devtrackers.gg
about 1 month
ago -
magazine2
-
Direct link
In this article, our naval designers will be talking about the new modeling of fragmentation damage from naval projectiles that were introduced to the game with the release of the Alpha Strike major update.
In summary: Naval projectiles have had their fragmentation damage changed in the Alpha Strike update, where before the maximum number of fragments produced was smaller than it could’ve been due to performance constraints. Additionally, before we also had to manually configure the damage fragmentation of each naval projectile. We’ve now developed a new system for this that has given us better flexibility when calculating the number of fragments without a loss of performance, allowing us to also increase the number of fragments produced. Upon doing this however, some ships have been taking increased damage due to geometry issues, where many have been fixed already. We’ll be keeping an eye on your feedback for any other problems that may occur, and fix these issues where needed.
Types of damage and what they depend on
All naval projectiles in the game cause three types of damage: kinetic (depending on the mass and speed of a projectile), high-explosive (depending on the mass of the explosive), and fragmentation (depending on the speed and mass of the projectile, the mass of the explosive, as well as the filling coefficient of the projectile, such as the ratio of the mass of the projectile body to the mass of the explosive). In War Thunder’s naval battles, projectiles of different types are used which as a rule, cause combined damage of several types.
The changes to the damage model of naval projectiles has only affected fragmentation damage, where it’s this damage that we’ll be talking about today in this article.
Why did we make changes?
Until the Alpha Strike update, the calculation of fragment damage had two important shortcomings. Firstly, due to a problem with reduced performance when calculating the formation of a fragmentation field, the maximum number of possible fragments for each projectile was greatly reduced. Let’s say that in reality, a shell is capable of generating up to 3,000 fragments, in the game it produced no more than 200. Calculating even this volume of fragments from shells from several salvos at once at each given moment of a naval battle meant sharp drops in FPS, which we want to avoid.
The second problem follows on from the first, where in these conditions we were forced to configure each projectile manually. This means that each naval shell model was configured according to several parameters such as armor penetration of fragments and the maximum distance of their scattering across all areas of the formation of the fragmentation field. Doing this is a long and labor-intensive process, and very vulnerable from a balance point of view, as it’s easy to inadvertently strengthen or weaken a projectile without meaning to.
How did fragment damage work before the Alpha Strike update?
Let’s take a look at how fragmentation damage affected ships before the recent update. Take for example a high-explosive fragmentation projectile with an instantaneous fuse. When hitting an object, the fuse detonates the explosive charge, which leads to an explosive sphere. As noted earlier, medium and large-sized shells produced about 200 fragments. This explosion sphere has the following parameters: armor penetration (breaking), size and damage. If the force of the explosion is sufficient enough to penetrate through armor, high-explosive damage is inflicted on all modules located behind it and falls within the radius of the sphere of destruction. Additionally, fragments from part of the nose cone also rush into the gap that was formed — about 10-12% of the total number of fragments that the projectile is capable of forming (only about 20 pieces) go through the penetrated armor. If the force of the explosion is not enough to break through armor, the fragments will not form behind it.
Once behind armor, fragments of the nose cone can strike either a ship module, the next armor layer or part of the structure. Having their own armor penetration parameters, fragments can penetrate the next armor layer or be stopped. At the same time, penetration decreases depending on the distance and previously broken armor.
When the armor is penetrated and a breach is formed, fragments scatter in a cone with an opening angle of 60 degrees. They hit anything in their path at different angles of deviation from the normal. Some of them are unable to break through more armor, and some do not hit it at all. As a result, any functional module of a ship that found itself opposite of the shell explosion receives less damage than the entire nose cone could inflict. Of the 20 fragments that penetrated for example, less than half could have hit a module, say about 8.
Overall, we’ve had to manually adjust the parameters of the fragmentation effect. With a known, small number of fragments that could cause damage to modules, the damage from one fragment could be high in order to maintain an adequate ratio of fragmentation damage to all other types of damage that a projectile has. However, the maximum damage of one fragment wasn’t made too great so that the total fragmentation damage of any of the cones did not exceed the high-explosive or kinetic damage of the entire projectile.
Is it right now?
Yes, fragment damage is now working correctly. We’ve developed a new system for calculating fragments, which allows us to give each projectile as many fragments as it’s supposed to according to its performance characteristics and do this not only without loss of performance, but in some cases even with an improvement.
A couple of years ago, we transferred the calculation of armor penetration shells to the Jacob De Marre formula, making it possible to obtain a uniform approach to calculating armor penetration. With this new system for calculating fragments, we get exactly the same automated approach to each projectile.
The maximum number of fragments, their armor penetration, damage and dispersion range are now calculated automatically. The difference in these parameters between for example 100 mm and 305 mm shells have become much closer to reality and in general have become much more accurate.
