Jump to content

Rac: Russian Roulette Of Mwo - Rac Statistics


19 replies to this topic

#1 The6thMessenger

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Nova Captain
  • Nova Captain
  • 8,045 posts
  • LocationFrom a distance in an Urbie with a HAG, delivering righteous fury to heretics.

Posted 30 June 2017 - 06:02 AM

Rotary Autocannons:

Russian Roulette of MWO

Posted Image




Quote

Introduction:

Rotary Autocannons are one of the new Ballistic Weapons introduced in the upcoming Civil War Tech, and is accessible during the PTS. Lovely for those who enjoyed pounding enemies with shells, especially beautiful equipped on an Urbanmech.

This analysis is meant to review the RACs, and provide recommendation in how to improve such weapons.

Quote

Methodology:

Using nothing more than a Rotary Autocannon, I continuously shot from beginning to end at a target upon the Academy. Suffering the full extent of the Spin-Up, completely filling up the Jam Meter, and still shooting until the weapon Jams. Then total damage done is recorded.

After the Jam has been cleared, and the Jam Meter empty, the Rotary Autocannon was shot again from Spin-Up to Jam, and then recording the total damage done. Repeated 24 times, recording 25 instances, and done so on each weapon system.

Quote

Quote

Discussion:

Posted Image

The AC5, weighs 8 tons, shoots shells dealing 5 damage, every 1.66s, at 3.012048192771084 DPS.

The AC10 weighs at 12 tons, shoots shells dealing 10 damage every 2.50s, at 4 DPS.

The UAC5, weighs 9 tons, shoots shells dealing 5 damage, every 1.66s. It can double-shot, at 15% chance of jam, for 6.0s. Effectively, it can double-shot 6.666666666666667 times on average, dealing 66.66666666666667 damage, at a cumulative 11.06666666666667 CD + 6.0s cooldown, effectively doing 3.90625 DPS.

For the RAC2, as shown by the data of 25 tries of average total damage done incurring 10 second jam. It has an average of 29.04 damage done, 7.26 seconds of average continuous shoot time. Does a minimum of 1.22 EDPS, and maximum of 2.22 EDPS, with an average of 1.68 EDPS.

For the RAC5, as shown by the data of 25 tries of average total damage done incurring 10 second jam. It has an average of 75. damage done, 7.26 seconds of average continuous shoot time. Does a minimum of 2.68 EDPS, and maximum of 6.20 EDPS, with an average of 4.24 EDPS.

The E-DPS is calculated by means of dividing the Average Damage Done by the added Average Shoot Time and the Jam Duration:

Quote

EDPS = Average Damage Done / (Average Shoot Time + Jam Duration)


The Maximum and Minium is calculated by means of dividing the Damage Done by the added Shoot Time and the Jam Duration, with the Shoot-Duration calculated by diving Damage/done with the Burst Damage/Second provided by the weapon system:

Quote

Max/Min EDPS = Damage Done / ((Damage Done / Burst DPS) + Jam Duration)


Shown in the scatter plot, the RAC2 have a fairly consistent range around 18 to 35 total damage done with only 3 instances out of 25 of above 40 total damage done.

While the RAC5 has a consistent range around 36 to 60 total damage done. Nearly reaching nearly 180 total damage done, with 10 instances out of 25, of above 60 total damage done. Both RACs show wide ranges of total damage done.

Quote

Limitation:

> I ******* hate math, no really I do. One of the worst time of my child-hood is suffering failing to do math like Trigonometry. That i was only passed at high-school math out of pity. However, I need data and math to communicate with objectivity. I cannot say that my data is correct, I only relied on MS-Excel to do the math for me. But if my math is correct -- and hopefully it is -- this is how RACs are objectively is.
> "10s Jam Duration" is taken from different community source.
> During the course of testing, there were times i think i may have missed and started from the beginning of one of the sessions.
> There was a time I missed the screenshot of the damage done, I simply ignored it and moved on.
> I only did 25 instances of full shooting, which may not be completely representative of the weapon considering full service along with many other users.
> This only assumes a complete trigger-happy shooting sequence of holding down the button till it jams while shooting, but does not account stopping when the gauge is full, let it go down and shoot back again. However like UACs, the jamming portion is where it gets it's extra DPS as how PGI did it. But it does not account of what could be, prescriptively if ever, the "proper" use of such weapon system.

