Jump to content
Sign in to follow this  
X3M

Looking for; anti infantry units

Recommended Posts

See title.

Anti infantry units that fail in doing their job.

Not yet telling why I want to know this.

 

But I am curious in which games, that you know of, a group of infantry units actually win against so called anti infantry units.

 

An example in C&C Dawn, 2 grenadiers might blow up a NOD buggy. (320 vs 300).

An example in starcraft, a very small group of marines dropped on top of a siege tank in siege mode. (250 vs 250).

Share this post


Link to post
Share on other sites

Cut the buggy some slack. Buggies and humvees are really manoeuvrable, and with some micro you can easily take out those grenadiers with them.

As for the siege tank... uh, that's not really an anti-infantry unit. It's artillery. Sure it may work well against infantry, but technically it's anti-armour, and like any artillery it needs an escort to make sure enemy units can't come too close to it. In defensive situations, I always use deployed siege tanks in combination with bunkers with 3 marines + 1 firebat combo, and preferably with some towers around for detection and extra AA.

Really, there's no real 'failing' involved; they're just not meant to be used one-on-one. This is classic rock-paper-scissors stuff. The games have more units for a reason.

Also, if you have cheap enough / fast enough anti-infantry units that do their job properly, you simply end up with a game where infantry is useless. See also, RA1, and its pillboxes / yaks / light tanks / dogs. And Rangers in RA1 seem more effective than C&C1 buggies/hummers, too. There's just way too much anti-infantry there.

Share this post


Link to post
Share on other sites

Cut the buggy some slack. Buggies and humvees are really manoeuvrable, and with some micro you can easily take out those grenadiers with them.

That is why I used the word "might". I knew I would anger some with that :D. The thought behind it is, it is possible that not all players know how to do this when they just start the game.

Same goes for the Vulture in SC1. Without hit and run or a wall in front. The Vulture sucks against any infantry. While it is meant for anti infantry.

For both cases, when they are back upped by plenty other units, they can't micro since they are pressed against a wall behind them.

(Corner and Conquer)

 

Of course there is no real failing involved. But a situation might arise where they do fail. That is what I am searching for. Because I want to take note of the why and how? (Since I want to make precautions with one of my designs).

 


Also, if you have cheap enough / fast enough anti-infantry units that do their job properly, you simply end up with a game where infantry is useless. See also, RA1, and its pillboxes / yaks / light tanks / dogs. And Rangers in RA1 seem more effective than C&C1 buggies/hummers, too. There's just way too much anti-infantry there.

Good argument. The answer here is due to the density of; durability OR weapon strength OR both.

 

Of course you don't pit infantry against anti infantry weapons. And most players go mass tanks any way. The 3 problems in the old C&C games are (1) density issue's, (2) limited production (click click click) and (3) the fact that tanks can crush infantry without taking the balance costs for this into account. Basically, any tank is anti infantry there. SC1 has no crushing ability, nor a fixed size for any unit.

I estimate that any tank in C&C dawn/RA1 that has the crushing ability, should have about +150 to + 450. The speed of the tank is the cooldown/ROF for the crushing.

 

Crushing is actually the best combination of weapon strength (instant death) and durability (able to run over in time).

Another thing against infantry is, they don't heal. In Tiberian sun, infantry are so much better. Even in KKND after some training, they own the grounds.

 

In C&C 3 the size issue is fixed. Actually, the Pitbull can get even an advantage against a swarm of mammoth tanks. Simply by the fact that they are very small and thus packed.

The crushing issue is fixed as well, tanks are limited in crushing infantry. One at a time, even if there are 100 in one spot. Try it out ;).

Share this post


Link to post
Share on other sites

And most players go mass tanks any way.

That was kind of my point. No one uses infantry in RA because they're useless.

But, eh... RA1 has medics too, you know.

