World of Tanks

World of Tanks: Developers Q&A

We all love Q&A from World of Tanks and it has been a while since we had one. Reddit NA Family Community Coordinator, TollhouseFrankhas been doing a long Q&A with WGNA and in his last batch of questions, he got quite a few replied by the developers Audrey and Anton B.

I will be publishing the majority of the questions, but instead of making a single massive article, they will be split into multiple ones with the most relevant information, so in result this article only as questions and answers from developers. Massive thanks to TollhouseFrank, for letting me share these.

Source: World of Tanks Reddit

Q&A

What is the reason only console WoT (Console) gets special single-player modes? Is there a plan for these modes to come to PC?

Andrey: Yes, there are. We started experiments with cooperative PVE the last year with the newcomer’s Bootcamp and Halloween special event, and we are planning some more PVE-oriented activities soon. Ultimately, our goal is to build a technical base for WOT PC, like high-performance server-side systemic AI solution, and prove it before taking further steps into a permanent single player or cooperative experiences.

Can spectator mode be added for Strongholds and Clan Wars so that clan mates can watch and learn?

Anton B: A few months ago we tested elements of a new spectator solution during our Frontline event. Once done, we will consider this application. For me this idea sounds worth trying.

How exactly is it decided by WG that a tank is overperforming or underperforming? Why is it that it can seem to take a year or more to make this determination whenever the player base is calling for action sooner?

Anton B: We try to make our decisions based on hard data, rather than on feelings. I mean, that it seems very easy to judge a vehicle from a subjective point of view. But, it easily can become a rushed, wrong judgement. Once a new vehicle is added, it’s quickly adopted by the most active, and often the most skilled of our players. Plus, others have a very limited knowledge on how to combat this new enemy – what are its weaknesses, where to shoot and so on. So, for the first several months any statistics related to it are distorted because of who is playing it, and how people play against it. After a while, usually in about half a year give or take, stats stabilize and we can separate the human factor from vehicle power to draw any conclusion on how the vehicle actually performs.

Would it be possible to bring all chat back with Alive/Dead players kept in separate channels?

Anton B: For now, we are not planning to do anything with cross-team chat.

With regards to the maps, will there be any effort put forward into making maps more ‘fluid’ for matches?

Anton B: Not sure what this question is about. Match fluidity is a combination of multiple factors, including the layout, tier the match is being played, the combination of vehicles in both teams and their exact composition, its breakdown, players’ personalities, individual tactical preferences and multiple other factors. Maps that are designed to reduce the variety of these factors are generally unpopular, and sooner or later are called off for the redesign. Unfortunately, not every design, or redesign, is equally successful.

Would it be possible for us to get more detailed feedback from the validation team at WGMods, as well as an understanding of how exactly they test the mods we upload?

Anton B: There are two stages of testing. 1. Every new mod of new version of existing mods does not become available for download instantly after uploaded. First, it’s pre-moderated by a special team (currently it is volunteers who are good with mods and have a lot of experience working with them) – they run basic smoke tests and make sure there’s no inappropriate content in the mod. As soon as the mod is checked, it becomes available to players. 2. (Optional step) Upon achieving a certain number of downloads, a mod undergoes more thorough stability and performance testing by our QA team. If any issues arise, it’s sent back to its creator for improvement. There’s a hidden feature in the backend to mark some of the mods as ‘trusted’. These mods will skip the usual testing procedure and become available instantly. We use this feature for our official mods and super-trusted mod makers whose mods are checked by our QA team prior to WoT updates anyway.

Are Devs looking at RNG and possible changes to it?

Anton B: There are multiple RNG elements in the game. Generally, when players speak about RNG, they mean damage and penetration distribution. These rules are in the game to represent armor and shell quality variance and other variables too complex to calculate in real time. For example, many cast turrets had varied armor thickness due to their complex shape, and even hundreds of armor groups won’t reflect it with any grade of accuracy. So instead we use randomization to emulate the influence of such factors on an individual shot, and we are quite happy with how these two work. However there is one RNG element we are looking into right now. It is a random function governing how shots are spread within the reticle. I can’t share any specific info on what are we planning yet, but we aim on making it more predictable and better reflecting what a particular gun is good (or bad) at.

What about +1/-1? Wouldn’t that be an easy solution to Matchmaker?

