Body balance

Other talk that doesn't fit elsewhere.
This is for General Discussion, not General chat.
Post Reply
User avatar
NoQ
Special
Special
Posts: 6226
Joined: 24 Dec 2009, 11:35
Location: /var/zone

Body balance

Post by NoQ »

Just curious. Looks like nobody uses light bodies at all. There are some trivial exceptions like the very early game when no better bodies are avaliable, scout vtols, and sensor hovers, but most of the fight is done with the medium and heavy bodies.

Looks like collective light and medium bodies are worse than mantis in all senses except size, but body size is a very subtle aspect of the game, so they are mostly unused.

Or am i just a noob?
3drts
Trained
Trained
Posts: 379
Joined: 01 Aug 2007, 03:50

Re: Body balance

Post by 3drts »

For combat purposes, bigger is better, yep.

I have seen, in limited circumstances, spamming of light bugs when cobras were available, as part of an MG rush.

Also, medium scorpion bodies are often used even when pythons are available, but not much after Mantis is available.

there is a good incentive to use light body vtols, as rearming time is proportional to vehicle weight, light body vtols reload really fast- this makes them good for use if you catch your enemy without AA - you can get more vtols out faster, and they can strike more often.

This is particularly effective on low oil maps where AA wont be spammed so bad - you can take it down as fast as it goes up with lancer bug vtols.

Collective bodies just suck, especially for tanks:
Leopard: slower, more expensive, weaker than Scorpion
(at least for VTOLs, leopard has a lower weight)
Panther: Slower, heavier, more expensive, weaker than Mantis.

The ability to build a stronger body with a light factory, might give the leopard some use, or the ability to build a stronger body with only 1 factory module might give the panther some use.....
But collective bodies come so late, all the factories will have had plenty of time to be upgraded to heavy factories, the "advantage" is nigh useless.
It would have to be a really really low oil map for one to have to consider the cost of upgrading a factory- after just a few units produced, the more cost effective medium bodies would make the extra $100 spent to upgrade the factory well worth it.

In 1.00-1.10, I liked the panther, as it was the strongest VTOL body you could get that would still get a speed of 700 with most weapons (except heaps and thermite)- until the Nexus bodies, that doesn't seem so any more, as heavy bodies seem faster now.

Tigers aren't competitive in terms of cost vs hitpoints, at least without factoring in weapon turrets (so assuming a turretless body).
Even so, I wouldnt accept a body with 2x the HP for 2x the cost, if it still holds the same weapon and has the same firepower -I could just build two of the weaker body, and they would absorb as much damage as the one more expensive body, while doing 2x the damage.

Tiger is only worth it for campaign when you have very experienced units (experience gives a speed boost that the tiger desperately needs) that you want to be harder to kill, or *MAYBE* (that is a big maybe) in MP if you are at the unit cap, but don't have nexus bodies yet.

Nexus bodies are rarely worth it on low oil, and their worth is conditional on high oil, given their longer production times.


As a side note, production times stats and formulas should be listed in the guide.
Peyzoc
New user
Posts: 4
Joined: 14 Sep 2010, 22:55

Re: Body balance

Post by Peyzoc »

I think, now good time to improve hi-level body balance for most usefull. I sad, why developers make high prioritet on vtol balance. Aa/Vtol's have been always good in previous versions, and dont have high populating during mp game.
User avatar
NoQ
Special
Special
Posts: 6226
Joined: 24 Dec 2009, 11:35
Location: /var/zone

Re: Body balance

Post by NoQ »

For combat purposes, bigger is better, yep.
I'm still curious about how the size itself matters. It's obvious that smaller size allows more fire density, i.e. more cannons per tile, which means your enemy will die faster. Is it worth it?
In 1.00-1.10, I liked the panther, as it was the strongest VTOL body you could get that would still get a speed of 700 with most weapons (except heaps and thermite)- until the Nexus bodies, that doesn't seem so any more, as heavy bodies seem faster now.
Hmm. As for VTOLs, medium scorpions and retributions are the fastest bodies ever.
User avatar
Zarel
Elite
Elite
Posts: 5770
Joined: 03 Jan 2008, 23:35
Location: Minnesota, USA
Contact:

Re: Body balance

Post by Zarel »

hao wrote:I'm still curious about how the size itself matters. It's obvious that smaller size allows more fire density, i.e. more cannons per tile, which means your enemy will die faster. Is it worth it?
Well, on a proper low oil map, most of the time you won't have enough units for it to matter.

But smaller bodies does mean pumping out units faster, and getting more damage per power you spend.
themousemaster
Regular
Regular
Posts: 611
Joined: 10 Nov 2006, 16:54

Re: Body balance

Post by themousemaster »

3drts wrote:Even so, I wouldnt accept a body with 2x the HP for 2x the cost, if it still holds the same weapon and has the same firepower -I could just build two of the weaker body, and they would absorb as much damage as the one more expensive body, while doing 2x the damage.


