439 replies to this topic

[Rebas Kradd]

Roadbeer, on 07 October 2014 - 07:58 AM, said:

Enough of using New players as the excuse for everything.

It's a factor, at the very least.

Sure, the sniper mechanic is undesirable (though there are other ways to fix that). But one experiment I'd rather not run is whether new players will be an even bigger fan of all their weapons scattering everywhere for no apparent reason.

[Roadbeer]

Rebas Kradd, on 07 October 2014 - 08:54 AM, said:

It's a factor, at the very least.

Sure, the sniper mechanic is undesirable (though there are other ways to fix that). But one experiment I'd rather not run is whether new players will be an even bigger fan of all their weapons scattering everywhere for no apparent reason.

News Flash:
It's not just the sniper mechanic.
You have weapons doing 2.5-4x more than TT values, but have only doubled armor values
You have heat generation at 4x TT values, but dissipation equal to TT values so it actually ENCOURAGES Alpha shooting.
You have a voodoo heat mechanic that penalizes everything but the edge cases it was meant to penalize. (see above)
The New Player doesn't have time to experience their weapons scattering everywhere because they're put in a match with a floating red ball over their head screaming "Shoot Me".
The New player can't figure out why other mechs are capable of doing things that they aren't because their arms are locked in position (which subsequently becomes a desired mechanic for those who AREN'T new players).

The new player is hamstrung because there is no documentation about how to play the game, what the actual mechanics are, and (iirc) a broken tutorial.

Convergence is the least of their ******* problems.

[Eddrick]

Look at how Armored Core did before Armorred Core 4. The games have a BIG learning curve that is unfriendly to new players.

It is possible for a game to thrive and be unfriendly to new players. You just have to keep your core audiance happy.

[CompproB237]

Rebas Kradd, on 06 October 2014 - 12:30 PM, said:

Are we STILL suggesting that we should remove this dynamic?

Players don't want their fire to go anywhere but where they're pointing. This is a problem unique to MechWarrior, because very few other games have multiple weapons being fired from different places on their avatar.

The moment you remove pinpoint convergence from the game, you remove the ability to target specific components of an enemy mech. Players will NEVER be able to target components if that's removed, at least not efficiently. The result is that "shoot the CT" becomes the metagame even more than it already is, and a lot of MW nuance is involved.

I get that convergence makes no sense, I get that it tends toward a meta, but the alternatives are even less optimal.

I just wanted to point out that "Shoot the CT" or 'aim for a body shot' is pretty common in most FPS games. They still have head hit boxes and they're still hittable even with the spread that every FPS has. Most FPS use a CoF for all weapons. CoD sniping notwithstanding but BF3/4 Sniping even had things factored in like Gravity, Friction from air, Weight of the round, velocity of round. I can't remember if they also took in to account Wind, Ambient Temperature and Humidity. Either way, these factors modify where the round actually goes. This mimics reality. This is one of the reasons why we're suggesting CoF or something similar to it.

I actually had an idea that blended roadbeer's, Koniving's, and a few others ideas. Those of my Organization have had lengthy (and somewhat heated) discussions on this and have come up with some interesting ideas. I'd post them once I've had more time to put the ideas to paper. I can say that, the idea we had involved Roadbeer's convergence system but having it deteriorate given certain conditions. Full speed run adds a longer time to it (as per Roadbeer's) but Jump Jetting literally reduces the convergence over time to none. The numbers would need work but the intention is so that a HGN/CTF/VTR using Jump Jets to 'Poptart' would be so that at the point in there jump where they clear their 'Mechs height they have no convergence at all (parallel weapons). Think of the computer struggling to maintain accuracy as the uneven thrust from the Jump Jets propels the walking death machine into the air.

Don't get me wrong here, the weapons would still fire where they're pointing (not your crosshair) but you'd now have to use more skill to aim and fire those weapons. In the end, you'd get the 'poptarts' only able to fire one, maybe two weapon system locations at a time before falling too far. Even then, these shots would not be in the same place (but they would be where you were aiming). This would not only spread the damage but also require considerable skill to even pull off. Granted, the player would need a graphic to give them an idea of where the shots are aiming. Think the distance markers on sniper scopes, but in a line from weapon mount locations in relation to when they're parallel. This can also have the benefit of allowing Jump Jets to return to normal.

Now here's where this convergence method gets interesting though. Notice that I said it needs to be so that convergence goes away once you're nearing a certain height? Imagine a light/medium using jump jets. Usually they're burning the jets briefly to gain mobility or hop over shorter obstacles. They're also usually medium to short range skirmishers. The convergence for lights would be much less a problem for them. They could also balance it by giving lights/mediums a quirk of some sort that allows them to jump 1.5x or 2x their height before convergence becomes parallel.


