100 replies to this topic

[Dlardrageth]

GaussDragon, on 26 December 2011 - 03:48 PM, said:

The 'force' number I used was arbitrary, and was independent of actual weapon damage. If that's what you mean? It still does affect light mechs most, but I added the "mass softener" because I felt they were getting knocked down too easily lol.

Just out of curiosity, how exactly will that formula of yours factor in the "impact angle"? I mean, taking the speed into the equation is a good first step for sure, but if you want to go for that detailed damage modeling, you would need to take the impact angle into account, won't you?

It should IMHO make a difference if the Mech you just hit/strafed at full speed was coming at you dead-on (0° angle?), if you hit him in a "driveby" fashion (90° angle?) if he's moving away from you (180° angle?) or anything in between. I would expect knocking down a Mech with a perfect side shot hit while he's on the move should be easier than doing the same while he's actually charging head-on at you.

You made an allowance for terrain slope already, think impact angle would be the next logical step. IMVHO.

[CeeKay Boques]

Do... do you think Physics could handle this? If a gauss rifle and LRMS each give 10 "points" or degrees of impact (1 point per missile, 10 hits say) and the torso is pushed over 45 degrees back, or left side rear, and right side front get hit simulateneously, that would be quite an impact and spin.

Do we have to Pilot Roll? is that the only way? I guess. Its probably easier or something.

Will getting hit in the legs as you run slow you down? can you get tripped with a slug to the foot while its in the air?

ah well.

[GaussDragon]

Dlardrageth, on 27 December 2011 - 11:50 AM, said:

Just out of curiosity, how exactly will that formula of yours factor in the "impact angle"? I mean, taking the speed into the equation is a good first step for sure, but if you want to go for that detailed damage modeling, you would need to take the impact angle into account, won't you?

It should IMHO make a difference if the Mech you just hit/strafed at full speed was coming at you dead-on (0° angle?), if you hit him in a "driveby" fashion (90° angle?) if he's moving away from you (180° angle?) or anything in between. I would expect knocking down a Mech with a perfect side shot hit while he's on the move should be easier than doing the same while he's actually charging head-on at you.

You made an allowance for terrain slope already, think impact angle would be the next logical step. IMVHO.

I accounted for this very thing. What you describe is pretty much exactly how the calculator models it. If a light mech is charging straight at you at a fast speed, it's pretty much impossible to get knocked down. If it's circling you at a perfect strafe (a perfect strafe is a value of 0.9 in the calculator) then it's easier for him to get knocked down. If he's passing you at an angle of 60 degrees (relative to you) then his speed and his angle completely offset each other (thus the 0.6 as a threshold). I made the angle variable only matter as much as the speed variable did, otherwise a mech standing still would have a way higher chance of being knocked down if he was rotated 90 degrees away from you as opposed to staring directly at you.

I quoted my original post explaining the 5 variables. The fourth one is impact angle. In the formula display, it would be the variable that shows up in the part of the formula "(0.6 - strafing angle)" however I pointed out one caveat to it. For this calculator, basically assume you're a mech that's standing in one spot and you're always turned towards your attacker, but he's the one moving around you at various angles. In order to refine it a bit further, I'd have to add a sixth variable which would take into account the orientation of the shooter relative to the attacker (the shooter being you, which is the location the formula is based upon, it's sort of the inverse of the strafing angle variable) - think of this theoretical sixth variable as the degree to which the shot itself is a direct hit or glancing blow - aka the direction you're moving at relative to your attacker when you shoot him. I could probably find a way to work it in, it'd probably be as simple as adding it in the form of an addition or subtraction to the strafing angle variable. At the moment I don't have time to really sit down and spend a few hours to tweak it out but I could probably refine it sometime after Jan 6th.

GaussDragon, on 09 December 2011 - 07:22 PM, said:

Knockdown Calculator

Spending all day with a securities book in front of you gets you re-acquainted with math fairly quickly (I'm studying for a securities designation, that's why I'm on here posting at all hours of the day). What I made here is a calculator for knockdown. Why did I make this thing? To illustrate the interplay that the slope of terrain has with other mechanics in contributing to a greater chance of being knocked down, therefore making the player more mindful of where he/she is running around. I made a few key benchmarks and then tweaked the formula from there. There are a few basic criteria that I tried to meet when making this formula (that I've now more or less forgotten now that I actually finished the calculator). The first premise was that 3 LBX 20s would have an arbitrary 'force' value of 135 (3x45), not damage, force. I wanted it so that a 100 ton mech standing still would not get knocked down, but a 45 ton mech standing still, would. I also wanted a 100 ton mech standing still to get knocked down while standing still but on a hill of roughly 40 degrees of incline or greater. From there I spent most of my time calibrating the thing so that 5 variables in the formula worked in fairly balanced way. As you'll see in the formula, the value of 17000 is the treshold for whether a mech gets knocked down or not. The Force of the shot and the final number are relatively immaterial and far more arbitrary. They were used more as guidelines to determine what factors were over or under-represented in the formula. It still isn't perfect.

