137 replies to this topic

[Jabilo]

Renthrak, on 11 May 2013 - 01:41 PM, said:

I voted YES, but I don't think the AC/5 is as useless as people seem to think it is.

I actually ran the numbers for ballistic weapons, comparing damage per ton of ammo (shot damage X number of shots in a ton), damage per ton of the weapon itself including heat sinks required to dissipate all the heat produced within the recycle rate of the weapon, and numerous other factors to determine the overall 'efficiency' of the weapon. The results were rather surprising.

The top two most efficient ballistic weapons are the Gauss Rifle (no surprise here), and the AC/5 (???????). It turns out that based on the tonnage and ammo required to use ballistics, the AC/5 is only second to the Gauss Rifle because the GR produces so little heat.

This doesn't seem to make sense based on the mediocre stats of the AC/5 itself, but basically, carrying an AC/10 or /20 costs more weight(weapon+ammo+HS) than the comparable increase in damage. One of the biggest things holding the AC/5 back in actual gameplay is the current prevalence of massive Alpha Strikes as the primary means of offense.

For reference, the AC/2 has higher flat damage output, but requires so many heatsinks that it ends up heavier than the AC/5 to achieve that level of performance.

Interesting! Supports my calculation above that on paper at least the AC5 after the proposed change could be a balanced weapon.

I think the argument here is one of "feel" against maths.

In the current meta the AC5 is less effective not because of stats, but because range (AC2), higher burst (UAC5) and higher alphas on a single location (AC20) are more effective in game play terms right now.

The problem is you can not balance weapon stats on the current meta flavour of the month as this is always liable to change.

Example: If greater heat penalties are introduced then suddenly PPCs and AC2s fall of the radar. If you have buffed the hell of of AC5s in the meantime then you have a problem...

Edited by Jabilo, 11 May 2013 - 01:55 PM.

[Trauglodyte]

The AC5 rof needs to be tweaked to be equivalent to that of the UAC5 (taking into consideration the jam rates). I'm fine ignoring the jam rate and just dropping it to 1.25 so it has the same 4 DPS as all of the other autocannons. I find it rather strange that PGI is messing with the weapon ranges when they haven't done such a thing thus far in the name of balance. The ROF is where it is getting hosed!

[megoblocks]

Huntsman, on 11 May 2013 - 08:40 AM, said:

While the ac20 is impressive for its massive single hit, and it certainly a good weapon, put 4 ac5s onto a 4x and in alphastrike you're dishing out that 20 damage every 1.7 seconds rather than every 4 seconds and you can stick 2 ml in there to boot along with max engine.

That 4 ac5 setup also weighs 32 tons and takes up 16 slots as opposed to 10 slots and 14 tons of the AC20. DPS / ton for the ac5 is .3675 and DPS / ton for the AC 20 is .35. Not much of an edge IMHO, and its better to slam 20 pts (or 40) into one location all at once.

[Renthrak]

megoblocks, on 11 May 2013 - 03:18 PM, said:

That 4 ac5 setup also weighs 32 tons and takes up 16 slots as opposed to 10 slots and 14 tons of the AC20. DPS / ton for the ac5 is .3675 and DPS / ton for the AC 20 is .35. Not much of an edge IMHO, and its better to slam 20 pts (or 40) into one location all at once.

If you include the number of shots that you can make with 1 ton of ammo, the gap between AC/20 and AC/5 widens considerably.

[Strum Wealh]

Levi Porphyrogenitus, on 11 May 2013 - 08:11 AM, said:

See the AC section of my post here: http://mwomercs.com/...ool-down-rates/

AC dps should be regularized around the 5 dps point. AC20s are there already with 20 damage every 4 seconds. AC10s would need to be sped up to 10 damage every 2 seconds, and AC5s moved to 5 damage every 1 second. The AC2 would be the odd weapon out, at its current 0.5s cool down, giving it only 4 dps, but considering its absurd range profile and super high velocity projectiles I'd be fine with that slight discrepancy.

Of course, all this assumes that ACs are being balanced not just with each other but with other weapon systems, with the idea being to give each class of weapon its own role and flavor.

