Jump to content
Forum Shutdown 28/7/2023 Read more... ×
Sign in to follow this  
killengage

RNG....what is the mechanic behind it ?

36 comments in this topic

Recommended Posts

Players
3 posts
4,972 battles

I'm just curious about this hole rng thing.

 

Examples:

 

- In my cleveland i shoot another cleveland. 51 hits no fire > i get hit with 3 shells fire. The game after i shoot another cleveland 4 hits 3 fires.....

 

Why the inconsistency ? Whats behind this formula. It feels like paying premium for a random game experience. 

  • Cool 1

Share this post


Link to post
Share on other sites
[SCRUB]
Beta Tester
4,811 posts
13,808 battles

You're making the incredibly common but also so so wrong assumption that "random" in any way means "consistent".

  • Cool 1

Share this post


Link to post
Share on other sites
Players
68 posts
6,483 battles

the mechanic is randomness ...

you could just roll dice instead 

e.g. Cleveland has 7% chance to set fire ( i honestly have no idea how high it actually is but it does not matter)

 

so for each shot there is a dice rolled with 100 sides from 1-7 a fire is set from 8-100 it is not 

so what is the chance that in 51 shots you dont set a fire? 

0,93^51=0,02469 

or as we would call it : 2,5% 

what is the chance to set fire in three shots? 

1-0,93^3=0,195643 

or as we call it 19,5%

Share this post


Link to post
Share on other sites
[SLOTH]
Players
3,041 posts
5,653 battles

It is basically like rolling a dice. Any further explanation would probably be too long and complicated to explain.

 

well if they cram it down our throats at school and we take it it'll be ok here
  • Cool 1

Share this post


Link to post
Share on other sites
Beta Tester
2,287 posts
11,047 battles

 

well if they cram it down our throats at school and we take it it'll be ok here

 

I think it is a bit different when you do it on college. We had a couple of hours of classes just about that topic.

 

Plus there are so many factors involved which we do not even know that I could not even explain it properly and on top of that I am too lazy to reopen my scrips for this :D

  • Cool 1

Share this post


Link to post
Share on other sites
[THESO]
[THESO]
Players
991 posts
12,433 battles

 Whats behind this formula. It feels like paying premium for a random game experience. 

 

RNG = Random Number Generator 

 

I don't know how it can be more descriptive than it already is :)

  • Cool 1

Share this post


Link to post
Share on other sites
Beta Tester
847 posts

 

62

Well, rng is actually pseudo random, a function over the microseconds of your computer clock, but I doubt you wanted to go into this detail.

Share this post


Link to post
Share on other sites
[SCRUB]
Players
8,460 posts
13,076 battles

You're making the incredibly common but also so so wrong assumption that "random" in any way means "consistent".

 

RNG is consistent, provided the pool of total results isn't infinite (which afaik is impossible to code with current technology). You are however correct in the way that the small handful of RNG calls for a single in any given game is by far too insignificant to approach the statistical consistency.

 

Just wanted to expand on that statement. Nothing else to see here, move along citizen.

Share this post


Link to post
Share on other sites
Players
2,928 posts
6,549 battles

To be honest, random is not consistent but it is frustrating. Most games that have some form of RNG deliberately limit the randomness so that it's not possible to get completely screwed one way or the other. Simplest way to do it is cap it e.g. guarantee a fire on the 20th hit if none have happened randomly before then. I actually consider this sort of fake randomness with a bit of forced consistency more fair and less frustrating in gameplay terms.

Share this post


Link to post
Share on other sites
[TTTX]
Players
1,841 posts
7,432 battles

 

RNG is consistent, provided the pool of total results isn't infinite (which afaik is impossible to code with current technology). You are however correct in the way that the small handful of RNG calls for a single in any given game is by far too insignificant to approach the statistical consistency.

 

Just wanted to expand on that statement. Nothing else to see here, move along citizen.

 

The pool of total results CAN'T be infinite, because the amount of possible numbers that can be stored on any medium is finite.

Share this post


Link to post
Share on other sites
Players
3 posts
4,972 battles

Ok guys,

 

Thanks for the added comments :)

 

I know the concepts about RNG (long time wot player), but whats behind it ? Longer rounds , more balanced teams , better game experience for everybody (from bad to unicum players). There must be a concept behind it when the developers
added this to the game. Will WOT and WOWS complete suck without it ? They have implemented a damage model into the game so why need this rng.

Share this post


Link to post
Share on other sites
Players
340 posts
134 battles

I know the concepts about RNG (long time wot player), but whats behind it ? Longer rounds , more balanced teams , better game experience for everybody (from bad to unicum players). There must be a concept behind it when the developers added this to the game.

 

If you're talking about random mechanics generally, then it's too large a subject. Each one has different reasoning and tradeoffs. For example, the random dispersion in WoWS allows ship & class balance to be tuned, and generates a spread of results rather than all-or-nothing salvos. The game wouldn't work without it. You can argue the magnitude and distribution, but not its existence.