For the record: I agree that I don't like "collective" bodies in MP/skirmish.


That said, if you are going to make the "2X the HP for 2X the cost for just 1X the firepower" agrument, don't forget to qualify the facts that:

A) With the tiger, if you lose exactly 1/2 the hitpoints, you don't lose 1 of the 2 firing cannons
B) You don't have to divide up the EXP for 1 unit

which are universal, and also in the map-specific realm, you have

C) If you are fighting in narrow confines, there are only so many "droids" you can pack per square area

And lastly, depending on how good you are at force commanding (AKA micro-nuts :P ), there is:

D) Go ahead and build 16 units of medium body high-power attackers for cost purposes, but build 4 Tiger-Cannons to put in front as damage sponges, allowing your 16 units to live, and fire, for a longer period.
3drts
Trained
Trained
Posts: 379
Joined: 01 Aug 2007, 03:50

Re: Body balance

Post by 3drts »

A) With the tiger, if you lose exactly 1/2 the hitpoints, you don't lose 1 of the 2 firing cannons
Right, when over 50% of the hitpoints are gone, you have the same firepower, assuming all the damage was focused on one unit.
When less than 50% of the hitpoints are gone, you have double the firepower.
Also I should have "a body with 2x the HP for 2x the cost" to be "a body that when a given weapon and propulsion is added to it, has 2x the HP for 2x the cost"

Also, Splash damage has not been mentioned, which does shift the balance to fewer, stronger bodies.

b) Exp in MP is almost meaningless

c) this also is a reason why cyborgs and medium/light bodies may be preferable.
Go against some MG hardpoints with 10 heavies able to squish together into firing range... or 15 mediums, or 20 lights, or 40 cyborgs.... sometimes maybe the quicker kill may be worth the reduced HP, but tigers dont pack more firepower into a given area, just more HP over pythons and Mantis.

d) or put 6-8 mantis hmgs (or some other unit like that) for the same amount of HP damage sponges, for a lower cost, or a higher sponge for the same cost.

Its really hard to think of situations in MP, where high attrition is expected and necessary, that Tigers pay off.

e) I'd like to see all collective bodies given a 2nd hard-point with a MG (the most basic, light one), and no other weapon can be put on that second hardpoint.
It would at least up their "coolness factor", even if by the time you get them, a MG is a weak peashooter relative to other weapons and enemy armor and HP.

I've never seen a MP situation where they are useful.

If their HP to cost ratio was improved, I could see using the tiger for mobile artillery, to better survive enemy CB fire.
Or defensive tanks at a chokepoint.
They should be strong but slow, and used in roles where that is good.
But their strength isnt cost effective, and their slowness is strategically crippling in most cases.
Furthermore, the advantages of heavier bodies, and their placement in the tech tree, ensures that leopard will never see effective use, even as a slow but strong body.
Jorzi
Regular
Regular
Posts: 2063
Joined: 11 Apr 2010, 00:14

Re: Body balance

Post by Jorzi »

My brother once played a game agains his friends. It was an artillery battle, so he simply assumed that the others were also putting their whole budget on artillery, and didn't care about AA. That was until one of his friends suddenly wiped the whole map with a tremendously huge heap of bug vtol:s. It was quite spectacular, he actually managed to fill the whole sky with VTOL:s, and since they reloaded so quickly, they simply never stopped coming.
That said I never use the leopard body, it simply lacks engine power for most stuff and it's not even cheap.
Python and bug is good for budget stuff, though, and retaliation works well as a tech advantage.
ImageImage
-insert deep philosophical statement here-
3drts
Trained
Trained
Posts: 379
Joined: 01 Aug 2007, 03:50

Re: Body balance

Post by 3drts »

Not just python and bug, but also scorpion and mantis.

I often see people use half tracks to counter the python's lackluster speed.
I think its worth noting that in a battle with HMGs, scorpion tracks beat python half tracks, and are cheaper, and faster. They can also pack a little tighter, and if your enemy isn't using high splash weapons, this is a good thing.

I wish we had build time stats, as on many maps, build time is as important as power.

New paradigm units simply cannot be beat for the cost, unless you are using some very expensive weapons which marginalizes the body+propulsion cost, and one instead looks at how many HPs they can pack into the unit using the expensive weapon (such as a plasma cannon)

Note that a tracked vengeance, costs 292.5 power, and gives 1110*4 = 4440 HP with tracks and full upgrades.

