Mythbusters ArmA edition.

walker · February 28, 2002

Hi all

There are lots of myths in ArmA started often as not by people who never tested what they write on the forums in the editor.

I got so sick of refuting them that I decided to collect them in one thread.

I will try and keep myths that are confirmed or busted in this first post of the thread.

Thread Contributors

Please Post any videos confirming or denying a myth along with the example mission so others may test you conclusions.

Rules

A Myth can only be confirmed with a test mission.

A Myth may remain plausible if it is only a video.

A proper explanation or repeatable experiment can bust a myth.

The Myths

ArmA has no Anti TK functions

This one was going the rounds due to some of the new server admins failing to turn on the functions or run scripts to prevent TK and supplemented by a myth that the anti TK enabling script had been removed in a patch.

http://community.bistudio.com/wiki/serverCommand
It was in the patch notes that it was disabled in 1.59

and a comment on the wiki page confirms it. I used to have an anti-teamswap script in warfare and since serverCommand was disabled, it stopped working.

Meanwhile an examination of the actual Patch notes provides no support for this. In fact its only mention of this command in the patch notes is one stating that the command has been fixed to prevent use by none admins. http://community.bistudio.com/wiki/A2OA:_Patch_v1.59

Fixed: Possible ServerCommand abuse by non-admin attackers.

I am guessing here, but an explanation for some peoples scripts not working; is because they are not being run on the server as admin.

The Myth was added to by a post by Piotr in the Wiki

http://community.bistudio.com/wiki/serverCommand

Piotr serverCommand (probably) has been disabled in 1.59

Hence the need to follow the Policy for all Editors of Scripting Command Pages to

Only post proven facts here.

As the Myth came up, several server admins tested it and pointed out it was working in this thread:

http://forums.bistudio.com/showthread.php?t=121294

MYTHBUSTED!

MYTH: Camp fires do not show up in FLIR

But camp fires do not show up :p, quite annoying that bug is. If you're going to employ a FLIR system do it properly right?

This is how Myths get started! :mad:

It is a simple one for any one with ArmA to refute in the Editor On Takistan.

Find a runway or other flat place and insert:

One: BLUEFOR, USA, Men, Sniper (TWS)

An: EMPTY, Objects, Fireplace (Burning) about 50m from soldier

Set time to ~02:00 am

Preview look at fire bring up sights

The fire clearly has an animated FLIR lod.

MYTHBUSTED!

MYTH: The AI sees through bushes!

Edit Added this Test mission

Test mission with which to test the myth that AI can see through bushes.

http://www.downloads.thechainofcommand.net/zips/X_Ray_AI_Mythbusted.zip

MTHBUSTED!

MYTH. The AI Cheats and can see me when it shouldn't and when I would not be able to see it

First up: No AI in any game sees anything.

They never have and it will be some time, involving masses of work by serious computer scientists and professors, until they do. Along the way several Nobel Prizes will be given to the computer scientists that achieve each bit of it.

NB You are not concealed by the fact you are behind something!

Following on from this, the excuse of: I ran behind the tree/bush/house(insert object) and the AI still new where I was, so it must be cheating; is equally lame.

Shooting, making a noise, such as rustling in the bushes, or stomping your boots around, moving about and sky lining your head to take a peak when in the AI's or players arc of view, or trying to be stealthy with the telegraph poll of an AT weapon on your back; all bust concealment.

Concealment does not magically reappear because you want it to.

The peekaboo behind the blankie effects!

NB You are not concealed by the fact your eyes are behind something!

This is the lesson you learn after many games with your parent or guardian of peekaboo behind the blankie! You remember this game The adult holds the blankie up in front of your eyes so you cannot see them. They watch your little legs kick in joy and listen to your little giggles of anticipation; because this the great all time bestseller, Game of the Eon (GOTEee) classic, "Peekaboo behind the blankie" game; and suddenly they pull the blankie away from in front of your eyes and say Peekaboo and you laugh and they laugh and much merriment is had by all!

One Day after much merriment and mirth playing "Peekaboo behind the blankie" you realise that the adult is in fact behind the blankie and that they can see you all the time; at this point the merriment becomes less and the joke appears lame, much to the distress of all concerned, because the little baby is growing up.

The grass may be covering your eyes but it ain't concealing your ass or your legs; because your fat ass lying down in it flattened it! Ever see those gaps in the corn where two people have been eh? nods as good as wink to the blind soldier with a grass blankie in front of his eyes!

