Jump to content

Looking At The Uac5 Through The Lens Of Probabilities.


57 replies to this topic

#21 MadIrish

    Member

  • PipPipPipPipPip
  • The 1 Percent
  • 152 posts

Posted 19 December 2016 - 01:17 PM

View PostYeonne Greene, on 19 December 2016 - 01:14 PM, said:

UAC/5 aren't supposed to be as good at 800 meters as the ERLL. That is kind of the point.


Exactly so why do they need to add the random jam, if you wanted to more true to lore then the UAC5 should only have a 1 in 36 chance per shot of jamming (2.7%) not a 1 in 6 chance. Yes a 15% chance is a 1 in 7 (1 in 6.67 to be more exact) per shot of having a jam.

Edited by MadIrish, 19 December 2016 - 01:53 PM.


#22 Y E O N N E

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPip
  • The Nimble
  • The Nimble
  • 16,810 posts

Posted 19 December 2016 - 02:29 PM

View PostMadIrish, on 19 December 2016 - 01:17 PM, said:


Exactly so why do they need to add the random jam, if you wanted to more true to lore then the UAC5 should only have a 1 in 36 chance per shot of jamming (2.7%) not a 1 in 6 chance. Yes a 15% chance is a 1 in 7 (1 in 6.67 to be more exact) per shot of having a jam.


The jam is there to limit your output. When you pay extra tonnage, you are paying it to have the potential to surge your DPS. The jam prevents you from surging it indefinitely, and the RNG nature of it compels you to use your judgement on when it is appropriate to attempt a surge.

The Ultra mechanic is not there to guarantee you several goes with the click-spam. If you were always guaranteed a surge, then it would be flat-out superior to the AC/5. The fact that it is already near-universally preferred to the AC/5 even in its current implementation tells you all you really need to know.

I really don't know why you bring the ERLL into this. They are not comparable weapons. They don't operate in the same bracket, normal do they even deal damage in a similar fashion.

Edited by Yeonne Greene, 19 December 2016 - 02:30 PM.


#23 MadIrish

    Member

  • PipPipPipPipPip
  • The 1 Percent
  • 152 posts

Posted 19 December 2016 - 04:33 PM

View PostYeonne Greene, on 19 December 2016 - 02:29 PM, said:

The jam is there to limit your output. When you pay extra tonnage, you are paying it to have the potential to surge your DPS. The jam prevents you from surging it indefinitely, and the RNG nature of it compels you to use your judgement on when it is appropriate to attempt a surge.

The Ultra mechanic is not there to guarantee you several goes with the click-spam. If you were always guaranteed a surge, then it would be flat-out superior to the AC/5. The fact that it is already near-universally preferred to the AC/5 even in its current implementation tells you all you really need to know.

I really don't know why you bring the ERLL into this. They are not comparable weapons. They don't operate in the same bracket, normal do they even deal damage in a similar fashion.



1. The potential to surge should be consistent based on heat management as it is with energy weapons and again I want to point out that people seem to forget that 4 UAC5s puts out alot of heat when spammed so your argument is flawed.

2. The UAC5 is supposed to be flat out superior to the AC5 and again you guaranteed a surge with comparable energy weapons.

3. You'll need to explain why you think an ERL boat can't surge with click spam (I do it all the time) and why you think they are not comparable.

#24 Y E O N N E

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPip
  • The Nimble
  • The Nimble
  • 16,810 posts

Posted 19 December 2016 - 05:07 PM

View PostMadIrish, on 19 December 2016 - 04:33 PM, said:



1. The potential to surge should be consistent based on heat management as it is with energy weapons and again I want to point out that people seem to forget that 4 UAC5s puts out alot of heat when spammed so your argument is flawed.


The potential is consistent. You have a 15% chance that each surge attempt will cause a jam. You know this, you can work around it. The game is not surprising you with something out of the blue.

Also, you can't surge energy weapons. You have one shot, and then you wait the full cycle. You also get more damage per unit of heat with a UAC/5 than you do with any energy weapon, so yours is the flawed argument.

Quote

2. The UAC5 is supposed to be flat out superior to the AC5 and again you guaranteed a surge with comparable energy weapons.


Can't have weapons that are supposed to be flat-out superior in a PvP game, they all have to have niches where they are strongest otherwise there is no point to their existence. Also, again, energy weapons can't surge.