 

Dice-rolling's just a default option for fire though. If you don't want every round starting a fire, and you don't want to put hundreds of fire-triggering hitboxes onto every ship, then the obvious solution is for each shell to have an independent percentage chance of starting a fire. The developers may not even be aware of the alternatives.

  • Cool 1

Share this post


Link to post
Share on other sites
[-SBG-]
Players
38,559 posts
19,178 battles

Ok guys,

 

Thanks for the added comments :)

 

I know the concepts about RNG (long time wot player), but whats behind it ? Longer rounds , more balanced teams , better game experience for everybody (from bad to unicum players). There must be a concept behind it when the developers

added this to the game. Will WOT and WOWS complete suck without it ? They have implemented a damage model into the game so why need this rng.

 

Because fires do not start every time.

Share this post


Link to post
Share on other sites
[SCRUB]
Players
8,460 posts
13,076 battles

 

The pool of total results CAN'T be infinite, because the amount of possible numbers that can be stored on any medium is finite.

 

That's why I had added the technological limitation bit.

 

Though if one were to discard the necessity to predetermine any result for a random roll, medium storage limits would cease to factor in as the only data you'd need to store would be the code for the random number generator.

Then again, a true random number generator isn't applicable for anything else than just generating random numbers (duh) without any context, so a randomly generated 1739252,83642 means nothing if there isn't a predetermined result that would be initiated upon drawing that number and obviously a true RNG (i.e. infinite number of potential results) can't be matched with such (for various reasons, mentioned finite medium storage being one).

 

The short of it:

RNG in WoWS is consistent, but a sufficiently large pool of results paired with an artificially restricted sample size results in enough statistical variation that said consistency can only be traced in all-time statistics and not for each single match.

 

Enough technobabble from me now.

Share this post


Link to post
Share on other sites
Beta Tester
2,119 posts
5,245 battles

Including (pseudo-)randomness is a question of game design and philosophy.

 

Some games have no randomness. Others are pure randomness. Players enjoy different things and different levels of randomness alter the game's skill requirement and cap (though not always in the same way). Generally speaking, randomness makes a game less forseeable and thus less strategic although a very high density of random events can make it forseeable again if the player is good with stochastics. Low randomness makes a game strategically demanding because the lack of randomness means the player can anticipate many things correctly. That's why most board games have a random element to make sure strategic geniuses don't completely overpower their opponents. In a game with low or no randomness, a superior player will win almost every time even if the skill difference is relatively moderate. Take chess: there's no randomness and you will almost always lose against even a slightly stronger player.

 

In a modern game aimed at many people, including no randomness is usually a very bad idea.

 

Generally there are too ways to reduce the impact of skill difference: Randomness and fog of war. A combination of these two is chosen in WoT and WoWs. The team has respectable vision across most of the map but large parts are still invisible to you. That's why you can get away with less randomness than a board game where everything is visible to all players. A game with more fog of war can get away with even less randomness because the players know much less about the state of the game.

Share this post


Link to post
Share on other sites
Weekend Tester
1,677 posts
20,290 battles

Ok guys,

 

Thanks for the added comments :)

 

I know the concepts about RNG (long time wot player), but whats behind it ? Longer rounds , more balanced teams , better game experience for everybody (from bad to unicum players). There must be a concept behind it when the developers

added this to the game. Will WOT and WOWS complete suck without it ? They have implemented a damage model into the game so why need this rng.

 

That is more of a subjective question (vs mechanics question) so it is to a certain extent a matter of opinion.

So my opinions are:

- As you can never be 100% sure of the outcome, it adds suspense.

- It allows occasionally "David vs Goliath" situations, i.e. that a weaker opponent may win against a stronger one with a "lucky shot"

- As RNG is integral part of real life, having it included makes the game feel more "lifelike"

- Now an opinion a lot of people will find controversial: Due to game specifics of WoWs (huge hitboxes, low relative speed of targets...) it adds to shooting skill requirement. To explain, if there was no RNG (i.e. dispersion) the ships would be much easier to hit, i.e. the room for aiming and still hitting the opponent would be something like: ----------x----------. With dispersion, which basically reduces the available margin for "error" in aiming, it looks like: -----x-----

 

It should be noted that there are huge differences in RNG implementation between WoT and WoWs. WoWs has much less RNG compared to WoT as there are no RNG rolls in penetration value of shells, damage dealt by shell...

 

So while you can never be 100% certain of outcome of an engagement (like in RL), RNG allows skilled players to "stack odds" in their favor and hence to perform significantly better on average than less skilled players. In case of shooting skills, this is why we see for example some players with say 40% accuracy and others with say 20% accuracy for the same ship. If there was no RNG, we would see accuracy of say 95% for skilled players and say 92% for less skilled ones as it is not that difficult to reliably hit a several hundred meters long target moving at a low speed without gun dispersion - hence skilled players would not be able to significantly outperform less skilled ones.

