The following offers will be available in the Premium Shop from 27 September at 07:00 until 4 October at 07:00 CEST (UTC+2).
Special Offers |
||
|---|---|---|
| 8,8 cm Pak 43 Jagdtiger Standard
€36.80 |
8,8 cm Pak 43 Jagdtiger Ultimate
€49.99 |
8,8 cm Pak 43 Jagdtiger Supreme
€79.99 |
|
|
|
Comes with:
Link (2 days left)
With developer Sub_Octavian.
Q1:
HI Sub_Octavian!
You caught my interest when you said you were an analytics person. Can you say more about the kinds of data and analyses that WG uses to determine whether gameplay changes are needed? Nothing proprietary, of course. For example, it is a very common practice among players to compare average winrates among ships in order to judge their relative effectiveness. However, I would imagine that with the full data set, you could do separate comparisons of effectiveness when, e.g. playing as the lowest-tier ship, or in CV vs. non-CV games. Do you use an internal regression of some sort to account for player skill when looking at balance? Do third-party ratings like WTR come close to approximating the kinds of things WG takes into account? Or do you do mostly analytics on the customer base, rather than the ships?
Regarding gameplay mechanics, I am curious about the penetration mechanic for multiple layers of armor, e.g. turtleback, front bulkhead, citadel wall. I usually see armor described as Layer1(mm) + Layer2(mm) + Layer3(mm), sometimes with effective thicknesses given due to different angling of the layers (i.e. perpendicular to Layer1 will hit Layer2 at an angle). My question is regarding overmatched armor layers — if this occurs, will they still reduce further penetration, and if so, at what effective angle are they assessed? Is it zero (overmatched armor ignored)? Is it assessed at 28mm @ 60 degrees (autobounce angle)? Is it 28mm @ true angle (perhaps as steep is 80+ degrees)? This is of interest currently, because the proposed 28mm bow/aft armor changes to high-tier battleships mean that they could be taking a lot of shells in an overmatched section that subsequently challenges the (angled) citadel wall, and I am interested in understanding what kinds of angling will be viable on different ships (e.g. NC and Iowa have an extremely large bow section, and a very small, thin belt for battleships).
A1:
Helloes. There are tons of data avaliable, and we have a small team that is constantly working to implement new types of server logs, so we always can see what’s up. In addition to different absolute (like avg.damage, win rate, life time) and relative (% damage structure) values, we also have “hands dependance” graphs which allow us to check ship effeciency among players of different skill. We also have some interesting stuff like heat maps for maps balancing purposes. However, we cannot rely solely on statistics and server data, no matter how accurate it is. We also monitor community feedback and try to combine “technical” and “humane” approaches.
Looks more like a movie trailer. The Mark IVs are good though.
Patch 9.16 is scheduled for October 4, 2016. If it is not broken until then.
The patch is scheduled for 29 September and will be applied on the servers from 05:30 CEST until 07:30 CEST (UTC+2) during which the game server will be offline.
The overall logic behind granting Credits and XP per battle was reworked to include the above changes. Consequently, if a player carries out useful actions (causing damage, fighting for control of key areas, supporting allies etc.), their average rewards will not change compared to Update 0.5.11. However, players whose behaviour is mostly “one-dimensional” (for example, battleship players that stay away from the heat of battle and try to only cause damage to the enemy rather than lead the attack, or destroyer players that focus on capturing key areas but do not perform any reconnaissance) will be less generously rewarded.
Update 9.16 includes optimisation of the visibility system, which is an important component of the game. Now, spotted vehicles will be displayed much quicker on screen – the difference will be particularly noticeable when spotting enemy vehicles at a distance of more than 300 metres.
In this article, we will explain how we managed to achieve such a result, and how it will affect the game.
Vehicles that spot enemies from a long distance will have a little more time to make a shot right after spotting an enemy vehicle.
Spotted vehicles will have a harder time crossing short, open areas, since they will be immediately displayed on the screens of their opponents. They will have to either perform active manoeuvres to make it difficult for the enemy to target them or choose a different route.
Achieving this improvement to the spotting system required solving two problems: firstly, how to rework the server calculations and, secondly, how to speed up the display of spotted vehicles in the game client.
Our server programmers re-wrote the visibility-system code, keeping all previous operating rules. The transfer to a more effective programming language allowed them to reduce the server load and increase the performance of the visibility system. In other words, the system started working more effectively while using the same rules and consuming less resources.
Detection Check is a server request that determines whether an enemy vehicle is within the visibility range and whether it can be spotted. The visibility check is performed constantly, with its interval depending on the distance from the enemy.
The interval between visibility checks depends on their distance: checks are more frequent at a short distance, and as the distance increases the interval increases as well. The optimisation performed allowed us to increase the number of visibility checks two-fold at nearer distances and adds additional simplified checks for further distances, leading to a three-fold increase in the total number of visibility checks in comparison with the previous version. The increased frequency of visibility checks in turn allows us to reduce the delay when displaying spotted vehicles.
Client-related improvements have saved about 60–70% of the time required to render a vehicle model. Now, when entering a battle, the vehicle models and map objects are all loaded in together. This solution allowed us to reduce the rendering time upon spotting as well as the possibility of “freezes” when enemy vehicles are spotted.
Roll out!
Tank guns have several features fitted to their externals that improve certain characteristics of the gun or the tank in general; or serve to protect the gun and crew. Let’s have a look at three features to see what they do in the coming days. Let’s start with the muzzle brake.
