Jump to content
You need to play a total of 50 battles to post in this section.
Fat_Maniac

Krupp and Sigma Explained

24 comments in this topic

Recommended Posts

[POMF]
Players
104 posts
6,514 battles

VL;DR - Very long, DID read

 

Very good post mate, very well put together. Knew a lot of this already but really helped understand it all better.

 

If only this was explained in game. Maybe a toggle to switch "Advanced ship description" on and off or "Simplified description".

Share this post


Link to post
Share on other sites
[HOO]
Players
1,312 posts
2,871 battles
13 minutes ago, valoaa said:

VL;DR - Very long, DID read

 

Very good post mate, very well put together. Knew a lot of this already but really helped understand it all better.

 

If only this was explained in game. Maybe a toggle to switch "Advanced ship description" on and off or "Simplified description".

The detail  is all from Little White Mouse, all I have done is copy and paste, as it's so clear hopefully the majority of players interested will understand it!

Share this post


Link to post
Share on other sites
[HOO]
Players
1,312 posts
2,871 battles
15 hours ago, El2aZeR said:

It's funny how many people still believe sigma = accuracy when it has been explained half a year ago that sigma is essentially better RNG.

Source:

https://thearmoredpatrol.com/2017/06/15/wows-qa-15th-june-2017/

Think they might be the same people that think camping at max range is also good.

  • Funny 2

Share this post


Link to post
Share on other sites
[-T-O-]
Alpha Tester, Players, In AlfaTesters
2,185 posts
6,367 battles

An excellent explanation m8! Even I learned smth new! Thank you!

Share this post


Link to post
Share on other sites
[BUSHI]
Players
3,461 posts
17,922 battles

Russian site http://proships.ru/stat/ships/ made by Z1000 which have very close connection with WG, has very interesting data of different stats of all ships. Here ex. comparision of accuracy of Warspite. % accuracy (as written in info) is based on horizontal, vertical dispertion (angle of shell falling) and sigma, calculated when firing to target size of stationary broadside DD.

 

As I understand, vertical dispertion is derived from angle of shell falling, which is straight connected to shell travel time, which is connected to air drag, shell weight and initial shell speed. So the shorter time has shell to travel to given point, the vertical dispertion is worse, becouse slight deviation of angle, makes shells landing far away vertically from targeting point.

 

So ex. slow Warspite shells and her good 2.0 sigma makes her very accurate BB, even slightly more then Iowa.

 

http://proships.ru/stat/ships/303,248/

DW.thumb.png.4f2585f3a60438a0da6a073534b825b8.png

 

http://proships.ru/stat/ships/160,248/

wa.thumb.png.f093190eb3675436532e318a5dcabcb4.png

 

http://proships.ru/stat/ships/248,157/

iowa.thumb.png.0a1b662875d00c8d66bc89efbacc5365.png

 

Here we see how Nagato is less accurate then Gneisenau till 13 km, and slightly more past that ?!. If they calculated it right of course. http://proships.ru/stat/ships/599,336/

 

nagato.thumb.png.f53177dc6f02334d3aa74a062cbd39d6.png

  • Cool 2

Share this post


Link to post
Share on other sites
[TOXIC]
Players
4,508 posts
11,466 battles
On 1/30/2018 at 9:06 PM, El2aZeR said:

It's funny how many people still believe sigma = accuracy when it has been explained half a year ago that sigma is essentially better RNG.

Source:

https://thearmoredpatrol.com/2017/06/15/wows-qa-15th-june-2017/

If shells tend to land closer to the middle of the dispersion ellipse (better sigma), it means that the guns are more accurate. Therefore the people who believe that sigma = accuracy are right (obviously it's not the only factor but I don't think anybody claims/believes that).

Share this post


Link to post
Share on other sites
[BLITZ]
Players
2,641 posts
9,896 battles

Great post.

 

Now if there only wasn't that grey font colour.

- A dark theme user

Share this post


Link to post
Share on other sites
[PARAZ]
Beta Tester
10,066 posts
15,733 battles
25 minutes ago, eliastion said:

If shells tend to land closer to the middle of the dispersion ellipse (better sigma), it means that the guns are more accurate.

 

First of all

ThreadNecro.jpg

 

Second, no, the guns are not more accurate. The guns are more reliable, but overall accuracy (aka dispersion) remains the same.

Although really, that's semanthics and up to personal viewpoint.

Share this post


Link to post
Share on other sites
[BUSHI]
[BUSHI]
Players
132 posts
7,577 battles
59 minutes ago, El2aZeR said:

Second, no, the guns are not more accurate. The guns are more reliable, but overall accuracy (aka dispersion) remains the same.

Although really, that's semanthics and up to personal viewpoint.

 

I mean when people talk about accuracy they mean average accuracy aka reliability aka shells go where you aimed.

Example: Two ships have the same dispersion, one with a sigma of 1,4 , but the other is a premium russian BB with a sigma of 1000.

The first struggles to hit anything, but the Emperor Putin I never misses.

Can you really, "REALLY" say that both ships have the same accuracy?

