189 replies to this topic

[The Lighthouse]

In one of the threads, Tarogato and I discussed about UACs and stuffs. I can't point out exact post because the search function of this forum is non-functional, but it was about Jaggermech with 4 x AC2 and 4 x UAC2.

He pointed out 4 x AC2 do not bring enough firepower and thus it is an invalid example. So I thought "well... sure yeah" and kind of forgot about it.


Then I read another thread, and MischiefSC pointed out AC2 is actually better than UAC2, even with Dragon 5N (again cannot find exact post as well).



Well, two different claims. So I did some test. Part of the reason is because MischiefSC did test this in one of the hottest maps, which naturally favor normal ACs.




So here's what I did:


1) Unskilled Jaggermech DD, which has 20% Jam reduction chance and 10% ballistic cooldown. Light 250 with no other external heatsinks.

2) With same mech, 4 x AC2 and 4 x UAC2.

3) Testing Grounds, Crimson Strait. The position where the mech can shoot Atlas at front and Catapult at behind and can hit Cicada's side torso.

4) Shoot Atlas til it dies, then go for Catapult, then Cicada until its side torso is destroyed.

5) Mech should never go over 100%+ heat.

Due to jamming nature, I did several runs... but the numbers were quite consistent.

Stopwatch activated at the same time I hit the mouse button to Atlas, and stopped right after side torso of Cicada was destroyed.

Avg with UAC2 : 40~55 seconds.
Avg with AC2 : almost exactly 30 seconds.

Duh... So the it is not even about jamming. UAC2 are simply way too hot that the mech is already going overheat during Atlas phase.


....

Wow....

Ok, then how about UAC5? So I grabbed my Cyclops hero mech (Sleipnir). Unlike Jaggermech, it is fully skilled with UAC jam duration reduction. Ran same test...


Avg with UAC5 : ~30 seconds. Actually it overheats around the time Catapult is dead. Super heat efficient indeed.
Avg with AC5 : ~30 seconds!!


I must remind all of you that UAC5 is the most heat efficient UAC. So I guess this time jam duration is doing its work... way too much, though the mech eventually goes overheat during Catapult phase.


......I am not going to bother with UAC10 and UAC20.



And think about it, I didn't add any extra heatsinks for this testing. Yes, both UAC and AC testings, mechs had same engine, no external heatsinks to make the testing fair.

So in real situation you probably could get far more from normal ACs because you will have more slots and most importantly, more tonnages for bigger engines, heatsinks or other weapons.

So what, only thing that UAC can do well is burst power for poking? Even that is arguable because of jamming and projectile speed, and the most of the time dakka mechs are used are for sustain fire for brawling.

And if I were to poke, I'd rather get Gauss Rifle or just laser weapons.





......







SIGH






......





I am not going to get rid of those UACs from my mechs, I have 280 mechs now, and I am too lazy to get rid of UACs and do all optimizing all of the mechs AGAIN.


But if you are truly going for optimizing mechs, and you truly care about doing well on this game and care about winning, I suggest do not even bother about ANY of UACs until some changes happen to ballistics.

Or unless you really love double-tapping. But I don't think feeling of double-tapping can overcome all of these issues UACs have.

[lazorbeamz]

UAC 2 is useless.

with double tap and jamming and without quirks/skills:
UAC5, 10 give ~+20% dps and double the alpha
UAC 20 gives ~+40% dps and double the alpha.

UAC have ~1.5-1.8 times less damage per heat unit when compared to AC and UAC will actually overheat you. The biggest disadvantage of UAC is their comparative heat inefficiency.

Conclusion:
if you want to DPS for a short amount of time before you overheat or shorter, then UAC (especially c-UAC) are much better. Also better for alpha strike.
If you want to DPS non stop until your mech gets destroyed then you want to take AC instead (especially clan AC 10 which is super heat effective). Also better for torso twisting

AC have endurance, UAC have burst

However laser alpha strikes will give you much more advantages then either

END

Edited by lazorbeamz, 18 November 2017 - 02:32 AM.