 

Lastly, I see a lot of people overestimating the impact of RNG mechanics in the game. While shooting ships is an obvious major component of the game, it is neither the only one nor that decisive (as in WoT for example). Positioning, target selection.... i.e. various tactical skills have an even bigger impact on the match outcome that shooting by itself. One could say that WoWs is basically a mix of chess and poker. Well placed fleet on the map (with classes of ships doing their role) almost assures the victory. At the same time, while single engagements on the battlefield cannot be reliably predetermined, skilled WoWs players will like professional poker players know much better how to stack odds in their favor, despite having an element of luck (or better say chance).

 

Edit: To answer your last question, yes I believe WoWs would suck big time without RNG :)

Edited by daki
  • Cool 4

Share this post


Link to post
Share on other sites
Players
9 posts
1,994 battles

Last match with New York killed 5, next killed 0, randomness is more prevalent in low rate of fire ships, that's all, accuracy of weapons is an integral part of engineering, and that is translated into a broad number in WoW called accuracy, there's really no debate to have since without RNG it would be quite the dumb game, and there's really not much to learn about it, shot is fired, random number is generated, shot may hit or may not, that's it.

Share this post


Link to post
Share on other sites
[SCRUB]
Players
8,460 posts
13,076 battles

 

- Now an opinion a lot of people will find controversial: Due to game specifics of WoWs (huge hitboxes, low relative speed of targets...) it adds to shooting skill requirement. To explain, if there was no RNG (i.e. dispersion) the ships would be much easier to hit, i.e. the room for aiming and still hitting the opponent would be something like: ----------x----------. With dispersion, which basically reduces the available margin for "error" in aiming, it looks like: -----x-----

 

 

Your personal preferences on having RNG aside, it doesn't add to shooting skill requirements. Either you can shoot, or you can't. What RNG adds is luck (and lack thereof): you can aim correctly, yet RNG causes most or all shots to miss, despite having ranged and lead the ship correctly. Or you can aim incorrectly yet still hit with a couple shells because the large deviation just happened to place some shells so far ahaed/behind your point of aim that even if you over-/undershot the target, you still hit.

 

There is no skill added in either case. Please don't mistake RNG with skill. RNG is the anti-thesis to skill!

Share this post


Link to post
Share on other sites
[SLEEP]
Players
127 posts
30,109 battles

What hurts with the RNG is when you get it good, compared to when you get it bad. For instance. get 50 hits no fire, next game 1st salvo gets fire GREAT!!! then a CV drops torp pattern on target and kill it, you get 250 fire damage. 

 

RNG is a necessary evil, but it does have to be programmed so follows a pattern which means it is not totally RNG, unlike rolling a dice. Flawless RNG is not and getting the balance right is something I have yet to see. As you said 50 hits and no fire then 4 hits and 3 fires are 2 complete ends of the ball park. With most RNG models it is possible to roll a 1 a hundred times in succession as well as rolling 6 100x too. 

 

I expect it is possible to have RNG between 1-100 with a limiter put at the extreme ends, so you cannot hit below 10 4x in a row or above 90 4x in a row. Not intending to say this is how it should be done, because fine tuning RNG has to be a real big nightmare.

Share this post


Link to post
Share on other sites
Beta Tester
1,023 posts

the mechanic is randomness ...

you could just roll dice instead 

e.g. Cleveland has 7% chance to set fire ( i honestly have no idea how high it actually is but it does not matter)

 

so for each shot there is a dice rolled with 100 sides from 1-7 a fire is set from 8-100 it is not 

so what is the chance that in 51 shots you dont set a fire? 

0,93^51=0,02469 

or as we would call it : 2,5% 

what is the chance to set fire in three shots? 

1-0,93^3=0,195643 

or as we call it 19,5%

 

Sorry, this is wrong. Each shell hit is statistically Independent. Meaning not setting a fire with a shell does not increase your chance to set a fire with subsequent ones. With a 7% chance to start a fire EVERY shell has a 93% chance to NOT set a fire. 

Your calculation is a common misconception of people who - No offense- do not know how propabilities really work mathematically.

Share this post


Link to post
Share on other sites
[SCRUB]
Beta Tester
4,811 posts
13,808 battles

 

Sorry, this is wrong. Each shell hit is statistically Independent. Meaning not setting a fire with a shell does not increase your chance to set a fire with subsequent ones. With a 7% chance to start a fire EVERY shell has a 93% chance to NOT set a fire. 

Your calculation is a common misconception of people who - No offense- do not know how propabilities really work mathematically.

 

Err.. that's exactly what those calculations show. He didn't say "the odds for the 51th shot not to set fire" but to "not get a single fire in 51 shots".

Share this post


Link to post
Share on other sites
Sign in to follow this  

×