Now look at the list of weapons that cost 250 or more (aka, approx. the price of a tracked vengeance or retribution):
All command turret variants:
Heavy Cannon - 250
Twin Assault cannon - 250
Plasma cannon - 750
Pepperpot mortar - 300
Groundshaker - 350
Hellstorm - 400
Ripple rockets - 300
Scourge - 300
Seraph - 400
Archangel - 500
Needle gun - 250
Railgun - 300
Gauss Cannon - 400
Heavy Laser - 600
Nexus link - 400
Flak Cannon - 250
Vindicator - 300
Stormbringer - 500

Considering the cheapest possible body+ propulsion is about 37.5 power, and the most expensive Nexus body+ propulsion is 292.5, all the above weapons you would only be able to roughly half the cost of the unit at best by using the cheapest body+ propulsion.
Aside from the command turrets (which can get ridiculously expensive), and the HC, twAC, Gauss, and Rail, all the weapons have HP of 300 or less. Weapon HP does not upgrade.

For the biggest illustration of a Nexus body being cost effective, lets choose a plasma cannon (cost 750, hp 50), and a bug wheels vs tracked vengeance:
bug wheels plasma cannon, full upgrades: power cost 787.5, hitpoints: 456
veng, tracks, plasma cannon, full upgrades: power cost: 1024.5, hitpoints: 4,490
Power ratio: 1.32 to 1, hit point ratio: 9.84 to 1
Clearly nearly 10x the Hp is going to be worth 32% more power

Lets compare a mantis tracks plasma cannon, vs a vengeance tracks plasma cannon:
Vengeance tracks HP: 4440, + 50 weapon hp = 4490
Mantis tracks HP: 2664, + 50 weapon hp = 2,714
Vengeance tracks cost: 292.5 +750 weapon cost = 1042.5
Mantis tracks cost: 117 + 750 weapon cost = 867

Power ratio: 1042.5/867 = 1.20
HP ratio: 1.65

So for every 5 Vengeance tracks Plasma cannon tanks, you can build 6 Mantis plasma cannon tanks.
For the firepower it takes to kill 6 mantis plasma cannon tanks, you can only kill 4 Vengeance Plasma cannon tanks.

One side has 1.65/1.2 = 1.365 = 37.5% more hitpoints, the other side has 6/5 = 20% more firepower.
In a fight between plasma cannon equipped mantis vs vengeance tanks, the vengeance is the victor.
This gets even more pronounced if the vengeance are able to be massed sufficiently, or use a chokepoint, such that the numbesr of units actually firing at the other side are equal, and the only difference is how many "reserve units" there are at the back of the group not firing, trying to find space to get closer to the front.

If we switch to Seraph Missile, the costs reduce to 692.5 and 517,
The power ratio goes to 1.34.
The HP ratio stays the same.
Now if both sides spend X money on tanks, one side has 1.65/1.34 = 23.1% more aggregate HP
But the other has 34% more aggregate firepower.
If the fight is such that both sides have an equal # of tanks firing at once (chokepoint scenario, or vengeance's manage to keep their distances with their equal speed), losses are such that the Vengeances should come out ever ahead - the kill ratio will be higher, which might give some units a rank of experience-> basically a 6% HP boost.
If both sides reach the unit cap, the replacement costs after the battle will favor the Vengeance bodies.
If all units in the fight are able to engage each other at once, the Mantis tanks will win.

If we switch to gauss cannons, the power ratio is the same as with Seraphs, but the HP ratio skews toward the mantis tank.
the new HP ratio (due to the higher gauss hp) is
4490+ 650 vs 2714+ 650 = 5140 vs 3364 = 1.53
So the aggregate HP ratio advantage of the vengeance tanks is 1.53/1.34, or 14.2% more HP for the same cost.
But the aggregate firepower for mantis, is 34% higher for the same cost, mantis tanks will smash the Vengeances, unless the vengeances can fight in equal numbers (unit cap reached, chokepoints, enemy units strung out)
*note, factoring in armor, makes the Vengeance look even better.

For scourge, with its further decreased cost, I think it will barely come out ahead if the vengeances fight in equal numbers.
For T2 weapons and such, I'd say pick the Mantis.
For T3 weapons - only use the Vengeance if you have time to build up to reach the unit cap, or at least a unit blob so thick that you'll have units in the back not able to fire, or its a map like squared, and you anticipate fighting in a chokepoint.