Keep in mind here that this only a small piece of the grand idea I had. I'm sure there's something that I missed or didn't think of.

I'm also aware that the logistics of this idea are probably impossible due to HSR. Most of the things to make this work the servers already get. It's basically only changing the aiming point for the player. It already takes in to account positional data, trajectory, point of impact, and rewinding the server for determining where the shots are hitting. Basically all the convergence change does is change the server authorizing the change in convergence. It then pushes the 'current convergence state' to the client. That would be another 8 bytes of data to transfer the values 00 to FF (-127 to 127 or int8_t for those uninitiated). This gives 256 values to represent 0 to 100% giving you an accuracy down to 0.4%. That is nothing. You could bump it up to 16 bytes of data to transfer the values of 0000 to FFFF (-32,768 to 32,767 or int16_t) giving you an accuracy down to ~15 ten-thousandths of a percent or ~0.0015%.

[C E Dwyer]

Chronojam, on 02 October 2014 - 06:27 PM, said:

I'd like to make a modest proposal. Remove the Stalker from the game entirely. That would immediately get rid of the 6PPC Stalker threat, which was (and remains) the critical impetus for Ghost Heat. After all, that's the rationalization PGI has used on the forum, on Twitter, and even today on the Twitch town hall.

Let's look at the advantages of removing the Stalker from the game over Ghost Heat. Removing an entire mech to "solve" 6PPC Stalkers would:

• Take no programmer assets, allowing mech quirks to be added sooner
  
• Require no balance changes, resulting in fewer variables to manage
  
• Have no impact on close-range and DPS-focused builds
  
• Not invalidate counter-sniper large laser builds
  
• No risk of unintuitive "40LRMs are cooler than 30LRMs" behavior
  
• No need to use third-party website calculators
  
• Bypass need to rationalize AC/2 ghost heat chainfire glitch as addressing imagined "griefing"
  
• Never worry about finding a purpose for the worthless Stalker 4N variant

Killing the 6PPC Stalker was super important, enough to justify the side effects. After all, we all recognize that Ghost Heat, by its design intent, purposely reduces player agency; it actively reduces build variety by making several builds less viable.

We all recognize that, mechanically, Ghost Heat did absolutely nothing to impact the dominant 2PPC1Gauss metagame loadout(a popular build compatible with many jump-sniping chassis variants).

We all recognize that "laserdrill" builds used to counter instantaneous, pinpoint-damage builds (relying on PPCs, Gauss Rifles, and Autocannons) and jumpsnipers were directly impacted by the "2" limit on large lasers for Ghost Heat. This reduced playstyle options and hard counters for long-range high-pinpoint builds

We all recognize that Ghost Heat impacted brawlers and mechs that relied on sustained DPS, such as the Swayback, or SRM-heavy builds. This reduced playstyle options and soft counters for long-range high-pinpoint builds.



To conclude:

Outright removing the Stalker chassis from the game to "solve" the 6PPC Stalker would have had much less negative impact across the game than implementing Ghost Heat did. Ponder this for a moment.

But maybe, we should all realize that the 6PPC Stalker is a red herring used to justify an opaque, poorly-communicated mechanic that punishes a number of stock builds, paid Hero Mechs, and hard/soft counters for the dominant competitive builds to which Ghost Heat did absolutely nothing.

Why not leave it in game and just simply not allow it to fit PPC's of any kind, as I believe there was one single custom mech in the entire canon piloted by one person that did fit them, so is hardly a sound basis for allowing ppc's on stalkers in the first place.

My bad one variant that did but not until 3075, which is still outside the games timeline unless its going to magically jump 5 years each real year

Edited by Cathy, 07 October 2014 - 10:28 AM.

[CompproB237]

Cathy, on 07 October 2014 - 10:24 AM, said:

Chronojam, on 02 October 2014 - 06:27 PM, said:

But maybe, we should all realize that the 6PPC Stalker is a red herring used to justify an opaque, poorly-communicated mechanic that punishes a number of stock builds, paid Hero Mechs, and hard/soft counters for the dominant competitive builds to which Ghost Heat did absolutely nothing.

Why not leave it in game and just simply not allow it to fit PPC's of any kind, as I believe there was one single custom mech in the entire canon piloted by one person that did fit them, so is hardly a sound basis for allowing ppc's on stalkers in the first place.

My bad one variant that did but not until 3075, which is still outside the games timeline unless its going to magically jump 5 years each real year

I think you missed the point of the OP.

Edited by CompproB237, 07 October 2014 - 10:38 AM.

