534 replies to this topic

[Allekatrase]

MustrumRidcully, on 22 November 2012 - 02:56 AM, said:

I don't agree that you can ignore Rate of Fire differences. DPS is important, but how you deliver your DPS is also important.

But there are ways to try to weigh DPS based on the rate of fire or rather the single shot damage of a weapon.

For example, you could use somethnig like this:
DPS_{alphamodified} = (Damage_Shot + 5 * DPS) / 6.
Now weapons with a low alpha value but the same DPS will have a lower value, and weapons with a high alpha value will have a higher value.

Yes but is that something you'd want to do? In some cases higher alpha is more desireable but in others it's not. In our arbitrary examples of two mechs standing and shooting each other it's likely that the weapon with a higher rate of fire will win given both weapons have the same dps because the more steady application of damage will hit the kill mark first whereas the high alpha weapon is likely to have more overkill damage and be between shots. For example, the AC/2 vs the AC/10 (not to say that these two weapons are all that comparable).

The ROF difference makes their use different, but I don't think it makes one better than the other overall.

[MustrumRidcully]

Allekatrase, on 22 November 2012 - 04:48 AM, said:

Yes but is that something you'd want to do? In some cases higher alpha is more desireable but in others it's not. In our arbitrary examples of two mechs standing and shooting each other it's likely that the weapon with a higher rate of fire will win given both weapons have the same dps because the more steady application of damage will hit the kill mark first whereas the high alpha weapon is likely to have more overkill damage and be between shots. For example, the AC/2 vs the AC/10 (not to say that these two weapons are all that comparable).

The ROF difference makes their use different, but I don't think it makes one better than the other overall.

It depends on many factors, but it could also very well be that even ina s traight brawl with no firing pauses ever the low ROF weapon will prevail. Low ROF enables you to "front-load" your damage as well.

After 3.5 seconds, an AC/2 will have dealt 16 damage, and a Gauss will have dealt 15 damage. So at this point, the AC/2 has gained the advantage. But 0.5 seconds later, the AC/2 will have dealt 18 damage, but the Gauss suddenly jumped up to 30 damage! If both mechs have equal (internal) armour at the start of this engagement, suddenly the Gauss has the advantage if it you needed more than 18 but equal or less than 30 damage to kill the target.

Basicaly, the minimum total damage damage advantage the AC/2 must have accumlated to leave the Gauss rifle behind forever would need to be 15 damage. With a DPS of 3.75 vs 4, that would require 60 seconds of uninterrupted fire. That's quite long, and would equal 240 damage from the AC/2 and
225 damage from the Gauss Rifle. That basically means you need to kill your enemy twice with an AC/2 before its full DPS advantage is realized...

Edited by MustrumRidcully, 22 November 2012 - 05:25 AM.

[Apoc1138]

Allekatrase, on 22 November 2012 - 02:32 AM, said:

The mathematical models ARE the game. Computers are just calculators. It's all just math. When looking at balance you have to simplify it to as few variables as possible. Your supposed examples of cover and ROF are also strawmen and your anecdotal evidence that it works for you is irrelevant.

How do you factor cover into balance? What is the value, given equal cover opportunities, of increased rate of fire and how do you balance that? It's an unnecessary complication with weapons that opperate in a similar fashion. If you were comparing to LRMs then the mechanics surrounding them would be a bigger element. But we're not. We're comparing direct fire weapons.

PGI's mathmateical models DO include these other factors though, based on live server telemetry.
How about EMP effects or screen shake, how do you balance these? These are also removed from the OP's and Vapor Trails models. Mathematical models are great IF they are accurate... however the vast majority of the "maths" being thrown around on this forum are fundamentally flawed and don't account for many important gameplay factors, or try to force energy weapons to be heat neutral over arbitrarilly long time periods purely due to the prejudices of the person who wrote them.

Ignore them when choosing weapons if you so choose, but don't expect PGI to ignore them as well.

MustrumRidcully, on 22 November 2012 - 05:20 AM, said:

It depends on many factors, but it could also very well be that even ina s traight brawl with no firing pauses ever the low ROF weapon will prevail. Low ROF enables you to "front-load" your damage as well.

