Jump to content

APDS (presumably) tweaked to not pen angled surfaces


Best answer

Happy New Year!

 

As you can imagine, Moscow is heavily into the New Year holiday at the moment, but I have been able to have a chat with BVV, now please understand, that translating a conversation where I speak English and he speaks Russian about such technical details does have drawbacks, however, here is the gist of it - at the moment information about the parameters of the shells is stored in the file damageModel.blk, the conclusions drawn based on other files may be erroneous. As for the connection of the inclined armor of the IS-6 and the fall of the penetration at an angle for similar projectiles, it does not exist, these settings for the armour angle of 29 degrees were in the file and were used one year before the IS-6 was added to the game. These values were based on the effect of a hit against the sloped front on a T-54, but it is also relevant regarding the Soviet 100mm or 122mm APDS shells in striking the American M60, where there is also an angle of inclination on the upper plate at an angle slightly greater than 60 degrees.

 

This adjustment is based on several facts known to us - test reports of US and British sub-calibre projectiles, where it becomes clear that the barrier equivalent to the frontal armor of the T-54 body (120mm / 60 degrees, tested on a 5 inch plate) was particularly difficult in penetration for sub-calibre shells of similar design, and increasing the angle of shell vs armour greatly increased the probability of non-penetration.

 

One source readily available is "Armor and Gunnery - Lorrin Rexford Bird & Robert D. Livingston" , where it shows relevant graphs of the angle of the meeting - the effect of tilt for different types of projectiles. This also demonstrates a sharp increase in the protective effect when the slope angle of the plate is reduced to less than 30 degrees.

 

image.png

 

Now I want to make a couple of things clear, our development team do not go out of there way to implement  any kind of bias, they strive to get hold of documents and publications from across the globe, some are difficult to obtain, some are easy, but one thing is clear, few show the same results accurately. If material can be obtained via sourcing with help from you guys, then we will use it. It is also clear that bias seems to operate from the players side too, the observation made by some here, is that this is deliberately done to disadvantage British vehicles, no, it really isn't as it also affects Soviet vehicles too, not to mention that the mechanics were in place long before the IS6 arrived. Data mining is great fun isnt it? But it only reflects data being interpreted in a player viewed manner, the problem with that is that it is a tool where relationships with other data is specifically tasked by our developers. Additionally, to those howling about having their posts warned/hidden - follow the rules, stay on topic and discuss it like civilised human beings and they wont be warned/hidden. Simple.

 

If there is relevant data, then pass it to the development team via the reporting procedure. Now chill out and enjoy the holiday period.

Share this post


Link to post
Share on other sites

I quote from a great Reddit post made by @Sonoda_Kotori.

 

"TLDR: Gaijin set up the normalization and penetration curves of APDS in a way that it would cleverly nerf its penetration against most Soviet tanks that has a UFP with a construction slope of 60-65 degrees. However, the stat card only talks about the "60 degrees" in a perfect world condition, so when it looks like an APDS can marginally penetrate, it often cannot.

DATAMINE TIME

First of all, projectiletypes.blk defines what presets are used by a type of shell.

"apds_tank": { "normalizationPreset": "apcr", "ricochetPreset": "apds", "slopeEffectPreset": "apds",

Therefore, an APDS follows the terrible normalization like APCR shells.

But is that it? No. Most Soviet UFPs were built with an angle of 60 degrees (or 30 degrees, which was used in Gaijin's code.), for example, .

According to slopeeffect.blk, which defines the slope effect encountered by every type of shell. Most shells' penetration penalty curve goes up by increments of 5 degrees, plotting a smooth curve. HOWEVER, for APDS' slope effect, there's a sudden BUMP at 29 degrees (61 degrees construction angle):

"caliberToArmor": 1.0,

"slopeEffect0deg": [0.0,20.0],

"slopeEffect5deg": [5.0,17.0],

"slopeEffect10deg": [10.0,15.0],

"slopeEffect15deg": [15.0,10.24],

"slopeEffect20deg": [20.0,7.2],

"slopeEffect25deg": [25.0,3.72],

"slopeEffect29deg": [29.0,2.76],

"slopeEffect30deg": [30.0,2.46],

"slopeEffect35deg": [35.0,2.24],

"slopeEffect40deg": [40.0,2.064],

"slopeEffect45deg": [45.0,1.8],

"slopeEffect50deg": [50.0,1.4],

"slopeEffect55deg": [55.0,1.2],

"slopeEffect60deg": [60.0,1.024],

"slopeEffect65deg": [65.0,1.0],

"slopeEffect70deg": [70.0,1.0],

"slopeEffect75deg": [75.0,1.0],

"slopeEffect80deg": [80.0,1.0],

"slopeEffect85deg": [85.0,1.0],

"slopeEffect90deg": [90.0,1.0]

