On normalization and penetration loss over distance of gold shells

Hello!

Recently there was a gold ammo related thread, and in the comments somebody mentioned that gold ammo does also have drawbacks and requires some knowledge to use. I decided to do some research to see how much those drawbacks actually matter, with a focus on normalisation and penetration loss over distance.

1. Mechanics

First, a primer on the relevant game mechanics.

Penetration loss over distance – AP and APCR shells lose some penetration as they fly. They start losing penetration at 100m, and reach a minimum value at 500m. The exact amount of penetration lost differs from shell to shell.

Angled armour is harder to penetrate than flat armour. The harsher the angle of impact, the stronger this effect is. Typically angle of impact is given as angle from a line perpendicular to the armour plate, so for a vertical plate it's the angle between a horizontal line and the path of the shell. Angling is why the IS-7s pike nose can bounce shots despite being only 150mm thick.

Normalization is a mechanic that reduces the effective angle of impact by a few degrees, depending on the type of shell used. For AP that is 5 degrees, so a hit at 60 degrees is treated as if it was at 55 degrees, which limits the effectiveness of angled armour somewhat.

Ricochet happens when a shell hits armour at an angle above a certain threshold, and it means that the shell automatically bounces off, no matter its penetration value. For AP this threshold angle is 70 degrees, and this angle is checked before normalization. Ricochet is why the high tier Swedish TDs can bounce shots despite having extremely thin armour.

Overmatch is actually 2 mechanics: 2x overmatch and 3x overmatch, and both apply only to AP and APCR. 2x overmatch happens when the caliber of an incoming shell is more than 2 times larger than the nominal thickness of a hit plate, and it causes the shell to receive much more normalization, depending on its base normalization and the ratio of caliber to armour thickness. 3x overmatch happens when the caliber of the shell is more than 3 times larger than armour thickness, and it prevents the shell from ricocheting. This is why those aforementioned Swedish TDs crumble under fire from high-caliber guns.

Ammo-specific mechancis – both common types of gold ammo, APCR and HEAT, have mechanical differences compared to the common AP. Those differences are as follows:

APCR:

  • Less normalization – 2 degrees versus the 5 degrees of AP, which also lowers the impact of overmatch.

  • Loses more penetration over distance.

  • Higher projectile velocity.

HEAT:

  • No normalization.

  • Does not lose penetration over distance

  • Lower projectile velocity.

  • Increased ricochet angle – 85 degrees versus 70 for AP and APCR.

  • Can't overmatch.

  • If it strikes spaced armour, it starts losing penetration as it flies further – 5% nominal penetration lost per 10 cm of flight.

2. Data

What I have prepared is a comparison of the impact of switching from silver to gold shells for nearly all tier VIII and X vehicles based on the penetration increase, the different amounts of normalization, and differences in penetration loss over distance. I have disregarded different projectile velocity, overmatch, and the spaced armour interaction of HEAT since putting those mechanics into numbers is a much more complex task. All data was taken from tanks.gg, for tanks in their "top" configuration as defined by that website. The only tanks that are missing are the ones that have the same type of shell for silver and gold, or have gold HE.

Here it is.

The link leads to a Google Sheets document containing four sheets, one for each vehicle class except artillery. Each sheet is further divided into 3 tables, one for vehicles that shoot AP silver and APCR gold, one for AP silver and HEAT gold, and one for APCR silver and HEAT gold.

Each table has a column for the tank's name, and then two similar groups of columns that hold the penetration and derived values of shells at 100m and 500m of distance to target. The "Multiplier" column holds the multiplier to penetration you get for switching to gold, calculated as the ratio of gold penetration to silver penetration. The most important column is the "breakeven angle" column, which says at which angle the bonus to penetration of gold shells is finally overcome by their worse normalization. I have marked the cells where this angle reaches 70 degrees of more in red, since by that point the silver shell or both would have ricocheted anyway. The tables for AP vs APCR also have two extra columns that compare the penetration loss of AP and APCR over distance in millimetres and proportionally.

3. Breakeven angle

The angle was calculated using custom functions that compare the multiplicative bonus to penetration you get for switching to gold with the increased effective armour thickness multiplier of your target due to lowered normalization of gold shells. If the armour thickness multiplier is higher than the penetration multiplier, that means that worse normalization has a higher impact than better penetration. The angle specified is the last full degree where gold has a better chance of penetrating, meaning it's always rounded down, never up.

4. Observations

AP vs APCR:

  • The common understanding that APCR shells lose more penetration over distance than AP is mostly false or insignificant. There's a large number of tanks for which APCR loses less millimetres of penetration than AP at 500m. For about half of tanks APCR is proportionally better at 500m than at 0m. Tier VIII Tank Destroyers are the only group where APCR tends to lose significantly more penetration over distance than AP, but this effect fades at tier X. Other than them, there's a handful of tanks with unusually low AP penetration loss and high APCR penetration loss, but they are outliers.

  • The lessened normalization of gold APCR is for the most part not an issue, indicated by the breakeven angle being at or above 70 degrees, which means that the shells would sooner ricochet than for APCR to lose its edge.

  • For all tanks that have both of their breakeven angles marked in red, firing APCR has no downside, which is a large majority of tanks. There's a few tanks for which while this is not true, like the 53TP, their breakeven angle is very close to 70, meaning its impact is limited.

AP vs HEAT and APCR vs HEAT:

  • The results for those two groups are similar to the first one, but more pronounced. Very few tanks are worse off firing HEAT than silver, and those that are tend to not be impacted much.

5. Conclusion

For a vast majority of tanks of tiers VIII and X the lowered normalization values of gold shells and their loss of penetration over distance are not significant.

For most tanks, firing APCR is strictly superior to AP in all situations, with the potential exception of overmatch, and the advantage is further exacerbated by its higher projectile velocity. Overmatch is related to normalization, but too complex for this analysis. From what I've toyed with on tanks.gg, overmatch is a guaranteed penetration or close to it no matter if the shell is AP or APCR, unless at a truly extreme angle, though this is hardly a conclusive analysis.

HEAT is a more complicated issue. Judging just by normalization and penetration loss over distance it's also strictly superior for most tanks, even more so than APCR. Similarly to overmatch, considering lower projectile velocity and HEAT-specific mechanics is too complex for me to do here.

6. A note on wotinspector.com vs. tanks.gg

I've used both tools in the past, but now that I was paying close attention to penetration mechanics I've noticed some discrepancies compared to how the game is supposed to work. For one, tanks.gg does not simulate HEAT loss over distance after striking spaced armour, while wotinspector.com does. On the other hand, wotinspector.com seems to be calculating overmatch wrong, as it merely doubles normalization, which runs counter to how WG explains the mechanic. It also allows HEAT to 3x overmatch, which should not be possible.

~Atloas, EU

Source: https://www.reddit.com/r/WorldofTanks/comments/p8bbj8/on_normalization_and_penetration_loss_over/

leave a comment

Your email address will not be published. Required fields are marked *