Quote

Conclusion:

The RAC2 is in an abhorrently bad spot. Consider that it's DPS is between 1.22 to 2.22 at an average of 1.68, but with a weapon of similar tonnage, the AC5 that does 3.01, its not as good. Factor in the immense face-time to do damage, and such damage means that it would be spread ALL OVER both by inherent cone of fire, and torso-twisting that can be done by competent pilots.

The RAC5 seems to have good damage output. At 4.24 average EDPS, at a maximum of 6.20 DPS, but at a low 2.68, resulting in a fairly inconsistent and erratic instances of total damage done. For comparison, the AC10 at 10 damage/shot and 2.50s interval, does 4 DPS, sitting at 12 tons. At 4.24 average is a massive boost of DPS, for a 10-ton weapon.

The RACs takes advantage of a very high damage per second, done so at an immense amount of face-time. Realistically, this is not good considering the meta, as staring people to death means better window of time for retaliation.

At nearly 7-8 seconds to maximize damage output, this duration is completely devastating. To put that into perspective, the old C-ER-LL does 11 damage for 1.5s, but it's already widely regarded as hard to use, at the same duration the RAC2 only dealt 6 damage, despite 8 tons of weight, and the RAC5 only dealt 14.4 damage which is good relatively but consider that C-ERLL is just 4-tons and is hitscan which already improves it's success rate, as opposed of RAC5's slow projectile that you need adequate lead, has to suffer spin-up time, requires ammunition, and 6 tons more heavy. Realistically, you might as well load up two ER Large-Lasers.

The RACs may also do large amount of damage, but that is spread everywhere due to the nature of stream-firing shells. Like lasers, they can be spread by torso twisting, aggravated by the fact that it has cone of fire, requires spin-up so it has the same barrier as the Gauss Rifle, but instead of a 15 PPFLD quick moving projectile that you can land fairly easily, instead it's a risky and deathly stare-down contest that does so much damage but spread all over.

Quote

Recommendation:

> The RAC5 has an erratic damage output, even if capable of high damage output. It would be best if the RACs -- even the RAC2 -- could be more consistent.
> The RAC2 needs immense damage boost to normalize it's damage output for it's weight, else it won't be worth to be used over better choices like the AC5.
> The RACs should immediately jam after filling up the Jam Meter, this makes the weapon predictable and opens up for more control, then balance the weapon around that which is much easier than relying on chance.
> Following the predictable Jam, it should have a larger duration before jamming.
> The Jam Meter shouldn't be filled while just winding up. If people worry about macros, another angle can be approached, like heat generation whenever spinning up or spinning down as well, as keeping it spun just builds up heat anyways.
> RACs spread an enormous amount of damage all around the mech, than focused on one spot. Making more powerful shells, and less amounts of shot would fix that. But also increase ammo count and/or damage per ammo/ton.
> RACs are hard to hit, immense velocity buff should be considered. 2000 for the RAC2, since it's AC2 caliber, and 1650 for RAC5 cause it's AC5 caliber.

Here's a complete set of suggested changes:
Spoiler

Quote

Sources:

Spoiler

Edited by The6thMessenger, 16 July 2017 - 01:12 AM.


#2 Linkin

    Member

  • PipPipPipPipPipPip
  • Bad Company
  • Bad Company
  • 284 posts
  • LocationCA

Posted 30 June 2017 - 09:54 AM

Heck of an effort, great write up!

I can't fully comment on all of the detail you put in, but I will toss in one of the glaring issues I noticed with the RAC's in the brief time I spent in PTS testing grounds. As you mention- the jam bar shouldn't fill until it starts spitting rounds. Having the jam bar fill as the weapon spins up is no good.