What does that mean? Well, when you are shooting at a plate with 60 degrees installation angle (30 degrees in Gaijin's code), your penetration at 90 degrees (303mm for L7's APDS) undergoes a decay of 2.46 times, resulting a final penetration at such plate of 123mm only. You can find this data on L7's stats card.

What the stats DID NOT show was what will happen when that angle is pushed over to 29 degrees, which it always happens to an APDS (since it has the terrible normalization of an APCR), your glorious 123mm at 30 degrees is nerfed down to 109 degrees. Making the matter worse, the curve SKYROCKETS at 65 degrees construction angle (25deg for Gaijin), with the shell undergoing a huge 3.72 times penalty, resulting in a penetration of merely 81.5mm for an L7 APDS, a postwar shell that's meant to penetrate most things.

The best part is, that the IS-6's UFP has a construction slope of... You guessed it, 65 degrees. With a slight angling, most Soviet 60deg UFP can easily become 65 degrees too.

P.S. A chart of IS-6's armor values

Hull:


100 mm (62-65°) Front glacis / 120 mm (53°) Lower glacis / 100 mm (65°) Driver's port

Sides:


100 mm (45°) Top / 100 + 10 mm Bottom

Rear:


60 mm (60°) Top / 100 mm (59°) Top sides / 60 mm (30°) Bottom"

 

Pfantom made some great graphs showing how ridiculous the slope modifier gets:

 

0y71gnd.jpg

 

arL9hez.jpg

 

My thoughts:

 

So, it appeared that projectiletypes.blk and slopeeffects.blk are not used anymore, instead damagemodel.blk is used. This file however contains the exact same data, so this find stays valid. Thanks to @_mike10d

 

Now, what does this mean in the game? Well, it means that whenever a T-54 (except 1947 model), T-62 or T-55 angles very slightly, 105mm APDS won't pen. Even though it clearly should. But that's just for the L7. Many more of such examples could probably be made for various calibres of APDS shots.

 

I don't want to start any conspiracy theories, but it's really hard to believe this is not intentionally done so. It's such a very specific value that perfectly fits these Russian tanks, and highly reduces the effectiveness of L7 APDS and other shells. I don't really want this kind of stuff to be in the game I love.

 

"But Mustang, why didn't you just put it in Bussian Rias thread?"

 

Because I feel this topic is worthy of a thread on its own. I don't believe in Russian Bias, but this is something that really boggles my mind.

 

"But Mustang, why didn't you just make a bug report?"

 

Because I don't have the documents which show how it should be, but we can hopefully agree on the fact that it shouldn't be like this. If someone wants to make a bp out of this, feel free to do this. I'm only posting what I've read so far.

 

Your thoughts?

 

Link to Reddit post: https://www.reddit.com/r/Warthunder/comments/7ngxcp/why_apds_is_bad_against_soviet_armor_exposed

 

Edited by LordMustang
  • Like 5
  • Thanks 21
  • Haha 1
  • Confused 3
  • Upvote 110

Share this post


Link to post
Share on other sites

Hmm, that would explain my L7 apds bouncing that leo twice yesterday where it formerly would've penned. And today I bounced an IS6 twice with my Conway, while shooting that 100mm plate that's supposed to be a weakspot, point blank. I used to reliably kill IS6s with my long 88 that way and today 120mm APDS wouldn't pen it? Other than this it seems business as usual

  • Upvote 3
medal medal medal medal medal medal medal

Share this post


Link to post
Share on other sites

This is great news!! Especially because I recently acquired the object 120 and I'm currently grinding towards the T-54. But speaking of the object 120, I've noticed I've been getting one shotted way too often which is unacceptable for a glorious soviet vehicle so I propose you nerf all rounds so they can't pen the 30mm armor!

  • Haha 27
  • Sad 1
medal medal

Share this post


Link to post
Share on other sites

Sorry if I wasn't clear, I edited OP. This applies to all APDS calibres. It's a shared characteristic. The L7 is a noticable gun/example suffering from this.