Quote

3. You'll need to explain why you think an ERL boat can't surge with click spam (I do it all the time) and why you think they are not comparable.


You don't do it all the time, because it is physically impossible in the game's current mechanics. You are compelled to sit through the entire 4.5 seconds it takes for the weapon to fire and cycle every. single. time. The weapon does its damage in one burst, but It's not a surge. You can't boost your output beyond what that initial burst will do, and your maximum DPS possible is static to one value....unlike a UAC, which gives you the option of taking off-cycle shots at your own risk.

And they aren't comparable simply because A.) one is front-loaded, B.) one has slow projectile flight time, making them unwieldy at 800 meters and even somewhat clumsy at their rated 600, C.) One has a max range of 660 after module and the other is 742.5, and D.) one of them has a damage-to-heat ratio of 3.33 to the other's 1.125. You don't bring a UAC/5 for extreme range work.

Edited by Yeonne Greene, 19 December 2016 - 05:07 PM.


#25 El Bandito

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPipPipPip
  • Big Daddy
  • Big Daddy
  • 26,736 posts
  • LocationStill doing ungodly amount of damage, but with more accuracy.

Posted 19 December 2016 - 05:29 PM

View PostSnowbluff, on 19 December 2016 - 11:26 AM, said:

Didn't you used to use obnoxiously large text? >:l


Large texts are at least easier to read. Simple.

Edited by El Bandito, 19 December 2016 - 05:43 PM.


#26 FupDup

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 26,888 posts
  • LocationThe Keeper of Memes

Posted 19 December 2016 - 06:55 PM

View PostTristan Winter, on 19 December 2016 - 11:37 AM, said:

It's almost as bad as balancing the gauss by increasing the chance of a catastrophic explosion that rips your mech apart. If you're going to give RNGeesus that much power, then you may as well introduce Cone of Fire.

That's not a good example since the Clan Gauss now has a 100% chance to explode. It's not RNGesus anymore. It's a consistent guarantee. The old 90% explosion rate was the real RNGesus system.

Edited by FupDup, 19 December 2016 - 06:55 PM.


#27 Tristan Winter

    Member

  • PipPipPipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 2,530 posts
  • LocationNorway

Posted 19 December 2016 - 07:12 PM

View PostFupDup, on 19 December 2016 - 06:55 PM, said:

That's not a good example since the Clan Gauss now has a 100% chance to explode. It's not RNGesus anymore. It's a consistent guarantee. The old 90% explosion rate was the real RNGesus system.

This is not really a fruitful debate at all. We can hash out this argument, but where will that leave us?

Gauss rifle is still being balanced by the chance of explosion. It's just a guaranteed explosion if the weapon is critted, but we're still using RNG to determine whether the gauss rifle is hit and critted. It's not like Star Citizen, where you can disable a weapon by aiming your gun directly at that weapon. That's skill based. In MWO, we have RNG determining whether the weapon was hit. And unlike the AC10 or the PPC, the gauss rifle actually explodes.

But this is all beside the point I was making. I want more skill-based weapons, fewer auto-hit weapons and fewer weapons that rely on RNGeesus to determine whether they do 50 damage or 500 damage in a short period of time.

#28 FupDup

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 26,888 posts
  • LocationThe Keeper of Memes

Posted 19 December 2016 - 07:16 PM

View PostTristan Winter, on 19 December 2016 - 07:12 PM, said:

This is not really a fruitful debate at all. We can hash out this argument, but where will that leave us?

Gauss rifle is still being balanced by the chance of explosion. It's just a guaranteed explosion if the weapon is critted, but we're still using RNG to determine whether the gauss rifle is hit and critted. It's not like Star Citizen, where you can disable a weapon by aiming your gun directly at that weapon. That's skill based. In MWO, we have RNG determining whether the weapon was hit. And unlike the AC10 or the PPC, the gauss rifle actually explodes.

But this is all beside the point I was making. I want more skill-based weapons, fewer auto-hit weapons and fewer weapons that rely on RNGeesus to determine whether they do 50 damage or 500 damage in a short period of time.

That's a problem with the entire critical hit system, not the Clan Goose Waffle.

I myself would like a more consistent crit system like having all guns always crit, but most guns do piddly crit damage. Only "crit seekers" do good crit damage. In a perfect world we could even aim at specific components for our crits to go there (e.g. aim at the enemy Gauss Rifle geometry to damage or destroy it). Stuff like that. Won't ever happen though.