I don't think the concept of 'infantry' really applies to Starcraft, though. There are so many diverse units that it becomes useless to think of some as an 'infantry' class, especially since the game has no crushing anyway. Pretty much everything Zerg that's ground based could be seen as 'infantry' :P

Share this post


Link to post
Share on other sites

Small was seen as infantry. Medium as vehicles and Large as tanks.

The Hydralisk was a vehicle, no matter how silly it sounds. And the Ultralisk a tank.

 

The sizes also had influence on damage effects and splash effects.

Terrans had the most diversity in weapons, thus had the most weapons against infantry. Vulture, Firebat and Ghost, most?, but just 3 for all I know :D. There was also the concept of organic/mechanic. But that only applies to healing. When using broodlings, all the terran ground units suddenly became organic.

 

If it comes to crushing, I believe that in C&C Tiberian Sun, the Mammoth MK3 could crush Nod buggies and Wolverines (or only just the buggy) And let's not forget, the Harvester in C&C Dawn could crush the Recon Bike. :)

 

The only reason that I used infantry in RA was when I had some medics as well. And then I only used the Rocket Soldiers against the Heavy Tanks. But hey, AI is stupid. In online play, this would never work unless you wall in squads of Rocket Soldiers behind some turrets. There comes the V2-Launcher.

Why did they fail? Overkill. They almost never had a chance to return fire. And they could easily get killed by splash. (I think they really miscalculated on that one too when they did balancing)

 


 

I found another one that fails against infantry. In KKND the Flamer ATV's (550 each) do not stand a chance against Snipers (200 each). (Nor would any tank) If it is just 1 against 3 Snipers, it will prevail. But as soon as it is 2 or more against 6, 8, 11 etc Snipers. The Snipers win by simply taking out tank by tank. The strength density is to high for these Snipers.

 


 

Any way, the reason why I asked in the first place is because a friend of mine told me that the tanks in my game actually loose against Infantry. Even though that was my intention in the first place. He said that any tank looses against Grenadiers. Turns out that he miscalculated by doing it randomly with dice -.-

A one time only attempt. With just 4 Grenadiers against 1 Assault Tank (is an anti infantry tank)

The Grenadiers got really lucky?

However, with my "average" calculation, the Assault Tank indeed has 25% health left. Which is rather low.

 

I need to learn how to program a simulator on the pc. Since my Texas Instrument is to slow to do 1.000.000 simulations.

I am curious how big the chances are for 4 Grenadiers to win against 1 Assault Tank. The Grenadiers have a lot of situations where a bit of bad luck would ruin their entire battle.

Or how any configuration would work.

Each configuration would be needing their own simulation?

 

What language do you recommend for that?

Share this post


Link to post
Share on other sites

Small was seen as infantry. Medium as vehicles and Large as tanks.

The Hydralisk was a vehicle, no matter how silly it sounds. And the Ultralisk a tank.

I doubt the game actually has any distinction between the types internally, though.

 

If it comes to crushing, I believe that in C&C Tiberian Sun, the Mammoth MK3 could crush Nod buggies and Wolverines (or only just the buggy)

Nope. It shows that in one video, but the unit has no such ability as far as I know. Also, it's Mammoth Mark #2, not #3 :P

And let's not forget, the Harvester in C&C Dawn could crush the Recon Bike. :)

Actually... ALL vehicles marked as 'crusher' can crush bikes and velociraptors, because they are both vehicle types which have the 'crushable' flag enabled. There's no specific vehicle-crusher option though. Any crushable vehicle can be crushed by the vehicles that are set as able to crush infantry.

 

Why did they fail? Overkill. They almost never had a chance to return fire. And they could easily get killed by splash. (I think they really miscalculated on that one too when they did balancing)

Well, at least in the C&C engine, units can't get damaged by their own splash damage. Heck, they can't get damaged by their own weapons at all. Fun experiment, in that aspect: fire an ion cannon at your advanced comm center. 0 damage :P

 

I found another one that fails against infantry. In KKND the Flamer ATV's (550 each) do not stand a chance against Snipers (200 each).

