- For PC
- For MAC
- For Linux
- OS: Windows 7 SP1/8/10 (64 bit)
- Processor: Dual-Core 2.2 GHz
- Memory: 4GB
- Video Card: DirectX 10.1 level video card: AMD Radeon 77XX / NVIDIA GeForce GTX 660. The minimum supported resolution for the game is 720p.
- Network: Broadband Internet connection
- Hard Drive: 17 GB
- OS: Windows 10/11 (64 bit)
- Processor: Intel Core i5 or Ryzen 5 3600 and better
- Memory: 16 GB and more
- Video Card: DirectX 11 level video card or higher and drivers: Nvidia GeForce 1060 and higher, Radeon RX 570 and higher
- Network: Broadband Internet connection
- Hard Drive: 95 GB
- OS: Mac OS Big Sur 11.0 or newer
- Processor: Core i5, minimum 2.2GHz (Intel Xeon is not supported)
- Memory: 6 GB
- Video Card: Intel Iris Pro 5200 (Mac), or analog from AMD/Nvidia for Mac. Minimum supported resolution for the game is 720p with Metal support.
- Network: Broadband Internet connection
- Hard Drive: 17 GB
- OS: Mac OS Big Sur 11.0 or newer
- Processor: Core i7 (Intel Xeon is not supported)
- Memory: 8 GB
- Video Card: Radeon Vega II or higher with Metal support.
- Network: Broadband Internet connection
- Hard Drive: 95 GB
- OS: Most modern 64bit Linux distributions
- Processor: Dual-Core 2.4 GHz
- Memory: 4 GB
- Video Card: NVIDIA 660 with latest proprietary drivers (not older than 6 months) / similar AMD with latest proprietary drivers (not older than 6 months; the minimum supported resolution for the game is 720p) with Vulkan support.
- Network: Broadband Internet connection
- Hard Drive: 17 GB
- OS: Ubuntu 20.04 64bit
- Processor: Intel Core i7
- Memory: 16 GB
- Video Card: NVIDIA 1060 with latest proprietary drivers (not older than 6 months) / similar AMD (Radeon RX 570) with latest proprietary drivers (not older than 6 months) with Vulkan support.
- Network: Broadband Internet connection
- Hard Drive: 95 GB
As all our players know, we never adjust fixed vehicle characteristics to improve balance. All vehicles in the game are made to be as close as possible to their real life characteristics, to the extent it is possible with the level of detail of our physics model and the documents that we have. |
This approach has several advantages for players, including the chance to use the game as special interactive library and the opportunity to check ‘whether a whale is stronger than an elephant’, an opportunity to compare vehicles in battle that never faced each other in real life combat.
If the players are justified in thinking that a particular vehicle was different in real life to the way that it is portrayed in the game, then the game will be changed to match, sooner or later (if the evidence they present is correct and reliable).
We are constantly working on ensuring that the performance and special features of all vehicles are well detailed both in terms of the setup of individual vehicles as well as the overall model.
Some of our recent additions include the introduction of advanced thermodynamics as well as improving the shell explosion model.
Naturally, some allowances do have to be made as long as the game stays a game, but we still try to bring everything that you can experience in a 5-30 minute battle into the game. The main allowances are related to the crew and to vehicle repair, which affects in-game tactics in certain ways, but does not change what will happen to your opponent straight after you fire or the way they move into firing position.
The balance of battles in the game is adjusted using the Battle Rating — a measure that affects who will oppose you in battle. |
There are many that believe that the developers consider vehicles with the same Battle Rating to have the same combat effectiveness, but it does not work that way at all. This rating only determines the possible opponents (this becomes obvious when you look at bombers and fighters). The basic premise of Battle Ratings is that they affect the tables of possible meetings in battle (see the related development blog).
Before Battle Ratings were introduced, aircraft were balanced according to their ranks. There was approximately the same number of ranks as there are battle ratings now and they were used in exactly the same way to determine possible opponents (but in those faraway times the fork was much wider, with +\-5 ranks). The availability of different game modes with different types of tasks and special features required different tables to be used for each mode (i.e. their own Battle Ratings).
Any changes to the width of the fork of the Battle Ratings that appear in battle and the Battle Ratings of the vehicles themselves are based on statistical data, analysis of player opinion and by considering which vehicles can meet in battle and the way it can affect the results of the battle (as opposed to thinking that because two vehicles are approximately equal, they should have the same BR—sometimes this is the case, but not every time).
We often hear that the balance should be adjusted in some other way, that it should be ‘historical’ or based on technical characteristics. There is no such thing as historical balance in any way, shape or form—there were few battles that were fought on equal terms as strategists always try to have the greatest advantage even before the battle. Another consideration is that the two sides of most battles would have different objectives—the mission of one side might be to occupy a staging area, while the other side could be trying to inflict the maximum possible losses on the enemy, or to delay them until the reserves arrive.
Vehicles were never created equal—most vehicle models were designed for a specific, non-ambiguous purpose. |
For example, fighter aircraft could be designed to fight bombers (in this case, with more powerful armament and greater speed), or to escort long range bombers (here they would have to be able to achieve long ranges and be able to operate effectively at the heights that bombers fly, while powerful armament is not as important). There are other typical tasks—anti-aircraft defence, patrolling, reconnaissance, assault, coastal defence and night missions. All of these applications affect the technical characteristics of the vehicles. Their historical effectiveness was affected by both their characteristics compared to other vehicles and the number and specialization of possible opponents.