[KraftySOT]

CompproB237, on 07 October 2014 - 10:19 AM, said:

I just wanted to point out that "Shoot the CT" or 'aim for a body shot' is pretty common in most FPS games. They still have head hit boxes and they're still hittable even with the spread that every FPS has. Most FPS use a CoF for all weapons. CoD sniping notwithstanding but BF3/4 Sniping even had things factored in like Gravity, Friction from air, Weight of the round, velocity of round. I can't remember if they also took in to account Wind, Ambient Temperature and Humidity. Either way, these factors modify where the round actually goes. This mimics reality. This is one of the reasons why we're suggesting CoF or something similar to it.

I actually had an idea that blended roadbeer's, Koniving's, and a few others ideas. Those of my Organization have had lengthy (and somewhat heated) discussions on this and have come up with some interesting ideas. I'd post them once I've had more time to put the ideas to paper. I can say that, the idea we had involved Roadbeer's convergence system but having it deteriorate given certain conditions. Full speed run adds a longer time to it (as per Roadbeer's) but Jump Jetting literally reduces the convergence over time to none. The numbers would need work but the intention is so that a HGN/CTF/VTR using Jump Jets to 'Poptart' would be so that at the point in there jump where they clear their 'Mechs height they have no convergence at all (parallel weapons). Think of the computer struggling to maintain accuracy as the uneven thrust from the Jump Jets propels the walking death machine into the air.

Don't get me wrong here, the weapons would still fire where they're pointing (not your crosshair) but you'd now have to use more skill to aim and fire those weapons. In the end, you'd get the 'poptarts' only able to fire one, maybe two weapon system locations at a time before falling too far. Even then, these shots would not be in the same place (but they would be where you were aiming). This would not only spread the damage but also require considerable skill to even pull off. Granted, the player would need a graphic to give them an idea of where the shots are aiming. Think the distance markers on sniper scopes, but in a line from weapon mount locations in relation to when they're parallel. This can also have the benefit of allowing Jump Jets to return to normal.

Now here's where this convergence method gets interesting though. Notice that I said it needs to be so that convergence goes away once you're nearing a certain height? Imagine a light/medium using jump jets. Usually they're burning the jets briefly to gain mobility or hop over shorter obstacles. They're also usually medium to short range skirmishers. The convergence for lights would be much less a problem for them. They could also balance it by giving lights/mediums a quirk of some sort that allows them to jump 1.5x or 2x their height before convergence becomes parallel.


Keep in mind here that this only a small piece of the grand idea I had. I'm sure there's something that I missed or didn't think of.

I'm also aware that the logistics of this idea are probably impossible due to HSR. Most of the things to make this work the servers already get. It's basically only changing the aiming point for the player. It already takes in to account positional data, trajectory, point of impact, and rewinding the server for determining where the shots are hitting. Basically all the convergence change does is change the server authorizing the change in convergence. It then pushes the 'current convergence state' to the client. That would be another 8 bytes of data to transfer the values 00 to FF (-127 to 127 or int8_t for those uninitiated). This gives 256 values to represent 0 to 100% giving you an accuracy down to 0.4%. That is nothing. You could bump it up to 16 bytes of data to transfer the values of 0000 to FFFF (-32,768 to 32,767 or int16_t) giving you an accuracy down to ~15 ten-thousandths of a percent or ~0.0015%.

This a million times.^

Anyone I bring from the FPS world to this game, notices were the only totally accurate weapons system out there, and most of them hate it. For the same reason COD goes to great map making lengths to try and balance its insane sniper rifles, no one wants to be one shotted, even in a game where dying and respawning happens every few seconds anyways.

Waiting 2 minutes for a match, 2 minutes for it to start up and get going, a minute to get to the fight, and 1 second to die is not acceptable and it drives away players. Thousands of them.

For every guy here saying he'd quit if this game stopped with the pinpoint accuracy, theres a thousand more who would happily play this, that is deep, over any other mech or sci fi vehicle combat game, (excluding the flight sims) if we werent so glaringly amateur.

Were this:

[Tombstoner]

Eddrick, on 07 October 2014 - 09:47 AM, said:

Look at how Armored Core did before Armorred Core 4. The games have a BIG learning curve that is unfriendly to new players.

It is possible for a game to thrive and be unfriendly to new players. You just have to keep your core audiance happy.

EVE + WOT = MWO or how i see this games potential.

[Rebas Kradd]

KraftySOT, on 07 October 2014 - 10:58 AM, said:

For every guy here saying he'd quit if this game stopped with the pinpoint accuracy, theres a thousand more who would happily play this, that is deep, over any other mech or sci fi vehicle combat game, (excluding the flight sims) if we werent so glaringly amateur.

That's a pretty convenient claim for you to make.

