Login

russian armor

RNG has too much of an impact on games

  • This thread is locked
PAGES (4)down
16 Aug 2013, 03:58 AM
#21
avatar of Nullist

Posts: 2425

Permanently Banned
Can small arms fire such as Cons/Grens roll crits?
16 Aug 2013, 10:48 AM
#22
avatar of Naeras

Posts: 172



Missing two snipes in a row was insanely uncommon.

Nope. Snipers had a 50% miss chance against other snipers unless the other sniper was stationary.
16 Aug 2013, 11:34 AM
#23
avatar of alcoholic
Patrion 15

Posts: 93



The problem isn't necessarily that these RNG-rolled events happen - it's their frequency. In COH1, there were few moments more intense than a Sniper vs. Sniper fight, where they both missed, or where that one Vet 3 rifleman dodged two mortar shells. These created incredibly interesting situations; however, they weren't common. Snipers almost always hit, 1-man rifle squads usually died to the first mortar shell, and AT guns almost ALWAYS hit. However, in CoH2, it appears that Relic thought that the easiest way to increase the "drama" in the game, was to increase the amount of RNG-based systems.


I dont get that statement. I think it contradicts your argument.

If i think about it i would say, that the more often a random event occurs, the less it influences the outcome of a game.

Example: Imagine 2 players competing in a dice game. player 1 wins if the once rolled dice shows a 6. player 2 wins if any other number comes up.

In this scenario player 1 has a 1/6 chance to win although the odds are against him.
Now if you alter the game to make it a "best of 10" player 1 cant expect to win at all.

With coh and rng that means: if there are only a few moments over the course of a game it may well be that most of them favour 1 player and he gets the win although he might be the worse player.

if random moments occure more frequently it becomes highly unlikely that one player benefits more than the other.

Now your statement was that in coh2 random events occure more frequently than in the original game. if that is true, this should make the game it less luck based over all.

one question:
you mentioned the chance of rolling criticals on full health infantry.
do you or anyone know how high this chance would be and if its still more likely to instantly kill a already wounded soldier?
16 Aug 2013, 13:18 PM
#24
avatar of Cyridius

Posts: 627


dice game. player 1 wins if the once rolled dice shows a 6. player 2 wins if any other number comes up.

In this scenario player 1 has a 1/6 chance to win although the odds are against him.
Now if you alter the game to make it a "best of 10" player 1 cant expect to win at all.



Doesn't work like that. Each event is independent and they don't effect each other. So instead of having a 1/60466178 chance of getting it, you have 10 1/6 chances.
16 Aug 2013, 13:34 PM
#25
avatar of Budwise
Admin Red  Badge
Donator 11

Posts: 2075 | Subs: 2

I always hated SOME of the RNG factors and I expressed it often to Relic devs and it was pretty clear that they wont budge on it so...

16 Aug 2013, 13:48 PM
#26
avatar of cr4wler

Posts: 1164

i don't think people that complain about RNG know what "accuracy" means.

and, most ironic of all, OPs post is full of "inaccuracies".

also funny: with flame weapons having an accuracy of 1100%, they are the ones that are least prone to the RNG (even most the criticals for these weapons are just custom death animations).
16 Aug 2013, 13:50 PM
#27
avatar of Spanky
Senior Strategist Badge

Posts: 1820 | Subs: 2

Naeras, i fookin lol'd at that pic because sometimes you could just loose an entire game only cause you got frustrated after the following.
16 Aug 2013, 13:50 PM
#28
avatar of Z3r07
Donator 11

Posts: 1006

The RNG in the game is what attracted me (and still attracts) me to the series. If I wanted intensely calculated mathematics I'd just play a different rts


same here, and 90% of the time it's not a game changer imo and it does create WOW or WTF moments that again, imo, is good for creating exciting moments giving even more reasons to watch a game or tournaments.
17 Aug 2013, 10:14 AM
#29
avatar of Doomlord52

Posts: 960

jump backJump back to quoted post16 Aug 2013, 13:48 PMcr4wler
i don't think people that complain about RNG know what "accuracy" means.

and, most ironic of all, OPs post is full of "inaccuracies".

also funny: with flame weapons having an accuracy of 1100%, they are the ones that are least prone to the RNG (even most the criticals for these weapons are just custom death animations).


If you feel I'm wrong about something, it would be nice to explain what exactly I'm wrong about - that way we can at least have some sort of debate, or I can at least provide some counter evidence. Just saying I'm wrong isn't really helpful.

As for 'Accuracy'; unfortunately RNG can still dramatically screw with what seems like a good chance. Let's say an AT gun has a 75% accuracy; in theory, that means it'll miss 1/4 shots (in ideal chances). That would be fair; the first shot might miss, but the next few would hit, killing the target.