#29 Tristan Winter

    Member

  • PipPipPipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 2,530 posts
  • LocationNorway

Posted 19 December 2016 - 07:43 PM

View PostFupDup, on 19 December 2016 - 07:16 PM, said:

That's a problem with the entire critical hit system, not the Clan Goose Waffle.

It's the only weapon that explodes and potentially kills you when you lose it. This fact is part of the equation when PGI decides how to balance it. It was what prompted their recent decision to up the explosion chance from 90% to 100%. It was an attempt to balance the gauss rifle. They used the chance of random death instead of, say, modifying the weapon's cooldown value. My argument, obviously, is that this is not how they should go about it.

#30 Y E O N N E

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPip
  • The Nimble
  • The Nimble
  • 16,810 posts

Posted 19 December 2016 - 07:49 PM

To be fair, removing the explosion on the IS Gauss does let it get taken in a huge number of 'Mechs without confining them to running slow STD engines. That would be a pretty significant balance change.

#31 FupDup

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 26,888 posts
  • LocationThe Keeper of Memes

Posted 19 December 2016 - 07:50 PM

View PostTristan Winter, on 19 December 2016 - 07:43 PM, said:

It's the only weapon that explodes and potentially kills you when you lose it.

There are also ammo explosions, which are a 90% chance to feel nothing at all and a 10% chance to receive a massive 150-200 damage (guaranteed death unless you have Clan CASE). Ammo explosions have a much greater pendulum swing from one side to the other than the Goose Waffle.

I kind of wish that those ammo explosions were also 100% chance, but only dealt a tiny fraction of the damage they do now (somewhere around like 20%). I like consistent outcomes.


View PostTristan Winter, on 19 December 2016 - 07:43 PM, said:

This fact is part of the equation when PGI decides how to balance it. It was what prompted their recent decision to up the explosion chance from 90% to 100%. It was an attempt to balance the gauss rifle. They used the chance of random death instead of, say, modifying the weapon's cooldown value. My argument, obviously, is that this is not how they should go about it.

The idea of the IS having a more durable Gauss (or more durable equipment in general) has floated around the forums since the early days of the Clam invasion.

I will again say that on the Clan side, the amount of randomness involved was reduced by pushing the chance up to 100%. The result of what happens after your gun is critted out is more certain (now completely certain) than it was before. Their is less variability.

You can ask for lower explosion damage or something as compensation, but having a chance below 100% makes it more subject to RNGesus than not.

Also, PGI did try a slower cooldown, but it ended up making the weapon's damage over time dip too low and thus cause it to fall out of use.

Edited by FupDup, 19 December 2016 - 07:51 PM.


#32 Y E O N N E

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPip
  • The Nimble
  • The Nimble
  • 16,810 posts

Posted 19 December 2016 - 07:56 PM

View PostFupDup, on 19 December 2016 - 07:50 PM, said:

Also, PGI did try a slower cooldown, but it ended up making the weapon's damage over time dip too low and thus cause it to fall out of use.


You know there are values between that one and what it is now, right?

#33 Tristan Winter

    Member

  • PipPipPipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 2,530 posts
  • LocationNorway

Posted 19 December 2016 - 08:08 PM

View PostFupDup, on 19 December 2016 - 07:50 PM, said:

I kind of wish that those ammo explosions were also 100% chance, but only dealt a tiny fraction of the damage they do now (somewhere around like 20%). I like consistent outcomes.

I also like consistency in a game like this. If we're playing WoW or something, then I don't mind RNG, because it permeates the whole game. In MWO, it's much more specific to certain things.

View PostFupDup, on 19 December 2016 - 07:50 PM, said:

You can ask for lower explosion damage or something as compensation, but having a chance below 100% makes it more subject to RNGesus than not.

That's true. But it's now more of a factor in the game, because it happens more often. I think it's increasingly difficult for PGI to balance weapons when they add factors like jamming, crit explosions or easy-mode missile locks. In this particular game, I would like to eliminate as many such factors as possible, as much as possible.

View PostFupDup, on 19 December 2016 - 07:50 PM, said:

Also, PGI did try a slower cooldown, but it ended up making the weapon's damage over time dip too low and thus cause it to fall out of use.

When was the gauss rifle at its least popular? I personally always thought it should have been something of a niche weapon, being at the very extreme end of the scale when it comes to projectile speed and range. But most of the time that I've played MWO, it's been one of the more popular weapons.

