chance of double jams (virtually never happen), and a overall improvement to sustained fire since you are almost always firing 1 of the guns if not both. Just make sure to not click at the same time and put a bit of stagger between the triggers (for example bind left UAC10 to left mouse, right UAC10 to right mouse, and never click at the same time unless alfa strike is paramount say target is not noticed you yet). On that regard, if you are using a mixed build (say 2 UAC5s and 2 UAC10s), you will get less jams overall if you bind half of each weapon to a separate trigger, and as before, dont click the simultaneously. Prolly best example of this is MCII-B, have 1 of each cannon per arm, and fire each arm on a different trigger for maximum reliability (and if you are close to cooking yourself, just fire one of the arms until it jams up, then swap to the other, dont doubletap all of them at once).
I will do a bit more testing on this later (maybee more in actual games and not testing grounds), but sofar ive had phenomenal in game results by replacing the classic dual UAC10 combo with a hybrid 1 UAC10 and 1 UAC5 setup, which not only is lighter and gives me more tonnage to devote elsewhere, but comes with increased effective DPS due to a lack of triggering that bug which seems to make my guns double jam like they are on crack, and provided better sustain due to my weapon firing non-stop until i either cook myself (which happens alot), or the enemies all drop dead. If you want to add more guns, make sure they are fired separately with a different trigger so that you decrease the chances of dreaded double/triple/quad jams, and allow your DPS to remain on the field.
#2
Posted 13 August 2018 - 12:27 PM
#3
Posted 13 August 2018 - 12:38 PM
#4
Posted 13 August 2018 - 12:44 PM
#5
Posted 13 August 2018 - 12:44 PM
Finally something i forgot to add, doubletapping incredibly quickly leads to slightly higher jam chances then clicking at a more leisurely pace, say halfway around when guns reloaded and not right after the 1st shot flies out the barrel. There are many odd occurences, so no way to truly come up with a meta around what to not do with UACs to avoid/limit jams, but i will say i still stand behind my never bring 2 UAC10s into a game that are linked to the same trigger, they doublejam way more often the statistics say they should accounting for jam chance (its like a 3% chance of a doublejam on 1st doubletap, so why does it happen multiple times a game even sometimes in a row).
#6
Posted 13 August 2018 - 12:50 PM
I have not tried putting each UAC5 on a separate trigger. Your results are interesting and I will have to give that a try. Thanks for that suggestion.
I have frankly given up hoping that the UACs will be looked at and they will ever again be useful for anything other than full boating.
Edited by Rampage, 13 August 2018 - 12:51 PM.
#7
Posted 13 August 2018 - 01:05 PM
I've never done specific testing myself, but from an observational standpoint I've certainly noticed the jam chance on UAC's seem proportionate to how many are on the mech.
I enjoy a 6xUAC5 Dire Wolf from time to time, and it's not uncommon for all 6 to be jammed at the same time, rendering my 100-tonner harmless until the weapons come back online.
Similarly, my Warhawk has on occasion emptied its 1xUAC20 ammo bins without so much as a single jam.
I usually attribute it to trigger discipline. Packing one UAC as a primary weapon means my attention is glued to timing the shots perfectly. Whereas the UAC5 boat firing is more akin to button-mashing. I suppose once one or two UAC's start to jam, they fall out of sync with the others and are perhaps more likely to be untimely fired resulting in yet another jam.
Opinion only, perhaps someone can correct me?
Edit: Clarification
Edited by Flyby215, 13 August 2018 - 01:07 PM.
#8
Posted 13 August 2018 - 01:16 PM
#9
Posted 13 August 2018 - 01:25 PM
jamming should be a predictable skill based mechanic (i.e. jambar)
not one that rewards or punishes you randomly
they should also get rid of burst fire. that mechanic is atrocious. and makes it completely dumb to aim CUACs against moving targets. autocannons should fire one shot and only one shot per trigger pull. because that gives you the best tactile response. instead of burst fire just have the CUACs do less damage per shot but with a faster cooldown.
Edited by Khobai, 13 August 2018 - 01:29 PM.
#10
Posted 13 August 2018 - 01:46 PM
Khobai, on 13 August 2018 - 01:25 PM, said:
jamming should be a predictable skill based mechanic (i.e. jambar)
not one that rewards or punishes you randomly
they should also get rid of burst fire. that mechanic is atrocious. and makes it completely dumb to aim CUACs against moving targets. autocannons should fire one shot and only one shot per trigger pull. because that gives you the best tactile response. instead of burst fire just have the CUACs do less damage per shot but with a faster cooldown.