In the same situation as described above, when a shell explodes on armor, 50-100 fragments go through the penetrated armor, rather than 10-20. There’s now more chances to hit a module located behind armor, as more fragments that went through armor will hit that module.
The parameters of armor penetration, damage and range themselves have changed unevenly. For some shells, they’ve increased a little, for some they’ve decreased a little, but the maximum number of fragments for all naval shells has increased proportionally.
The fragment sections of a naval projectile has also changed somewhat. There’s three sections that can be distinguished: the front, middle and rear end. The nose cone at the front which has an angle of about 60 degrees contains up to 15% of all fragments that a projectile can produce. In the mid section, there’s up to 75%, and the remaining amount is at the rear end.
All other types of damage remain unchanged: kinetic, explosive damage and pressure damage has not changed. HE shells with a flash fuse no longer penetrate more, instead they begin to cause more fragmentation damage.
Addressing your feedback: with the new damage, can a ship be destroyed in one shot?
After the release of the Alpha Strike update, we’ve received feedback that a large ship could be destroyed by obviously weak shells, for example 100-120 mm. There was also feedback that some ships were destroyed after being hit by one shell.
When we saw this, we immediately began inspections and identified several ships that had errors in their geometry. For example, the Arizona and Hyuga could have their magazines detonated by a hit from almost any projectile in the area of the elevator under the main caliber turrets. The damage passed through the elevator, causing a fire that spread to the ammunition cellars. Mutsu’s bow magazine could detonate from a shell burst on the hull near the waterline. Similar errors were observed in several other ships. We’ve carried out prompt work to correct the errors that have been found, meaning all known issues are now fixed. We’re continuing to monitor your feedback and will try to promptly fix any other issues found. If you’ve experienced some issues, creating a report on our bug reporting platform will greatly help us reproduce them and apply a fix. Please make a report here 3.
Is it true that auxiliary caliber shells have become more effective than main caliber shells?
A ship’s armor works properly and does not allow the shells which cannot penetrate or break through to do so. HE shells have not received any more armor penetration.
Having said this, we found a bug where in one very specific combination of events, a projectile could hit the underside of a ship, where there’s little to no armor. A fix for this was released in a very recent update, so this should not be occurring again.
We’re going to be reviewing the algorithm for generating kill-log messages and the set of information in it. We’re also planning to provide more detailed operational information about the damage a player inflicts on an opponent during a match, and we hope this confusion will soon be corrected.
Thanks for reading,
The War Thunder Team
In summary: Naval projectiles have had their fragmentation damage changed in the Alpha Strike update, where before the maximum number of fragments produced was smaller than it could’ve been due to performance constraints. Additionally, before we also had to manually configure the damage fragmentation of each naval projectile. We’ve now developed a new system for this that has given us better flexibility when calculating the number of fragments without a loss of performance, allowing us to also increase the number of fragments produced. Upon doing this however, some ships have been taking increased damage due to geometry issues, where many have been fixed already. We’ll be keeping an eye on your feedback for any other problems that may occur, and fix these issues where needed.
Types of damage and what they depend on
All naval projectiles in the game cause three types of damage: kinetic (depending on the mass and speed of a projectile), high-explosive (depending on the mass of the explosive), and fragmentation (depending on the speed and mass of the projectile, the mass of the explosive, as well as the filling coefficient of the projectile, such as the ratio of the mass of the projectile body to the mass of the explosive). In War Thunder’s naval battles, projectiles of different types are used which as a rule, cause combined damage of several types.
The changes to the damage model of naval projectiles has only affected fragmentation damage, where it’s this damage that we’ll be talking about today in this article.
Why did we make changes?
Until the Alpha Strike update, the calculation of fragment damage had two important shortcomings. Firstly, due to a problem with reduced performance when calculating the formation of a fragmentation field, the maximum number of possible fragments for each projectile was greatly reduced. Let’s say that in reality, a shell is capable of generating up to 3,000 fragments, in the game it produced no more than 200. Calculating even this volume of fragments from shells from several salvos at once at each given moment of a naval battle meant sharp drops in FPS, which we want to avoid.
The second problem follows on from the first, where in these conditions we were forced to configure each projectile manually. This means that each naval shell model was configured according to several parameters such as armor penetration of fragments and the maximum distance of their scattering across all areas of the formation of the fragmentation field. Doing this is a long and labor-intensive process, and very vulnerable from a balance point of view, as it’s easy to inadvertently strengthen or weaken a projectile without meaning to.
How did fragment damage work before the Alpha Strike update?
