Does 2 minutes of research time matter?

Oh lawdy, I forgot about this thread. Ugh...

I know I wasn't really the person that cared about armor regen, but I appreciate you looking this up and I'm sure they appreciate it as well!

There's no need to have units if we have comments like your blurb above.

However, if comments are lacking, brain-dead bare-minimum units like "health regenerated per time" or "distance per (time)^2" help minimize testing requirements.

Given unlimited testing capacity, then this isn't a problem at all. But I'm sure Wealthy & Friends might enjoy having a time-saving starting point for their testing.

I could think of some other single-parameter regen methodologies that would need more testing to understand:

• "health regenerated per (time)^2"
  
• "time per health regenerated"
  
• "time^2 per health regenerated" (lol, mindfuck)

But just having units only provides you a "category" of methodologies. There are a bajillion simple scalings and shiftings that would show up in testing. Lazybum pointed out one alternative transformation that he initially assumed:

That seems like a big fucking difference even though we're still working in units of "health regenerated per time". But those kinds of simple transformations would get normalized after testing, so it's not a big deal.

I was saying that you have no proof that you've "tried everything" in the scripts and therefore no one can improve on your efforts.

Maybe you did try everything. Maybe you did a fuckton of testing and mapped out what each script parameter did. But you apparently didn't record it (or haven't shared it in the many times I've called you on it). You're human. Humans forget stuff and fabricate memories.

So whether you did or didn't, it doesn't matter. You didn't write it down so you might as well have not done it at all.

This wasn't what I was talking about in my post, but you seem confused about this and I love you bunches, so whatevs.

Examples, hoooo! This concept isn't tied to armor regen, but I gotta pick something.

What if that single armor regen parameter measured acceleration instead of velocity?* Same single parameter, same lack of comments, different strategy/gameplay.

Strategic difference: Now if you're spinning your regen tank in a fight, then it's more advantageous to let a side loose as many plates as possible before you turn away and then let that same side completely regen before spinning to it against the enemy.

Regen isn't special at all. If you'd like me to think up a different example for a different uncommented script, just say so.

In mathy terms, this is about linear and nonlinear transformations. That's why I asked about units. Kilometers to meters is a linear transformation. You're just multiplying or dividing a constant 1000. Units of distance to units of distance. No biggie. But kilometers to kilometers^2 is not linear. That's units of distance to units of area. We're measuring completely different things. The units matter.

*Fun fact: I honestly thought regen accelerated before this thread because compo's regen felt really really slow compared to regen, but I knew compo's parameter was relatively close to regen's. I'm also wondering if there's any acceleration to the sprint recharge or calc recharge. Details like that are always on my mind.

I swear to god, I wasn't even trying to troll Trickster. It just comes naturally. ove:

By "source", I was wondering how the game code uses those script parameters. As I mentioned above, there are some off-the-wall ways that you can transform a single seemingly simple parameter.

EDIT It just dawned on me that my responses to candles & tricksty poo are the same concept in different words. If both ways make no sense, then you guys need to pay attention in linear algebra class.

 

That map is very unique with that res flag. You don't "lose" 3/4 if the map if you have mid. It always makes me nervous. BE did a great job keeping mid.

Early on, I was not confident about getting turrets on that map because I know the res can allow you to rush heavies. Brentor was cool with it and no one spoke out, so I went for it. I don't normally ask to get turrets. I typically have enough confidence to just get them.

And I did a really really shitty job plugging mid. I should've put a farm on both ends of it.

All around, my comming was mediocre at best. Part of it was my inexperience with leading a sizable team on forest and part of it was my lack of communication (I didn't use my mic).

But I'm not put off, I'd love to try it again on that map. Contrastingly, I don't want to play fucking money ever again. I can't com my way out of a paper bag on that damn map.

But is there a static wait time before starting the recharging? If so, then there's a slight nonlinearity when you look at the effective time to recharge.

Also, is there a wait time for armor regen? Brentor got regen tonight and I was watching for the regeneration and it didn't feel like one plate every second.

It's measured in plates per second, so compo should regen at exactly 30% the speed of regen. I still thinks it feels more like 10-20% of the speed, but perceptions can lie.

We already have that. Reactive is great in early tanks (especially lights) and compo is a better fit in heavy tanks. I don't know if that's a good thing, but that's how it is.

 

The bad news is that I haven't taken linear algebra. The good news is that I'm taking it next term. The better news is that I understood and already knew everything you said.

For the sake of completeness, if the regen rate was measure in time per health, it wouldn't make much a difference, but if it was time[sup]2[/sup]/health, then the strategic choice would be to rotate your plates the moment they get damaged.

Also, when in doubt, assume that anything that can be considered a continuous rate is linear unless evidence exists to the contrary.. Stamina/calc charge, regeneration, bio damage, heat dissipation (Okay, that one isn't *entirely* linear, but linear enough), crosshair size (that one's linear, but it isn't smooth except on small, closed intervals) sabotage damage, repair speeds and so forth.