Generally agree... but, why not normalize to 4 DPS?
AC/2 recycle @ 4 DPS: 0.50 seconds
AC/5 recycle @ 4 DPS: 1.20 seconds
AC/10 recycle @ 4 DPS: 2.50 seconds
AC/20 recycle @ 4 DPS: 5.00 seconds

This puts the AC/5 very close to the base ROF of the current UAC/5 (1.2 seconds, vs 1.1 seconds; the 0.10 second difference is smaller than the average reaction time of Olympic-caliber sprinters, at 0.16-0.19 seconds, so no player who isn't specifically looking for it - and employing chronometers - is going to notice a substantial difference in a small enough timeframe to matter), leaves the AC/2 and AC/10 and LB 10-X as they are (which seem to be reasonable states, in terms of theoretical DPS), and imparts a negligible change on the AC/20, while also keeping the potential DPS of boated ACs at a more manageable level.

Notably, normalizing the ACs at 4 DPS still results in all of the AC models having a theoretical DPS capability that exceeds that of the vaunted Gauss Rifle (at 3.75 DPS), Standard PPC (at 3.33 DPS), ER-PPC (also at 3.33 DPS), and Streak SRM-2 (at a mere 0.86 DPS).

Also of note is the distinction between "theoretical DPS" (what the weapon itself is capable of, under ideal conditions) and "actual DPS" (which is affected by the capability - or lack thereof - of the individual user).

Lefty Lucy, on 11 May 2013 - 08:39 AM, said:

Why should an 8 or 6 ton weapon have the same DPS as an AC20? I understand that faster firing rate is not necessarily a straight-forward buff because alpha damage is often more important than damage over time, but I also don't see any reason why an 8 ton weapon with much better range and shot velocity should have the same DPS as a 14 ton weapon. It does need a faster firing rate, but IMO not that much.

Strictly-speaking, an AC/2 should have a DPS of 0.2 (2 units of damage per 10 second interval (the duration represented by a standard TT "turn")), with the AC/20 having a DPS of 2.0 (20 units of damage per 10-second interval) - that, in fact, is essentially what the numbers used in the designation actually mean.

However, the Devs implemented it as "per-salvo damage" rather than the original "damage per unit time", which meant establishing a new metric for the latter.

Giving the AC/2 anf the AC/20 the same theoretical DPS merely means that two players with perfect aim that are within the effective range of both weapons on open terrain should take approximately the same amount of time to defeat one another, with victory being a 50/50 probability.
In practice, of course, the two weapons fill very different niches - the AC/2 is meant to sand the armor off of an opponent bit-by-bit (preferably from long range), while the AC/20 is meant to cause concern for losing limbs in one or two salvos (albeit only at short range); it's somewhat analogous to the difference in threat between a single piranha and a single great white shark - both will make you just as dead just as quickly they hit/bite the right spot(s), but the degree of damage done in one shot/bite differs wildly.

[Levi Porphyrogenitus]

Strum Wealh, on 11 May 2013 - 06:25 PM, said:

Generally agree... but, why not normalize to 4 DPS?
AC/2 recycle @ 4 DPS: 0.50 seconds
AC/5 recycle @ 4 DPS: 1.20 seconds
AC/10 recycle @ 4 DPS: 2.50 seconds
AC/20 recycle @ 4 DPS: 5.00 seconds

This puts the AC/5 very close to the base ROF of the current UAC/5 (1.2 seconds, vs 1.1 seconds; the 0.10 second difference is smaller than the average reaction time of Olympic-caliber sprinters, at 0.16-0.19 seconds, so no player who isn't specifically looking for it - and employing chronometers - is going to notice a substantial difference in a small enough timeframe to matter), leaves the AC/2 and AC/10 and LB 10-X as they are (which seem to be reasonable states, in terms of theoretical DPS), and imparts a negligible change on the AC/20, while also keeping the potential DPS of boated ACs at a more manageable level.

Notably, normalizing the ACs at 4 DPS still results in all of the AC models having a theoretical DPS capability that exceeds that of the vaunted Gauss Rifle (at 3.75 DPS), Standard PPC (at 3.33 DPS), ER-PPC (also at 3.33 DPS), and Streak SRM-2 (at a mere 0.86 DPS).

There are a couple of reasons why I think that the higher dps number is better for AC. First, dps isn't everything. It is a useful metric, but it must be informed by a whole slew of other factors, such as range, heat, ammo consumption, weight, critical space, etc.