Also we probably work out the adult is behind said blankie because they giggle too, and we saw them put the blankie up, which when you think about it later is a dead give away.

The AI is affected by Peekaboo Behind the Blankie same as you!

The following test using a team switch to set up an AI behind an obscuring grass object proves they too are affected by it!

What AI does to those still playing Peekaboo Behind the Blankie is the same as what happens to a noob in PvP

Following on from that, this is what the AI does when it realises you are still playing and giggling at the Peekaboo Behind the Blankie game!

As you can see the AI is reacting almost exactly the way a human MP player does to any noob who does not realise they have been seen. I must embarrassingly admit that this still happens to me sometimes in PvP. :(

MTHBUSTED!

MYTH: I should not have got shot because I was behind a bush, tree, wall, insert object

The difference between: Cover and Concealment.

Being in concealment does not mean you are in cover! and of course vis a versa. Just to remind everyone of the difference between: Cover and Concealment.

Concealment does not stop bullets, and things that amateurs think are cover, are often just concealment. And concealment once busted does not magically re-appear because you want it too. ;)

Cover is something that stops bullets. Cover often also can mean concealment but not always, you may be covered inside M1A1 but you re not concealed, everyone knows precisely where you are as it makes a lot of noise has an IR signature bigger than a bill board, produces masses of smoke even when it is not firing, and stirs up clouds of dust that can be seen for miles.

MTHBUSTED!

MYTH: I was behind a tree so I was in cover, the tree is foot thick that will stop a bullet!

A foot wide tree is not cover or concealment. Trees that develop arms and shoulders draw attention to them selves. While a well camouflaged rucksack sticking out of the side might be confused for an outgrowth, if still; moving growths are unusual. Trees with AT weapons sticking jauntily out of them will almost certainly draw strange looks followed by a bullet.

Any tree less than 2 feet wide offers little cover and no concealment for the average human. Average human shoulders are between 16 inches and 26 inches wide, very fit people or those with a large frame, or playing sports or engaged in activities that require lots of upper body strength; can and do exceed this considerably. Soldiers may well fall in to the latter category.

Consider the circle to be a tree trunk in cross section: If a foot of wood will stop some bullets there is 4 inches of the width of 1 foot wide tree in cross section that is about 1 foot thick. At the edges it is zero inches thick. So out of its twelve inches 8 inches offer zero to very little cover. Since we already know humans can be more than two feet wide we must accept the fact that a foot wide tree is not and never will be cover.

MTHBUSTED!

MYTH: I was behind the corner of a wall so I was in cover!

Corners of walls are also not true cover. At its apex a corner offers zero cover. After this it offers a maximum of the square root of the 2n^2 inches of cover where n is the distance in inches you body is back from apex. Ah the joys of Pythagoras.

I point these things out because there are some new guys here who have playing other games where they negatively train you into thinking such objects offer cover. This is incorrect and can result in short games for the aforesaid miss trained individuals.

The situation with this negative training as a result of playing silly games like COD4 CS etc. is that I have even seen people "Taking cover" I kid you not, behind a 4 inch wide post and, I kid you not, peeking out from behind it.

All New guys please reassess the situation when we meet in MP I do not want to spend hours of game play dragging or carrying you bloody ass from one medic to the next.

MTHBUSTED!

MYTH: We were on the top of the hill. I crawled over the crest of the hill the enemy should not have been able to beat us

NB DONT CRAWL OVER THE CREST AND THINK YOU ARE IN COVER!

Or a hill side is not a tacically superior position to being at the bottom of the slope

There was a time when I used to make this mistake. It comes from seeing all those films where they take the hill to be in a tactically superior position. A case of Hollywood negative training as it were.

In the old days of land wars, of cavalry charges and short range missile weapons like spears and bows, the gravity well afforded by a hill gave you a big advantage, With high powered rifles and machine guns, rockets and artillery that has disappeared.

When you crawl over the crest of a hill you are just as exposed as the person at the bottom of the hill. Depending on steepness perhaps you are in a worse position! If it is very steep and your enemy is far enough away he can stand up and take shots at you when you are laid down and he is the smaller target! Think about it. Worse if you stand up you do not get the same target size reduction he does. And it is impossible to duck while being on your belly.