Link to Excel document: http://www.usaupload.net/d/p54but1vdqe

The formula takes into account 5 variables:

1. Force - An alpha from 3 LBX 20 is equal to 135, it will not knock a 100 ton mech down standing still, but it will knock a 45 ton mech down that is standing still.
2. Mass - The more massive the mech, the less likely it is to be knocked down.
3. Speed - The faster the mech is moving towards you, the shooter, the less likely it is to be knocked down because the force from your shot is countered by its movement towards you. A 45 ton mech moving relatively slowly will no longer worry about having to get knocked down as long as it is moving straight.
4. Strafing angle - I made the strafing angle subordinate to speed so to speak. If a mech is strafing you at an angle within 60 degrees or less, the greater the speed, the less likely it is to get knocked down. However, as the mech strafes at an angle greater than 60 degrees, it is more likely to get knocked down, compounded with speed. That's right, this formula has a built-in penalty for, you guessed it, circle strafing.
5. Terrain slope - This formula treats sloping terrain as a hazard. The greater the slope of terrain after all other factors (except for the force of the incoming shot), the greater the risk of being knocked down. The hazard is greater for smaller mechs because they reach the 17 000 value threshold soooner.

This formula isn't perfect. It assumes that only the strafing mech is at an angle, and that the shooter's shot isn't coming from any offsetting/exacerbating angle. Secondly, if a mech is moving backwards and gets shot in the face, it is more likely to fall down, as it should. However, this formula does not take into account where the shot came from. A shot coming from directly behind a reversing mech would have the exact same effect as a shot coming from directly in front of it according to the formula. This formula isn't perfect and I had to spend a decent amount of my time learning how to use Excel so that people could actually play with the variables themselves. One little bonus that you may notice if you **** the spreadsheet off is that I prevented people from entering in slope and shot angles greater than 0.9.

Edited by GaussDragon, 27 December 2011 - 01:36 PM.

[GaussDragon]

VYCanis, on 26 December 2011 - 04:24 PM, said:

here's something to consider.

yeah, lights should be the easiest to knock down, i think we are all in agreement on that.

but what about making them some of the fastest to recover and get back up and running? (provided they survive of course)

That makes logical/balanced sense. I'm for it.

[verybad]

Slightly off topic, but I would like for larger ballistic weapons to have some recoil that affects targeting and balance a bit.

[Nik Van Rhijn]

Looks like this one reached a reasoned consensus in only 2 pages. Thanks for all the comments and especially GaussDragon for the mathematical basis.

[GaussDragon]

verybad, on 27 December 2011 - 01:54 PM, said:

Slightly off topic, but I would like for larger ballistic weapons to have some recoil that affects targeting and balance a bit.

Hmmmmm. I'd have to see how it'd play out in an actual beta to have an opinion on it since I don't have any experience to compare it to in MW4. I can't really say how it'd affect gameplay because it could factored in to have major or minor recoil in conjunction with reload times etc etc. Again, I wouldn't have an opinion until I tried it.

Nik Van Rhijn, on 27 December 2011 - 01:56 PM, said:

Looks like this one reached a reasoned consensus in only 2 pages. Thanks for all the comments and especially GaussDragon for the mathematical basis.

TY. I think we're all in favour of complexity so long as it adds to overall fun factor and it seems consistent to implement it. I don't think there was anything here that was ever really controversial or that the community got divided over in the past (unlike everything else lol).

[Dlardrageth]

Just as a minor note - the whole recoil issue will be left hanging somewhat in the air as long as PGI doesn't adress the issue of how ACs work exactly in game. I mean, we do have already two schools/traditions regarding this. One - the ACs basically being single-shot sluggers (mostly from some MW gameplay of past), the other them being burst-fire weapons which unleash a couple of shells in a burst at the target (classical lore). IMHO it would make a major difference for the influence of a recoil factor, which way PGI will go. And of course it would affect all ACs, not only the bigger ones, there is only so much gyrostabilization can do to mitigate recoil.

Edited by Dlardrageth, 28 December 2011 - 01:43 PM.

[AJC]

something that surely should happen is heavy hits like a AC-20 hit should knock you around a bit at least throw your aim off by some bit.

[Volume]

Range maybe should be taken into account (for, say, LBX), and, yeah, like you all said - PHYSICS, point of impact, degree of impact, etc, etc, it's not 1999, we don't need to limit our amount of hitboxes - angles and sloped armor and such things can be modeled.

[GaussDragon]

Volume, on 28 December 2011 - 01:52 AM, said:

Range maybe should be taken into account (for, say, LBX), and, yeah, like you all said - PHYSICS, point of impact, degree of impact, etc, etc, it's not 1999, we don't need to limit our amount of hitboxes - angles and sloped armor and such things can be modeled.