After 3.5 seconds, an AC/2 will have dealt 16 damage, and a Gauss will have dealt 15 damage. So at this point, the AC/2 has gained the advantage. But 0.5 seconds later, the AC/2 will have dealt 18 damage, but the Gauss suddenly jumped up to 30 damage! If both mechs have equal (internal) armour at the start of this engagement, suddenly the Gauss has the advantage if it you needed more than 18 but equal or less than 30 damage to kill the target.

Basicaly, the minimum total damage damage advantage the AC/2 must have accumlated to leave the Gauss rifle behind forever would need to be 15 damage. With a DPS of 3.75 vs 4, that would require 60 seconds of uninterrupted fire. That's quite long, and would equal 240 damage from the AC/2 and
225 damage from the Gauss Rifle. That basically means you need to kill your enemy twice with an AC/2 before its full DPS advantage is realized...

and the AC/2 weighs 6 tons and the Gauss weighs 15 tons, so you could easily fit 2 AC/2's, and ammo AND a couple of heatsinks for the same cost of a single gauss... and then you've got 8DPS which would very quickly overtake the gauss and may well be a good choice for a brawler and harraser, you also get sustained shake on target with the 2's

[Terror Teddy]

Apoc1138, on 22 November 2012 - 06:14 AM, said:

PGI's mathmateical models DO include these other factors though, based on live server telemetry.
How about EMP effects or screen shake, how do you balance these? These are also removed from the OP's and Vapor Trails models. Mathematical models are great IF they are accurate... however the vast majority of the "maths" being thrown around on this forum are fundamentally flawed and don't account for many important gameplay factors, or try to force energy weapons to be heat neutral over arbitrarilly long time periods purely due to the prejudices of the person who wrote them.

Ignore them when choosing weapons if you so choose, but don't expect PGI to ignore them as well.



and the AC/2 weighs 6 tons and the Gauss weighs 15 tons, so you could easily fit 2 AC/2's, and ammo AND a couple of heatsinks for the same cost of a single gauss... and then you've got 8DPS which would very quickly overtake the gauss and may well be a good choice for a brawler and harraser, you also get sustained shake on target with the 2's

By then it comes down to movement and battlefield conditions and a difference between alpha strikes and sustained damage over time.

[MustrumRidcully]

Apoc1138, on 22 November 2012 - 06:14 AM, said:

PGI's mathmateical models DO include these other factors though, based on live server telemetry.
How about EMP effects or screen shake, how do you balance these? These are also removed from the OP's and Vapor Trails models. Mathematical models are great IF they are accurate... however the vast majority of the "maths" being thrown around on this forum are fundamentally flawed and don't account for many important gameplay factors, or try to force energy weapons to be heat neutral over arbitrarilly long time periods purely due to the prejudices of the person who wrote them.

Ignore them when choosing weapons if you so choose, but don't expect PGI to ignore them as well.



and the AC/2 weighs 6 tons and the Gauss weighs 15 tons, so you could easily fit 2 AC/2's, and ammo AND a couple of heatsinks for the same cost of a single gauss... and then you've got 8DPS which would very quickly overtake the gauss and may well be a good choice for a brawler and harraser, you also get sustained shake on target with the 2's

If you account for heat sink and ammo requirements so that you can actually fire just for 20 seconds, the AC/2 is less Damage to Tonnage efficient than the Gauss Rifle:



That means - you will spend more weight to run 2 AC/2s than you would spend to run 2 Gauss Rifles for 20 seconds. In other words, your "couple of heat sinks" is actually quite a lot. A AC/2 after all produces 2 heat per second, while a Gauss produces 0.25 per second.

(If you want details on how these charts are assembled, follow the weapon balance link in my signature).

Edited by MustrumRidcully, 22 November 2012 - 07:04 AM.

[Apoc1138]

Terror Teddy, on 22 November 2012 - 06:28 AM, said:

By then it comes down to movement and battlefield conditions and a difference between alpha strikes and sustained damage over time.

. I agree, and it is difficult to model mathematically... luckily pgi don't need to think for hours on how to model it as they have their own private army of lab rats randomly testing for them to account for all these extra, difficult to model variables... hence why they are making changes that they think will lead to both better balance and also niches for weapons to fill

MustrumRidcully, on 22 November 2012 - 07:03 AM, said:

If you account for heat sink and ammo requirements so that you can actually fire just for 20 seconds, the AC/2 is less Damage to Tonnage efficient than the Gauss Rifle: That means - you will spend more weight to run 2 AC/2s than you would spend to run 2 Gauss Rifles for 20 seconds. In other words, your "couple of heat sinks" is actually quite a lot. A AC/2 after all produces 2 heat per second, while a Gauss produces 0.25 per second. (If you want details on how these charts are assembled, follow the weapon balance link in my signature).