Second, role matters. I know plenty of people will disagree with this, but I see different weapons as having distinct and separate roles. For instance, PPCs are medium-long range sniper weapons. ERPPCs are long range sniper weapons. The Gauss is a long range sniper weapon. SRMs are a close-range brawling weapon. LRMs are a medium-long range fire support weapon capable of indirect fire but at its best when used to hit somebody who is too busy fighting to evade. Streaks are a close-in backup weapon that serves either as an efficiency choice (very accurate, so doesn't waste ammo) or as a close-in-weapon-system to provide point defense against nimble enemies who get too close. Standard lasers are a workhorse option for short to medium range, while er lasers are medium- or long-range direct fire support option and pulse lasers are a short- to medium-range brawling weapon.

Autocannons have their own role too. The light ACs are direct fire support weapons. The AC2 is an extreme-range annoyance that really adds up the damage over a sustained barrage. The AC5 is a long-range fire support option not meant for point-damage sniping but rather for sustained damage over the course of an engagement. The heavy ACs are brawling weapons. The AC10 should be a short- or medium-range brawling weapon that gives decent point damage and decent sustained damage with an ok ability to add some reach to your arsenal. The AC20 should be the premier brawling option, with unmatched point damage and strong sustained damage but at a huge cost in ammunition and range.

For ACs to fulfill the roles that I see them as having, they need higher dps than other options. They are restricted pretty heavily with ammo (I've seen Jagermechs with 750 AC2 rounds run out with half the enemy team still in the game, and I've run out of AC20 ammo many times while packing three tons or 21 shots), and the ballistic trajectory means they aren't going to be as accurate as many other options (even the AC2 can be harder to aim than the PPC, since PPC blasts don't drop while AC2 rounds do). Also, the ones with ranges that can compete with the PPC, erPPC, and Gauss completely lack point-damage capability. They don't do lump-sum damage, they do sustained damage, and they very much reward people who are able to keep the shots hitting on the same target consistently (adding to the skill cap).

The PPC meanwhile does less damage over time, but it does its damage all in a lump, has great range, and can fire forever (unless you overheat and your engine explodes). Plus, it's super accurate with its very high projectile speed and lack of bullet drop. The erPPC amplifies the range advantage and removes the minimum range disadvantage. It does run very hot, though, but given how light it is compared to ACs (an AC10, which matches the point damage but is far shorter ranged, with three tons of ammo is 15 tons, for which you can run an erPPC and 8 DHS, though the critical space might be an issue) and considering how quickly it recycles when fired it is a strong brawling option under the current game balance. Its advantages against enemy ECM are icing on the cake.

Compared to the AC5, which is what the PPC should really be held up to, it does double the damage per shot, is more accurate, and has no ammo consumption. The AC5 should get an advantage to compensate, and aside from being much cooler to operate I think that rate of fire is the best place for it to get it. Since almost nobody uses standard AC5s currently (unless they're doing a dakka build with a pair combined with a pair of AC2s or UAC5s, or maybe quad AC5s), it seems like the current rate of fire must therefore be too low, taking all the other factors into account.

TLDR - ACs should normalize around the higher dps number because dps is better for brawling and for direct fire support while point damage and alpha performance is better for sniping. Plus, lots of ancillary issues need to be considered, such as ammo, weight, critical space, etc.

[Anyone00]

Well from what I understand the problem stems from the TT: an AC is suppose to be largely obsolete when the Ultra or LB-X (or Rotary) version of that damage class comes out (until specialized autocannon rounds come on the scene). Ultra and LB-X are suppose to be hard to come by in TT but this isn't table top. Plus LB 10-X can't fire single slug rounds like it's suppose to so it doesn't make AC/10 obsolete and AC/2 have been buffed to half way to being a RAC/2 w/o the jamming (although you can argue this is to make up for the lack of through the armor criticals and can function as a long range suppression weapon).

[MustrumRidcully]

Renthrak, on 11 May 2013 - 01:41 PM, said:

I voted YES, but I don't think the AC/5 is as useless as people seem to think it is.

I actually ran the numbers for ballistic weapons, comparing damage per ton of ammo (shot damage X number of shots in a ton), damage per ton of the weapon itself including heat sinks required to dissipate all the heat produced within the recycle rate of the weapon, and numerous other factors to determine the overall 'efficiency' of the weapon. The results were rather surprising.