While crawling to the crest is correct for observation (preferably near some bushes or trees so you do not skyline), going on to the forward slope is inadvisable in almost all cases.

The correct way to use the crest of a hill is to use it as cover. Eg standing up so that only your head shoulders and gun are presented to the enemy as a target then ducking down when under fire. When you receive fire change firing position. Better still change position after every (shot/burst)

For MGs this is all different, as they have to lay down to fire accurately the key protection to using an MG is primarily distance. Ideally the MG should be situated just on the crest or even reverse slope and used as an indirect fire weapon using plunging fire, with a spotter on the forward slope, calling in your round impacts and tracer same as a sniper consequently all MG assistants need binoculars! Where an MG must fire from the forward slope, then hard cover such as MG nests, sandbags, walls etc. is advisable

MYTHBUSTED!

MYTH: The AI saw me when I was down in the grass, it should not have been able to.

THE AI DOES NOT SEE ANYTHING IN ANY GAME!

For the AI Grass clutter, smoke, fog rain and perhaps dust is a statistical factor this is true in all games not just BIS's.

This is what Marek the CEO of BIS has said about the grass layer concealment:

As a matter of fact and unlike popular belief, the AI is affected by grass in Arma 2. There is approximation for clutter surrounding every unit affecting how well the AI can see it.
As always, there is a room for improvements in fidelity, so feel free to complain (but probably with better words, can you?)

That "lame" layer as present in Arma 1 was not very compatible with new rendering system and we are trying to come up with a better and more accurate solution but that of course is only visual effect important for human players, AI is unfair and doesn't use image recognition from the real rendering of game world :p

All discussed in more detail in this thread

http://forums.bistudio.com/showthread.php?t=81935

Factors involved in AI awareness:

1) Afore-knowledge, if your knowsAbout value has increased due to the AI seeing you already

2) Activity:

a) Shooting and making a noise

b) Movement including speed and size of movement, a head turn is more stealthy than running down the street.

3) Distance

4) Camouflage values (in your units config)

a) Are you person or a vehicle?

b) Are you classed as a sniper and thus wearing a ghillie suit?

c) Straight lines are a big give away in spotting, so an AT tube on your back reduces your camouflage value.

5) Acuity:

a) The crewmen of a vehicle (they seem to be treated as group X) looking through the vision slots of an armored vehicle in a urban setting are not as good at spotting things as an officer or sniper on a hill looking out over a plane.

b) Optics, binoculars and sights improve spotting range.

It is not known if Optics reduce AI peripheral vission (the test for this would be how close to an AI sniper on a watch direction you can sneak compared to a soldier with iron sights, that said the woodcraft of a sniper, exhibited as general skill, would be higher than that of a normal soldier so a better test is the same soldier, first equipped with iron sights and then with a sniper scoped rifle.)

6) Environmental factors:

a) Weather

b) Smoke

7) Direction of facing, beware of heads turning, peripheral vision is simulated in ArmA

8) Communication, units in a group X know about entities as a group and may pass the information to other groups, particularly the famous ArmA Guard waypoint does this where the threat is tank y and unit x or group X has no AT capability it will call on unit z or Group Z to fulfill this.

9) Distraction; if AI x is more concerned with a higher threat y2, or is being suppressed and concerned with getting out of fire.

10) Fear the AI can in some circumstances be "Fleeing" then x is not concerned with y.

11) Awareness; an AI tank buttoned up but in safe mode can be sneaked up on from behind and a satchel charge placed under it. Try doing this with an aware tank and bye bye.

12) Intervening objects and terrain, of course how much of you is visible affects whether you are seen, but it is the last thing in the algorithm! (Ask any decent programmer why!)

And the test that proves the AI are affected by the distant grass layer and assault and or shoot at last known position is here:

Here is a teleport mission with which to test the AI knowsAbout values and how they affect the enemy AI searching for you.

http://www.downloads.thechainofcommand.net/zips/AI_See_You_in_Grass_Mythbuster.Chernarus.zip

The mission is based on one by NeMeSiS but slightly altered by me to make it easier to see what is going on and to remove extraneous data in the form of an civilian used as a teleport mark. Thank you to NeMeSiS for his hard work trying to prove the myth. You can read our discussion on the subject in the later pages of this thread.

MYTHBUSTED!

MYTH: It takes more direct hits from a T55 than from a T34 to destroy an M1A1.