. and what about if you include the 20 free engine heatsinks? I prefer your maths to vapour trails, however it is still incomplete... 20 seconds of firing twin ac2s would be 160 damage by the way, vs 75 for the gauss

Edited by Apoc1138, 22 November 2012 - 07:09 AM.

[Allekatrase]

Apoc1138, on 22 November 2012 - 07:06 AM, said:

. I agree, and it is difficult to model mathematically... luckily pgi don't need to think for hours on how to model it as they have their own private army of lab rats randomly testing for them to account for all these extra, difficult to model variables... hence why they are making changes that they think will lead to both better balance and also niches for weapons to fill

. and what about if you include the 20 free engine heatsinks? I prefer your maths to vapour trails, however it is still incomplete... 20 seconds of firing twin ac2s would be 160 damage by the way, vs 75 for the gauss

That was firing two gauss rifles meaning 150 damage for the gauss.

[Indoorsman]

Asatruer, on 21 November 2012 - 10:09 AM, said:

Cover does indeed happen, but cover is not a factor of a weapon, nor a factor of a mech, Cover is a factor of the map, and the use of cover is a factor of the Pilot. When trying to balance weapons, or mechs, you have to grant cover equally to all parties, just like you have to assume equal pilot skill, because otherwise what you are doing is confounding the results of weapon balance with the factors of map or pilot balance. If you runs a series of "tests" where both parties have no cover, equal cover, one has cover, and then the other has cover you can get actual results related to weapon and mech balance. Running the tests where only one mech has cover skews the results to show that having cover is better than not having cover.

The whole reason I brought cover up is because that is the only mathematical way that PPC can defeat Gauss. To ignore that seems stupid. To ignore ECM/increased screen shake seems stupid too. The balance between these 2 weapons isn't a flat out 3.75>3.33 comparison of dps, or a tonnage efficiency comparison either. There are many factors at play and when everyone is calling for PPC buffs based soley on the current implementation of it w/o ECM then the current ONLY way it is better than Gauss needs to be considered. Which is if it has cover and Gauss does not, or RoF. As you said that's not a fair comparison by itself, hence the other 3 tests.

[MustrumRidcully]

Apoc1138, on 22 November 2012 - 07:06 AM, said:

. I agree, and it is difficult to model mathematically... luckily pgi don't need to think for hours on how to model it as they have their own private army of lab rats randomly testing for them to account for all these extra, difficult to model variables... hence why they are making changes that they think will lead to both better balance and also niches for weapons to fill

. and what about if you include the 20 free engine heatsinks? I prefer your maths to vapour trails, however it is still incomplete... 20 seconds of firing twin ac2s would be 160 damage by the way, vs 75 for the gauss

I am thinking about more ways to model weapon balance. But nothing specific yet.

But one thought just in: THe Gauss also gets the free heat sinks, but doesn't need them itself- for just one or two meager tons, you could add additional weapons that would utilize those heat sinks as well - and now you suddenly have to consider scenarios where these weapons also come to bear... Of course, on a K2, you#re strapped for everything if you equip 2 Gauss, but on a heavier mech (Cataphract?) you could more easily afford this weight.

One alternate scenario I had in mind though was to actually consider something like "loadouts" with a weapon that try to hit certain minimum DPS requirements and then compare the weight cost for that. This would allow accounting for the (more or less) free engine heat sinks.

Edited by MustrumRidcully, 22 November 2012 - 02:00 PM.

[Apoc1138]

Allekatrase, on 22 November 2012 - 07:57 AM, said:

That was firing two gauss rifles meaning 150 damage for the gauss.

you are comparing 2 AC/2's at 12 tons to 2 gauss rifles at 30 tons? umkay...

MustrumRidcully, on 22 November 2012 - 01:59 PM, said:

I am thinking about more ways to model weapon balance. But nothing specific yet.