  • Cool 1

Share this post


Link to post
Share on other sites
[TOXIC]
Players
4,508 posts
11,466 battles
1 hour ago, El2aZeR said:

 

First of all

ThreadNecro.jpg

 

Second, no, the guns are not more accurate. The guns are more reliable, but overall accuracy (aka dispersion) remains the same.

Although really, that's semanthics and up to personal viewpoint.

I wonder how I ended up here. Must've been some stealthy link that directed me here and then I missed the fact that I ended up six feet under :Smile_teethhappy:

Also, the maximum dispersion is just a mechanical part of the system to model the accuracy. It's like claiming that better fire chance doesn't make guns better at setting fires since you either set a fire or not anyway, you just get better RNG on the roll :Smile-_tongue:

The gun is more accurate if it gets better (tighter) grouping of shots. In WoWs grouping of shots is determined based on dispersion elipse and sigma. Hence, better sigma (assuming the same dispersion elipse) gives you better gun accuracy. It's not a matter of opinion or personal viewpoint.

Although you certainly are right that it's a matter of semantics, I'll give you that much :Smile_trollface:

semantics
1 : the study of meanings

(after Merriam-Webster dictionary)

  • Funny 1

Share this post


Link to post
Share on other sites
Players
74 posts
5,136 battles

Now that i see this thread, the explanation of Sigma is not entirely right.

 

When talking of about ship accuracy. We say that the ship has "1.0 σ" and not "σ=1.0". It's nitpicky but prevents misunderstanding when looking at the graphs.

 

For every shell fired a random number with the result corresponding to the normal distribution function(Gauß distribution with σ=1 and μ=0 propably) is being generated. Lets say that the ship has 1.0 σ. This value "cuts off" the normal distribution function as seen above at -1.0σ and +1.0σ. Then these cut-off points correspond to the maximum dispersion. In case of Bismarck at max range, - 273m and +273m. Keep in mind that selecting an enemy decreases the dispersion by 1/2 which means the dispersion, while the target is selected, goes from -136.5m to +136.5m which equals the 273m horizontal dispersion.

 

Now here comes the kicker: If you look at the graphics with the σ=1.0 then you will notice the percentages. Those are the propabilities that the random number will fall into this region. This means that 34.1%+34.1%=68.2% of the shells will correspond to the graph shown above. But what happens if the random number being generated is outside of this region? What happens with the other 100%-68.2%= 31.8% of the shells? Those shells will be equally distributed instead of normally distributed accross the aiming circle. Which means that 31.8% of the shells will fly absolutely random. A higher sigma means that more shells will be normally distibuted.

 

MfG Boom

Share this post


Link to post
Share on other sites
Players
440 posts
8,131 battles

Great work especially since you cover not only sigma, but also hori/vert dispersion, which is crucial. Krupp value explanation is also nice. Both should be semi-explicit stats in the game IMO (at least on wiki or whatever. If someone has an overview i can add them).

 

On Sigma however; sigma affects the tendency of a shell to fly closer to center. True. For BBs, the fun thing is you often have three-gun turrets, so that three sigma calculations are done at once. Chance of these all being very bad becomes lower as sigma increases.

 

Now, if you fire a salvo, you get the same calculations as when you do separate shots. But, now all shells fly as one group. And statistically, a combination of high sigma and high number of barrels will have a good few shells in the inner bit (say 15%) of the dispersion field.

 

This phenomena is called shell grouping, and the difference can be observed really well in the Yamato- in fact, its what makes it so potentially devastating even at ~23km.

 

How do you "proof" shell grouping? 

Shell grouping can be demonstrated by comparing single firing vs salvo firing using any ship, but the effect is most evident in ships with high base dispersion.

 

Take a Yamato, go in a training room firing at yamato broadsides and you'll find youll score far more than 3x as much citadels using salvo fire on a decent range, since youd only be able to score lucky cits using regular sequential firing

Share this post


Link to post
Share on other sites
[TEACH]
Alpha Tester
2,455 posts
9,547 battles

image.png.ee9b80ef1d3f506455fe5ac93a8f0e91.png

 

When the hell did this game move beyond just pointing and clicking, life was so simple then.

Share this post


Link to post
Share on other sites
Players
344 posts
3,095 battles
6 hours ago, Shaka_D said:

image.png.ee9b80ef1d3f506455fe5ac93a8f0e91.png

 

When the hell did this game move beyond just pointing and clicking, life was so simple then.

Well, something along these lines eventually happens to every game, as the hardcore player base improves their understanding of under-the-hood mechanics and how to abuse them. I would presume that WOT went through this same process at one point in the distant past.

  • Cool 1

Share this post


Link to post
Share on other sites
Players
440 posts
8,131 battles
8 hours ago, who_dares_wins said:

Well, something along these lines eventually happens to every game, as the hardcore player base improves their understanding of under-the-hood mechanics and how to abuse them. I would presume that WOT went through this same process at one point in the distant past.

WoT actually is moving the opposite way

Share this post


Link to post
Share on other sites

  • Recently Browsing   0 members

    No registered users viewing this page.

×