I still think a lower cooldown is preferable to decreasing max range or projectile speed or anything like that. It should be a niche sniper weapon. But PGI doesn't see things that way, so now it's just a compromised multi-purpose ballistic weapon. It's not as popular for brawlers as it used to be, sure. But it's not like the Double Gauss, Double PPC Kodiak is helpless at close range either. Not as much as I'd like it to be, anyway.

#34 MadIrish

    Member

  • PipPipPipPipPip
  • The 1 Percent
  • 152 posts

Posted 21 December 2016 - 11:52 AM

View PostYeonne Greene, on 19 December 2016 - 12:57 PM, said:

Dice are fine, you need to manage your situational awareness. You know the capability is unreliable, so use your brain to avoid situations where you'll be SOL if your gun jams up.

It isn't like you can't fire the UAC with perfect 100% reliability if you need it.


I'm sorry I'm reading your last sentence as though you are trying to imply that the UAC5 doesn't jam so long as I use my latent telekinetic abilities to prevent it from jamming? Actually the UAC5 is never 100% reliable but I never asked for 100% reliability only for 100% consistency to allow the pilot to manage the fight rather than random O* S**t! moments. You will notice that in the test I fired 4 UAC5s 1 time with 1 click and jams unpredictably occurred, these jams last 6 seconds, so yes in a brawl it actually is never 100% reliable but it is also totally unmanageable.

#35 Levi Porphyrogenitus

    Member

  • PipPipPipPipPipPipPipPipPip
  • Mercenary
  • Mercenary
  • 4,763 posts
  • LocationAurora, Indiana, USA, North America, Earth, Sol, Milky Way

Posted 21 December 2016 - 03:14 PM

PGI is riding a very fine edge for balancing Ultras v standards, and I'm not quite sure that they get it right in all cases. That line is that standard ACs are utterly reliable, but that over sufficient time Ultras will have a higher net DPS (taking into account time spent jammed). This gives them distinct roles. Ultras are about two things: sustained output over the course of a match and burst output. Standards are about two other things: lower resource investment and guaranteed damage output.

If Standard ACs exceeded overall DPS on Ultras, then that would make standards the only reasonable choice. If Ultras guaranteed that they would always outperform Standards in a given moment then it would make them the only reasonable choice.

You can see some practical effects of this state when looking at the AC5 and UAC5. For single hardpoints, the AC5 is always the right choice. For numerous hardpoints, the UAC5 is always the right choice unless it would require excessive resource investment relative to the ability of the chassis to accommodate them.

The biggest problem we have right now is with Clan Ultras, since the Clan Standard AC is a place-holder for cLB-AC slugs and the relationship between cUACs and cLB-ACs is not the same as that between the AC5 and UAC5. This throws off the balance between the two, and I don't think that PGI has found the right relative performance values to make them work properly together. Of course, if they could come up with a way to add ammo selection to LB-ACs it might well make the problem much easier to solve.

#36 Corrado

    Member

  • PipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 817 posts
  • Locationfinale emilia, italy

Posted 21 December 2016 - 03:36 PM

View PostTristan Winter, on 19 December 2016 - 11:37 AM, said:

I think the whole x% chance of weapon not working for an extended period of time is a terrible way of balancing the UACs. It's almost as bad as balancing the gauss by increasing the chance of a catastrophic explosion that rips your mech apart. If you're going to give RNGeesus that much power, then you may as well introduce Cone of Fire. The difference between vomiting 500 damage in a few seconds and all guns jamming within the first few shots is just dumb luck.

It would be nice if the UAC jammed based on how hot your mech is, or based on a cooldown bar such as the one used for the flamer. This makes skill a greater factor and limits the importance of luck. In my mind, that's a good thing.


the purpose of RNG is to use the random factor to reduce the gap between good players and bad players. since the above average player numbers are way smaller than the below average, this helps the newcomers and casuals. asymmetric maps, RNG jams, RNG crits, RNG overheat damage. i don't mean we're close to WoT's disgusting levels but the road goes there.

#37 LordMelvin

    Member

  • PipPipPipPipPipPipPip
  • Urban Commando
  • Urban Commando
  • 567 posts

Posted 21 December 2016 - 03:40 PM

View PostLevi Porphyrogenitus, on 21 December 2016 - 03:14 PM, said:

PGI is riding a very fine edge for balancing Ultras v standards, and I'm not quite sure that they get it right in all cases. That line is that standard ACs are utterly reliable, but that over sufficient time Ultras will have a higher net DPS (taking into account time spent jammed). This gives them distinct roles. Ultras are about two things: sustained output over the course of a match and burst output. Standards are about two other things: lower resource investment and guaranteed damage output.