[Spheroid]

Dumb to test it on Slepnir. On a Jagermech UAC-5 outperforms AC-5 on many builds despite being well past the golden age of UAC.

[Khobai]

IMO they need to remove the double tap and jamming mechanics completely and normalize UACs. RNG is crap and doesnt belong in the game.

UACs should work like standard ACs but with 20% higher DPS, 40% lower damage per shot, 20% lower range, and 20% lower velocity

For example:

AC2 = 2 damage, 0.72 cooldown, 840m range, 2000m/s velocity
UAC2 = 1.2 damage, 0.36 cooldown, 700m range, 1800m/s velocity (never double taps and never jams)

AC5 = 5 damage, 1.66 cooldown, 720m range, 1550m/s velocity
UAC5 = 3 damage, 0.83 cooldown, 600m range, 1300m/s velocity (never double taps and never jams)

Edited by Khobai, 18 November 2017 - 03:49 AM.

[The6thMessenger]

But it's not like that's the only way to see the value in UACs. Personally, I value the UAC5's double-tap. Sure you could do the same damage overtime due to Jamming. But then on peekaboos, AC5 bursts 5, when UAC5 bursts 10.

It's nice to have them literal ACs with just faster rate of fire and only overheats overtime; predictable down-time.

[Seranov]

How many times did you repeat this test, to iron out inconsistencies that were not related to the control groups you were attempting to test?

There is also the fact that being able to double-tap is infinitely more important than sustained damage, because sustained damage comes with the caveat of you staring at the target and hoping they don't kill you before you kill them.

[Muujig612]

I prefer AC2s over UAC2s. But that's just me.

[Nema Nabojiv]

Those triple uac2 Dragons people use sometimes are the evidence of the contrary. They are able to do some decent 800-1000 damage per match with what seems to be an underwhelming loadout, so there's that.

[Khobai]

Quote

There is also the fact that being able to double-tap is infinitely more important than sustained damage, because sustained damage comes with the caveat of you staring at the target and hoping they don't kill you before you kill them.

Poking is what standard autocannons should be for

DPSing while facetanking is what UACs should be for

different purposes ensures both have a use in the game

Quote

Those triple uac2 Dragons people use sometimes are the evidence of the contrary. They are able to do some decent 800-1000 damage per match with what seems to be an underwhelming loadout, so there's that.

yeah but how much of that damage is actually lethal? and how much of it is trash damage?

im willing to bet at least half of it is trash damage if not more

match score should really weight damage based on location. damage done to locations like arms and legs should matter less than damage done to CT. and headshots and backshots should be worth the most. getting legged also shouldnt kill you just reduce your max speed to like 15% but thats another topic entirely.

Edited by Khobai, 18 November 2017 - 05:28 AM.

[Bud Crue]

If the basis of the test is standing still and killing a non-moving target in the open as fast as possible,I wonder how the test would be doing two things:

1: Not double tapping the UACs. Just hold the button down. To me in the test conditions presented that would seem to be the most efficient way to kill the target in question: a solid stream of projectiles without extra heat. These are UAC 2s after all...its gonna take a while; why risk the heat or losing output?

2: Test against dual gauss. I mean again if the test is you just standing there not taking fire, to see how long it takes to kill something at range, I think on the mechs mentioned it makes a hell of a lot more sense to skip auto cannons of any kind and go gauss.

I think of AC2s and UAC2s as annoyance weapons mainly. They get better players attention and make bad players duck when hit. In no situation, that I've ever encountered does my target just stand there unresponsive. I run a 4UAC2 4ML Jagger S and a 4UAC2 2ML DD and both are fun and can be effective but they play very differently, alas I have yet to have a target just stand there and let me feed shots in to it at range (though if you are on escort and get a clear line on the stupid Atlas, its kinda close to the OP's test, until the other reds shoot back).