But one thought just in: THe Gauss also gets the free heat sinks, but doesn't need them itself- for just one or two meager tons, you could add additional weapons that would utilize those heat sinks as well - and now you suddenly have to consider scenarios where these weapons also come to bear... Of course, on a K2, you#re strapped for everything if you equip 2 Gauss, but on a heavier mech (Cataphract?) you could more easily afford this weight.

but equally... with 20 free in engine DHS', the 2 AC/2's are MASSIVELY out DPS'ing the single gauss for the same weight... you can still add more weapons and heatsinks to either mech but the one with 2 AC/2's instead of the gauss is ALWAYS going to out DPS the single gauss plus the same type of weapons - even if the AC/2 boat has to use less of that weapon to fit heatsinks to maintain fire for 20 seconds

the gauss is better for a long range sniper, the 2 AC/2's are better for a brawler but can also be used for plinking at longer ranges (actually longer ranges than the gauss even)

in this scenario the gauss actually benefits from cover reducing the time the AC/2's have to be landing blows while the gauss reloads

something like the cataphract could fit 4 AC/2's easily, and enough heatsinks to last longer than 20 seconds of sustained fire... or it can fit 2 gauss rifles... again, horses for courses, the gauss is a better sniper, but 4 AC'2s will be chucking out 16DPS vs. 7.5DPS... even allowing for gauss being easier to hit the same location with more damage, a more than double the DPS makes up for that, IMO

this is called balance... the gauss is not OP because there are plenty of weapons that are better than it in situations other than where the gauss is best... some weapons are currently UP (but not by much) and PGI have a proven track record and statements that weapon balance is on going and each adjustment they make gives each weapon a new twist

I don't want all weapons to be balanced on a damage per ton basis, there are too many gameplay factors involved that would end up making those weapons OP if everything was balanced based on too simple of a factoring

Edited by Apoc1138, 22 November 2012 - 05:51 PM.

[Vermaxx]

DPS is a terrible measure for anything other than rote math. I hate people using DPS in games like this, instead of the mmo games where it really matters - burning down a boss before the hard wipe timer.

DPS is only good for comparing things on a set plane. It has its place in 'gross' balancing changes, but fine tuning shouldn't worry about DPS over all other factors.

The AC2 has amazing "DPS" as compared to a gauss, but a single AC2 is not by any means a stellar performing gun. It certainly can pump out more raw damage than a gauss - but you miss some and definitely hit all over, and suddenly the math is less impressive. The gauss (when it hits, but that's the same issue as an AC2) applies 15 damage to one location.

[Asatruer]

Vermaxx, on 22 November 2012 - 05:56 PM, said:

DPS is a terrible measure for anything other than rote math. I hate people using DPS in games like this, instead of the mmo games where it really matters - burning down a boss before the hard wipe timer.

DPS is only good for comparing things on a set plane. It has its place in 'gross' balancing changes, but fine tuning shouldn't worry about DPS over all other factors.

I guess you hate the Devs then, because DPS is going to be one of their main ways to communicate item information to players.

[Apoc1138]

Asatruer, on 22 November 2012 - 07:55 PM, said:

I guess you hate the Devs then, because DPS is going to be one of their main ways to communicate item information to players.

for brawlers, a dps of 8 vs 4 would be a very important factor... I would hope that players building a more balanced build would be intelligent enough to work out for themselves what actually works in a particular role

[MustrumRidcully]

Vermaxx, on 22 November 2012 - 05:56 PM, said:

DPS is a terrible measure for anything other than rote math. I hate people using DPS in games like this, instead of the mmo games where it really matters - burning down a boss before the hard wipe timer.

DPS is only good for comparing things on a set plane. It has its place in 'gross' balancing changes, but fine tuning shouldn't worry about DPS over all other factors.

The AC2 has amazing "DPS" as compared to a gauss, but a single AC2 is not by any means a stellar performing gun. It certainly can pump out more raw damage than a gauss - but you miss some and definitely hit all over, and suddenly the math is less impressive. The gauss (when it hits, but that's the same issue as an AC2) applies 15 damage to one location.

DPS matters, but also how well yo ucan deliver it. That#s why I suggest using a weighted DPS value that takes into account damage per shot respectively rate of fire. 4 DPS delivered in 0.25 seconds packages is worse than 4 DPS delivered in 2 second packages.