If Standard ACs exceeded overall DPS on Ultras, then that would make standards the only reasonable choice. If Ultras guaranteed that they would always outperform Standards in a given moment then it would make them the only reasonable choice.

You can see some practical effects of this state when looking at the AC5 and UAC5. For single hardpoints, the AC5 is always the right choice. For numerous hardpoints, the UAC5 is always the right choice unless it would require excessive resource investment relative to the ability of the chassis to accommodate them.

The biggest problem we have right now is with Clan Ultras, since the Clan Standard AC is a place-holder for cLB-AC slugs and the relationship between cUACs and cLB-ACs is not the same as that between the AC5 and UAC5. This throws off the balance between the two, and I don't think that PGI has found the right relative performance values to make them work properly together. Of course, if they could come up with a way to add ammo selection to LB-ACs it might well make the problem much easier to solve.

I was under the impression that standard ACs were for sustained dps while Ultras were for burst dps. In the time frame that Ultras would operate they would deal more damage than standard ACs but the engagement time had a cap. Whereas standard ACs could fire indefinitely (outside of heat and ammo constraints).

Pretty sure this has been suggested before but what if Ultras always had a faster fire rate but had an internal "jam bar" like miniguns tend to have in other FPSs? So you could rattle off shots faster than the standard AC (could be 7 ultra shells for 4-5 standard) but were limited on max duration (maybe 3-4 seconds sustained fire). The MASC system could be repurposed for this as a bar that increases as you uses it and decays when not in use. In a fight where you can stand on a hill and fire unmolested standard ACs will be better but in a brawl or hit-and-run where every second counts Ultras could burst out shots.

Of course the downside is this version of Ultras promotes more "peek and poke" gameplay where you rattle off a full jam worth of damage and then hide until you can do it again.

#38 Tristan Winter

    Member

  • PipPipPipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 2,530 posts
  • LocationNorway

Posted 21 December 2016 - 03:54 PM

View PostCorrado, on 21 December 2016 - 03:36 PM, said:

the purpose of RNG is to use the random factor to reduce the gap between good players and bad players. since the above average player numbers are way smaller than the below average, this helps the newcomers and casuals. asymmetric maps, RNG jams, RNG crits, RNG overheat damage. i don't mean we're close to WoT's disgusting levels but the road goes there.

My objection is mostly that they're doing it wrong. It's very hard to have the feature of internal components being taken out unless there's any random factors involved at all. I mean, what decides if it's your DHS or your 1 ton of ammo that is being critted? What bugs me is just the way some of these things are handled. Same as asymmetric maps. Asymmetric maps are not necessarily terrible, because you can have asymmetric maps that are balanced and are symmetrical in a kind of rock, paper, scissor situation. I.e. position A is vulnerable to attacks from B, position B is vulnerable to attacks from C, and position C is vulnerable to attacks from A. Of course, you don't want to make it a circle (yay Nascar), but you want to encourage the teams to move around the map to gain the advantage, whilst fighting. PGI's maps don't generally do that.

#39 MadIrish

    Member

  • PipPipPipPipPip
  • The 1 Percent
  • 152 posts

Posted 24 December 2016 - 12:34 AM

I'm strictly talking about the IS UAC5, the only UAC the IS has in its arsenal and the only one it will ever have from what I've heard. The Clan UACs can spit out non-stop projectiles and with multiple UACs equipped so many so that you can't see...extremely imbalanced. I spectated many Clan Mechs with 4 UACs spitting out enough projects to make an A10 jealous.

Edited by MadIrish, 24 December 2016 - 12:35 AM.


#40 The Lobsters

    Member

  • PipPipPipPipPipPip
  • The Clamps
  • The Clamps
  • 269 posts
  • LocationLocation Location.

Posted 24 December 2016 - 02:55 AM

One thing to say in support of the RNGjam is that it prevents fire control macros from giving a jam advantage. I too think that jam chance as a product of some other mech value would create more rewarding play. It would get ruined by macro players though, so RNG is the lesser of two evils in this case.





33 user(s) are reading this topic

0 members, 33 guests, 0 anonymous users