  • Upvote 23

Share this post


Link to post
Share on other sites

Warthunder is a completely unbiased game and all findings in OP must be coincidences.

 

You agree, comrade? (commissar with 4 armed guards behind him)

  • Like 1
  • Haha 6
  • Upvote 53
medal medal medal

Share this post


Link to post
Share on other sites

That's interesting. 

Thx for sharing.

 

Now we just need to give this more attention so the devs have to fix it.

Simply as that.

Edited by Iron_physik
  • Haha 5
  • Upvote 12
medal medal medal medal medal medal medal medal

Share this post


Link to post
Share on other sites

Welp, if this isn't how the APDS should work, and if this was intentional, then maybe they'll change it with community outcry much like the IS-6. I mean, APDS already has some pitiful damage from most angles, adding this crutch seems like kicking someone while they're down.

  • Like 2
  • Upvote 21
medal medal medal medal medal medal medal

Share this post


Link to post
Share on other sites

Darn, if this was brought this to their attention sooner it could have been fixed already.

 

Oh wait, I guess it's already been fixed.

Thanks to reddit user Curanthir for digging that up.
 

I guess this should have been bug reported before the UK tree was even introduced then? Someone want to go ahead and bug report the entire Italian GF tree preemptively?

 

Maybe we could crowdfund a firing demonstration of an L7 demolishing a T-54 for Gaijin to watch? Seems like they've made up their mind and are stuck in their own reality, and the only thing we haven't tried yet is a physical demonstration to snap them out of it.

 

On a serious note, the current values aren't bugs, mistakes, errors or misunderstandings. They are exactly the way Gaijin wants them. Gaijins number one priority is to make money (and in my personal opinion to push the mythos of Russian Supremacy second), the statements about seeking historical accuracy are just lip service to attract customers. The only thing that will change their mind is a threat to their bottom line, which bug reports and reddit threads do not do.

Edited by ADrunkMoth
  • Like 1
  • Thanks 1
  • Upvote 40
medal medal medal medal medal medal

Share this post


Link to post
Share on other sites

On a related note, in game APBC out-performs APCBC at every angle and caliber-to-armor ratio. APCBC being the standard for post-war shells because it was an objectively better type.

 

Guess what nation uses a lot of APBC in game.

 

 

  • Thanks 1
  • Upvote 28
medal

Share this post


Link to post
Share on other sites
8 minutes ago, IJN_DD_Yuudachi said:

On a related note, in game APBC out-performs APCBC at every angle and caliber-to-armor ratio. APCBC being the standard for post-war shells because it was an objectively better type.

 

Guess what nation uses a lot of APBC in game.

 

 

wait  germans ? or is it the baguettes ? maybe the uk ?  

  • Haha 6
medal medal medal medal medal medal medal medal

Share this post


Link to post
Share on other sites

Well that's kinda interesting. Very unsurprising, only, lets call them "hardcore fans", were convinced that the game isn't rigged. 

 

To be clear. That's rigging. Not making an honest mistake, but deliberately skewing the odds in your favour. 

  • Thanks 2
  • Upvote 34
medal

Share this post


Link to post
Share on other sites
5 minutes ago, Sonoda_Kotori said:

APBC and APHEBC use the same calculations, and guess who use APHEBC now)))))

APBC is the single most over-performing and illogically powerful round in the game. Does anyone remember the uproar when people pointed out APFSDS can penetrate better than LOS (which it actually can)? APBC not only does that against several angles, but in many cases it completely ignores the angle. Gaijin has programmed a shell in the game which can impact a plate sloped at 45 degrees and not lose a single mm of penetration compared to a vertical plate. Does anyone else realize how insane that is? And what do you know, APBC (by in game definition) was a purely Russian design used exclusively by Russian tanks (with the exception of a few US APBC designs which are pegged to RUS APBC standards because Gaijin can't admit US APBC was better against slopes in reality). So we have a game where guns designed and tested for the specific purpose of reliably defeating certain tanks are in no way able to, and on the flip side tanks who had quite poor anti-tank performance are praised and portrayed as war-winning wonder weapons.

  • Like 4
  • Thanks 1
  • Upvote 36
medal medal medal medal medal medal

Share this post


Link to post
Share on other sites
Guest
This topic is now closed to further replies.
  • Recently Browsing   0 members

    No registered users viewing this page.

×
×
  • Create New...