I still maintain it's a fallacy to assume that MMORPGs don't care about such distinctions, either. When that MMORPG boss is running away from you (or throws you away from him) it matters quite a lot fhwether you deal your damage every 0.5 seconds or every 4 seconds. And we're not even talking about the healfests Startrek Online has in PvP, where it's always crucical to deliver a lot of burst DPS so you can break through the healers...
Or how the TOR designers noticed that "on paper", some classes seemed balanced, but the usability of the potential damage was lower for one class, making it more difficult to play or less likely to actually reach its designed performance level.

But yet, in all these games... DPS isn't a useless statistic. It has quite an impact. But when you get within 10-20 % of the same DPS, it really becomes more useful to compare damage per shot and err on the lower recycle time weapons.

Barring extreme differences, like one weapon fires every 0.5 seconds and the other every 20. (Say, the Clan Uber-Duber-Heavy Gaus Rifle that deals 60 damage ever 20 seconds...)

Edited by MustrumRidcully, 22 November 2012 - 11:47 PM.

[Indoorsman]

MustrumRidcully, on 22 November 2012 - 11:44 PM, said:

4 DPS delivered in 0.25 seconds packages is worse than 4 DPS delivered in 2 second packages.

Is this objective or subjective? One of the reasons I personally don't like the AC20 much is because if you miss your DPS spikes DOWN. The effect of missing w/an AC2 is hardly noticeable though. 4 DPS being better in longer/larger packets is an assumption/preference. So balancing the game that way would be biased torwards those who prefer their weapons to fire slow.

like you kinda said:

MustrumRidcully, on 22 November 2012 - 11:44 PM, said:

it's a fallacy to assume

:-p

[MustrumRidcully]

Indoorsman, on 23 November 2012 - 06:43 AM, said:

Is this objective or subjective? One of the reasons I personally don't like the AC20 much is because if you miss your DPS spikes DOWN. The effect of missing w/an AC2 is hardly noticeable though. 4 DPS being better in longer/larger packets is an assumption/preference. So balancing the game that way would be biased torwards those who prefer their weapons to fire slow.

like you kinda said:

:-p

It is a bit situational. At long range my experience suggests that you're better off with slow firing weapons. Unfortnately, let's face it - with the current net code, we probably all hit worse with ballistics than we would if this game was not an online game. Do we need to account for this when balancing weapons?

I don't know. In my opinion, if we do, we've basically given up on MW:O as a good online game. They must fix their net code.

Okay, but that aside - yes, I think in general there are objective measurements to consider when it comes to ROF.

Take your AC/2 vs AC/20. When you get your target in your cross-hair the first time, and hit, you deal 20 damage in one blow. It will take the AC/2 fired at the same time 4.5 seconds to deliver the same damage - and at the 4tth second, you are already firing your second AC20 shot, so now it's another 4.5 seconds to catch up. Of course, the AC20 has a higher DPS than the AC/2, so it's not quite fair.

Let's imagine an AC/16 fired at the rate of the AC/20.
It will take 3.5 seconds for the AC/2 to catch up to the AC/16's first shot. And 0.5 seconds later, it will have to play catch up again.

UNless you simply miss more often with the AC/16 than the AC/2, the AC/16s is better.

Now ,this "Unless" could be something very subjective, in which case I wouldn't consider it much for discussing weapon balance - but it could also be an objective element. For example AC/2s currently fly faster than AC/20s projectiles. That gives the AC/2 an advantage in hitting, as you have to lead less, which means any errors in your judgement on the enemies movement between time of shot and time to target will have less effect.

I leave quantifying all this as an exercise to the reader.

Edited by MustrumRidcully, 23 November 2012 - 07:02 AM.

[Indoorsman]

MustrumRidcully, on 23 November 2012 - 07:01 AM, said:

It is a bit situational. At long range my experience suggests that you're better off with slow firing weapons. Unfortnately, let's face it - with the current net code, we probably all hit worse with ballistics than we would if this game was not an online game. Do we need to account for this when balancing weapons?

I don't know. In my opinion, if we do, we've basically given up on MW:O as a good online game. They must fix their net code.

Okay, but that aside - yes, I think in general there are objective measurements to consider when it comes to ROF.

Take your AC/2 vs AC/20. When you get your target in your cross-hair the first time, and hit, you deal 20 damage in one blow. It will take the AC/2 fired at the same time 4.5 seconds to deliver the same damage - and at the 4tth second, you are already firing your second AC20 shot, so now it's another 4.5 seconds to catch up. Of course, the AC20 has a higher DPS than the AC/2, so it's not quite fair.