Well, that's because it's 10% of a plate per second, not a full plate per second. The regen value is "0.1", not "1.0". But there's no delay on regen, it's always going.

I'd definitely try to go off a timer, I've tried estimating rates very often in other situations, and every time I tend to be wrong.

 

The cycle time for the calculator is how often it can be "fired". In the particular case of the calculator, it's the delay after placing a building that you can't place another one, and it's the delay between repair "ticks" when repairing or recycling a building/vehicle.

Also, I found this in the PrimaryAttack() for the calculator.

 

Jephi'es passive aggressive nature is revealed!!

 

I know we're getting off topic a second time, but if regened hp is an increasing concave up function of time, then you'd maximize total regened hp by maximizing the time that it can regen. You do that by giving each side the maximum hp to regen. That's accomplished by waiting until a side is almost completely damaged.

Granted, that's equilibrium behavior at full tilt. If you have three other pristine plates in the beginning, maybe you don't want to be that aggressive. But you'll pay for it if you eventually need to get back to that first plate and it's not fully regened.

"Linear enough?" Now you're sounding like an engineer!

But seriously though, some of that stuff shouldn't be linear. Any time you're regenerating something like hp (they player doesn't choose to lower it and they want to keep it as high as possible), you need to reward the player for not taking a hit during a long stretch of time. Cod does it with their bleeding eyes regen and it "feels" right (because it is). Dem detailz...

Nevermind, I was thinking regen worked at one plate per second and compo was .3 plates a second. Maximum derp, engaged.

 

You can't decalculate the CV anymore?

 

(pPlayer->m_bCommander) is checking if the player who is using a given weapon is currently the commander. The implication of this is that at some point, Trickster found a way to use the engineer calculator while in the CV, so it instantly returns from the attack function.

That's health/time[sup]2[/sup]. I said time[sup]2[/sup]/health, which would put health in terms of the square root of time. It would be a terrible idea to do it that way, but it's still a possibility.

 

Since Spartacus doesn't only post in Off Topic, sometimes I click "Find all posts"

I think a lot of what he's saying (also) relates to calculus, with the squares and the convexity and the acceleration. Linear algebra in 1 dimension (transformations from R to R) isn't very interesting.

I must admit, the "kilometers to kilometers[sup]2[/sup]" confused me for a surprisingly long time (and yes I have used the metric system frequently before :p ). I was thinking of a linear transformation of the form

f : R → R[sup]2[/sup].

Functions like that could take a length x and give you an ordered pair representing the side lengths of a 2:1 rectangle with area x[sup]2[/sup]. But y kilometers[sup]2[/sup] isn't an ordered pair, it's a single number, so we are actually dealing with a function from R to R. So a (hypothetical) function that calculates the health of a wall based on damage received d and wall height h could be

g : R[sup]2[/sup] → R defined by g(d, h) = h - d.

g is linear as it stands, although one would have to modify it if one wants to avoid giving the wall negative health.

 

I think my lazy notation was confusing.

I meant to communicate that if you have two regen parameters that are both measured in something like HP/time, then they are linear combinations of each other.

But if you have one measured in HP/time and another measured in something like HP/tim^2, then they aren't linear combinations of each other.

When Trickster & Friends were saying that units didn't matter because they were "arbitrary", they meant that each way of measuring the regen parameter in the set of all ways to measure the regen parameter were linear combinations of each other. But they were missing that the set of all ways to measure the regen parameter is a lot bigger.

 

10 pages of discussion, still can't be arsed to play the game.

Also, 2 mins of research time doesn't matter.
</thread>

 

You're waaaaay overanalyzing it. It's not that units don't or do matter because of the formulas used. Units don't matter because code doesn't differentiate between units (Assuming that you don't use some hackjob form of custom data types, which Empires doesn't do). If I want to, I could say that the regen rate is the sum of the rotational velocity of the tank's turret and the angle formed by drawing lines from the two nearest buildings to wherever the emp_params entity is, minus the tenth root of the cycle time of the player's primary weapon. Does it make any sense from any point of view? No, but it'd compile and it'd run just fine. A float is a float and an int is an int, regardless of what they actually represent.

Now you can say that certain values are implicitly and not explicitly defined, like regen rate would be in 1/(health*time) by taking the formula and solving for the units of whatever you want. But you might as well just ask for the formula then anyway.

 

2.12, hop in com, type inv_next in console, switch to next weapon. Weapon fires. This is how barracks on the hill in crossroads and 9 mining the enemy com from com view came about.

was patched out with impulse 101

 

That is beautiful.

 

I can haz bio armor that regenerates dead plates next to it once its full?

 

I got lost with "weapon fires". Are you firing your weapon while driving the CV?