1126 replies to this topic

[GreyNovember]

I've never really had a problem with pinpoint. If I got splashed by a huge alpha, good shot to the guy aiming at me.

That being said, I always wondered how torso mounted weapons converge when they don't rotate in their mounts. Same with arms that have no lower actuators.

I guess you could set it so that you set your ideal convergence range in the mechlab (i.e. I want all my guns to converge at 500m in front of me), so you know where you can reliably shoot and do pinpoint damage. Anything beyond or below that has impacted accuracy. Anything on the torso is always perfectly converged.

Arm weapons though have drag depending on what you've mounted on them. An Atlas with a small laser on his arms will converge much faster than one carrying a PPC, for example.

[Ralgas]

the only way to fix it is to take aiming out of the hands of the shooter, full stop. (Think kotor in permanent realtime but location instead of hit/miss)

As i keep saying anything else will just usher in a new lrmaggedon or god mode lights (if not both).


Edit: Not even going anywhere near the chassis unbalancing it will do with grouped hardpoints, you think a 9s is deadly now? imagine one with 3 erppc in it's LT ignoring any convergence mechanic that's not cof based.

Edited by Ralgas, 20 January 2015 - 02:49 AM.

[norus]

If the way to fix the ttk is to make the game entirely RNG I don't really want any of that. Especially if it went lore bad for hit ratio. What I imagine happening is a sudden HUGE surge of light mechs with long range weapons running around with an impenetrable RNG shield on top of how hard they are to hit already, all while firing off their weapon. Seriously can you imagine trying to hit a light at 700m away running 150kph horizontally to you with a ppc? Between travel time, lag, and horrible convergence? Back in the beta I remember using gauss rifles and it was near impossible to hit a non stationary target at range because the convergence was so bad at times if there wasn't a wall immediately behind where the target would be running.

Edit: Removing extra word

Edited by norus, 20 January 2015 - 02:48 AM.

[Alistair Winter]

Apocryph0n, on 20 January 2015 - 02:27 AM, said:

Removing convergence on anything but Arms would be cool, i guess, that's what arms are for, right? Also makes people place their most precious weaponry in the low armored arms again, which prevents "shielding" sides to a certain degree. Also you could make them converge slowly, and make the "Pinpoint" skill accelerate arm convergence.

I agree with both points here.

First, it's kind of ridiculous how superfluous arm-mounted weapons are on many mechs. The TDR-9S is a prime example, but you also see Timber Wolves and Dire Wolves mounting their main weaponry in the CT and STs, while using their arms as shields with little or no armament. This is not a new phenomenon, it goes back to the AS7-D-DC with 2LBX10, 3SRM6 and no energy weapons. We're also seeing a lot of AWS-8Qs with 4 PPCs in the STs and no arm-mounted PPCs. And the Centurion is perhaps the most classic example of people defiling a classic design and making it a zombie SRM-boat.

Second, I also like the idea of slowing down convergence for the arms. Right now, we have a useless skill and PGI has said they're replacing the convergence speed skill with weapon hitpoint skill, which makes no sense and serves only to effectively nerf crit-based weapons. Slowing down convergence on arms would be a nice way to avoid mechs like King Crab and Dire Wolf dominating a game without convergence.

I'd also like to add that several mechs in this game are effectively equipped with lower arm actuators allowing them to aim their AC20s with greater dexterity than they should. If one arm has a lower arm actuator, it seems the opposite arm counts as having one as well. Which is bullcrap.

[MeiSooHaityu]

It's a fine line to walk.

Many don't want pinpoint weapon damage, but at the same time, many people wouldn't like not hitting something they were clearly aiming at.

I personally would prefer to actually aim and hit what I was aiming at. I don't want to pull the trigger and see a shotgun laser pattern or shots clearly missing at range due to some dialed in varience.

The current frustration with pinpoint is far less than the frustration brought on by lazy aimed weapons fire.

That's just me though.

[Kmieciu]

The easiest option would be turning off convergence for torso weapons and arms that don't have lower arm actuators.

A more advanced option would be a manually set convergence, like in Warthunder.