Another significant factor that affected the popularity and effectiveness of vehicles was the ease of crew training, their predictability in the hands of the pilot and ease of servicing and repair. Any vehicles that had not been fully developed and were unreliable or overly complex were rarely the pilot’s favourites and it was rare for them to display their powerful sides.
Land vehicles tasks could be even more varied: assault vehicles could penetrate well reinforced enemy positions; anti-tank vehicles would destroy enemy tanks (preferably at a long range), reconnaissance, infantry support, anti-aircraft, etc. The majority of ground vehicle models were also designed to support or attack infantry.
In every situation, most vehicles could also be used against other vehicles—this flexibility is another positive aspect. However, every vehicle also had its main role and purpose. Quite often, the same vehicle model was re-purposed or modified to fulfil new objectives when they appeared as well as when it became obsolescent.
Sometimes, vehicles were made to be used under the conditions of numerical or air superiority, sometimes only for defence and sometimes only as part of units that required support from another type of vehicle and other specific uses.
Neither their years of service, the years they were designed and accepted for armament nor the battles in which various types of vehicles fought each other show that they were equal.
Balancing by using technical characteristics is a tool that can only be used for duels, where none of the other objectives and tasks that make War Thunder stand out among other games are considered. These unique features allow us to create and use different vehicle classes and unite ground and aerial vehicles in one battle. |
The use of tables instead of Battle Ratings provides a possibility of using more precise settings (at least, it allows for different tables to be used for different countries) but also makes them harder to understand and interpret. It then becomes impossible to identify who you will ‘meet’ at a glance, as tables of changes during upgrades would take up many times more text than what is used for all the changelogs at the moment.
This is why there is no reasonable alternative to our Battle Ratings and most players have acclimatised to the current system which leads us to discussing the road ahead.
To be continued...
Kirill Yudintsev
Creative Director of Gaijin Entertainment
Comments (126)
I thought they were announcing new BR system or MM, this is completely waste of time! We all know we need BR system, but is it working properly? No! That's why people asking for other systems, stop excusing and fix your MM or change BR system completely...
I believe real problem is MM which puts 12 6.7 against 12 7.3, nobody can say this match is balanced at all. QUE TIME QUE TIME QUE TIME, if matches aren't fun we won't get more players and que time will be always a problem. Priority should be balance so matches can be truly fun..
It all sounds like a "too late justification" for ignoring the playerbase. Just take a look at this: https://forum.warthunder.com/index.php?/topic/232765-lower-the-br-mm-spread-from-10-to-07/ "...the Battle Ratings of the vehicles themselves are based on statistical data, analysis of player opinion and by considering..." "PLAYER OPINION"?? . What a joke!
Lol I was thinking the same thing
What statistical Data, to this point Gaijin never said anything about the calculation. There is no reason to assume that this is done correctly. There is no sign that robust statistics were considered, or that it was done with any expertise in trend estimation theory. In one of the first threads about this, there was an example that average time in battle is used (i guess to get the survivability). It was done as an example what the BR ist.
Anyone with experience in statistics/trend estimation would be warned imidiatley. Wich Average? the mean? the median? WHAT? the most commonly kown average is the mean, so there is a good chance it was used... it's also the worst average to use in trend estimators (Red Flag No1) This is an extremely important question... so i asked it noone could answer (Red Flag No2).
A reason given was that they can't tell me because players will mess with the system. BUT trend estimators are designed to iron out this flaws, so it shouldn't be an issue. (Red Flag No3).
One thing about all those different roles however, is that they are often limited in their application in WT - a long range fighter in an Air battle is no different o a short range fighter - its range is irrelevant. It also precludes playing vehicles in the manner in which they were used historically - a Tiger 1 is not a "heavy tank" when all it faces are other vehicles that can destroy it with some degree of ease. So important historical differences become irrelevant. I hope you can fix this
This BR system was very fair ,Hmmm...In 6.7 battle,BR system set one 6.7 tank to germany fight to 6.7 enemy tank like this:" https://imgur.com/6Tr5YLp "
"Ease of servicing and repair" - doesn't apply to gameplay, where every vehicle begins "factory fresh", fully supplied and crewed "Assault vehicles could penetrate well reinforced enemy positions" - There are no reinforced positions in game. "two sides in most battles would have different objectives" - In game, both teams have the identical objectives. "conditions of numerical air superiority" - such conditions never occur in game Don't let factors outside the game's scope influence gameplay
Ground forces is honestly a pain to balance, but at this point, they're pretty balanced (save top tier, and I can understand why it's a mess up there). I honestly think the best way to fix most balance issues is really at this point map design; they need to get bigger, with some urban and loads of open space, but with varied terrain. Russians are urban, Germans wide open spaces, Americans and Brits flank hard and wide. The mm seems to favor the narrow boxed in maps that favor russian tanks.
I first started playing War Thunder in 2013. I took a break for 3 years then came back. In 2013, I saw the world war thing. Now its 2016 and it still isn't available?
Dear Gaijin Thanks for being so out there, maybe its just my personal impression but I think since you brought out the Russian bias advance pack you really became much cooler as devs. And speaking about BRs, I dont know about the majority, but most (if not all) people who I know would rather have more spread out BRs (this is mostly for ground forces, even though late tier planes could do with a few more .3 and .7) although it means increased queue times. What do you think of that?
Good morning everybody!!!!WW2 it's next big update like 1.57? Including vehicle and airplane and othersee?
Submit a complaint