Well, that's logical. One is close-range anti-infantry, the other is long-range, and clearly also anti-infantry. Pitting a unit against its specific counter does not make it a 'failed' unit :P

Though from your description vs tanks it does seem like the kind of unit a player might prefer to just nuke from orbit :P

 

I need to learn how to program a simulator on the pc. Since my Texas Instrument is to slow to do 1.000.000 simulations.

I am curious how big the chances are for 4 Grenadiers to win against 1 Assault Tank. The Grenadiers have a lot of situations where a bit of bad luck would ruin their entire battle.

Or how any configuration would work.

Each configuration would be needing their own simulation?

 

What language do you recommend for that?

Um. If I'd program it, I'd do it in c#, but that's just because I'm good at C#, and I like how quickly you can whip up small desktop apps with that.

Still, all of this stuff can be calculated with statistics without requiring actual simulations...

Share this post


Link to post
Share on other sites

The game did have distinction between types.

Vultures, ghosts and firebats only do 25% damage against large. This is a very extreem factor.

The fact that you felt as "not. Is because the rest almost had no effects. Most weapons where normal. And those 3 where hardly used.

 

Mammoth MK2, oops :)

Well, I did experience that myself once, bye bye buggy. It was during game play, not the movie. That asks for testing though, it could have been a mod back then.

All units could crush bikes? Are we talking about the DOS/Dawn version?

 

 


Well, at least in the C&C engine, units can't get damaged by their own splash damage. Heck, they can't get damaged by their own weapons at all. Fun experiment, in that aspect: fire an ion cannon at your advanced comm center. 0 damage

I'll try that one. Ima firing my lazor!

Another fun fact! The rocket launcher in Dune2 can accidently hurt itself. I can know :D.

 

Regarding the ATV, it is a heavy vehicle. Not yet a tank, but lets say, it burns a lot!

Except certain infantry -.- Therefore failure.

 

Still, all of this stuff can be calculated with statistics without requiring actual simulations...

Wanna take on a challenge? I am one that tries to calculate everything instead of simulating. Thus when I got challenged by my buddy, I decided to calculate.

If you take on this challenge, I will create one topic, especially for that one.

 

Strangely enough, I did post a challenge on this forum in the past. Which clearly needed simulation instead of calculation :D.

But still, the one that I am talking about now is a lot easier in comparrison.

The other one with simulaiton has been solved. The one that I work on now has not yet been solved. But I do know the number of simulations that are needed to make sure the numbers are sort of correct. :)

Share this post


Link to post
Share on other sites

The game did have distinction between types.

Vultures, ghosts and firebats only do 25% damage against large. This is a very extreem factor.

The fact that you felt as "not. Is because the rest almost had no effects. Most weapons where normal. And those 3 where hardly used.

Interesting. Didn't know that :)

 

All units could crush bikes? Are we talking about the DOS/Dawn version?

No.. all crushing units can crush bikes.

And there is no difference whatsoever in internal game logics between the DOS and Win95 version of Tiberian Dawn. I even played a multiplayer game between the DOS and the Win95 version once (before I started patching the W95 version, that is. With my additions to C&C95 they're obviously not compatible anymore :P)

Another fun fact! The rocket launcher in Dune2 can accidently hurt itself. I can know :D.

Really? Wow. Didn't know that.

 

Wanna take on a challenge? I am one that tries to calculate everything instead of simulating. Thus when I got challenged by my buddy, I decided to calculate.

If you take on this challenge, I will create one topic, especially for that one.

 

Strangely enough, I did post a challenge on this forum in the past. Which clearly needed simulation instead of calculation :D.

But still, the one that I am talking about now is a lot easier in comparrison.

The other one with simulaiton has been solved. The one that I work on now has not yet been solved. But I do know the number of simulations that are needed to make sure the numbers are sort of correct. :)

Well, if you give me the actual stats and formulas I can whip something up, sure.

Share this post


Link to post
Share on other sites

Regarding the rocket launcher in Dune 2. The chance that this happens is very very slim. But possible.

