38 replies to this topic

[focuspark]

Sable Dove, on 21 February 2013 - 04:06 PM, said:

It has a huge effect on the MG's gameplay. It's just a matter of the MG itself having a negligible impact on gameplay in the first place. Also, having to hit one piece of equipment 20 times is most definitely a large improvement over having to hit it 250 times. It's still terrible, but it is significantly better.

And I'll admit the AC20 was a bad example; I mistakenly thought that crit damage also applied to the component itself. Apparently I was giving the MG too much credit. So let's say the AC2 instead. The ability to randomly destroy any internal but an AC20 in one shot is a pretty significant effect.

Today I stripped the armor of a disconnected mech with my PPC and then emptied 1000+ round of MG ammo into the torsos. I managed to destroy 1 (yes only 1) component before the mech tipped over and died.

Something isn't working as intended.

Edited by focuspark, 21 February 2013 - 04:16 PM.

[Stringburka]

Yes, I'm not saying it isn't a buff - I'm saying it isn't much of a randomness involved because the chance is so high and the actual game effect is so low. If you fire the machine gun one shot at a time until an internal blows you'll note that nearly always will it have taken close to 20/(slotstakenbyobject/totalslotsindepartment). For an AC-10 in an arm, it will almost always take close to 34 shots. Sometimes 31, sometimes 37, but even though it COULD be destroyed in 7 shots the probability of that happening is insanely low. I don't have any software to do that calculation here and now for different chances, but let's just say if you did this experiment a hundred times I'd put fifty bucks on that it wouldn't happen below 20 shots even once.

So we don't misunderstand each other: I'm not saying the machine guns critical damage modifier is irrelevant to how the machinegun is used or it's effect. I'm saying that the destruction of items with miniguns/flamers isn't very random at all (possibly with the exception of Gauss rifles which have low HP and high slots enough that the random aspect is actually relevant, there's something like a 0.2% chance of blowing it in two shots).

With the larger hitters such as AC-10 and above however, critting is quite random - whether one likes that or not is another matter.

[focuspark]

The other problem with the current mechanic is that the devs are likely to say "what, the MG can do X damage - it's fine" when in reality it's doing X / 50 damage.

Better to just have flat numbers - they're easier to balance and easier to understand.

[Stringburka]

focuspark, on 21 February 2013 - 04:13 PM, said:

Today I stripped the armor of a disconnected mech with my PPC and then emptied 1000+ round of MG ammo into the torsos. I managed to destroy 1 (yes only 1) component before the mech tipped over and died.

Something isn't working as intended.

No, currently the MG buff is probably not working as intended, if one looks at data smurfy has made available. It seems like the damage bonus is currently inactive for both the MG and flamer, and that they have insane crit chances (as in, far over 100% crit chance)

[Sable Dove]

Yeah, I think we're kind of arguing the same side, or just that we're arguing two different things. I know that in effect, it's less random than it seems because of the high volume, but there's still the chance that someone gets lucky and gets 10 triple crits in a row, or unlucky and never crits the whole game (an exaggeration, but still technically possible). If it averages out to a certain multiplier, why not simply give the weapon a flat multiplier and take the random element out of it.

It's the same reason I don't like the headshot mechanics in this game because it's about 99% luck. Sure, you might try to hit the head, but unless both mechs are stationary, latency and tiny hitboxes makes headshots primarily luck-based. I've gotten a lucky headshot before, and I bet you the Atlas who went from undamaged to dead instantly didn't have a very fun game.

I just don't feel that a luck-based mechanic has any place in a competitive game. And I'd rather MGs and Flamers deal increased damage to the component itself. Until engines can be destroyed independently of the torso, no crit-seeking weapon will be very useful for its crit-seeking capabilities.

[focuspark]

If the numbers on Smurfy's are correct we should be seeing something like


14% 0.48 = 0.067
80% 0.96 = 0.768
30% 1.44 = 0.432
per shot = 1.267

AND given the MG's rate of fire that would be 12.67 DPS and we're NOT seeing damage like that from the MG. My Cicada has 4 of them and they still feel like I'm throwing dust at people.

