Op fire weirdness and the "hot or cold" effect.

Merr · Post by **Merr** » Tue Sep 21, 2010 1:53 am

Just my understanding ... (so far)

Another factor regarding Opfiring is the target's chance to Kill your Opfiring unit ...
If distance > 3 and chancetokill your unit is < 10% if won't fire ??
But, if your chancetokill target >= 0-15% based on range, it will fire.

Lets look at the reactionChance alone ; starting with target in front arc ... Tank vs Tank
1. Chance = 100%
2. target < 2 .... chance now 200%
3. target sneaking .... chance now 100%
4. target sneaking is an elite tank hunter, reduce chance to react (if it has not fired yet this turn) ... chance now 50%
5. if reacting unit moved and is not sneaking reduce chance to react ... chance now 25%

If target range > 2 ... just halve the % in 3-5 above respectfully ... 50,25,12.

Basically - if your opponent sneaks around .... you have 50% or less to react.

ssouthrey · Post by **ssouthrey** » Tue Sep 21, 2010 2:12 am

I understand that units will run into intangibles that might cause them to op fire but that rate should only be about 10% to 5%. I don't mind missing. Heck, if I plan out a great ambush and my unit has a 95% to hit and I miss so be it. That's the way it goes. But to have continuously great ambushes and have my units not op fire at all, that begins to be a problem. Especially when I don't understand why that is happening. I'm playing a game right now:
2 M-10's, 1 Sherman 76 vs. 2 StuG's. My units were lined up about 7 squares away on an open field facing the StuG's. The StuG's are facing 90' away at an M-5 I have on a hill. I fire 6 times at a StuG and he returns fire 6 times and then is suppressed. On the computer turn, the remaining StuG turns toward my 3 tanks and fires twice. None of my 4 (including the M-5 on the hill) tanks op fires back. Why? I would love to understand why the computer StuG counter fires 6 times and my 4 tanks doesn't return fire once.

ssouthrey · Post by **ssouthrey** » Tue Sep 21, 2010 2:19 am

Merr wrote:This is telling me that if I have 1 Shot and my morale is >= 80 ... unit won't fire?

Wow. I need to start looking at the code. I didn't realize it was that open.

LOGAN5 · Post by **LOGAN5** » Tue Sep 21, 2010 2:29 am

Merr:
srand ( time(NULL) );
that should prevent the rand from picking the same sequence of numbers each time right?

Merr · Post by **Merr** » Tue Sep 21, 2010 2:34 am

ssouthrey wrote:
Wow. I need to start looking at the code. I didn't realize it was that open.

Wide open

ssouthrey ... Have a look and comment please... Everyone too! ... if you can stand it without going blind

Anyway, I'll stop babbling ....
IMO .... I think it's all rotten luck

if you can't hit anything despite good tactics. Perhaps too much luck? That's what we all want to know.

Luck is in all games we play ...
Look at a 1d6 .... If you had to roll a 6 to hit, we might assume that after rolling the die 6 times we'd get a 6 !!!
Every midshift at work, when on break, I watch others play Risk ... I

when one guy defends with 1 unit and consistently rolls a 6 whilst his opponent attacks with 3d6 and rolls all 1's !
Of course, there's a trick to rolling dice too!

Opfire in other games seem absolute .... If within range and can kill ... SHOOT! ... but is that a good thing?
I don't know whats worse, shooting too much (reveal hidden) or too little (are you blind!).
Alas ... to get our men to follow orders ... Heck, the gunner could have sneezed and missed his op.

BTW, consecutive shots appear to only grant an additional 10% chance to hit .... I think it should be 30% ... undershot, overshot, spot on!

EDIT : Of course, a 30% bonus would work if we had 3 shots ... maybe we need more shots?

Merr · Post by **Merr** » Tue Sep 21, 2010 3:23 am

LOGAN5 wrote:Merr:
srand ( time(NULL) );
that should prevent the rand from picking the same sequence of numbers each time right?

Logan5,

I don't know

... I'm a noob ... If it is correct, next question is where would the code go?
In the Ticks?
In the StartTurn?
In the StartTurnPost?

Good news is that if it isn't a random thing and it's something in the code, anyone can fiddle about with it in a homebuilt scenario to test outcomes.