The top two most efficient ballistic weapons are the Gauss Rifle (no surprise here), and the AC/5 (???????). It turns out that based on the tonnage and ammo required to use ballistics, the AC/5 is only second to the Gauss Rifle because the GR produces so little heat.

This doesn't seem to make sense based on the mediocre stats of the AC/5 itself, but basically, carrying an AC/10 or /20 costs more weight(weapon+ammo+HS) than the comparable increase in damage. One of the biggest things holding the AC/5 back in actual gameplay is the current prevalence of massive Alpha Strikes as the primary means of offense.

For reference, the AC/2 has higher flat damage output, but requires so many heatsinks that it ends up heavier than the AC/5 to achieve that level of performance.

I did such calculations a long time ago, too.

I ultimately found this way a bit too... naive. Or not practically as relevant as I hoped.

A big factor is that we have a high heat capacity in MW:O. That means even with just 10 standard heat sinks (and you can't have less), you can fire for a while before you overheat. In practice, you don't need to compensate all the heat weapons produce, only have enough heat cap and enough dissipation that it lasts for a typical engagement. For a sniper, such an an engagement can last only 5 seconds or so, before it is back into cover. For a brawler, it might better be measured in damage - the fight lasts as long as it takes to kill the enemy or be killed by him.

Then, there is another factor. If we're really only talking about engagements lasting 5 to 30 seconds, DPS becomes can become very imprecise. In 5 seconds, a Gauss Rifle delivers by DPS 18.75 damage, and an AC/2 deals 20. In truth, however, the Gauss Rifle deals 30 damage, and the AC/2 22 damage.

Of course, the problem is, once you get this detailed, you notice the numbers of one weapon don't scale up linearly.
A single AC/2 might be able to fire for 40 seconds uninterrupted with just 10 standard heat sinks. But that doesn't mean 2 AC/2 would just need 20 standard sinks to last 40 seconds. With 20 heat sinks, two AC/2s could fire "only" for 25 seconds. And 4 AC/2 with 40 heat sinks would last only 17.5 seconds.


So, what I eventually went for as analysis was:
Set a target damage value to be reached, with a target engagement time, and set how often this would need to be done (to get a feel for how much ammo you'd need). Then, calculate how many instances of a particular weapon you need to get that damage goal in the alloted time, calculate how many heat sinks it would require to not overheat and how much ammo it consumes per engagement and multiply by engagement numbrs, and then compare the total damage this build would do to the total weight that build would need.
I did treat the first 10 engine heat sinks as free and all that jazz, too, to get a few reasonable values.

These are two charts I came up with, the first for every short engagement times (4-12 seconds) with high burst damage:


and a second one for somewhat longer engagement times with less burst damage.





A Google Document containing the spread sheet can be found here: https://docs.google....cXc&usp=sharing
(The original is an excel document, which is also what I used to generate the charts.)

I find the charts illustrating the current meta quite well - PPCs rock for short-term damage, but quickly drop off if the engagement starts to last any time.

With high burst situations, the AC/5 can do relatively okay compared to the UAC/5, but it's quickly outperformed.

Edited by MustrumRidcully, 12 May 2013 - 01:38 AM.

[Jabilo]

Wow, great analysis.

So Mustrum, throw your hat in the ring and give us your opinion,

If YOU could suggest a balance change to the AC5 what would it be?

Do you like the proposed range change or would you like to see something else?

Pretend that all other weapons have to stay the same as they are now except for the changes already slated for the next patch.

[ArmageddonKnight]

I find it amusing that some people like to compare the damage of a AC5 vs a medium laser.. Considering a medium laser is only 1 crit slot, and 1 ton + very few DHS vs 4 crit slots and 8 tons + ammo.

[Lefty Lucy]

MustrumRidcully, on 12 May 2013 - 01:31 AM, said:

I did such calculations a long time ago, too.

I ultimately found this way a bit too... naive. Or not practically as relevant as I hoped.