To be perfectly honest, i actually like UACs just as they are, they are different enough from RACs that each is its own separate weapon system (and not just copy/paste with different rates of fire, DPS, and damage per pellet).
My issue with UACs is the stupid amount of inconsistency with them, specifically certain combinations that are almost always feast or famine with nothing inbetween. Ive had games where i dont recall jamming a single time using a dual setup UAC10+5, and ive also had games where i had in excess of 10 double jams on the 1st doubletap (granted games like that i did terribly). UACs need to have burst capability to make them the poky cannon weapon (which neither ACs short of quad AC10s nor RACs are), i just want them to jam more consistently. To that end, id love to see something akin to certain limits added, say never more then half your guns can jam at a single time (to prevent quad jams which make 0 sense statistically), and also mechanics that force a jam if you are on a roll and for some reason doubletap to insanity.
I think the best way to do that would be a dynamic jam chance, which starts off at 0 (so you ALWAYS get 1 doubletap regardless), and go up to 100 at say 6 doubletaps so you are 100% guaranteed to jam the guns when you get to 6 doubletaps without any waiting in between. Essentially, its still a bit of a lottery and you can still get bad/good luck, but its nolonger so broken with excessive jamming/not jamming.
#11
Posted 13 August 2018 - 05:37 PM
#12
Posted 13 August 2018 - 11:17 PM
You will get more double jams in UAC5 than UAC10s because they fire more often, so you get more "dice rolls". This is also why you can get 4UAC2s jammed. Also, to get 2 jammed they dont have to jam simultaneously- one can jam and the other to jam after another shot during the jam duration of the 1st. Again, this affects the fast firing UACs more.
#13
Posted 14 August 2018 - 12:57 AM
it could also be that when multiple guns are fired the rng is only called once and the same value is being used to determine whether all the weapons jam. in other words they are using the rng wrong. each weapon needs its own discreet rng call.
im all for changing the jam mechanic for something that better emphasizes fire discipline. like if the jam chance decays the longer you space out your shots. so if you wait 2 seconds for the second tap on a uac20, you get half the jam chance as opposed to double tapping via button mashing. id also throw on some heuristics in order to prevent cold jams. there could be decreased jam probability if the guns have been idle for some time or if they have never been fired. if you could eliminate cold jams and button mashing that would be great.
Edited by LordNothing, 14 August 2018 - 01:09 AM.
#14
Posted 11 February 2021 - 04:16 AM
Is there any further info on what the OP mentioned?
Also curious about the jam chances when fast doubletapping vs slow doubletapping, is there a real difference?
#15
Posted 11 February 2021 - 04:20 AM
#16
Posted 11 February 2021 - 04:51 AM
One cause for inconsistent perceptions is that if your client gets out of sync with the server, there is a chance your client will fire a shot, and then be corrected by the server that "no, you were actually already jammed", but this only happens with desync. Desync is normally rare unless you have severe internet connectivity issues, and causes a bunch of other oddities too if or when it happens.
Edited by Gagis, 11 February 2021 - 04:52 AM.
#17
Posted 11 February 2021 - 05:13 AM
I´m only on the fence about the double jam thingy, but if I have the patience i´ll come back with some numbers.
#18
Posted 11 February 2021 - 05:32 AM
Went into testing grounds with the TBR-S with 2 UAC5 and did a sequence of 100 double taps.
In those 100 double taps i got:
8x single jams (1 UAC jams alone)
12x double jams (both jam)
its a small sample size yes, but something appears to be at work in this particular case...
So this is a 16% chance of a single jam, and a whopping 12% chance of a double jam.
Edited by Absaint, 11 February 2021 - 05:40 AM.
#19
Posted 11 February 2021 - 09:05 AM
Only issue I ever saw with UACs is if you double tap really fast, the second tap is before the CD bar gets to halfway, you have increased jam chance.
I learned how to time shooting based off sounds of ammo firing how to lower jam chances with UAC5 and UAC10. Rarely run UAC20 only on a IS Hunchie once, UAC2 is debatable depending on IS or Clan.
#20
Posted 11 February 2021 - 09:10 AM
Wildstreak, on 11 February 2021 - 09:05 AM, said:
This is an old myth often repeated on the forums, but it doesn't actually happen.
Testing grounds is really weird sometimes, so to be sure its best to test stuff in a private lobby.
2 user(s) are reading this topic
0 members, 2 guests, 0 anonymous users