And I only noticed it when I attacked one with my trike.

Thus far, I only reproduced it 2 times. Maybe the chance can be calculated by reading the code?

 


 

Regarding the challenge. If you are absolutely sure to take it on. I suggest you do it in steps. It is harder than any one of us thought.

 

The back story:

My buddy told me that there is a good chance that 4 Grenadiers can defeat 1 Assault Tank.

Even though he says that. When I calculate an average match, the health remaining on the Assault Tank is 25%.

With a basic calculation I can say that the chances are still pretty good there for the Grenadiers to win. But it doesn't say how big the chance is.

There might be more, or perhaps less, situations where one Grenadier survives long enough for taking cover. And thus prolonging it's death.

To make matters worse, The average calculation path did not show a moment for the Assault Tank to spend gained XP (Right before the very last round). After killing 2 Grenadiers, it can spend it only on the damage. Making the damage 16,7% stronger.

If the fight is prolonged, the Assault Tank will benefit from this stronger weapon.

 

The challenge:

He challenged me to calculate. It was only 4 against 1, right?

For simplicity, I removed the actual damages and reduced it to smaller numbers. Thus 9 damage on 108 health has been simplified to 1 damage on 12 health.

 

Well now, the numbers that your need are:

 

Grenadier

health = 3

agility = 4, meaning that each bullet has a chance of 4 out of 6 to hit the Grenadier. (66,7%)

accuracy = 3 out of 6 (50%) for hitting an opponent

range = 1, meaning that they shoot first in each round (very important effect). It also means that they can only take cover behind one other Grenadier. When the front line is smaller than the support line, the enemy may shoot on an unit in the support line.

 

Assault Tank

health = 12

durability = 100%. So this one can be ignored.

accuracy = 100%. So again this one can be ignored. However, each projectile has to face the agility of the grenadier.

multiplier = 4, meaning, this tank shoots 4 bullets. Each bullet has it's own dice roll.

range = 3, this tank shoots after when the Grenadiers have shot. That is the only effect for now. (with 5 tanks or more, it is a different story)

 

Another major part of the game is;

when a projectile is sure to hit. Another die roll is done to determine if the projectile does 0, 1 or 2 damage.

Rolling 1/2/3/4/5/6 results in 0/0/1/1/2/2

After spending XP on the damage, the roll can go differently:

The results are the same, however, if 1 is rolled (0 damage) than the die may be re-rolled 1 more time.

 

When a Grenadier has only 2 or 1 health, it has to take cover behind another Grenadier. But only if it is possible.

Thus 3 health protects 1 other Grenadier with 2 or 1 health.

And 2 health protects 1 other Grenadier with 1 health.

The reconfiguration takes place after one round, not during a round.

 

So, I wonder what you would like to calculate first.

My check list is almost complete, the hardest part is still to come though.

But perhaps if I and you can't solve it, we can make a simplified simulation program with the numbers that I have. :).

Done:

- Chances Assault Tank doing damage on Grenadiers. Including the agility of the Grenadier. 1 list.

- Same chances from the Assault Tank, but then 1 XP level added. The numbers show a big land slide now. 1 list.

- Chances Grenadier doing damage on the Assault Tank. Including the accuracy of the Grenadier. This for 4, 3, 2 and 1 Grenadier separately. 4 lists.

Still to do:

- Creating the paths that show the losses on both sides. Meaning that if the Tank starts with 12 health, the next round it has 12 to 4 health. And the Grenadiers start with 4, each 3 health. The next round they can be reduced to less, and take on a different configuration of the meat/support line. I also have to take into account that once 2 Grenadiers die, the Assault Tank becomes stronger.

 

Good luck,

and the simplest list to start with is the one with just 1 Grenadier trying to hit something. A hint, this list has 3 chances.

 

PS.

Playing and calculating this game are 2 completely different worlds. ;)

Share this post


Link to post
Share on other sites

Well, at least you have the idea of how complex the math is behind the game.