However, just increasing the numbers can make the same situation horrible: 10/40. It's still the same accuracy, but just with a bigger sample size. With that larger sample comes a lot worse possibilities; the AT gun might miss 5 times in a row; by which time the tank has flanked it, killed it, or gotten away. Now, let's imagine that the game worked on a fixed statistic; if it missed 5 shots in a row, it will, without a doubt, hit the next 15. Unfortunately, the AT gun will never get a chance to fire the remaining 15; so now I've fired 5 times, and hit 0 times; 0% accuracy.

This is the problem; on paper, the statistics look fine; flamers have a 15% crit chance; which means that a flamer HT vs. a conscript squad will only crit 15% of it, or one guy (and a bit). However, as everyone has seen, that rarely happens. Either the flamer will roll up, crit half the squad instantly, or it will roll up, do next to zero damage, and then die to AT/AT nades, etc.

If i think about it i would say, that the more often a random event occurs, the less it influences the outcome of a game.


Unfortunately, that would only be true if every single RNG roll had the same importance; and it doesn't. A lucky roll with a flamer can mean either half your squad dies instantly; meaning a large MP drain, as well as a strong push for map control. Multiple poor RNG rolls can mean that my ANTI-TANK gun misses a T34 5 times in a row, does zero damage, and your T34 kills my AT gun, and then proceeds to kill a bunch of other stuff.

If it were tiny stuff, like single-man crit kills, it would be fine. We get into a fight, one of my grens goes down instantly, and one of your conscripts goes down slightly after that - that's fine. It's when it becomes horribly lopsided (3 of my grens go down, for zero conscripts) that RNG overtakes skill.

17 Aug 2013, 11:16 AM
#30
avatar of Daaaave

Posts: 12

I wouldn't play COH or COH2 if it weren't for the random factor. It's what keeps the game interesting. It's silly to say that games are decided by random factors - this is extremely rare, and thanks to the law of large numbers the meta always evens out to have the better players winning.

If I wanted to play a spreadsheet I would play Starcraft or one of the other yawnful RTS games that I have no interest in.


+1
17 Aug 2013, 13:06 PM
#31
avatar of cr4wler

Posts: 1164



If you feel I'm wrong about something, it would be nice to explain what exactly I'm wrong about - that way we can at least have some sort of debate, or I can at least provide some counter evidence. Just saying I'm wrong isn't really helpful.

As for 'Accuracy'; unfortunately RNG can still dramatically screw with what seems like a good chance. Let's say an AT gun has a 75% accuracy; in theory, that means it'll miss 1/4 shots (in ideal chances). That would be fair; the first shot might miss, but the next few would hit, killing the target.

However, just increasing the numbers can make the same situation horrible: 10/40. It's still the same accuracy, but just with a bigger sample size. With that larger sample comes a lot worse possibilities; the AT gun might miss 5 times in a row; by which time the tank has flanked it, killed it, or gotten away. Now, let's imagine that the game worked on a fixed statistic; if it missed 5 shots in a row, it will, without a doubt, hit the next 15. Unfortunately, the AT gun will never get a chance to fire the remaining 15; so now I've fired 5 times, and hit 0 times; 0% accuracy.

This is the problem; on paper, the statistics look fine; flamers have a 15% crit chance; which means that a flamer HT vs. a conscript squad will only crit 15% of it, or one guy (and a bit). However, as everyone has seen, that rarely happens. Either the flamer will roll up, crit half the squad instantly, or it will roll up, do next to zero damage, and then die to AT/AT nades, etc.



Unfortunately, that would only be true if every single RNG roll had the same importance; and it doesn't. A lucky roll with a flamer can mean either half your squad dies instantly; meaning a large MP drain, as well as a strong push for map control. Multiple poor RNG rolls can mean that my ANTI-TANK gun misses a T34 5 times in a row, does zero damage, and your T34 kills my AT gun, and then proceeds to kill a bunch of other stuff.

If it were tiny stuff, like single-man crit kills, it would be fine. We get into a fight, one of my grens goes down instantly, and one of your conscripts goes down slightly after that - that's fine. It's when it becomes horribly lopsided (3 of my grens go down, for zero conscripts) that RNG overtakes skill.




you say in first post that missing two countersnipes was extremely uncommon... the chance was 25% for two consecutive snipes to miss against a moving sniper.

on the other hand you are claiming that 15% chance crits for flamers (btw, 15% is not true, grenadiers for example only have a 5% chance to get instagibbed and i'd bet money that it is never more than 5% for any entity) happen far too often.

as for the pak with 75% accuracy... even if 10 out of 40 shots are calculated misses, with paks/at guns probably at least 4 out of those 10 misses will hit anyway because of scatter.


as you can see your argumentation is full of inaccuracies, which is why you are drawing wrong conclusions.
17 Aug 2013, 15:45 PM
#32
avatar of Nullist

Posts: 2425

Permanently Banned
jump backJump back to quoted post17 Aug 2013, 13:06 PMcr4wler
on the other hand you are claiming that 15% chance crits for flamers (btw, 15% is not true, grenadiers for example only have a 5% chance to get instagibbed and i'd bet money that it is never more than 5% for any entity) happen far too often.