LOGAN5 · Post by **LOGAN5** » Tue Sep 21, 2010 3:25 am

shawnt63 wrote:Interesting thread, I have chatted with LOGAN5 before about this issue and I just ran into it again with a game of dday +3. Don't know if it is just randomness or if there is really something amiss, but I was sitting in ambush good fields of fire and advanced warning, but the allies came in and destroyed all my armour in two turns and I got a big fat zero on the kills.

That was what maybe a week ago that we played "beta clash of armor" you couldn't have played more than maybe lets say 10 games since... so if this phenomenon is so rare then why does it keep happening? I guess we are getting all the bad luck out of the way and for the next 6 months we will all be hot.. I really think something is fishy, I'm still having fun with the game but when you get on a cold streak you might as well just surrender because things will NOT turn around for you that match, don't know how to explain it but that is just what happens.

LOGAN5 · Post by **LOGAN5** » Tue Sep 21, 2010 3:29 am

Merr wrote:
LOGAN5 wrote:Merr:
srand ( time(NULL) );
that should prevent the rand from picking the same sequence of numbers each time right?
Logan5,

I don't know ... I'm a noob ... If it is correct, next question is where would the code go?
In the Ticks?
In the StartTurn?
In the StartTurnPost?

Good news is that if it isn't a random thing and it's something in the code, anyone can fiddle about with it in a homebuilt scenario to test outcomes.

Merr: I dunno i'm not familiar with this language, i got that from some C++ website, but i know there is a way to randomize the timer and it should be at the beginning of the program.. Fiddling with 1 or 2% here and there wont account for your entire army getting shredded to pieces while your opponent loses nothing, I'm convinced we got bigger problems, either that or I'm just off on some crazy conspiracy theory.

ssouthrey · Post by **ssouthrey** » Tue Sep 21, 2010 6:37 am

I'm a programmer but I'd rather play when I get home than do more programming.

I'm going to dig into the code. Didn't realize it was open.

In regards to random numbers in C you can use rand() but if you want to seed the generator then you would use srand(time(0)).

Post by **IainMcNeil** » Tue Sep 21, 2010 6:48 am

The random numbers are seeded and need to be for multiplayer so both results happen exactly the same on both machines.

This tells me that if you roll a 100, you won't fire .... should read (0,99)

If it said => then you would be right but it says > so it works fine.

The Op fire has other factors feeding in. E.g. a unit moving round a corner and hunting has a chance to get the first shot off. Units with 1 shot remaining don't waste it - they only shoot if they have a chance to hit (like hidden units) but the criteria changes as the enemy gets closer. You get less concerned about how much you can hurt the closer they get and when very close always fire. If your morale is low you do not hold fire because the chances are you'll get suppressed before you get the chance to return fire. There are lots of factors feeding in to model the real world behaviour.

cptkremmen · Post by **cptkremmen** » Tue Sep 21, 2010 3:46 pm

I was skeptical about this, but have to say I am becoming concerned about this. I am playing a multi player game right now, had good ambush positions as the germans. British attacked and i think destroyed 10 german armoured vehicles in the one turn (tanks) whilst suffering no effective return fire whatsoever. On my turn my last few tanks then popped out at slow speed 1 square and each and every one of them were destroyed immediately by op fire without getting a shot in. I reckon the combined score for the turn was about 14-0 when I as the German was in a good position and did nothing wrong, statistically that's about a million to one. I do not think the slitherine random number generator is working properly.

cptkremmen · Post by **cptkremmen** » Tue Sep 21, 2010 3:56 pm

I have never noticed this phenomenon in single player which i was exclusively playing. In multiplayer it seems very prevalent. My money is on a compromise that has had to be used for multiplayer that is screwed up. if you roll "lucky" it seems to be remembering the lucky roll rather than rerolling each time. Possibly not something as simple as that, but something is very wrong....

If this does not go away I am going back to single player

Post by **IainMcNeil** » Tue Sep 21, 2010 4:00 pm

There is no way for that to happen. Everyone uses the same code - the code doesn't know if it is you or them shooting. If the value were stored it would effect both sides.

I really think this is just random chance.

The issue is in multiplayer you notice it more as your opponent is much tougher and being unlucky matters much more.

Acererak · Post by **Acererak** » Tue Sep 21, 2010 4:15 pm

Could it be that the seeding created for making both machines have same results could be causing some kind of pattern repeating?