#3 Mike Barnes

    Member

  • PipPipPip
  • The God
  • The God
  • 50 posts

Posted 30 June 2017 - 10:19 PM

Really nice write up, thank you for the source, and really getting it together.

#4 Mike Barnes

    Member

  • PipPipPip
  • The God
  • The God
  • 50 posts

Posted 30 June 2017 - 11:22 PM

The 6th messenger, I think with the rework they nailed it with the RAC/5, it feels great, and in a very high risk high reward state, as now the minimum if you just get extremely unfortunate is about equal to an AC/5 and the max is about equal to a UAC/20, and averages out to a little better than a C-UAC/10. The 2 is better but still not great.

#5 MustrumRidcully

    Member

  • PipPipPipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 10,644 posts

Posted 01 July 2017 - 01:01 AM

I would also suggest removing the RNG aspect of the jam. It makes the weapon too unpredictable for balance. It should be designed so that skill in avoiding reaching the full bar is far more important than luck in firing when the bar is maxed out.

ADjusting DPS and Ammo/Ton sounds also like a good idea. The individual shots don't deal much damage, so clearly the ammunition must be much smaller and lighter than regular AC/2 or AC/5 ammo.

I would also recommend lowering the ramp up/down times further. The weapon is already harmstrung with needing a lot of facetime, it gets worse if you can't even react quickly with the weapon - especially since you won't be using them effectively at long range.

#6 Mike Barnes

    Member

  • PipPipPip
  • The God
  • The God
  • 50 posts

Posted 01 July 2017 - 01:09 AM

I actually like the RNG chance of the weapon, It feels unique in this way. As long as there is a decent bell curve for time to jam, I can work with that. Right now the boat I am planning, and have had some success with is a quad rac/5 boat, two at a time, I think once I have an Anni, this will go better, but basically if I play skillfully with quad RAC/5s the dakka party never stops, and with the RAC/5 damage where it is now I'm happy.

#7 MustrumRidcully

    Member

  • PipPipPipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 10,644 posts

Posted 01 July 2017 - 01:12 AM

View PostMike Barnes, on 01 July 2017 - 01:09 AM, said:

I actually like the RNG chance of the weapon, It feels unique in this way. As long as there is a decent bell curve for time to jam, I can work with that. Right now the boat I am planning, and have had some success with is a quad rac/5 boat, two at a time, I think once I have an Anni, this will go better, but basically if I play skillfully with quad RAC/5s the dakka party never stops, and with the RAC/5 damage where it is now I'm happy.

It already has a unique mechanic with the jam bar. And the RNG isn't a unique part of it - UACs have had that forever.

#8 The6thMessenger

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Nova Captain
  • Nova Captain
  • 8,045 posts
  • LocationFrom a distance in an Urbie with a HAG, delivering righteous fury to heretics.

Posted 01 July 2017 - 01:21 AM

The problem with the RNG mechanic is that it's already an additional risk over exposing yourself for the entire duration.

We just don't need any more risk aside from staring enemies to death. Come on.

#9 phoboskomboa

    Member

  • PipPip
  • Ace Of Spades
  • Ace Of Spades
  • 28 posts

Posted 01 July 2017 - 11:32 AM

Great write up! Amazing attention to detail.

I would like to play devil's advocate a bit, though, and ask if maybe the RAC-2 might be useful not as a dps weapon, but as a support tool. It's VERY difficult to return fire when receiving shots from the RAC-2. Could a potential use be to pair one with a bank of lasers, and use the RAC-2 as a shield?

I think I'm going to test this use a bit.

#10 SOL Ranger

    Member

  • PipPipPipPipPipPip
  • Shredder
  • Shredder
  • 232 posts
  • LocationEndor, exterminating little evil bear people for the Empire.