I thought flame crit chance was 15% per tick.

Explain how you came up with 5% on Grens, please.
17 Aug 2013, 15:53 PM
#33
avatar of cr4wler

Posts: 1164

weapons do not have crit chances.
entities have crit chances (which are applicable to certain weapon types).
the crit chance for instagibbing grens at 100% hp is 5% and is applicable only to flame weapons.



edit:
btw, how did you come up with 15%? explain that.
17 Aug 2013, 16:24 PM
#34
avatar of Nullist

Posts: 2425

Permanently Banned
Dev stated 15% somewhere.
17 Aug 2013, 17:41 PM
#35
avatar of cr4wler

Posts: 1164

jump backJump back to quoted post17 Aug 2013, 16:24 PMNullist
Dev stated 15% somewhere.


...which is exactly why people make threads like this in the first place:

they don't know how the game mechanics work, don't know what the real values are, and then "hear" or "read" in the ominous "somewhere" place about something that confirms whatever they were making up in their heads about how the game might work.

this is nothing against you personal, but with some of these numbers i've read on the forums, they just seem like someone pulled them out of their ass. and then, if someone like me comes along and claims that the numbers are bogus and digs through the game files for the real values those guys always are like "source?", when they never even had a source for their information in the first place (other than their ass), and would rather believe their own made up numbers than the ones that are actually in the game.

i hope you can see how this is frustrating.
17 Aug 2013, 17:53 PM
#36
avatar of Doomlord52

Posts: 960

jump backJump back to quoted post17 Aug 2013, 13:06 PMcr4wler



you say in first post that missing two countersnipes was extremely uncommon... the chance was 25% for two consecutive snipes to miss against a moving sniper.


I'm referring to CoH1, where it was extremely uncommon for you to miss multiple sniper shots in a row. Yes, it could happen, but it might be once a game - if that.

on the other hand you are claiming that 15% chance crits for flamers (btw, 15% is not true, grenadiers for example only have a 5% chance to get instagibbed and i'd bet money that it is never more than 5% for any entity) happen far too often.


As others have stated, it's believed that pqumsieh stated it as 15%. However, it was in a forum post, which makes it very hard to track down.

as for the pak with 75% accuracy... even if 10 out of 40 shots are calculated misses, with paks/at guns probably at least 4 out of those 10 misses will hit anyway because of scatter.

Unfortunately this is not the case. As in the Twitch thing, scatter only works when the model is big (it helps), and if the scatter is rolled so that it actually hits there. Unfortunately, because of this, scatter doesn't work against moving vehicles. It also doesn't (really) work if the target is small. That's why they made the box for the STUG smaller.


as you can see your argumentation is full of inaccuracies, which is why you are drawing wrong conclusions.

If you really feel this way, I could make a video showing how incredibly dumb the RNG system is.
17 Aug 2013, 17:55 PM
#37
avatar of Nullist

Posts: 2425

Permanently Banned
@cr4wler: yes, i can see how that is frustrating, but, as you stated, it has absolutely nothing to do with me.
17 Aug 2013, 17:59 PM
#38
avatar of ace4sure

Posts: 102

I can confirm cr4wlers statement.
Here is the crit table of Conscripts/Grenadiers in JSON (might be a little more readable than XML)

So for Cons and Grens it means:
  • Health <= 100% > 50% health: 5% crit chance for flame weapons
  • Health <= 50% > 0% health: 10% crit chance for flame and explosive weapons
  • Health = 0%:
    75% crit chance for small arms, ballistic (kill)
    25% crit chance for small arms, ballistic, explosive (casualty)
    5% crit chance for explsovies (kill death_intensity_100)
    90% crit chance for flame weapons (burn)
    10% crit chance for small arms and ballistic (out of control)
    50% crit chance for all weapon to explode flamerthrower


I am also wondering where you got these "15%" from.
The thing about flame weapons in general is, they are intended to never miss and cause splash damage, which makes them pretty reliable damage platforms.

Talking about the KV-8 in special, that thing simply does an insane amount of 40 health damage (for comparison: 13 dmaage for Egineer/Pioneer flamers, 2x10 damage for the Flammenwerfer HT), which combined with the genral characteristics of flame weapons and the resilient platforms makes this unit clearly OP in my eyes. I have always said this and gonna stick with it!