Really, i am still very skeptical about this but i am no expert and some guys with math and programming skills seem to be concerned about it, so the issue deserves some attention in my honest opinion.

On the other side, i just noticed that most people make the strongest complaints when the cold effect applies to them, but don´t see so many complaining when is the hot effect the one you are enjoying. Just a little psychological note on my side.

And again i repeat myself. I don´t know if there is some issue or not, but i don´t feel cheated by the game at all. I think i won the games i deserved to win, and lost the ones i did as well. And can really feel that my tactics and gaming decisions have an effect in the overall result. Very few games i played before made me feel it this deep.

Post by **IainMcNeil** » Tue Sep 21, 2010 4:21 pm

Not really - I can do some recording of the random number generator but it takes time to do when we'd rather be creating new content/features. I am certain this is fine so reluctant to spend time on it.

LOGAN5 · Post by **LOGAN5** » Tue Sep 21, 2010 5:00 pm

cptkremmen wrote:I was skeptical about this, but have to say I am becoming concerned about this. I am playing a multi player game right now, had good ambush positions as the germans. British attacked and i think destroyed 10 german armoured vehicles in the one turn (tanks) whilst suffering no effective return fire whatsoever. On my turn my last few tanks then popped out at slow speed 1 square and each and every one of them were destroyed immediately by op fire without getting a shot in. I reckon the combined score for the turn was about 14-0 when I as the German was in a good position and did nothing wrong, statistically that's about a million to one. I do not think the slitherine random number generator is working properly.

Strange isn't it... The slitherine guy keeps insisting that it is all in our perception, maybe he is right maybe he is wrong...1 or 2 bad game fine, but its really hard to watch your army get slaughtered over and over again no return fire and say that its all just coincidence

Post by **IainMcNeil** » Tue Sep 21, 2010 5:43 pm

Are we talking about chances to kill or chances to fire? We seem to flip flop and there are very different factors effecting them.

If you see a situation where a unit does not fire please tell us where. There are lots of factors feeding in to it and it is almost certainly just a lack of understanding of why it did not fire. If this is the case we need to look at ways to make it more obvious.

The same code is used in singler and multiplayer so if it doesnt happen in single player it can't in multiplayer. The only difference is at the start of the turn a number seed is passed to ensure you stay in synch with your opponent.

We have seen this issue about luck with other games and after checking it has always been down to player perception.

ssouthrey · Post by **ssouthrey** » Tue Sep 21, 2010 6:42 pm

With me, I see the problem in both single and multiplayer. I also don't think the random number generator is the problem. I feel pretty confident that is working. I'm concerned with the op fire. I think it's either one of two things:
1. The logic behind return fire is intended to work as we see now, which I don't feel is accurate and thus must mod to my liking.
2. There is a bug in the logic somewhere that is causing the weird behavior.

On a side note, I've noticed in single player, the computer op fires much more than I do. Over a 2 to 1 discrepancy. I didn't care much about it because I was playing the more offensive missions but lately I've been playing the defensive missions and my stationary, dug in units just don't fire as often as the AI units even in the open.

Iain, as a fellow developer, I don't want you to feel like I'm picking on your work. You guys have done a phenomenal job and I'm just trying to help knock out any possible problems (I admit there may not be one). Thanks for your work and time.

LOGAN5 · Post by **LOGAN5** » Tue Sep 21, 2010 7:38 pm

There is nothing in the code that would favor one side over the other at anytime... I am curious about the if 1 shot left and moral <80 maybe i am miss understanding the purpose of this line.

it seems to check if it is hit by moral it might just have low moral from last round, if moral last round is greater than this round
But anyway, im not trying to pick apart and have you defend every line of code, after a big loss you can't help but think the game was somehow bugged.

I have a few games going right now and its been pretty even, i killed a panther 1 shot, and got another 5% panther kill and i wasn't complaining about that lol
but i bet my opponent will be digging through the source code tonight to figure out how the hell that is possible hehe

Acererak · Post by **Acererak** » Tue Sep 21, 2010 10:11 pm

LOGAN5 wrote: I have a few games going right now and its been pretty even, i killed a panther 1 shot, and got another 5% panther kill and i wasn't complaining about that lol
but i bet my opponent will be digging through the source code tonight to figure out how the hell that is possible hehe

That´s exactly my point