Didn't the LBX have a damage drop-off in MW4? I'm almost positive it did. So perhaps have a drop-off for all weapons so that getting up close to an assault mech as a light mech is really risky behaviour. Thoughts?

[Nik Van Rhijn]

@Aegis "Basically, ignoring the fact that we're using a finite number derived from the tabletop version of the game; a version which the devs have stated cannot be transposed into the envisioned game on a 1:1 ratio," Actually what they said was that they would use BT except where it did not translate. Damage translates directly, as does the knockdown rule, and attempts to keep upright (allowing for terrain etc). If they followed your statement it wouldn't be MW.

[GaussDragon]

The exact numbers aren't what matters, it's the relationships. I used an arbitrary number of 135 to represent force, but it's how that number affects other mechs in the environment is what matters. And I agree with a lot of people here, porting the relationship 1:1 (so to speak) may not work for a computer game. It's a matter of making knockdown a balanced mechanic within the game as a whole.

[Sir Aaron]

i believe that reaction to damage take, and that includes movement, should be inside the game. It will make a huge different. For example it should be possible to knock down a Mech too if you hit the Mech a second/third/fourth time just as it bends back from the before hit and such things. It would make a incredible play although knockdown could too be quite an overkill because when you are on the ground you are a pretty easy target without being able to fire back for some seconds until you get your Mech enough under control to return fire.

[Nik Van Rhijn]

It's theoretically why the AC20 was such a feared weapon (in fluff) because it not only does such damage but has the possiblity of knocking any mech (even an assault) down. This was not modelled well in MW4 - hitting a light mech at point blank range didn't even put it off it's stride.
Personally I think ballistics (including Gauss) should have more "chance" of a knockdown than energy weapons, even if they apply the same damage. Being hit by 2 PPC's at the same time should make you judder and have a chance of being knocked down.Not so sure about 4 ML's, but they meet the TT rule requirement.

[Pht]

Yes, I can see this being in the game, and even having an avatar skill addressed to it.

As far as knock from incoming fire - 'mechs really don't experience knock on this level. If the target took enough knock from the 'mech firing on it to unbalance or knock it over ... the firing 'mech should experience MORE knock. Just bad mojo.

No, the reason there's a PSR (piloting skill roll) for taking lots of damage very quickly is because your 'mech has just lost a LOT of weight in armor virtually instantly; sometimes more than a ton. The 'mech has no way of predicting for this and it happens faster than it's gyro system can compensate for.

So you don't "absorb knock" - no, you *fall away* from where you just lost a bunch of weight.

[Nik Van Rhijn]

Not so convinced on that. 1)Recoil 9from AC'S) is (mostly) absorbed on fireing by dampers etc plus the mech "knows" it is firing and can counter the recoil. The impact however is "unexpected" and therfore will have an effect (applies to Gausss impacts as well. The effect of armour destruction would be an added bonus. I agree with what you say for energy weapons.

[verybad]

The reason the AC-20 was/is feared was because it was the first weapon that was an automatic kill if you hit the head. PSRs are a pain, but it was easy enough to generate them with multiple weapons even in the 3025 era.

[Pht]

Nik Van Rhijn, on 28 December 2011 - 05:57 PM, said:

Not so convinced on that. 1)Recoil 9from AC'S) is (mostly) absorbed on fireing by dampers etc plus the mech "knows" it is firing and can counter the recoil.

If it could not effectively control and counter that recoil it very well wouldn't be able to make viable shots with a good chance of hitting things miles away on the horizon - and LOS shots are doable with ac2s.

Long range shots while on the move from arm mounted weapons would also be impossible if the 'mech couldn't cope with recoil, as well, much less extreme range shots.

Quote

The impact however is "unexpected" and therfore will have an effect (applies to Gausss impacts as well. The effect of armour destruction would be an added bonus. I agree with what you say for energy weapons.

Again, if the incoming fire had enough energy at range to knock a target over or even significantly unbalance it, than the firing 'mech would experience MORE knock from recoil from firing those weapons than the target would experience.

[Pht]

Aegis Kleais™, on 28 December 2011 - 11:23 AM, said:

** VOTED OTHER **

Basically, ignoring the fact that we're using a finite number derived from the tabletop version of the game; a version which the devs have stated cannot be transposed into the envisioned game on a 1:1 ratio,...

Pardon me?

Why can't the armor values be ported over? And if you agree they can convert over to a realtime format, than why shouldn't they?

Quote

I do believe that Gyroscopic overload/failure should be in the game.

Gyros don't so much fail or overload as much as that they just can't react quickly enough for quite a few battlefield conditions.

We have a pretty good baseline of how capable a gyro in a 'mech is - pretty much anything having to do with piloting that requires a PSR is "too much" for a 'mech gyro to handle. That's when the piloting skill comes into play!

Edited by Pht, 28 December 2011 - 06:18 PM.