36 replies to this topic

[Baddicus Wolf]

My excursions into in the training grounds confirms the OPs results. I followed Lefty's lead and went with the legs as there are two of them and they are easy to single out.

Shooting an Atlas:
Optimal Range --> 13 shots
75M --> 17 shots
60M --> 26 shots
45M --> 48 shots
30M --> 110 shots
15M --> 280 shots (to get through armor, I stopped there)

Basically, at 60M you are at half damage, and at 30M you are doing a little over 1 dmg per PPC shot. Anything under 45M seems pointless to shoot at unless you have heat to burn or are desperate.

[Krzysztof z Bagien]

Sobakasu, on 20 March 2013 - 12:02 AM, said:

Granted this isn't close to a good test but in the same thread Thomas mentions it's linear some else claimed to test this in a match with a friend and it comes closer to what I get vs a linear decrease:

5 dmg @ 60m, 3.5 dmg @ 45m

I admit I haven't tested in an actual game, I'll try that when I have time.

I did those tests with Kmieciu. We set certain amounts of armour on our mechs and counted how many shots are required to remove that armour from different distance (can't remember exact values as it was 3 months ago). I was hoping for an answer from Thomas, but... no luck.

[Deathlike]

Have you tried 90m and 90m+ ranges?

It may depend on the mech and the distance because it could possibly be "skewed" by weapon location/convergence.

Edited by Deathlike, 14 April 2013 - 11:41 AM.

[Lefty Lucy]

Deathlike, on 14 April 2013 - 11:40 AM, said:

Have you tried 90m and 90m+ ranges?

It may depend on the mech and the distance because it could possibly be "skewed" by weapon location/convergence.

Not by this much though.

[Deathlike]

Lefty Lucy, on 14 April 2013 - 11:48 AM, said:

Not by this much though.

At this point, the devs ability to do math is... sub-par.. and we're not even talking about physics... just basic math.

Edited by Deathlike, 14 April 2013 - 11:50 AM.

[armyof1]

Lefty Lucy, on 14 April 2013 - 11:48 AM, said:

Not by this much though.

Yeah we're not talking 30 deg. inclinations here, something is clearly not linear in this equation. If anything it looks more exponential. And besides why would the distance meter not show actual point to point distance anyway? That's just another bug if it's so.

Edited by armyof1, 14 April 2013 - 11:58 AM.

[Khanahar]

Wow. Could we get more tests on various components for this to get a more complete data set?

Still, my brain has been blown.

[Deathlike]

armyof1, on 14 April 2013 - 11:53 AM, said:

Yeah we're not talking 30 deg. inclinations here, something is clearly not linear in this equation. If anything it looks more exponential. And besides why would the distance meter not show actual point to point distance anyway? That's just another bug if anything.

Sure... the only reason I suggested other mechs is because you'd never know the difference if one mech is "different" than another other than the assumptions.

There was a Hunchback thread that discussed how the reticule didn't match up with the weapons convergence. You would think it is a "fair" assumption that all things being equal that this wouldn't happen... but it did.

In terms of a "scientific" type of test, you have to compare against known values vs unknown values... so this is why you have to do something comprehensive in order to get good data. The testing grounds is good for this... even though it's imperfect like the devs have said. We have to have a starting point, so this research needs to be expanded further IMO.

Edit:
Link to hunchback info:
http://mwomercs.com/...est-short-clip/

Edited by Deathlike, 14 April 2013 - 12:09 PM.

[armyof1]

Deathlike, on 14 April 2013 - 12:01 PM, said:

Sure... the only reason I suggested other mechs is because you'd never know the difference if one mech is "different" than another other than the assumptions.

There was a Hunchback thread that discussed how the reticule didn't match up with the weapons convergence. You would think it is a "fair" assumption that all things being equal that this wouldn't happen... but it did.

In terms of a "scientific" type of test, you have to compare against known values vs unknown values... so this is why you have to do something comprehensive in order to get good data. The testing grounds is good for this... even though it's imperfect like the devs have said. We have to have a starting point, so this research needs to be expanded further IMO.