[Punkass]

One thought about convergence. It doesn't make a whole lot of sense that mechs have instant convergence at long ranges. Go to Alpine peaks and run a meta sniper build. You can snipe mechs that are well outside your targeting range (1200+ meters) and have all your shots converge to a single point at that range. This doesn't make sense because your targeting computer is having your shots converge onto a target that it doesn't even see.

So the first change to convergence rules could be zero convergence on targets that you are not actively targeting. This does a few things: Spreads damage at long range because the shots aren't converging on mechs that are not actively targeted. It also puts sniper builds on a more even playing field with LRMs since it would require active targeting in order to get your shots to converge. Finally, it puts further emphasis on the use of light mechs to act as forward observers for long range sniper mechs and LRM carriers.

[Rebas Kradd]

Punkass, on 07 October 2014 - 12:19 PM, said:

So the first change to convergence rules could be zero convergence on targets that you are not actively targeting. This does a few things: Spreads damage at long range because the shots aren't converging on mechs that are not actively targeted. It also puts sniper builds on a more even playing field with LRMs since it would require active targeting in order to get your shots to converge. Finally, it puts further emphasis on the use of light mechs to act as forward observers for long range sniper mechs and LRM carriers.

That is...interesting. New players could simply be taught that targeting gives you better convergence. And it wouldn't stop single-weapon sniping.

[Punkass]

Rebas Kradd, on 07 October 2014 - 12:26 PM, said:

That is...interesting. New players could simply be taught that targeting gives you better convergence. And it wouldn't stop single-weapon sniping.

Well, I suppose a single weapons still wouldn't have convergence. But, a single weapon that is firing straight out is a lot easier to compensate for than multiple weapons that are firing straight out. It would raise the skill ceiling on sniping blind with a single weapon, but only ever so slightly.

[Tombstoner]

Punkass, on 07 October 2014 - 12:35 PM, said:

Well, I suppose a single weapons still wouldn't have convergence. But, a single weapon that is firing straight out is a lot easier to compensate for than multiple weapons that are firing straight out. It would raise the skill ceiling on sniping blind with a single weapon, but only ever so slightly.

People need to accept that perfect convergence combines all the weapons in the group into one weapon with no consequences.

When playing alpha strike online all mechs have just one weapon.

[Roland]

Eddrick, on 07 October 2014 - 09:47 AM, said:

Look at how Armored Core did before Armorred Core 4. The games have a BIG learning curve that is unfriendly to new players.

It is possible for a game to thrive and be unfriendly to new players. You just have to keep your core audiance happy.

As someone who played the AC series a lot, I think it must be noted that the AC games are tailored pretty directly to a JAPANESE audience.

That audience has a much different set of expectations from western PC gamers.

[That Dawg]

Xarian, on 02 October 2014 - 06:34 PM, said:

OP: it is no longer 2013

THAT is exactly what I thought, I thought someone bumped a thread from two years ago.

oh, four ppc nova, I run it.

[Lily from animove]

Redshift2k5, on 02 October 2014 - 06:41 PM, said:

Having to use more than one weapon group or to stagger fire is not a "punishment".

Nobody likes getting shot in the face by 6 PPCs, 2 pinpoint AC20s, nine medium lasers, or any of a large number of other huge boated alpha builds.

and so the solution is a mechanic preventing to fire X weapons within X time. Ghost heta prevented some builds, but still dual gauss + X lasers is still easily possible.

30 max treshold, any heatsinks get true heatsink value. Double gauss can not be fired with any other wapons. Then max you will see is a is a dakkawolf spitting UAC 5's. because everything else reaches heatcap fast, But since heat dissipation is high things will now fire more often (many wapons still work), but with lesser alphas. damage will spread over time.

Edited by Lily from animove, 14 October 2014 - 07:11 AM.

[Chronojam]

That Dawg, on 14 October 2014 - 07:05 AM, said:

THAT is exactly what I thought, I thought someone bumped a thread from two years ago.

oh, four ppc nova, I run it.

I guess it's not that simple.

[Kirkland Langue]

I still believe the answer is to give each weapon an "accuracy" statistic - then have all weapons converge to a random spot offset from perfect convergence, based upon that particular weapon's accuracy. That way it would allow players to aim for specific locations on a mech - but even if they are boating 10 medium lasers, only some of them would hit that location with the others either hitting other locations or maybe even missing.

But the issue is less urgent than:
1. Releasing CW
2. Removing bad game mechanics (blinding shots as an example, there are others)

[Chronojam]

Ghost Heat is a bad game mechanic, though.

[Krivvan]

No one even cares anymore. Most are fine with ghost heat at this point.

Not to mention that PPCs in general aren't really meta at all right now. Even if ghost heat didn't exist anymore.