Sable Dove, on 21 February 2013 - 04:33 PM, said:

Yeah, I think we're kind of arguing the same side, or just that we're arguing two different things. I know that in effect, it's less random than it seems because of the high volume, but there's still the chance that someone gets lucky and gets 10 triple crits in a row, or unlucky and never crits the whole game (an exaggeration, but still technically possible). If it averages out to a certain multiplier, why not simply give the weapon a flat multiplier and take the random element out of it.

It's the same reason I don't like the headshot mechanics in this game because it's about 99% luck. Sure, you might try to hit the head, but unless both mechs are stationary, latency and tiny hitboxes makes headshots primarily luck-based. I've gotten a lucky headshot before, and I bet you the Atlas who went from undamaged to dead instantly didn't have a very fun game.

I just don't feel that a luck-based mechanic has any place in a competitive game. And I'd rather MGs and Flamers deal increased damage to the component itself. Until engines can be destroyed independently of the torso, no crit-seeking weapon will be very useful for its crit-seeking capabilities.

IMHO either everything should have a luck component or nothing should. Some with and some without doesn't really work well... again my opinion only.

[Stringburka]

If the numbers on smurfy are correct, we have no idea how it works really. Because what those numbers say is that the crit chance after increase is 39% for 1 crit, 94% for two crits and 33% for three crits - but since we (or I at least, please correct me if wrong) don't know how the actual rolling for crits works we can't know if they function at all, especially since the total percentage is a fair bit above 100%.

And since the damage increase doesn't seem to work (a typo in the line) the damage per crit seems to still be 0.04.

So honestly, at this time, I have no idea at all how MG crits work. This discussion has been from the perspective of "when it works", since the issues seem to be simple typos they should probably be fixed soon.

[StalaggtIKE]

Flame can never be effective at dishing out heat. Because a shutdown mech is a dead mech. With perfect weapon convergence and better netcode, head shots are ridiculously easy to do. I score a kill over 80% of the time when a mech shuts down in front me.

Edited by StalaggtIKE, 28 February 2013 - 11:52 AM.

[Stringburka]

StalaggtIKE, on 28 February 2013 - 11:52 AM, said:

Flame can never be effective at dishing out heat. Because a shutdown mech is a dead mech. With perfect weapon convergence and better netcode, head shots are ridiculously easy to do. I score a kill over 80% of the time when a mech shuts down in front me.

I survive over 80% of the time I shut down. And I'm in a light 'mech or cicada.

But yeah, I'm more in favor of reducing the heat cost rather than the heat effect of the flamer.

[rgreat]

I support the OP.

[StalaggtIKE]

Stringburka, on 28 February 2013 - 12:04 PM, said:

I survive over 80% of the time I shut down. And I'm in a light 'mech or cicada.

But yeah, I'm more in favor of reducing the heat cost rather than the heat effect of the flamer.

Lol. So do I. I guess that can be attributed to player ignorance. That will be the case as long as most are none the wiser. Either way I've been working on my heat management, so I don't become a victim.

[Stringburka]

StalaggtIKE, on 28 February 2013 - 12:32 PM, said:

Lol. So do I. I guess that can be attributed to player ignorance. That will be the case as long as most are none the wiser. Either way I've been working on my heat management, so I don't become a victim.

It also depends on loadout. If someone shuts down for like one and a half second, you have maybe two seconds to destroy their cockpit. If they shutdown in front of you and you're in something with a 33+ pinpoint alpha, they die. My primary 'mechs have 10/12/15 (raven varying), 20/25 (cicada varying) and 19 (spider) each; I need two to three alphas to kill someone even if their cockpit is in my face.

[StalaggtIKE]

Stringburka, on 28 February 2013 - 01:00 PM, said:

It also depends on loadout. If someone shuts down for like one and a half second, you have maybe two seconds to destroy their cockpit. If they shutdown in front of you and you're in something with a 33+ pinpoint alpha, they die. My primary 'mechs have 10/12/15 (raven varying), 20/25 (cicada varying) and 19 (spider) each; I need two to three alphas to kill someone even if their cockpit is in my face.

True. I didn't consider that, as all of my builds have the potential for 33+ alpha damage.

[Tesunie]

Funny. Since the changes to machine guns, I've lost component after component to a machine guns spider I was ignoring to deal with a larger threat. His machine guns ripped apart my LRMs and AC on breached sections. Just happy my ammo still had sections of armor over them.

Of course, I think you missed the memo that machine guns are anti infantry, not anti armor/mech. wait till they add in infantry and vehicles into game like they have announced... These weapons will be great and have purpose then.