A big factor is that we have a high heat capacity in MW:O. That means even with just 10 standard heat sinks (and you can't have less), you can fire for a while before you overheat. In practice, you don't need to compensate all the heat weapons produce, only have enough heat cap and enough dissipation that it lasts for a typical engagement. For a sniper, such an an engagement can last only 5 seconds or so, before it is back into cover. For a brawler, it might better be measured in damage - the fight lasts as long as it takes to kill the enemy or be killed by him.

Then, there is another factor. If we're really only talking about engagements lasting 5 to 30 seconds, DPS becomes can become very imprecise. In 5 seconds, a Gauss Rifle delivers by DPS 18.75 damage, and an AC/2 deals 20. In truth, however, the Gauss Rifle deals 30 damage, and the AC/2 22 damage.

Of course, the problem is, once you get this detailed, you notice the numbers of one weapon don't scale up linearly.
A single AC/2 might be able to fire for 40 seconds uninterrupted with just 10 standard heat sinks. But that doesn't mean 2 AC/2 would just need 20 standard sinks to last 40 seconds. With 20 heat sinks, two AC/2s could fire "only" for 25 seconds. And 4 AC/2 with 40 heat sinks would last only 17.5 seconds.


So, what I eventually went for as analysis was:
Set a target damage value to be reached, with a target engagement time, and set how often this would need to be done (to get a feel for how much ammo you'd need). Then, calculate how many instances of a particular weapon you need to get that damage goal in the alloted time, calculate how many heat sinks it would require to not overheat and how much ammo it consumes per engagement and multiply by engagement numbrs, and then compare the total damage this build would do to the total weight that build would need.
I did treat the first 10 engine heat sinks as free and all that jazz, too, to get a few reasonable values.

These are two charts I came up with, the first for every short engagement times (4-12 seconds) with high burst damage:


and a second one for somewhat longer engagement times with less burst damage.





A Google Document containing the spread sheet can be found here: https://docs.google....cXc&usp=sharing
(The original is an excel document, which is also what I used to generate the charts.)

I find the charts illustrating the current meta quite well - PPCs rock for short-term damage, but quickly drop off if the engagement starts to last any time.

With high burst situations, the AC/5 can do relatively okay compared to the UAC/5, but it's quickly outperformed.

Two things your charts don't take into account well are range and hardpoint limitations, which is why the small laser and medium laser don't dominate the meta as much as your efficiency analysis would predict.

[MustrumRidcully]

Jabilo, on 12 May 2013 - 06:48 AM, said:

Wow, great analysis.

So Mustrum, throw your hat in the ring and give us your opinion,

If YOU could suggest a balance change to the AC5 what would it be?

Do you like the proposed range change or would you like to see something else?

Pretend that all other weapons have to stay the same as they are now except for the changes already slated for the next patch.

If really nothing else is changed.... (I am more in favor of more wider changes to the heat system before we go back to finetuning), the AC/5s cooldown time should probably be lowered to about 1.25. At least for the 4 to 16 seconds engagement times, this will make the AC/5 competitive with the Ultra AC/5.
The AC/10 should probably get a recycle rate of 2 seconds, as should the LBX-10.

The efficiency changes for the AC/5 vs UAC/5 in Single Shot Mode vs UAC in Double Shot Mode:
4 Seconds/60DMG/40Engagements: 1.5 / 1.3 / 1.6
8 Seconds/120DMG/20Engagements): 2.9 / 2.6 / 2.9
16 Seconds/240DMG/10Engagements): 5.4 / 5.4 / 5.7

Note, that the UAC/5 efficiency is calculated based on a 25 % chance to jam and a jam duration of 3 seconds, and averaging the effective cooldown the weapon has with jam.

It's not perfectly balanced, but it seems reasonably close to me.

[MustrumRidcully]

Lefty Lucy, on 12 May 2013 - 07:17 AM, said:

Two things your charts don't take into account well are range and hardpoint limitations, which is why the small laser and medium laser don't dominate the meta as much as your efficiency analysis would predict.

Well, it only describes boats, obviously. A Gauss Rifle user could get a lot of extra damage output by equipping 2 Medium Lasers, without adding any serious heat issues to it, a PPC user couldn't. But hard points itself don't do much for the balance of weapons. The moment you have a hard point combination available for a good weapon, you've won, and if you don't, take the next best thing.

Range is the reason why I sorted the weapons like I did. The longer the range of a weapon ,the lower its overall efficiency should be.