Edited by Bud Crue, 18 November 2017 - 05:27 AM.

[Nema Nabojiv]

Khobai, on 18 November 2017 - 05:01 AM, said:

yeah but how much of that damage is actually lethal? and how much of it is trash damage?

im willing to bet at least half of it is trash damage if not more

After a certain amount of damage done to a single target it stops being trash damage as the target is so badly damaged that your more pinpoint teammates can pick the components they want in one or two shots.

Same goes for lrm damage btw, but lurms need a lot of time to soften the target enough to justify their use.

[Daggett]

I think the test scenario is heavily biased against UACs.

The fact that they generate more heat than ACs tells a lot about their role intended by PGI and to which weapons they should be compared. It is totally clear that UACs are not designed as cool sustained DPS brawling weapons so it does not make much sense to test them with the condition to kill 3 mechs as fast as possible where heat becomes an issue.

It is also questionable to do this with an unskilled mech because the two UAC nodes are mandatory when UACs are the main weapon like most laser-builds want their laser duration nodes. The fact that ACs can ignore those nodes and get others instead is not significant enough especially when considering most AC builds already want the nearby magazine capacity or many cooldown nodes and therefore don't save that much nodes anyway.

Last but not least the chosen mech also favors ACs in your test scenario, because the UAC jam chance reduction is weak when heat is the limiting facor while the cooldown reduction always works on cool ACs but not on jammed UACs.

It's no wonder that the standard ACs will always win such tests. Especially the heavier UACs are obviously designed to quickly kill or cripple a single target within a short window of opportunity and then retreat to cool down.

I have not tested the AC2s but you can test both AC5s again against the Atlas only and you will see that even an unskilled Mauler without quirks is at least 20% (about 3s) faster with 4x UAC-5s than with standard AC-5s.
Add in quirks and skills and some mechs like the Dragon 5N or your Jagermech can even increase this difference to 50%.

In conclusion they are simply quite different weapons with different roles despite their similar name and probably can't even be tested against without favoring one weapon in the scenario.

Edited by Daggett, 18 November 2017 - 07:23 AM.

[Muujig612]

Nema Nabojiv, on 18 November 2017 - 04:37 AM, said:

Those triple uac2 Dragons people use sometimes are the evidence of the contrary. They are able to do some decent 800-1000 damage per match with what seems to be an underwhelming loadout, so there's that.

That 40% jam chance reduction quirk might have something to do with that. Obviously.

[lazorbeamz]

Muujig612, on 18 November 2017 - 06:55 AM, said:

That 40% jam chance reduction quirk might have something to do with that. Obviously.

Just FYI this 30-40% jam chance reduction will only give ~15% more DPS in case of UAC5. Quirks have a small effect similar to that of other ones like ballistic cooldown -15% etc

Edited by lazorbeamz, 18 November 2017 - 07:16 AM.

[Muujig612]

lazorbeamz, on 18 November 2017 - 07:16 AM, said:

Just FYI this 30-40% jam chance reduction will only give ~15% more DPS in case of UAC5. Quirks have a small effect similar to that of other ones like ballistic cooldown -15% etc

40% less chance of jamming is not small at all, if you factor in the base DPS increase from double tap of UACs, and the 20% cooldown quirk for the UAC5 (10% for UAC2 spam). It all adds up to one big DPS increase.

[panzer1b]

I actually did some math theorizing a few months back and figured out that the IS/cUAC2, and the IS/cUAC10 actually resulted in less sustained DPS compared to the AC versions if you factored in the jam chance and duration relative to the actual sustained DPS of the non ultra model. Only the UAC5 or UAC20 came out ahead in its DPS stats.