[Tesunie]

To take some of the misinformation out, all weapons have a crit chance. The above stated weapons discussed above have increased crit chances over normal weapons, with I believe, increased crit damage as well. So the statement someone made about all weapons should more having luck involved, they all do to some extent. Some weapons are better at breaching armor, and others are better at wrecking internals. True even for a lot of today's weapons. You don't waste anti infantry weapons on a tank, and you don't waste anti track/armor weapons on infantry.

Edited by Tesunie, 28 February 2013 - 01:50 PM.

[Stringburka]

Tesunie, on 28 February 2013 - 01:49 PM, said:

To take some of the misinformation out, all weapons have a crit chance. The above stated weapons discussed above have increased crit chances over normal weapons, with I believe, increased crit damage as well.

The exact numbers are:

Standard crit chance (all weapons):
25% chance for 1 critical; 14% chance for 2 criticals; 3% chance for 3 criticals.

LBX, Machine Gun, and Flamer should also have:
+14% chance for 1 critical (39% total); +8% chance for 2 criticals (22% total); +3% chance for 3 criticals (6% total).

Currently there seems to be a bug, however, that makes MG's and Flamers have 39% for 1 crit, 94% for 2 crits and 33% for 3 crits - how, if at all, this works I have no idea.

For every crit a weapon inflicts, a random slot in that hit location (for example right torso) is selected and whatever is in there takes the weapon's damage.
LBX, Flamer and Machine Gun should all have this damage increased as a bonus, however; LBX critical damage is 2 rather than 1, Flamer critical damage is 0.044 rather than 0.04 and machine gun critical is 0.5 rather than 0.04.
However, yet again there seems to be a bug where the machine gun and flamer damage bonus isn't currently active.

[Tesunie]

Stringburka, on 28 February 2013 - 04:11 PM, said:

The exact numbers are:

Standard crit chance (all weapons):
25% chance for 1 critical; 14% chance for 2 criticals; 3% chance for 3 criticals.

LBX, Machine Gun, and Flamer should also have:
+14% chance for 1 critical (39% total); +8% chance for 2 criticals (22% total); +3% chance for 3 criticals (6% total).

Currently there seems to be a bug, however, that makes MG's and Flamers have 39% for 1 crit, 94% for 2 crits and 33% for 3 crits - how, if at all, this works I have no idea.

For every crit a weapon inflicts, a random slot in that hit location (for example right torso) is selected and whatever is in there takes the weapon's damage.
LBX, Flamer and Machine Gun should all have this damage increased as a bonus, however; LBX critical damage is 2 rather than 1, Flamer critical damage is 0.044 rather than 0.04 and machine gun critical is 0.5 rather than 0.04.
However, yet again there seems to be a bug where the machine gun and flamer damage bonus isn't currently active.

Thanks for the info. I didn't know the exacts on those weapons.

Just remember that they just made changes to the weapons, so some of the changes might not be working as intended yet. It might take some time for them to notice a problem, or keep in mind that it might be working, just you aren't seeing the effects enough right now. We don't have the back logs of combat to see it the weapons are criting properly.

I can say that an ignored spider with mgs stripped me of all components very quickly in a match. Someone else peeled the armor off, and they left me with no weapons left in those spots in very short order. Was in a Dragon, and lost my ct missiles and arm AC to those mg. They left me with a lone med laser to my name when they finished with me.

Edited by Tesunie, 28 February 2013 - 04:55 PM.

[MustrumRidcully]

It's like Artemis and Double Heat Sinks all over again...
Is there really no more reliable way to test this stuff before it goes "Beta"? Or is it some problem with their build process, that uses code or data from a different branch?

[Stringburka]

MustrumRidcully, on 01 March 2013 - 01:15 AM, said:

It's like Artemis and Double Heat Sinks all over again...
Is there really no more reliable way to test this stuff before it goes "Beta"? Or is it some problem with their build process, that uses code or data from a different branch?

I agree, this error should probably have been found before, especially if it really is just a typo.

That said, we have no hard proof or confirmation this is a bug - only a statblock which seem to have typos in it. It could be tested, of couse, if someone really want - just team up two people and start shooting at each other with MG's, counting the shots.

Edited by Stringburka, 01 March 2013 - 01:42 AM.