Posted 01 July 2017 - 12:19 PM

View Postphoboskomboa, on 01 July 2017 - 11:32 AM, said:

Great write up! Amazing attention to detail.

I would like to play devil's advocate a bit, though, and ask if maybe the RAC-2 might be useful not as a dps weapon, but as a support tool. It's VERY difficult to return fire when receiving shots from the RAC-2. Could a potential use be to pair one with a bank of lasers, and use the RAC-2 as a shield?

I think I'm going to test this use a bit.


A suppressive fire weapon needs to have higher up-time through the assault or defense you are trying to cover and needs to be reliable to the point it actually fires a while when you want it to, whereas the RAC's jam quickly and stay jammed for a long time and as such they have no place in that role. You are better off with AC/2's as a suppressive fire weapon right now.

It is a darn shame because I was looking to these weapons to be assault weapons of exactly that brutal brawling kind, spin up and spray at anything movine and keep going as until targets no longer appear or your 'Mech melts, whichever happens first. The premature jaming and especially the RNG element in the jam chance just kills it.

#11 phoboskomboa

    Member

  • PipPip
  • Ace Of Spades
  • Ace Of Spades
  • 28 posts

Posted 01 July 2017 - 02:21 PM

Yeah, I tested a single rac2 as a supressive fire weapon, and it can be useful in situations, but it's just so awkward to use, and totally not worth the weight. It wouldn't work long enough to be used through an entire brawl between 2 assaults.

#12 The6thMessenger

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Nova Captain
  • Nova Captain
  • 8,045 posts
  • LocationFrom a distance in an Urbie with a HAG, delivering righteous fury to heretics.

Posted 09 July 2017 - 12:36 AM

Posted Image

Here's re-adjusted data. RACs having buffed, but the chance retained, the distribution remains the same. However since the damage output has increased, we can simply add the damage buffed, and it should be the same.

-0.25s of the spin-time would mean that we would have done 0.25s worth of damage more.

So the each damage instances, all 25 of them, have been added (0.25 x new DPS) depending on weapon. And each already exisiting damage has been multiplied by the amount of DPS buff it received.

Like so:

Quote

RAC2: (0.25 * 5.6 DPS) + (Old Damage * 1.4) = Adjusted RAC2 Damage
RAC5: (0.25 * 10.8 DPS) + (Old Damage * 1.125) = Adjusted RAC5 Damage


I also sorted the data gathered.

Notice how at the RAC5, the 1st 14 shots were just right around 60. But then shoots up quickly to 200 damage.

Similarly, the RAC2 have a bit more consistent rise, but then the last 3 seems to shoot up as well, and following the trend if the sample-size was a lot more, it could just as well shoot-up as the RAC5.

I don't like this distribution at all, this tells me that the RACs are pretty unreliable. We're already investing long stare time to do damage, all of those damage are spread and we have to bother with all of the demand to land a projectile based-weapon, we have to bother with a spin-up time, and then there's no guarantee we can stare for longer for even more damage. The damage is satisfactory, especially with such a light-weight weapon system, but how it deliver that damage, and how unreliable the weapon is just kills the thing.

Edited by The6thMessenger, 09 July 2017 - 04:57 AM.


#13 Reno Blade

    Member

  • PipPipPipPipPipPipPipPipPip
  • The Blade
  • The Blade
  • 3,459 posts
  • LocationGermany

Posted 09 July 2017 - 01:55 AM

what if you can manage to never jam and just let the bar cool off when twisting or being in cover?
in most fights you will be more limited by the spool up time where the bar is rising.
I feel that a fast medium mech with one RAC5 like a wolverine or cent can engage in hit and run tactics to use most of the high dps.

#14 The6thMessenger

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Nova Captain
  • Nova Captain
  • 8,045 posts
  • LocationFrom a distance in an Urbie with a HAG, delivering righteous fury to heretics.