[Joseph Mallan]

kapusta11, on 20 January 2015 - 02:37 AM, said:

Bad because...? You have something better in mind that does not involve RNG stuff like cone of fire or inability to fire where you aim for which no one will accept? And delayed convergence is basically a gauss charge in nature. Face it it's not only bad for the majority of playerbase but is also bad for business as the result.

And btw, the idea could be expanded to insted of restricting fire of multiple weapons into penalizing it with your so desired cone of fire, everyone's happy.

There is nothing better and RNG is actually a good solution. IF GPS tracking cannot be more accurate than a 5 meter radius How is a ballistic trajectory supposed to be surgical accurate? RNG is more true than most suggestions I am reading.

[kapusta11]

Joseph Mallan, on 20 January 2015 - 02:55 AM, said:

There is nothing better and RNG is actually a good solution. IF GPS tracking cannot be more accurate than a 5 meter radius How is a ballistic trajectory supposed to be surgical accurate? RNG is more true than most suggestions I am reading.

RNG with current damage to armor ratios is a pure gamble.

[CocoaJin]

I'd like to remove instant pin-point convergence. I'd like there to be some reasonable level of dispersion and reticle bloom to indicate induced targeting/aiming error due to mech movement, recoil where applicable, collisions, etc. I'd like to explore limited to no tracking/convergence for torso mounted Ballistics/SRMs and PPCs. Reduced tracking/convergence of torso mounted lasers(considering they technically can have tracking lenses, but it should be less than arms ).

Meh, I ain't skerd of hardcore...

That being said, I predict an increase in dry humping, brawl builds that will basically require the enemy t!tty bumping you in order to fire their alphas.

Edited by CocoaJin, 20 January 2015 - 03:03 AM.

[van Uber]

Alistair Winter, on 20 January 2015 - 01:49 AM, said:

Right now, two torso mounted weapons will hit the same exact spot at any range. If you removed convergence, the torso mounted weapons would fire in parallell trajectories and it would be impossible for a CPLT-K2 with gauss to hit the same Center Torso with both weapons simultaneously, for example.

This is not necessarily a binary problem. You can have a solution in between these two states.

For example: The old Spitfire had four machine-guns as armament. They converged at 300m (iirc) to give optimal effect. This range could be adjusted, but not on the fly.

You could have something similar in MWO where weapons have a fixed range for convergence (say around 2/3 of their optimal range). Maybe you could adjust the convergence on the fly, say by holding down ctrl and using the scroll-wheel, by increments of 50m. If so, put a limit on how short a range weapons can converge, so that mixing long range with short range weapons becomes a bit of a challenge.

[Joseph Mallan]

kapusta11, on 20 January 2015 - 02:57 AM, said:

RNG with current damage to armor ratios is a pure gamble.

Yes it is. Its pretty true to life when both parties are moving at an average of 60-90 KpH.

[Apocryph0n]

MeiSooHaityu, on 20 January 2015 - 02:49 AM, said:

It's a fine line to walk.

Many don't want pinpoint weapon damage, but at the same time, many people wouldn't like not hitting something they were clearly aiming at.

I personally would prefer to actually aim and hit what I was aiming at. I don't want to pull the trigger and see a shotgun laser pattern or shots clearly missing at range due to some dialed in varience.

The current frustration with pinpoint is far less than the frustration brought on by lazy aimed weapons fire.

That's just me though.

thing is: pretty much every FPS has cone of fire. (Call of Duty, Battlefield. Even ARMA 2+3 which are Simulators, not shooters.) So i don't get why it is completely out of the equation for most people here.

I personally get that it sounds like "They are taking something away from me!" then. But actually: you get more precise weapons, less precise (Guass being rather precise then, and say as example: AC's being slightly off, Lasers kind of in the middle ground and missiles, well they always were random and will stay that way )

On the other hand: We could keep infinitely accurate weapons (which is 100% non realistic, every weapon has muzzle deviation, except for lasers.) and make the convergence thing change.