Essentially, back when, unless you absolutely needed burst damage, AC2 and AC10 ended up a much better weapon (not to mention ran colder). UAC5 and UAC20 ended up superior to the AC versions in most situations as well since they had higher burst and higher sustained DPS discounting the very minor increased heat generation (7 vs 6 on the 20, almost irrelevant unless u were trying to use it alongside lasers). Only exception is heavy jam chance quirks, which will result in the ultra always beating the regular cannon provided you actually have enough DHS to cool it.

Anyways, does anyone have a updated jam chance table for the various ultras at this point in time? Im asking since i cannot find this information ANYWHERE, isnt listed in game, and the jam chances in the spreadsheet i made a long time ago have prolly been altered as they are from before new tech became a thing.

Edited by panzer1b, 18 November 2017 - 07:25 AM.

[MischiefSC]

I did a ton of tests on AC2/UAC2/RAC2. 3 on a Marauder, same map, running the circuit. AC2 was the clear winner. UAC2 has a narrow window where it is *sometimes* a little better. If it doesn't jam. All the UACs are like that. As a given rule though for overall performance you're better off with regular ACs. Unreliable burst DPS when luck favors you in return for worse heat and on all but 2s and 5s speed damage is, on average, a bad trade.

UACs are a gamble and you're not the house.

[The Lighthouse]

Thank you guys for the all the suggestions and thoughts! Will do additional tests as we speak.

Seranov, on 18 November 2017 - 04:28 AM, said:

How many times did you repeat this test, to iron out inconsistencies that were not related to the control groups you were attempting to test?

There is also the fact that being able to double-tap is infinitely more important than sustained damage, because sustained damage comes with the caveat of you staring at the target and hoping they don't kill you before you kill them.

16 times for UAC2, 8 times for UAC5, and only 3~4 times for normal ACs because the difference was at best 1 second per trial.

In order to reduce inconsistencies, the test was done on...

1) Same map.
2) Same mech.
3) Same position.
4) Same targets.
5) Removed mech travel time / complete stationary.
6) For Jagger the mech is completely vanilla, not skilled.
7) Not sure making a control group is possible or needed when it is direct comparison with two.

Double-tap, in vacuum, would be good burst damage UNTIL the mech reaches its heat limit, that is IF UACs do not jam at all.

Bud Crue, on 18 November 2017 - 05:27 AM, said:

If the basis of the test is standing still and killing a non-moving target in the open as fast as possible,I wonder how the test would be doing two things:

1: Not double tapping the UACs. Just hold the button down. To me in the test conditions presented that would seem to be the most efficient way to kill the target in question: a solid stream of projectiles without extra heat. These are UAC 2s after all...its gonna take a while; why risk the heat or losing output?

2: Test against dual gauss. I mean again if the test is you just standing there not taking fire, to see how long it takes to kill something at range, I think on the mechs mentioned it makes a hell of a lot more sense to skip auto cannons of any kind and go gauss.

I think of AC2s and UAC2s as annoyance weapons mainly. They get better players attention and make bad players duck when hit. In no situation, that I've ever encountered does my target just stand there unresponsive. I run a 4UAC2 4ML Jagger S and a 4UAC2 2ML DD and both are fun and can be effective but they play very differently, alas I have yet to have a target just stand there and let me feed shots in to it at range (though if you are on escort and get a clear line on the stupid Atlas, its kinda close to the OP's test, until the other reds shoot back).

1. If hold the button down, then I assume the TTK would be same as AC2, but if we are going to do that, why not just use AC2 in the first place?

2. Will do.

AC2 and UAC2 are definitely not annoyance weapons when boated with more than 2. 4AC2/UAC2 is not something you can ignore.... Actually it works better because higher caliber indeed make people duck, while AC2/UAC2 does not create much shake so people think they can just ignore..... then they realize they have almost no armor left.

Daggett, on 18 November 2017 - 06:42 AM, said:

I think the test scenario is heavily biased against UACs.

The fact that they generate more heat than ACs tells a lot about their role intended by PGI and to which weapons they should be compared. It is totally clear that UACs are not designed as cool sustained DPS brawling weapons so it does not make much sense to test them with the condition to kill 3 mechs as fast as possible where heat becomes an issue.