Let's imagine an AC/16 fired at the rate of the AC/20.
It will take 3.5 seconds for the AC/2 to catch up to the AC/16's first shot. And 0.5 seconds later, it will have to play catch up again.

UNless you simply miss more often with the AC/16 than the AC/2, the AC/16s is better.

Now ,this "Unless" could be something very subjective, in which case I wouldn't consider it much for discussing weapon balance - but it could also be an objective element. For example AC/2s currently fly faster than AC/20s projectiles. That gives the AC/2 an advantage in hitting, as you have to lead less, which means any errors in your judgement on the enemies movement between time of shot and time to target will have less effect.

I leave quantifying all this as an exercise to the reader.

So basically AC2 is more accurate and should hit more often, negating its' lower deeps. It lacks the burst, and the spikeness when you miss, of the AC20. Also, the burst dmg of the AC20 is considered a positive, when you miss you not only decrease your dps but your burst too. AC2 has screen shake which can be its' counter-positive I guess. Miss once w/the AC2 and it's not really going to decrease your suppression any.

In conclusion: you can't, 100%, math out weapons balance. Yay for trial and error.

But! If you were going to do something where 4 dps every .25 seconds is worse than 4 dps every 2 seconds, then... Wouldn't either the 4 dps every .25 seconds weapon need higher dps to make its' "relative"/perceived dps the same? Or the 4 dps every 2 seconds weapon have its' dps decreased? That's the problem. And then you just buffed the 4 dps every .25 seconds weapon that some people already preferred or you nerfed the 4 dps every 2 seconds weapon that most people supposedly prefer. The result would affect popularity of the weapons, but may not actually balance them out. Of course if heat, weight, size, indirect/direct, screen shake, ammo/no ammo, instagib/dot, spread/pinpoint, ecm, crit seeking, heat generation for enemy, etc... aren't enough ways to diversify weapons then sure, try hidden variables/formulas/backend things to force people to use certain weapons.

[MustrumRidcully]

While I maintain my belief that same DPS is usually better with a lower rate of fire than a higher rate of fire -if you don't agree with that, maybe we should first get on balancing weapons based on less subtle differences, and focus on the big differences - like one weapon needing 15 tons of heat sinks in addition to 7 tons of its own weight to deliver 3 DPS for 30 seconds, and another weapon needing when another needs only 15 tons of its own weight and 3 tons for its ammo to fire 30 seconds for 3.75?

[Indoorsman]

MustrumRidcully, on 23 November 2012 - 08:22 AM, said:

While I maintain my belief that same DPS is usually better with a lower rate of fire than a higher rate of fire -if you don't agree with that, maybe we should first get on balancing weapons based on less subtle differences, and focus on the big differences - like one weapon needing 15 tons of heat sinks in addition to 7 tons of its own weight to deliver 3 DPS for 30 seconds, and another weapon needing when another needs only 15 tons of its own weight and 3 tons for its ammo to fire 30 seconds for 3.75?

It's not whether I agree that lower RoF is better or not, it's that even if I do it's preference.

[Asatruer]

Indoorsman, on 23 November 2012 - 08:06 AM, said:

If you were going to do something where 4 dps every .25 seconds is worse than 4 dps every 2 seconds, then... Wouldn't either the 4 dps every .25 seconds weapon need higher dps to make its' "relative"/perceived dps the same?

While thinking about how the MG can be made to be as worthwhile as it should, this has been something that I have thought. Even though in TT the MG does 2/3rds the DPS of the Small Laser (that would be 0.666 DPS here, rather than the 0.4 we have), I think here it should be doing closer to, if not the same, DPS as the SL. The SL is doing its 1 DPS over the course of 3/4th of a second once every 2.25 seconds (or does cooldown start after beam duration ends?) and with the MG needing the full attack cycle of the SL to do 2/3rds the damage, the MG is more likely to have its damage spread out over multiple locations of the target, or interrupted by cover or the inability to either leg or torso twist fast enough to keep on the target. Seems only fair to buff that to a close to equal amount of DPS. Maybe the zero heat thing is a big enough advantage, but I think the ammo explosion risks make up for that


That said, I do not think the AC/2 should come close to the DPS of the AC/20, nor should it be close to the DPS of the AC/10...