Both has pros and cons. Convergence approach keeps the accuracy but limits you to snipe/be accurate with either your arm weapons or 1 torso weapon only. while the rest flies off to godknowswhere.

CoF approach would mean: you can aim, but you have to choose your weapons for the job. No more "sniping someones cockpit out at 350meters with an AC20" (imagine a 150mm howitzer being used to fight accurately in a knifefight)

It's all a question of how they would implement each approach. You can overdo both things and make them too harsh/not worth considering.

That said: I do guess I'm more a fan of the convergence for build diversity, since CoF still would mean people have shield arms and stuff.

[Alistair Winter]

van Uber, on 20 January 2015 - 03:00 AM, said:

This is not necessarily a binary problem. You can have a solution in between these two states.

For example: The old Spitfire had four machine-guns as armament. They converged at 300m (iirc) to give optimal effect. This range could be adjusted, but not on the fly.

You could have something similar in MWO where weapons have a fixed range for convergence (say around 2/3 of their optimal range). Maybe you could adjust the convergence on the fly, say by holding down ctrl and using the scroll-wheel, by increments of 50m. If so, put a limit on how short a range weapons can converge, so that mixing long range with short range weapons becomes a bit of a challenge.

I would love the ability to manually adjust convergence. I tend to be in favour of just about any suggestion that makes it harder to manage both the piloting and the shooting in MWO. More controls = more fun for me. I just dont think PGI would be up for it.

Also, if you allow the player to adjust convergence in the middle of a game, then the difference between having convergence and not having it would be very small indeed. Especially for mechs with short range in the first place.

[van Uber]

Alistair Winter, on 20 January 2015 - 03:04 AM, said:

Also, if you allow the player to adjust convergence in the middle of a game, then the difference between having convergence and not having it would be very small indeed. Especially for mechs with short range in the first place.

Depends on the increments allowed, make it 100m and it will be fine-tuning for snipers and quite binary for brawlers. Or, you could have like three standard ranges (1/3, 2/3 and 3/3 of optimal). Still, I expect such a feature to be quite the brawler buff and make it even more polarising between sniping and brawling.

Edited by van Uber, 20 January 2015 - 03:14 AM.

[CocoaJin]

Hmmm, adjustable convergence? I'm on the fence.

[ArchSight]

Lily from animove, on 20 January 2015 - 02:29 AM, said:

people want the instant high alpha PP gone.

there are many ways to achieve this.
Two that I think can do it:

make it impossible to fire X weapons within the same timeframe. this would cause people fire in sequences ---> PP gone.

make a cone of fire

other possibilities seem hardly be feasable with the client server architecture and the engine we have atm.

I would choose the first one, because then pilot aim keeps a skill parameter in the game, especially when you fight heavily downsized light mechs a cone of fire will already untouchable lighs be able to tank 2x the firepower an assault can tank. A big issue. The cone of fire in general would again favor specific mechs over others by the geometry and hitboxes they have.

so limiting the amount of possible PP weapons is probably the most faires way to go. Can be done with limiting the amount of weapons fired, or limiting heat.

Evil Laugh* There's now another person that sorta agrees with me on the forums. ...


Convergence is not the answer to PPFLD due to hit registration being harder to track when not all weapons are pointed at the cross hair and the fact that some gamer's don't like it when what they aim at, they don't hit it at all dead center where the cross hairs are.

There's other ways to balance the game.
Changing weapons themselves so they easily spread damage on a moving target like what they did to the Clan AC's where they fire their damage in multiple bullets or Clan ERPPC splash damage to spread the damage to other area's.

In fact, the concept of forcing multiple shots being taken at a moving target works really well at spreading a players damage done to a mech. Heat scale doesn't do this however, it only punishes a few set amount of weapon groups from being fired together with either a small heat increase or large one depending on the weapon group used. It doesn't affect all groupings of weapons that still can shoot a large amount of damage into a single location.