Let’s take a look at how fragmentation damage affected ships before the recent update. Take for example a high-explosive fragmentation projectile with an instantaneous fuse. When hitting an object, the fuse detonates the explosive charge, which leads to an explosive sphere. As noted earlier, medium and large-sized shells produced about 200 fragments. This explosion sphere has the following parameters: armor penetration (breaking), size and damage. If the force of the explosion is sufficient enough to penetrate through armor, high-explosive damage is inflicted on all modules located behind it and falls within the radius of the sphere of destruction. Additionally, fragments from part of the nose cone also rush into the gap that was formed — about 10-12% of the total number of fragments that the projectile is capable of forming (only about 20 pieces) go through the penetrated armor. If the force of the explosion is not enough to break through armor, the fragments will not form behind it.
Once behind armor, fragments of the nose cone can strike either a ship module, the next armor layer or part of the structure. Having their own armor penetration parameters, fragments can penetrate the next armor layer or be stopped. At the same time, penetration decreases depending on the distance and previously broken armor.
When the armor is penetrated and a breach is formed, fragments scatter in a cone with an opening angle of 60 degrees. They hit anything in their path at different angles of deviation from the normal. Some of them are unable to break through more armor, and some do not hit it at all. As a result, any functional module of a ship that found itself opposite of the shell explosion receives less damage than the entire nose cone could inflict. Of the 20 fragments that penetrated for example, less than half could have hit a module, say about 8.
Overall, we’ve had to manually adjust the parameters of the fragmentation effect. With a known, small number of fragments that could cause damage to modules, the damage from one fragment could be high in order to maintain an adequate ratio of fragmentation damage to all other types of damage that a projectile has. However, the maximum damage of one fragment wasn’t made too great so that the total fragmentation damage of any of the cones did not exceed the high-explosive or kinetic damage of the entire projectile.
Is it right now?
Yes, fragment damage is now working correctly. We’ve developed a new system for calculating fragments, which allows us to give each projectile as many fragments as it’s supposed to according to its performance characteristics and do this not only without loss of performance, but in some cases even with an improvement.
A couple of years ago, we transferred the calculation of armor penetration shells to the Jacob De Marre formula, making it possible to obtain a uniform approach to calculating armor penetration. With this new system for calculating fragments, we get exactly the same automated approach to each projectile.
The maximum number of fragments, their armor penetration, damage and dispersion range are now calculated automatically. The difference in these parameters between for example 100 mm and 305 mm shells have become much closer to reality and in general have become much more accurate.
In the same situation as described above, when a shell explodes on armor, 50-100 fragments go through the penetrated armor, rather than 10-20. There’s now more chances to hit a module located behind armor, as more fragments that went through armor will hit that module.
The parameters of armor penetration, damage and range themselves have changed unevenly. For some shells, they’ve increased a little, for some they’ve decreased a little, but the maximum number of fragments for all naval shells has increased proportionally.
The fragment sections of a naval projectile has also changed somewhat. There’s three sections that can be distinguished: the front, middle and rear end. The nose cone at the front which has an angle of about 60 degrees contains up to 15% of all fragments that a projectile can produce. In the mid section, there’s up to 75%, and the remaining amount is at the rear end.
All other types of damage remain unchanged: kinetic, explosive damage and pressure damage has not changed. HE shells with a flash fuse no longer penetrate more, instead they begin to cause more fragmentation damage.
Addressing your feedback: with the new damage, can a ship be destroyed in one shot?
After the release of the Alpha Strike update, we’ve received feedback that a large ship could be destroyed by obviously weak shells, for example 100-120 mm. There was also feedback that some ships were destroyed after being hit by one shell.
When we saw this, we immediately began inspections and identified several ships that had errors in their geometry. For example, the Arizona and Hyuga could have their magazines detonated by a hit from almost any projectile in the area of the elevator under the main caliber turrets. The damage passed through the elevator, causing a fire that spread to the ammunition cellars. Mutsu’s bow magazine could detonate from a shell burst on the hull near the waterline. Similar errors were observed in several other ships. We’ve carried out prompt work to correct the errors that have been found, meaning all known issues are now fixed. We’re continuing to monitor your feedback and will try to promptly fix any other issues found. If you’ve experienced some issues, creating a report on our bug reporting platform will greatly help us reproduce them and apply a fix. Please make a report here 3.
Is it true that auxiliary caliber shells have become more effective than main caliber shells?
A ship’s armor works properly and does not allow the shells which cannot penetrate or break through to do so. HE shells have not received any more armor penetration.
Having said this, we found a bug where in one very specific combination of events, a projectile could hit the underside of a ship, where there’s little to no armor. A fix for this was released in a very recent update, so this should not be occurring again.
We’re going to be reviewing the algorithm for generating kill-log messages and the set of information in it. We’re also planning to provide more detailed operational information about the damage a player inflicts on an opponent during a match, and we hope this confusion will soon be corrected.
Thanks for reading,
The War Thunder Team