Anton B: Henry Louis Mencken once said, that for every complex problem, there is an answer that is clear, simple, and wrong 🙂 Speaking seriously, the current MM supports two-tier battles and we have no indication people are enjoying them more than three-tier ones. The issues with MM we are facing now (for example, with preferential MM vehicles or with adding extra strict matching by roles) are related to its code architecture and to the servers’ cores utilization, not with the principal matchmaking rules. After receiving really strong community feedback on our approach to preferential vehicles changes, we are now looking into rebuilding matchmaker’s core. Once done, it will naturally lead us to reviewing every MM rule once again to ensure none of the decisions for a new architecture are dictated by the old tech requirements. I promise we will take another look into +X/-X rule as well.

Why has there been a focus on map changes that seem to discourage ranged play?

Anton B: We don’t have such focus. Some of our new maps, like Glacier or upcoming Studzianki offer plenty of long-range gameplay. Also, I’ve heard complains that we have too much bush sniping in our game, y’know… 😉

Will there be updates or changes to how map rotation is handled, so that we do not see the same maps or similar style maps over and over in a row?

Anton B: Last year, we modified map pulling rules to lower the chance for each player to see the same map twice in a row or just too often, and based on players feedback and our data we have succeeded – with current maps selection the chances to play the same map twice in set of 7 matches are less than 1%. However, that solution does not cover “styles” of maps it pulls. First off, there are people who want style variety, and there are people who don’t, and by satisfying one group we’ll alienate the other; so leaving it to RNG looks like overall better idea. Then, the way servers handle arenas require us to maintain a certain quantity of maps for variance pulling to work. Thus to support style-based pulling in addition to the current one we will need LOTS of maps, many times more we have now.

When will we get British Light Tanks to tier 10?

Anton B: Unfortunately, not this year. We are about to start discussing the production plans for 2019 and I definitely will bring this question to the table.

Do you know when we will be getting the Chieftain Tank?

Anton B: That depends on what Chieftain you do mean. Right now we are considering adding a Chieftain and US T95 hybrid, known as Chieftain/T95 – the product of standardization and parts interchangeability projects initiated by the UK, United States and Canada in 1957 – 1959 when both the Chieftain and T95 were still in development; but it is unlikely we will add the Chieftain Mk I any time soon.

Will we see a shift towards selling things like preset camo and other visual-only premium content, and as a follow up, would such sales possibly allow premium tank prices to go down?

Anton B: Over the last Christmas, we offered the first out of four planned Visual Customization updates, one more is in production right now and we plan to have it live really soon. So, we indeed wish to offer to you guys better decorations for your favorite vehicles. At the moment, I can’t foresee how it will impact the pricing of the other premium options. Mainly we aim to have lots of customization pieces becoming achievements based rewards.

Was the account roaming idea completely scrapped? And if so, are other options considered to facilitate playing with friends from other servers?

Anton B: It wasn’t completely scrapped, but as it affects very few of our players, we decided not to look into it this year in favor of other features. I’m sorry about that.

When it comes to in-game physics, are there any current plans to revisit physics to deal with some of the weird behavior regarding rocks, buildings, and how tanks flip?

Anton B: Yes, both for particular vehicles physics tweaking and map objects collision shapes changes. Still, I want to stress that it is often a question of compromise as often tweaks that are addressing one weird behavior are also creating a new issue, or few issues, due to the almost endless amount of various situations a player can find himself at. So, sometimes we just have to stop when it is “good enough” not to make it worse.

Will an easier method of righting a flipped vehicle come to the game to replace simply pushing the allied tank?

Anton B: To the best of my knowledge, no.

Is it possible to change rocks back to the way they used to be pre 1.0?

Anton B: If you mean glitches with some objects collision, than yes and we are constantly working on it; I’m really sorry it is taking longer than we all wish for. If you mean changing the rocks models back to what they were before (or re-making them with new engine so they will match vertice-to-vertice of the old models), then I’m afraid my answer is no.

Are there any more regular branches in the works for existing nations? It feels like we’ve been too spoiled lately with new nations and premiums.

Anton B: Yes, but I won’t spoil.

What is the current status of armored cars, alternate hulls, multiple guns/turrets, and server roaming? I’m not looking for an ETA or anything, but I’m curious if these ideas are just shelved for time being or if they’ve been cancelled entirely.