I say the alpha strike's should be damage based on receiving a heat scale penalty instead of weapon grouped based. So if a mech fires more damage than the highest damaging IS weapon in the game (the AC 20) at the same time it would receive a heat penalty based on how much more damage it was going to do. Depending on how much heat the mech gains after the alpha strike would determine how many alpha strikes a player can do in their mech. The heat cost should be high to make alpha strike's risky and rare to do in a short amount of time in combat. That would enforce player's to shoot their weapons one after another instead of all at the same time every moment they can. Having player's chain fire would increase the chances of spreading the damage on a moving target and making the alpha strike into that special one time use for a skilled shot moment would be exciting for player's talk about in their stories of MWO combat. (Dont stand still or your going to die. [How many times do I have to say this])

Forcing multiple shots taken would likely not work well with current quirk amounts because quirks already make it possible to shoot many times to catch up with high alpha PPFLD builds. Giving a larger DPS advantage to those mechs if what I said was implemented. The quirks will have to be nerfed along with nerfing alpha strikes.

Now just need to put it on a test server to see if most of the community likes it enough to tolerate it.

When the alpha strike finally falls to the DPS warriors everything in MWO will change in a cataclysmic fury of forum warrior's now saying the dragon is OP pls nerf it naw!

[Karl Streiger]

Everything is better as the current system for sure - and if you really lead the crosshair - you may have realize that convergence is focused on the spot below the crosshair.
Thats why shooting from above is more precisely. (Convergence on the bottom - some meters behind the target) while its pure luck to hit a target when you shoot uphill or on the same level - when the "convergence" point is hundreds of meters behind.

You will have the same effect - when the "convergence" slow time is introduced but with more "negative" effect.
I can remember before engine rating was scaled down - i had a Catapult C1 armed with 2 ERPPCs moving almost 100kph - it was almost impossible to hit a target at 10 o'clock of your movement direction.

So yes - instant or automatically convergence have to go:
two ways to go - and to make it more simple:
Convergence is fixed on the effective range for each weapon. Aim for the 10 with an AC 5- and you hit the 10 only at 540m. Your shot fired from the left arm - may hit the short left 9 when closing or wide right 8 when increasing the distance.

Or you can "take" the range of your target ("lock on") - automatically - with small delay -or a nother button that uses the range under your crosshair.

The last will make "shooting" - and lead fire more simple and controllable.

[Lily from animove]

Karl Streiger, on 20 January 2015 - 03:20 AM, said:

Everything is better as the current system for sure - and if you really lead the crosshair - you may have realize that convergence is focused on the spot below the crosshair.
Thats why shooting from above is more precisely. (Convergence on the bottom - some meters behind the target) while its pure luck to hit a target when you shoot uphill or on the same level - when the "convergence" point is hundreds of meters behind.

You will have the same effect - when the "convergence" slow time is introduced but with more "negative" effect.
I can remember before engine rating was scaled down - i had a Catapult C1 armed with 2 ERPPCs moving almost 100kph - it was almost impossible to hit a target at 10 o'clock of your movement direction.

So yes - instant or automatically convergence have to go:
two ways to go - and to make it more simple:
Convergence is fixed on the effective range for each weapon. Aim for the 10 with an AC 5- and you hit the 10 only at 540m. Your shot fired from the left arm - may hit the short left 9 when closing or wide right 8 when increasing the distance.

Or you can "take" the range of your target ("lock on") - automatically - with small delay -or a nother button that uses the range under your crosshair.

The last will make "shooting" - and lead fire more simple and controllable.

the last will heaviyl interefe with latency and co and make ecm even more of a powerhouse.

the problem is nearly any of the systems have their edgecases where it is too extreme, not working or exploitable. and so together with this system other changes would have to be done too.

[EvilCow]

I really would like to try a different convergence scheme along the lines of:

arms: Automatic convergence but the 2 arms can be out of sync if the mech is running or under fire. The convergence "skill" would reduce the automatic convergence time.

torso/head: Fixed convergence set in mechlab.

Nothing would be random of course. TTK would go up and we could get rid of ghost heat and other "less than ideal" mechanics.

[Mudhutwarrior]

Okay I get the argument but as of yet no one has answered will it make the coding easier or harder. So many issues with hit registration my concern is will it make it better or worse.

Answer that question and I will decide then.