Frankly I'm not so sure the testing grounds are imperfect, the things I've tested with headshots and different parts that I expect will take a certain damage before being destroyed have all met my expectations. If anything I think it's more accurate since you won't have to worry about network issues because you're actually running it locally. SRMS were also found to be faulty in testing grounds even though it was first considered to be bugged only there. But later it was clear it was the same as in a real match. As long as you consider how things like where ammo is placed and might explode to ruin your test data, testing grounds seems to work fine.

[Wintersdark]

armyof1, on 14 April 2013 - 11:53 AM, said:

Yeah we're not talking 30 deg. inclinations here, something is clearly not linear in this equation. If anything it looks more exponential. And besides why would the distance meter not show actual point to point distance anyway? That's just another bug if it's so.

No, the distance meter shows the distance from the camera to the crosshair point. This will differ from the end of the barrel to the crosshair point. It's not unusual for arms to be much further forward than the camera.

(But not by so much as to cause these results, of course)

[Lefty Lucy]

Wintersdark, on 14 April 2013 - 12:43 PM, said:

No, the distance meter shows the distance from the camera to the crosshair point. This will differ from the end of the barrel to the crosshair point. It's not unusual for arms to be much further forward than the camera.

(But not by so much as to cause these results, of course)

I performed an in-game test with a CTF-3D, placing the PPCs in the shoulder hard points. They are actually very near the same height and depth as the cockpit, so I think it's a good test bed.

[Deathlike]

armyof1, on 14 April 2013 - 12:16 PM, said:

Frankly I'm not so sure the testing grounds are imperfect, the things I've tested with headshots and different parts that I expect will take a certain damage before being destroyed have all met my expectations. If anything I think it's more accurate since you won't have to worry about network issues because you're actually running it locally. SRMS were also found to be faulty in testing grounds even though it was first considered to be bugged only there. But later it was clear it was the same as in a real match. As long as you consider how things like where ammo is placed and might explode to ruin your test data, testing grounds seems to work fine.

They are imperfect with respect to module (pilot) bonuses and mech bonuses. IIRC didn't someone say that the firing rate with lasers is different in the training grounds than in the real game with respect to cooldown (or specifically, when the cooldown is applied). Lasers would fire and cooldown would apply after the lasers were fired, instead of applying them after the lasers finished their firing.

Edited by Deathlike, 14 April 2013 - 12:54 PM.

[Wintersdark]

Deathlike, on 14 April 2013 - 12:50 PM, said:

They are imperfect with respect to module (pilot) bonuses and mech bonuses. IIRC didn't someone say that the firing rate with lasers is different in the training grounds than in the real game with respect to cooldown (or specifically, when the cooldown actually happened).

Firing rate is definitely off in the training grounds.

I set up a mouse macro for my 5 ac/2 Jager. In the training grounds they work out perfectly firing continuously. In the proper game, it steps through the 5 then there is a noticeable delay, so it feels like burst fire instead of automatic.

That, incidentally,demonstrates how ac2's only get their 4dps in the training grounds,and more like ac5 DPS in the game proper.

[Krzysztof z Bagien]

ChartWarrior Online!



Tested in testing grounds (though I belive that it works like this in live game too).
shots - shots required to destroy component
dam/sh - damage dealt with single PPC shot at given distance. It is calculated by dividing component health (number in [bracket], it's armour + internal structure) by shots-1 (as if before last shot that actually destroyed a comopnent its health was almost 0, that is not true in most cases). I had no patience to test on Cat and Atlas at 30m.
It's definitely not linear.

Edit: somehow I posted this in the wrong post LOL
Above values are not incredibly accurate, but it shouldn't make that much of a difference even if actual distance projectile traveled was 55m instead of 60m for example.

Edited by Krzysztof z Bagien, 15 April 2013 - 03:13 AM.

[Deathlike]

It's CDA, not DCA or whatever. Otherwise, the plague of Garths will descend onto you.

[Krzysztof z Bagien]

Better now?

[Rippthrough]

Chavette, on 13 April 2013 - 12:48 PM, said:

Would be epic if they turned out to be wrong even with a simple calculation like this... lol

Can't say I'd be surprised though.

A dev, wrong about their own mechanics? Surely not....oh, wait....