One thing I have noticed is that I can take more direct hits from a T55 on a M1A1 than from a T34.
Is that accurate?

Just tested this in the editor on Takistan at the Airport at 1500 m.

T55 took ~15 to 14 rounds of HEAT,

Total rounds for a default armed T55 on average to blow up US Army M1A1 is 15 to 14 rounds.

T34 took all 11 of AP rounds plus a further ~15 rounds of its HE to blow up the same US Army M1A1.

Total rounds for a default armed T34 on average to blow up US Army M1A1 is ~26 rounds.

A true test would be to remove all HE on the T34 and replace with AP (armor piercing) rounds.

If you can come up with an example mission showing proof of your myth then fine untill then.

MYTHBUSTED!

Kind Regards walker

Edited June 30, 2011 by walker
Added Myth that Anti TK script enabling command was turned off

BeerHunter · June 4, 2007

Good ones.

Not to mention that high powered rounds will easily penetrate a 12" tree.

And who in their right mind would think that being silhouetted on the top of a crest doesn't make you an ideal target?

Me thinks that the vast majority complaining about how easy they die and blaming the AI garner their knowledge and tactics from Hollywood where cardboard boxes are used for cover , bullets won't penetrate a wooden picket fence and you really should be able to dive through the air blasting away with with 2 Desert Eagles and score 100% accuracy.:p

Edited July 24, 2010 by BeerHunter

Dead3yez · July 31, 2008

http://pictures.second-amendment.org/albums/userpics/10001/223-ballistics-2.jpg

I'd like to see those penetration tests done at 300 meters (328 yards). And, also in ArmA. I just had a quick play around with different weapons... And it's nice. :)

Edited July 24, 2010 by Dead3yez

NoRailgunner · January 23, 2007

But keep in mind sometimes strange things can happen. ;)

Guess if performance goes down and below 15fps AIs are screwing up fast.

Takko · April 10, 2009

I have never heard of most of the "myths."

nettrucker · December 27, 2004

Great thread Walker.

thanks for putting at rest all the doubts players might have, thinking the AI is cheating when they got their butt handed over and die often.

kind regards

Steakslim · February 23, 2009

I love you long time Walker, now I have a link for when I'm too lazy to explain one of these myths to someone.

2024-11-26T18:24:30Z

Awesome thread walker, even if it does show your mildly excessive obsession with Arma. :p

(Not a bad thing!) :D

Just a suggestion, maybe put each myth under a spoiler, with a title before the spoiler so people know what it will be.

Other wise the main post may end up being 3 miles long! :p

Thanks!

walker · February 28, 2002

Awesome thread walker, even if it does show your mildly excessive obsession with Arma. :p
(Not a bad thing!) :D

Just a suggestion, maybe put each myth under a spoiler, with a title before the spoiler so people know what it will be.

Other wise the main post may end up being 3 miles long! :p

Thanks!

Hi richiespeed13

Good Suggestion Done!

Kind Regards walker

Grizzle · July 29, 2007

Nice thread. Thanks walker!

cossack8559 · July 30, 2009

Interesting thread :)

cole · February 2, 2007

good idea for a thread, walker! Hopefully some of the smartasses will shut up now.

Heatseeker · April 10, 2003

MYTH. The AI Cheats and can see me when it shouldn't and when I would not be able to see it

Well that really depends....

If you see a civilian car 1 km away and you dont have binoculars or a scoped weapon can you identify if the driver is friend or foe? :p .

CarlGustaffa · March 13, 2007

Hehehehehe... I more than giggled :D Funny writeup :)

Big Dawg KS · December 14, 2004

MYTH: Walker does not have too much free time

:j:

MYTHBUSTED!

Evidence: he made this thread.

:rolleyes:

mosh · February 17, 2002

Good thread walker, very interesting. Thanks!

madrussian · June 19, 2006

Interesting approach Walker... appears this took you some time and effort to put together. :)

Although I haven't examined all of your alleged "myth busts", I take particular issue with this one regarding AI not being able to see through bushes.

MYTH: The AI sees through bushes!

<object width="480" height="385"><param name="movie" value="http://www.youtube.com/watch?v=n-oYzp2iyqQ&hl=en_GB&fs=1"></param><param name="allowFullScreen" value="true"></param><param name="allowscriptaccess" value="always"></param><embed src="http://www.youtube.com/watch?v=n-oYzp2iyqQ&hl=en_GB&fs=1" type="application/x-shockwave-flash" allowscriptaccess="always" allowfullscreen="true" width="480" height="385"></embed></object>

MTHBUSTED!