Anton B: Tricky one, but I’ll try to answer without too many leaks. We do not have any plans for now to introduce alternate hulls, and we are not working on server roaming at the moment. We are seriously considering at least one tank with two guns, namely ST-II, but we are not completely happy with multi-turret tests we made during the last Halloween event, and will not release anything unless we have a better controls for several weapons (in essence, two fully synchronized turrets are very similar to two guns on one turret, at least from input / controls perspective).

What are the developers’ opinions on recoilless rifles, high-tier autocannons, and low-caliber (50-75mm or so) smoothbore guns?

Anton B: These things do not fit into current World of Tanks paradigm. But who knows what the future holds? 😉

Is there any intent on doing a balance sweep back through old tanks that haven’t been updated since patch 9, give or take? And, to help provide insight to the development balancing, what is the team’s cadence on reviewing performance and balance on tanks? If you can speak to it, what is your +/- % performance gap (10%, 15%, etc.)?

Anton B: Yes, we are working on the rebalancing of mid and low-tiers right now. This is huge task as we are talking about making changes for up to several hundred vehicles in total. So we are starting with one nation only, to verify our design logic first. As for balancing cadence, we have a special team that is in charge of keeping everything in check, and they are working on it full time. Every change we are making, every new vehicle we are shipping is causing ripples through the whole ecosystem, changing over time what people are playing and how, what they are leveling and so on, so you can’t control it just with periodic checks.

There are automated tools that are running all the time and signaling when something is starting to get too far away from where it was intended to be. We also have a few predictive tools (not precise, to be honest), as well as all kinds of dashboards for periodic manual reviews. The main challenge for us is to withstand the pressure from our regions (sorry guys!) for as long as it is needed to collect enough good data to isolate human factors from vehicle power in the battle statistics. 😉

As for the performance gap, it is somewhat more complicated than a simple number. Making it really simple, if one would imagine that a “”balanced”” vehicle played by a perfectly average player should be at 49% of wins and 2% of draws, than a gap in performance we are fine with would be around +/- 3%. But this is too synthetic to actually use for balancing.

Advertisements

8 comments

  1. “However there is one RNG element we are looking into right now. It is a random function governing how shots are spread within the reticle.”

    “we aim on making it more predictable and better reflecting what a particular gun is good (or bad) at.”

    Looking forward to this the most.

  2. “it is unlikely we will add the Chieftain Mk I any time soon”

    Mk I? I want that sweet Mk. VI from the game files. They could nerf it to the stats of the T95/Chieftain hybrid to be released.

    Also give me roaming please, I would like to check out the Russian servers again.

    “ST-II”

    Easy: just copy the gun mechanics from the MTLS and give it the autoreloader mechanics from the pizza tanks and boom, done.

    “we are not completely happy with multi-turret tests we made during the last Halloween event, and will not release anything unless we have a better controls for several weapons”

    I had no issues with the halloween testing. Once I get the hang of it after a few battles it was so much fun to play with the multi-turreted tanks. I really don’t know how WG could improve on this. Even Warthinder is using a similar mechanic for multi-turret aiming and firing.

    “Yes, we are working on the rebalancing of mid and low-tiers right now.”

    RIP Leopard 1, JPZ E-100, E-100…

    1. MTLS… MTLS has no multi-gun mechanics. The shot literally comes out from the nothing that is between the 2 modeled guns.

      And no, twin gun should not be the same as auto-reloader. What’s wrong with you?

      1. “And no, twin gun should not be the same as auto-reloader. ”

        Who said that all multi-gun systems should work like this? You won’t have more loaders and gunners in the crew magically to operate the two guns seperately in the ST-II just because they shove 2 of the same guns into it. What’s wrong with you?

  3. Harkonnen, you are very welcome! I am glad that you found the information interesting enough to put on your site.

  4. It seems that the question of the rework of prem ammo mechanism missed.

    “Why has there been a focus on map changes that seem to discourage ranged play?”
    Because Russia!

    1. Can this question be found somewhere else or do you mean that they forgot to ask it? Asking because I’m very interested in that particular question 😉

  5. Some of these answers are just complete and utter bullshit. About rebalancing tanks and taking a year to rework them: What in the hell?! The Object V4 was CLEARLY overpowered. Just stats don’t say everything apperently. Otherwise it would’ve been nerfed much earlier. Or what about the Chinese mediums? These tanks have been here for years and still they haven’t received any love (except for the premiums… how ironic, huh?).

    Wargaming needs to get their shit together and work their asses off to make this game friendly for players who like to try all different tanks from all different tiers from all different nations.

Comments are closed.