One speciality about the Molotov: needless to say that the dot damage of it also does flame damage and therefore have the flame crit chance with every damage tick they do. So don't stay within the dot area!
And for all people complaining the Moltov damage is so unreliable, don't count on a crit, feel lucky if you get one, at least that's what I do.

Another importat thing i want to add, is that iIrc criticals are calculated AFTER the damage has applied. Otherwise a mortar or T34 shell (80 damage) would not be able to instantly kill off a soldier.
For the KV-8 flmaer this means with the first hit target is already at 50% health (40 damage), at 0% health with the next hit. Cause this happens within 1 second it basicly looks like the KV-8 would do crits all the time at 100% health targets.

What I am curios about is the crit chance for explosive weapons: only 30% combined chance against 0% health targets?
Also i am still guessing about the purpose of that "_no_critical" value.

Regards
ace
17 Aug 2013, 18:00 PM
#39
avatar of cr4wler

Posts: 1164

jump backJump back to quoted post17 Aug 2013, 17:55 PMNullist
@cr4wler: yes, i can see how that is frustrating, but, as you stated, it has absolutely nothing to do with me.


yeah, i just wanted to make sure you understood that i was trying to insult you with what i had written....


I'm referring to CoH1, where it was extremely uncommon for you to miss multiple sniper shots in a row. Yes, it could happen, but it might be once a game - if that.


i was referring to coh1 as well. believe me, i had many a sniper battle, and i think my record was 6 missed countersnipes in a row.

rest of my argument still stands, your numbers are wrong AND you interpret them wrong.
17 Aug 2013, 18:12 PM
#40
avatar of cr4wler

Posts: 1164

I can confirm cr4wlers statement.
Here is the crit table of Conscripts/Grenadiers in JSON (might be a little more readable than XML)

So for Cons and Grens it means:
  • Health <= 100% > 50% health: 5% crit chance for flame weapons
  • Health <= 50% > 0% health: 10% crit chance for flame and explosive weapons
  • Health = 0%:
    75% crit chance for small arms, ballistic (kill)
    25% crit chance for small arms, ballistic, explosive (casualty)
    5% crit chance for explsovies (kill death_intensity_100)
    90% crit chance for flame weapons (burn)
    10% crit chance for small arms and ballistic (out of control)
    50% crit chance for all weapon to explode flamerthrower


I am also wondering where you got these "15%" from.
The thing about flame weapons in general is, they are intended to never miss and cause splash damage, which makes them pretty reliable damage platforms.

Talking about the KV-8 in special, that thing simply does an insane amount of 40 health damage (for comparison: 13 dmaage for Egineer/Pioneer flamers, 2x10 damage for the Flammenwerfer HT), which combined with the genral characteristics of flame weapons and the resilient platforms makes this unit clearly OP in my eyes. I have always said this and gonna stick with it!

One speciality about the Molotov: needless to say that the dot damage of it also does flame damage and therefore have the flame crit chance with every damage tick they do. So don't stay within the dot area!
And for all people complaining the Moltov damage is so unreliable, don't count on a crit, feel lucky if you get one, at least that's what I do.

Another importat thing i want to add, is that iIrc criticals are calculated AFTER the damage has applied. Otherwise a mortar or T34 shell (80 damage) would not be able to instantly kill off a soldier.
For the KV-8 flmaer this means with the first hit target is already at 50% health (40 damage), at 0% health with the next hit. Cause this happens within 1 second it basicly looks like the KV-8 would do crits all the time at 100% health targets.

What I am curios about is the crit chance for explosive weapons: only 30% combined chance against 0% health targets?
Also i am still guessing about the purpose of that "_no_critical" value.

Regards
ace


well, if the target is already dead, what more would a critical do, right? (for the explosive weapons)

as for the _no_critical, i think it was intended to be there in order to have the chance for no critical to occur... i don't know exactly how the game works, but if you looked at the game files anyway, look at all the requirement_types.... THAT stuff doesn't make any sense at all.... so, the _no_critical is a minor inconvenience... ;-)
PAGES (4)down
1 user is browsing this thread: 1 guest

Ladders Top 10

  • #
    Steam Alias
    W
    L
    %
    Streak
Data provided by Relic Relic Entertainment

Replay highlight

VS
  • U.S. Forces flag cblanco ★
  • The British Forces flag 보드카 중대
  • Oberkommando West flag VonManteuffel
  • Ostheer flag Heartless Jäger
uploaded by XXxxHeartlessxxXX

Board Info

727 users are online: 1 member and 726 guests
aerafield
2 posts in the last 24h
10 posts in the last week
29 posts in the last month
Registered members: 50036
Welcome our newest member, Bendiger
Most online: 2043 users on 29 Oct 2023, 01:04 AM