As I clearly detail on this thread regarding the AI's Biotic X-Ray Vision through certain objects, the AI can indeed detect you through certain game objects, in this case certain bushes. The tests revealing this evidence are 100% repeatable, and 100% without ambiguity. If there are any disbelievers, please check out the demo mission I posted within that thread. Here's the demo mission again. The evidence inside speaks for itself.

In contrast, posting even a thousand videos of cases where AI-view-blocking is working correctly, does not in any way prove that AI-view-blocking works all the time and in all cases!

The undisputable truth: AI-view-blocking has repeatable issues in specific cases and needs some attention!

Cheers! :cool:

walker · February 28, 2002

Hi MadRussian

Mission includes none vanilla ArmA content.

Map.eu Mod to be precise.

Please redo the example mission without mod files.

Myth remains busted.

Kind Regards walker

ginger mcale · May 29, 2009

Excellent work. Thank you for that. Now if someone says one of this myths we can link here. Awesome! :yay:

ryguy · February 8, 2010

Cool! You should add to the original post if you discover more myths!

mosh · February 17, 2002

This might be hard to prove, walker, but what about the myth that the AI was better (particularly with driving) in Operation Flashpoint? I see this one at least twice a day on these forums. I also know it's false.

BeerHunter · June 4, 2007

Hi MadRussian
Mission includes none vanilla ArmA content.

Map.eu Mod to be precise.

Please redo the example mission without mod files.

Myth remains busted.

Kind Regards walker

Tend to agree Walker. I've found that if you don't run up behind the AI clanking and clunking , wait until they are looking AWAY from you and stay CONCEALED behind a bush , they don't know you're there until you swing your weapon around (instead of using free look) , lie prone with you're feet sticking out a mile or start shooting.

Then they tend to pepper the area until finally you are..demised. :p

walker · February 28, 2002

Hi Mosh

Apples and oranges.

Even asuming you used an ArmA II MOD of one the island such as Everon. Terrain Density in ArmA II is way higher than in OFP.

What vehicles are the same and have the same config as in BIS's OFP? UAZ would be your best bet.

And what useful information does it tell us? That OFP is simpler than ArmA II? Duh!

That the AI in ArmA II works in a more complex environment? Double Duh!

If you want to install OFP and ArmA II on the same machine, then do the example identical mission for both OFP and ArmA II; so it can be tested please do.

Kind Regards Walker

Edited July 24, 2010 by walker

Big Dawg KS · December 14, 2004

This might be hard to prove, walker, but what about the myth that the AI was better (particularly with driving) in Operation Flashpoint? I see this one at least twice a day on these forums. I also know it's false.

That's not a myth or a truth, that's just an opinion. If there's hard evidence that can be tested, then perhaps. But anyone can have personal preferences as to what they think is better.

r71 · July 22, 2001

Walker I have done my own test and the AI can see you through bushes very easy. I will try and get some screen shots to show you the test I ran.

This has been the same for the ai all the way back from 2001

Mythbusters ArmA edition.

Recommended Posts

walker 0

Share this post

Link to post

Share on other sites

BeerHunter 0

Share this post

Link to post

Share on other sites

Dead3yez 0

Share this post

Link to post

Share on other sites

NoRailgunner 0

Share this post

Link to post

Share on other sites

Takko 10

Share this post

Link to post

Share on other sites

nettrucker 142

Share this post

Link to post

Share on other sites

Steakslim 1

Share this post

Link to post

Share on other sites

Guest

Share this post

Link to post

Share on other sites

walker 0

Share this post

Link to post

Share on other sites

Grizzle 0

Share this post

Link to post

Share on other sites

cossack8559 10

Share this post

Link to post

Share on other sites

cole 0

Share this post

Link to post

Share on other sites

Heatseeker 0

Share this post

Link to post

Share on other sites

CarlGustaffa 4

Share this post

Link to post

Share on other sites

Big Dawg KS 6

Share this post

Link to post

Share on other sites

mosh 0

Share this post

Link to post

Share on other sites

madrussian 347

Share this post

Link to post

Share on other sites

walker 0

Share this post

Link to post

Share on other sites

ginger mcale 11

Share this post

Link to post

Share on other sites

ryguy 10

Share this post