Otherwise, Mantis rocks the cost effectiveness vs the Vengeance.
At less that full upgrades, body HP isn't as high relative to weapon HP, and the HP ratios are lower (power cost never changes), and mantis beats the Tigers in cost effectiveness for pretty much any weapon, all the while, being able to move faster.
Mantis rocks, and it takes really really expensive weapons to make higher cost, higher HP bodies cost effective to build

At the right weapon cost, hp, and upgrade level, I could potentially see python being preferable.
But as units you will mass through several research upgrades, you cant go wrong with mantis, and Tiger is a poor choice (not to mention, it just comes around late)

Though in general, the rockets, due to low HP and relatively high cost, would favor sturdier but more exepsive bodies, over weaker but cheaper bodies.
The problem is rockets are best on fast units to take advantage of salvo fire "dancing" in and out of weapons range, or keeping at long range if they have a range advantage (as they often do).
Tiger, being slow, just sucks. Vengeance, being fast, is OK.
Germanium
Trained
Trained
Posts: 94
Joined: 21 Jun 2010, 01:39

Re: Body balance

Post by Germanium »

Well written, 3drts! I have done the same calculations as well, but at some point you often have to consider unit production time which you also mentioned earlier. I would love to get to know the equation for build time of units. I "think" it is related to HP and weight but i can't get the correct factors for it (from testing), it seems. That leads me to the conclusion that i miss something there, but
a*HP + b*weight ~ build time
does not work out for me. Any ideas on that?
It is not knowledge, but the act of learning, not possession but the act of getting there, which grants the greatest enjoyment. - Carl Friedrich Gauss
3drts
Trained
Trained
Posts: 379
Joined: 01 Aug 2007, 03:50

Re: Body balance

Post by 3drts »

I am also puzzled by build time, and cannot do any calculations with it.

I just know that nexus bodies build slower than np bodies, can't really quantify it.
I thought it might be related to cost, or HP.

I assume a trucks "build points" stat relates to how fast it repairs, and how fast it builds, which suggests a 1:1 ratio if build time to hitpoints for structures, but I don't know.

I would *guess* structure build time is calculated in much the same way as unit build time, but that is only a guess.

This is a question for the developers I guess.


One thing that has disappointed me about RTS games in general, is balancing using build time, rather than build cost.

I'd love to see a game where if race A and race B are given unlimited resources, but only 15 minutes to build, then Race A kicks the crud out of Race B.
Likewise if Race A and Race B are given X amount of resources, but as much time as they want to build units, race B kicks the crap out of Race A.

In other words, on a cost basis, the units of race B are just plain better than race A- race A units build alot faster though.

Or perhaps not splitting based on race, but tech requirements - perhaps you have two roughly equivalent units, one costs more than the other, or one is stronger than the other, but takes longer to build.

I'd like Nexus bodies to be like that: very cost effective, but not "time effective".

Right now, the NP bodies are cost and "time effective".

Maybe reduce the build time of Collective bodies: so if you have a surplus of power, and you need units fast, you can crank out these somewhat crude but effective tanks really fast (look at alot of the equipment made in WWII- simplified - somewhat crude- for mass production - for the same amount of, or less, physical resources, you could have a more accurate, lighter weapon).
If your production is limited by power flow rate, you switch to more resource efficient tanks, that take longer to build.

Eh just a thought, but I suppose we could leave the current collective body stats along, and slash the build time, that should get people using those bodies in some cases, or on some maps.
Germanium
Trained
Trained
Posts: 94
Joined: 21 Jun 2010, 01:39

Re: Body balance

Post by Germanium »

I agree, reducing the build time for the "green" bodies as i keep calling the collective ones, could probably make it worth building them sometimes (for defense e.g.).

I think that the other part is already somewhat included. I did not write it down but in a few of my tests i have played around with producing stuff in limited amount of time/ limited amount of ressources. And the outcome was not always the same.

And then you miss the research part. Unless you have all techs or t3 weapons, there is already a time limit given by new weapons. Normally you cant do a tank battles with assault guns when the other has scourge tanks (in general at least). In that sense, for most weapons you already have a time limit given and the t3 game is not predominantly a tank battle currently. But basically a Twin Gauss Dragon Track should beat anything (tanks i mean) given unlimited time but i am quite sure that in the time you produce 5 twin gauss dragons you have have 10-15 scourge/seraph mantis tanks (just an estimation not a known number) and even if they wouldnt beat the twin gauss dragons they could be outmoved and threaten the base. So for a rts game i think its already quite "complicated" ;-) .... but then again it rarely comes to tank battles in that part of the game. The 4 hour games are (gladly) not overly common to me.

Just my thoughts though. And thanks for your ideas on what build time is based on.
It is not knowledge, but the act of learning, not possession but the act of getting there, which grants the greatest enjoyment. - Carl Friedrich Gauss
3drts
Trained
Trained
Posts: 379
Joined: 01 Aug 2007, 03:50

Re: Body balance

Post by 3drts »

If I give both side 10k power, and one side builds vengenace scourge, and the other twin gauss dragons, I think the scourge win.

I'm talking about trading manufacturing time for extra "resource value"

Have one tank body clearly be the best body for the money, but also the slowest producing body.
Right now, the best body for the money, is also the fastest producing body, which is why np bodies are so kickass.
Post Reply