Posted 09 July 2017 - 02:10 AM

View PostReno Blade, on 09 July 2017 - 01:55 AM, said:

what if you can manage to never jam and just let the bar cool off when twisting or being in cover?
in most fights you will be more limited by the spool up time where the bar is rising.
I feel that a fast medium mech with one RAC5 like a wolverine or cent can engage in hit and run tactics to use most of the high dps.


Not shooting it when red, is like not double-shooting UACs, that's how PGI designed that -- which i am protesting. It's pointless, as you are doing meager damage - with the minimum total damage - 1 instance of damage, 43.2 (32 shots, 4s) for the RAC5 and 25.2 (36 shots, 4.5s) for the RAC2, that's your only damage when you got it to max gauge but not let it jam. I don't know about the dissipation time, but i know that you're wasting your gauge when you're stopping shooting mid-way cause the jam bar fills even when you're just spinning up. And the jam-duration simmilar regardless of how much damage you did, you might as well go all the way to maximize your output.

Edited by The6thMessenger, 09 July 2017 - 02:16 AM.


#15 LordNothing

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 16,772 posts

Posted 09 July 2017 - 03:34 AM

despite this seemingly damning evidence to the contrary, i always considered the rac2 to be the more usable weapon due in part to its shorter spinup time. i think i made the point somewhere that the spin up time alone is long enough to be ganked in a fp match. bring both guns down to a quarter second spin up time, or better yet, make it ramp up its rof. i also support instant jam @ full bar and only filling up when a shell is fired.

Edited by LordNothing, 09 July 2017 - 03:35 AM.


#16 The6thMessenger

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Nova Captain
  • Nova Captain
  • 8,045 posts
  • LocationFrom a distance in an Urbie with a HAG, delivering righteous fury to heretics.

Posted 09 July 2017 - 04:38 AM

View PostLordNothing, on 09 July 2017 - 03:34 AM, said:

despite this seemingly damning evidence to the contrary, i always considered the rac2 to be the more usable weapon due in part to its shorter spinup time. i think i made the point somewhere that the spin up time alone is long enough to be ganked in a fp match. bring both guns down to a quarter second spin up time, or better yet, make it ramp up its rof. i also support instant jam @ full bar and only filling up when a shell is fired.


Yes, the RAC2 is kind of usable, since we're already used with Gauss Rifle having 0.75s charge time.

But realistically, that's just aggravating the already bad weapon system. Sure there can still have some spin-up time, but realisitically it should be just there to affect DPS calculations, like what i did. Because as the PTS demonstrated, it's not gonna fly in it's current state.

My best answer?

View PostThe6thMessenger, on 02 July 2017 - 04:14 AM, said:

Quote

General Changes:

- RACs Jam 100% when Jam Meter is filled.
- RACs do not fill Jam Meter while spinning, only while shooting.
- RACs Jam Meter does not go down until Spin-Down is achieved.
- Always during operation, the weapon generates heat. Be it during spin-up, spin-down, or firing, it will generate heat. Like Stealth Armor.

Quote

RAC2:

Damage/shot: 0.80
Projectile Speed: 2000
Shots/Sec: 10.00
Burst DPS: 8.00
Total Damage/Burst: 40
Total Shots/Burst: 50
Spin-Up Time: 0.50s
Spin-Down Time: 0.50s
Burst Duration: 5.00s
Jam Dissipation: 4.00s
Jam Duration: 4.00s
EDPS: 4.00

Quote

RAC5:

Damage/shot: 1.25
Projectile Speed: 1650
Shots/Sec: 8.00
Burst DPS: 10.00
Total Damage/Burst: 50
Total Shots/burst: 40
Spin-Up Time: 0.50s
Spin-Down Time: 0.50s
Burst Duration: 5.00s
Jam Dissipation: 4.00s
Jam Duration: 4.00s
EDPS: 5.00


To define our new values, lets start:

EDPS = Effective DPS, the dps of the weapon once the Jam sequence is taken account.
Burst Duration = How long can the weapon shoot continuously before Jamming.
Jam Duration = How long the weapon cannot be fired.
Jam Dissipation = How long the meter is cleared at full - that means if the gauge is completely filled and at some form of magic does not jam, it would take 4.0s for the meter to be cleared.

