Jun 9, 2015
APB Reloaded - Tiggs
Weapon Rebalance Description
We mentioned the Weapon Test District in last month’s Q&A post. We are gaining larger and larger amounts of data which is being gathered in order to solidify changes that may be made to some weapons in a future patch. For those interested in participating, head to the Advanced tab of the District Select Menu. Now that you've had some time to play with the pre-selected 10 weapons, let’s go into exactly what has changed, along with some insight into what data we've gathered, our conclusions from that, and where we are potentially going with the next set of changes.
Firstly, a couple of thoughts and caveats about the data below:
NTEC
What we've done and why
The current version of the NTEC is by far the most popular weapon in APB Reloaded. On a chart of combat events using weapons (total kills + deaths) we can see that the NTEC is used in 14% of cases:
Even against the STAR, the weapon used at one point by every character, the NTEC dominates in terms of use, and this is stifling weapon variety during missions. A major point of this ubiquity is its ability to tap fire at near full speed and not lose any accuracy. We don't feel this is the ideal use for an assault rifle, and the primary goal for our changes have been to reduce its ability to keep full accuracy under near-cyclic tap fire, while moving it more to a weapon that benefits from being fired in short bursts.
The first part of this was reducing its overhead. By reducing the damage down from 19% to 18.5% we remove the ability to use Heavy Barrel 2 without a major downside.
Secondly we reduced the initial accuracy loss and recoil when firing, and had it increase over time. This makes the weapon much more effective at burst firing (the second shot is almost as accurate as the first) while making tap fire less accurate as you need to wait longer to allow the accuracy exponent to reset.
The accuracy of the NTEC at 10m from the hip when firing fully automatically looks like the following (the Y axis is centimetres, X is bullet fired).
Aiming from Marksmanship mode is a multiplier on accuracy, so the graph will look the same but with a contracted Y axis.
What Effect has it had?
Above is the graph of percentage kills at various ranges taken from the last week's data we've been gathering. The prototype values are Orange lined. It's important to note that the data for the prototype weapons is not enough to draw any conclusions from yet, but it is useful for early thoughts. Of all the data gathered, the NTEC by far is the one we have the most data on, so we can at least make some conclusions. From the above graph, we can see that percentages of kills at 4-24m are down, and the percentage of kills from 30-60 are up, so it's losing out to weapons at closer ranges and/or winning more at mid-ranges. The Kill / Death ratio is also slightly down, though it's within a range that could be attributed to variance. Note that a K:D of less than 1 is perfectly acceptable for a non-situational weapon, as gravity, suicides, accidental team kills and vehicles take their toll.
Weapon Rebalance - HVR-762
Weapon Rebalance - VAS-C2
Weapon Rebalance - ISSR (A / B)
Weapon Rebalance - OCSP
Weapon Rebalance - FFA-BullShark
Weapon Rebalance - COBR-A
Weapon Rebalance - VBR
Weapon Rebalance - NFAS-Ogre
Weapon Rebalance - NCR-Anubis
We mentioned the Weapon Test District in last month’s Q&A post. We are gaining larger and larger amounts of data which is being gathered in order to solidify changes that may be made to some weapons in a future patch. For those interested in participating, head to the Advanced tab of the District Select Menu. Now that you've had some time to play with the pre-selected 10 weapons, let’s go into exactly what has changed, along with some insight into what data we've gathered, our conclusions from that, and where we are potentially going with the next set of changes.
Firstly, a couple of thoughts and caveats about the data below:
- Data only really becomes valid with tens of thousands of data points. While we get millions of kills each week, only a fraction of these are in Prototype Districts and as such this preliminary data will be somewhat incomplete. This can be clearly seen by 'Spiky' graphs.
- We're comparing ALL weapon data vs Prototype weapons in Baylan only. It's a lot faster and provides us more info gathering everything to compare against, rather than comparing against like weapons in Baylan Fight club. As we move on and get closer to our desired results we'll be running longer, slower comparisons that are more accurate. At this stage, 'fast' comparisons that let us know we're going in the right direction are good enough and let us iterate faster.
- We gather metrics based on the point of death. Specialized weapons, such as Rocket Launchers and Sniper Rifles for instance will have higher than average Kill Death Ratios since players will often have switched to their side-arm when caught unawares and thus more likely to die.
- A lot of APB kills across all weapons come from 0-4m, regardless of the weapon. Through both our environment and vehicles, opportunities arise for players to get close, and almost every weapon can be effective at point blank range.
- What's Next?
NTEC
What we've done and why
The current version of the NTEC is by far the most popular weapon in APB Reloaded. On a chart of combat events using weapons (total kills + deaths) we can see that the NTEC is used in 14% of cases:
Even against the STAR, the weapon used at one point by every character, the NTEC dominates in terms of use, and this is stifling weapon variety during missions. A major point of this ubiquity is its ability to tap fire at near full speed and not lose any accuracy. We don't feel this is the ideal use for an assault rifle, and the primary goal for our changes have been to reduce its ability to keep full accuracy under near-cyclic tap fire, while moving it more to a weapon that benefits from being fired in short bursts.
The first part of this was reducing its overhead. By reducing the damage down from 19% to 18.5% we remove the ability to use Heavy Barrel 2 without a major downside.
Secondly we reduced the initial accuracy loss and recoil when firing, and had it increase over time. This makes the weapon much more effective at burst firing (the second shot is almost as accurate as the first) while making tap fire less accurate as you need to wait longer to allow the accuracy exponent to reset.
The accuracy of the NTEC at 10m from the hip when firing fully automatically looks like the following (the Y axis is centimetres, X is bullet fired).
Aiming from Marksmanship mode is a multiplier on accuracy, so the graph will look the same but with a contracted Y axis.
What Effect has it had?
Above is the graph of percentage kills at various ranges taken from the last week's data we've been gathering. The prototype values are Orange lined. It's important to note that the data for the prototype weapons is not enough to draw any conclusions from yet, but it is useful for early thoughts. Of all the data gathered, the NTEC by far is the one we have the most data on, so we can at least make some conclusions. From the above graph, we can see that percentages of kills at 4-24m are down, and the percentage of kills from 30-60 are up, so it's losing out to weapons at closer ranges and/or winning more at mid-ranges. The Kill / Death ratio is also slightly down, though it's within a range that could be attributed to variance. Note that a K:D of less than 1 is perfectly acceptable for a non-situational weapon, as gravity, suicides, accidental team kills and vehicles take their toll.
Weapon Rebalance - HVR-762
Weapon Rebalance - VAS-C2
Weapon Rebalance - ISSR (A / B)
Weapon Rebalance - OCSP
Weapon Rebalance - FFA-BullShark
Weapon Rebalance - COBR-A
Weapon Rebalance - VBR
Weapon Rebalance - NFAS-Ogre
Weapon Rebalance - NCR-Anubis
Share: 