It is also questionable to do this with an unskilled mech because the two UAC nodes are mandatory when UACs are the main weapon like most laser-builds want their laser duration nodes. The fact that ACs can ignore those nodes and get others instead is not significant enough especially when considering most AC builds already want the nearby magazine capacity or many cooldown nodes and therefore don't save that much nodes anyway.

Last but not least the chosen mech also favors ACs in your test scenario, because the UAC jam chance reduction is weak when heat is the limiting facor while the cooldown reduction always works on cool ACs but not on jammed UACs.

It's no wonder that the standard ACs will always win such tests. Especially the heavier UACs are obviously designed to quickly kill or cripple a single target within a short window of opportunity and then retreat to cool down.

I have not tested the AC2s but you can test both AC5s again against the Atlas only and you will see that even an unskilled Mauler without quirks is at least 20% (about 3s) faster with 4x UAC-5s than with standard AC-5s.
Add in quirks and skills and some mechs like the Dragon 5N or your Jagermech can even increase this difference to 50%.

In conclusion they are simply quite different weapons with different roles despite their similar name and probably can't even be tested against without favoring one weapon in the scenario.

Actually it is biased against normal ACs for two critical reasons.

1) Firing mech was stationary, thus it can dissipate heat faster than when it moves. How many time do you see an competent player standing still unless during early game?

2) No change of mech customization for both UAC and AC, which means when mech was equipped with ACs, it was heavily under-tonned. Again how many times do you see mechs that are under-tonned in higher tier games?

Of course, the second argument only works when we do not talk about weapons in vacuum.

And not sure cooldown favors AC more than UAC, since BOTH AC and UAC does get cooldown reduction, just not jam duration reduction. But in the end cooldown quirks makes both weapons fire faster regardless.

I will do some additional testing. But thing is I do not have Mauler and have no intention of getting it until Black Friday / Cyber Monday sale arrives.

Edited by The Lighthouse, 18 November 2017 - 08:45 AM.

[panzer1b]

Does anyone know what the jam chance is for ultras in the current patch?

[The Lighthouse]

Did some more testing with Jager DD with 3 AC5 and 3 UAC5.

Best number I got from UAC5 : 32 seconds
Worst case : 40 seconds
Avg UAC5 : 35
Avg AC5 : 36 seconds.

They are essentially same for 30-ish sustained fire IF the mech is stationary (as soon as I simulate moving condition, UAC gets really, really worse).

....Except they should not be same, since UAC5 weights 1 ton more, and takes 1 more slot space than normal AC. With three of them, the difference is 3 tons and 3 slots. That's free external double heatsink plus 2 free tons right there.

For sustained fire and/or brawling purpose, UAC is just terrible deal folks. AND we are talking about UAC5, the most heat-efficient and most dps-efficient in terms of heat.


And we have things like poking, long-range stuffs that 'burst' nature of UAC can be used... but are you guys seriously going to use UACs for poking? or long-range stuffs? When you already have lasers for poking and PPC/ER Lasers for long range?

Actually, for Inner Sphere I'd chose normal ACs for poking since the low face-time and single-bullet nature work far better for poking.

Guys, more I test these, more I think UACs are just inferior weapons compared to ACs. As once MischiefSC explained, it's a trap.



A devious f***ing trap.



We 'feel' UAC seems stronger weapons due to double-tap nature (ooo MOAR bullets! So more damage?) and myth of burst damage stuffs... but in reality UAC just works so poorly that is actually inferior to ACs for almost all aspects.

panzer1b, on 18 November 2017 - 08:33 AM, said:

Does anyone know what the jam chance is for ultras in the current patch?

I believe it is still 17 percent, only the duration has changed in November 2016 patch.

Edited by The Lighthouse, 18 November 2017 - 08:44 AM.