The purpose of Jam Dissipation is to illustrate how fast the gauge is cleared. And with the Burst Duration at 5s, but the Jam Dissipation at 4.0s - WTF? The6thMessenger? Wouldn't that mean they could just not fill the jam meter completely, let it dissipate a few and then shoot again?

Well supposed that we only used up 2.5s or half of our total damage/burst, doing 20 for the RAC2 and 25 for the RAC5? It would take 2.00s for the jam meter to be cleared + 0.50s of spin-down time + 0.50s of spinup time for the next flurry. RAC2 doing only 20 damage, that is 3.636363636363636 EDPS. The RAC5, doing 25 damage, only does 4.545454545454545 EDPS.

You know what that means? To maximize EDPS you really need to use the weapon at it's fullest extent, else just stopping midway just penalizes yourself.

How about the spin-up? That short, and not filling the gauge when spinning up, wouldn't that make spin-up time useless? People could just practice preemptively spinning it up, or macroing it to do so. And what about poking?

I wouldn't worry about poking, it's only mostly effective if you can do a most of your damage, a lot of damage, at the shortest amount of time to minimize damage being received. Not only you're preemptively building up heat without doing damage as you maintain the spin-up time, you also need to stare enemies at an idiotic amount of time which is really just bad. To put that into perspective, the RAC5 needs 1.5s of stare time to do 15 damage, when a gauss rifle could do it in an instant it hits. That means that 1.5s from a gauss rifle to stare could be used for cover, either armor rolling or side shielding, or just hiding back into cover. So really, doing so wouldn't be that efficient, so i wouldn't worry about it.

Here's a different take for the RAC system.


Anyways, who's good with graphs. Could we get a PCA?

I don't know how to do PCA.

Edited by The6thMessenger, 16 July 2017 - 01:13 AM.


#17 Samedi Wretch

    Member

  • PipPipPip
  • 91 posts

Posted 10 July 2017 - 03:18 PM

View PostThe6thMessenger, on 09 July 2017 - 04:38 AM, said:


Anyways, who's good with graphs. Could we get a PCA?

I don't know how to do PCA.


I know how to do PCA, but I'm not sure how it applies here. What do you have in mind?

Normally it's done with multi-dimensional matrices of continuous variables. Maybe you're looking for something a little more straight forward, like best fit lines of scatter plots, or OLS regression? That's similar to PCA, but the point of PCA is dimensional reduction.

Good work btw. Thanks for the data!

Edited by samadhiVOID, 10 July 2017 - 03:23 PM.


#18 The6thMessenger

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Nova Captain
  • Nova Captain
  • 8,045 posts
  • LocationFrom a distance in an Urbie with a HAG, delivering righteous fury to heretics.

Posted 10 July 2017 - 03:20 PM

View PostsamadhiVOID, on 10 July 2017 - 03:18 PM, said:

I know how to do PCA, but I'm not sure how it applies here. What do you have in mind?


Well, it's nice to know what's the probability of getting which amount of damage before jamming. We could also compare the probability of getting the maximum damage, with the minimum damage.

#19 Samedi Wretch

    Member

  • PipPipPip
  • 91 posts

Posted 10 July 2017 - 03:29 PM

Send me your excel files, I'd be happy to play with it.

#20 D1G17AL

    Member

  • PipPipPipPipPip
  • Giant Helper
  • Giant Helper
  • 103 posts
  • Twitter: Link
  • Twitch: Link
  • LocationCalifornia

Posted 15 July 2017 - 10:12 AM

Wanted to mention that the ghost heat when firing a single RAC/2 and a single RAC/5 at the same time is absurd. That should be the least heat penalized combo. Firing a pair if either RAC/2's or 5's yields a very small spike in ghost heat. It's a shocking difference. It doesn't seem like it should be that way.





1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users