 

And I think there is no other way than to write a program regarding the set up of the Grenadier squad.

 

I think you have given up as well if it comes to math, right? No shame to it. Until now, no one could calculate this?

 

Back to topic:

The Artillery of C&C Dawn, with a bit of bad luck might loose against grenadiers. But chances are small.

The same goes for the MRLS.

What all units have in common is the fact that the tank is lonely and micro fails or the bullet misses by accident. Except for the Flame ATV from KKND, that one dies by certainty.

 

When I think about using 1 unit against infantry.

Could it be that the Trikes in Dune 2000 has a hard job against Troopers? I mean, 1 Trike has to face at least 3 Troopers for a fair fight. Given that the Troopers manage to fire all 3 on the Trike and they are enough scattered that the Trike can only take out 1 at a time.

When the numbers increase. The Troopers face the fact that they share targeted ground. And several Trikes are able to fire upon them. A dead Trooper has no chance in shooting back.

Share this post


Link to post
Share on other sites

Whoo man, that's a bunch of stuff. You might as well implement the whole thing as computer game xD

What's the actual effect of the "protect" logic, though? You never really say. Are they completely shielded that way? Also, can the player specifically choose which one to protect behind which other one, or are there "use the one with the highest hitpoints" kind of rules? Because such tactical choices make the result less reliant on the random values. For example, if you got 3 grenadiers, one with 5 health, one with 2 and one with 1, you could hide the 2-health one behind the 5 health to ensure it survives while probably sacrificing the 1-health one, or you can hide the 1-health one behind the 5-health one and hope the 2-health one will hold out on his own.

Also, does the tank's player actually aim at a specific grenadier, or are the grenadiers seen as one "squad", with the damage randomly going to one of them?

Share this post


Link to post
Share on other sites

Well, for now the issue about protection and support is open.

See the other thread.

Share this post


Link to post
Share on other sites

Never mind. The choice has been made.

 

When the support line can be breached. The attacker chooses where the remaining projectiles go. Thus the weakest units fall victim.

Of course the attack can abuse this and hurt several units instead of killing one.

 

If the defender would choose the victim, this rule could be abused. 1 wall in front, and one good armored unit together with the support. Would be taking all het damage. Not going to allow that.

 

Thanks to Comrad C to point this out.

 

At least now I can continieu callculating as well.

Share this post


Link to post
Share on other sites

Going to attempt to program a simulator on my TI92+.

 

Including the randomness of all the weapons.

The biggest obstacle will be sorting the Grenadiers into formation.

 

On further note, I did try in excel by expanding the tree. But the tree expands in 2 dimensions instead of just 1. Which is impossible to mind-map for me.

 

Edit:

Epic fail !!!

The biggest obstacle is to big for me.

It is like writing AI one and AI two and pit them against each other.

 

Sure I can say which ones are in the front line and which ones are in the back. For this one I can think of any combination of Grenadiers and simply write it down as meat and support. But how do I let the attacker divide the damage amongst them? Simply killing the weakest one and keep track of which of the Grenadiers is available?

 

Programming... not for me.

Share this post


Link to post
Share on other sites

That's what I said... AI behaviour and strategy can differ vastly. They make simulations a lot more than just random calculations :(

Share this post


Link to post
Share on other sites

Still it is a simulation that I need if I want to calculate the exact winning chance.

 

I am going to challenge my cousin. But I think that even he fails.

 

I do know however how 4 Grenadiers are positioned when some have damage or are dead.

And I do know how the attacker has to choose the target.

 

I am able to program this by simply telling the program where to put the units. In other words. Writing down the list for the simulator.

But the list is big. And I have to program a bit for recognizing what is left. For that, a little sorting is needed as well.

 

The up side is, that it is possible. Just a big program is needed.

The down side is, that you can only write the program for 1 simulation. Other situations can not be calculated.

 

Erm, maybe my cousin knows.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this  

×