But speaking of hard points, for the 4 second burst setup dealing 60 damage, you need:
2: AC/20, AC/10, LBX, Gauss
3: AC/5, Ultra AC/5, Ultra AC/5, Large Pulse Laser, PPC, ER PPC
4: AC/2, Medium Pulse Laser
6: Medium Laser
7: Large and ER Large Laser (why so many: 4 seconds is unfortunately longer than the recycle time of the Large Laser! Just half a second more, and you'd only need 4...)
10: Small Laser, Small Pulse Laser
17: Flamer
37: Machine Gun (Just in case we need a reminder how awfully, terribly bad the MG is.)

If you look at the Small Laser - there aren't any mechs in the game that could even fit 10 Small Lasers.
There are mechs that can fit 6 Medium Lasers in every class. But these mechs could also fit most of the longer range combos.

[Coralld]

I don't think the AC/5 needs a RoF increase, but I do believe that the projectile should get a range and speed increase with the gun getting an increase to HP. Where as its Ultra/5 brother gets an HP decrease.

[Saint Scarlett Johan]

Levi Porphyrogenitus, on 11 May 2013 - 08:11 AM, said:

See the AC section of my post here: http://mwomercs.com/...ool-down-rates/

AC dps should be regularized around the 5 dps point. AC20s are there already with 20 damage every 4 seconds. AC10s would need to be sped up to 10 damage every 2 seconds, and AC5s moved to 5 damage every 1 second. The AC2 would be the odd weapon out, at its current 0.5s cool down, giving it only 4 dps, but considering its absurd range profile and super high velocity projectiles I'd be fine with that slight discrepancy.

Of course, all this assumes that ACs are being balanced not just with each other but with other weapon systems, with the idea being to give each class of weapon its own role and flavor.

Awesome, I'm glad someone else said exactly what I was going to say.

[Cola]

sadly, anything not a PPC with jumpjets will be useless. so I don't care about the AC5.

[WolvesX]

arden, on 12 May 2013 - 11:41 AM, said:

sadly, anything not a PPC with jumpjets will be useless. so I don't care about the AC5.

But It would be nice if one would care about the AC5.

MOAR DAKKA, MOAR FUN!

[Sephlock]

Quote

Paul stated:

AC/5
After reading my last post, dunno where my head was when I said AC/5 damage would be changing. AC/5 damage is NOT changing. Their range however is increasing from 540m out to 620m.

Whatever it was, please up your intake.

[Karl Streiger]

Base on just DPS and Alpha the AC 5 didn't look that good...but really...do you really believe a weapon is that lackluster when dozen or hundred of people running Quad AC 5s?

You can compare the Dual AC 5 perfectly with the Gauss...Gauss does more Alpha and even the DPS isn't superior. But if you can keep in the fight for more as 7 secs....a time where a tripple AC 2 would have produced 42 heat...the Dual AC 5 starts to become better as the Gauss....the longer you keep in sight and firing the Gun the better the AC 5 become.

That is clearly a disadvantage too...staying in contact means you can not use cover...but if you have to hold a line...the AC 5 will allow you to keep firing...no break because of heat...or reloading....just keep firing.

The DualAC5 however works well on a Dragon with a Window of Engagement of 2 seconds too.
You can make a strafing run...and place two shots in a target ( in this time...hardly any other weapon but the AC 2 or the UAC5 can respond...and you deal in that 2 sec more damage as a dual AC 2 combo. If you can manage to fire you second salvo split second before you are back in cover you are even able to out gun a tripple AC 2 combo....

So considering the WoE the Light(Medium) AC is a fine weapon.

[armyof1]

The reason the dual AC5 seems kinda good is because of the hardpoint limitations of the specific mech. But when you have a mech that can carry either dual AC5 or dual UAC5 then I would never choose the AC5s. The clearly higher rate of fire and the possibility to double fire makes it great for short bursts of damage where you don't expose yourself for more than a short moment. If you don't want to double fire then use an external program with a macro that will give you the 1.1s firing rate that will never jam. Either way the AC5 is really inferior to the UAC5 whenever you have a choice, and a higher firing rate at around 1.3 makes lots of sense to make the AC5 worth using.

Edited by armyof1, 13 May 2013 - 02:15 AM.