80 replies to this topic

[VanillaG]

I am splitting this off from one of the LRM salt threads to get some feedback. One of the things that I realized in the LRM thread is that LRMs and Streaks are the only weapons that require locks to work which makes direct fire weapons more powerful because of pinpoint convergence.

I would be fine with pinpoint convergence if you had to have a target lock to get it, but being able to poke out of cover and blast an alpha with pinpoint accuracy has nothing to do with skill. Almost every FPS has three different types of weapons spreads; running is largest, walking is in the middle, and zoomed in/sighted is the smallest. All MWO has is free zoomed in pinpoint aiming while running which is somewhat broken. You should have to work/build to get that pinpoint convergence, not just assume that get it for free. I don't think that PGI can implement delayed convergence for technical reasons so I would be fine with reticle shake like when a mech jumps.

If I was king for a day, there would always be some reticle shake based a target lock, low heat, and lower speed. With a lock, low heat, and low speed you would get a perfectly still reticle when you aim at the targeted mech. Without those three conditions you would still hit a mech but your damage could be spread to different components or even miss base on the severity of the reticle shake. You could use the following to modify the shake so you are not totally subjecting yourself RNGesus:

• Heat - anything above 70% heat increases reticle shake
  
• Speed - anything above 66% max speed increases reticle shake
  
• Targeting Computers - decrease reticle shake and decrease lock time
  
• Target Info Gathering/BAP/cAP/Command Console - decrease lock time

Basically you require ALL weapons to have a lock to get max performance. Requiring locks gives more emphasis to the Infowar component and makes sharing locks important for all weapons, not just missiles. You can snap shoot without a lock but you will be spreading your damage and potentially missing. LRMs will still have indirect fire but they still spread their damage so they are not super overpowered. This would push MWO back towards more of a simulator, which I am not sure PGI wants to do.


EDIT:
I think that I have conflated accuracy with convergence. My solution would be to just implement the reticle shake and leave in the existing convergence mechanism in place. Having reticle shake is probably as far the engine can go in the effort to reduce the unlocked pinpoint accuracy advantage that PPFLD weapons currently have.

Edited by VanillaG, 08 June 2017 - 09:15 AM.

[Khobai]

PGI tried something like that already and they couldnt get it to work so they abandoned it. Its unlikely theyll ever try again.

The easiest and next best thing would be not allowing weapons to do any damage beyond optimum range without a target lock.

[Greyhart]

Oh this again. I agree the pin point accuracy and perfect convergence is a problem when the mech is divided into components.

I also cannot think of a single shooter that has the pin point accuracy when running. Hell a lot of then have a cone of fire when standing still and only certain weapons have pin point accuracy like sniper rifles.

But all this will be lost in the flood of people tell you that pin point accuracy and perfect instant convergence means they have more Skillz and they don't want any randomness at all. Despite the fact that it makes game play less fun from a casual point of view.

If we're lucky we will get people discussing the technical specification of current battlefield weapons (I like those because they're interesting). But ultimately they aren't about games so don't really count as reasons.

But no one will be able to provide another game that does pin point accuracy and perfect convergence.

Those advocating for the pin point accuracy and perfect convergence will not ask themselves why it only appears on MWO and not other games. They will obsess that their mad skillz will be undermined by having to deal with slight randomness that everyone else has to deal with. I am assuming that these are also the types of people who buy the most expensive stuff because it has a marginally better response time and they think that will make them better, because they can react to things on a microsecond basis. These are also the same people that will min max to perfection because they need that 0.01% higher damage rate as it makes all the difference.

I think there is a thread where a Dev goes into how the hit reg works and the issues that they have.

Edited by Greyhart, 08 June 2017 - 07:53 AM.

[Andi Nagasia]

they tried something like that in the Infowar PTS,
it was a Laser Lock Mechanic, where if you didnt have the Mech Targeted your lasers only had 60% range,
it didnt test well, and back lash was so bad that they dropped all of Infowar even though most of it was good,

[VanillaG]

Khobai, on 08 June 2017 - 07:52 AM, said:

PGI tried something like that already and they couldnt get it to work so they abandoned it. Its unlikely theyll ever try again.

The easiest and next best thing would be not allowing weapons to do any damage beyond optimum range without a target lock.

I am not trying to implement delayed convergence or cone of fire, just introducing the random reticle shake that you currently get when jumping. The main difference is that you would always have a reticle shake unless your crosshairs were over the targeted mech. I think that the reticle shake combined with your no damage past optimum would go a long way to making this more of a simulator than a FPS.

[El Bandito]

Anything to add immersion. Lock based accuracy is up in that valley. Same deal with progressive heat penalty, and knockbacks.

Edited by El Bandito, 08 June 2017 - 08:11 AM.

[Pariah Devalis]

Andi Nagasia, on 08 June 2017 - 08:09 AM, said:

they tried something like that in the Infowar PTS,
it was a Laser Lock Mechanic, where if you didnt have the Mech Targeted your lasers only had 60% range,
it didnt test well, and back lash was so bad that they dropped all of Infowar even though most of it was good,

To be fair, what they implemented was backasswards ********. Just like how they got detection mixed up. Instead of directly impacting the weapons themselves for range and damage, they should have just made weapons not converge at all without a target lock. Make that modified by a set amount for having ECM equipped as it reduces your ability to be detected. Would have been a heck of a lot simpler, IMO.

Also, that detection range thing still confuses me what they were thinking. Radar cross section is an actual thing. The bigger an object, the easier it is to detect. They wanted to base detection range based on the one doing the detecting, as opposed to basing detection range based on how detectable you are as a function of size/tonnage. Seemed backwards to me then, as it does now.

Edited by Pariah Devalis, 08 June 2017 - 08:20 AM.

[process]

I don't know why they can't just, by default, set the target-under-reticle range to something near infinite or maybe like 10,000 meters. Conceptually, all your guns would be parallel to one another. This convergence range would remain fixed until a target is locked, at which point the convergence range follows the target.

We already have this in practice -- see wide mechs shooting at close range.

I'm less fond of penalties to movement and heat, I feel like that just encourages camping, but I'm not opposed.

Edited by process, 08 June 2017 - 08:19 AM.

[Dread Render]

VanillaG, on 08 June 2017 - 07:39 AM, said:

, but being able to poke out of cover and blast an alpha with pinpoint accuracy has nothing to do with skill

I wish I had this... I run about 60% laser boats, some mixed some all the same.
Whenever I fire and do not have a lock I notice the beams fire off all over the place.
This is why I constantly Hammer the R key all game long. And I don't even think about it.
its so burnt in to my brain is a just a reflex now,

Now True, Sometimes it does seem to converge. But perhaps that is when I had the target marked before or something,
anyway
i am TOTALLY NOT getting convergence like you are describing.
Guess your just a lucky guy... or i'm not.

[Pariah Devalis]

Dread Render, on 08 June 2017 - 08:21 AM, said:

I wish I had this... I run about 60% laser boats, some mixed some all the same.
Whenever I fire and do not have a lock I notice the beams fire off all over the place.
This is why I constantly Hammer the R key all game long. And I don't even think about it.
its so burnt in to my brain is a just a reflex now,

Now True, Sometimes it does seem to converge. But perhaps that is when I had the target marked before or something,
anyway
i am TOTALLY NOT getting convergence like you are describing.
Guess your just a lucky guy... or i'm not.

The way the game is currently designed, if you have a reticule on an object, your weapons will all autoconverge to that specific range point. R key does not currently impact your targeting computer's convergence pattern. In other words, if your weapons are not converging on the same point, either your reticule is not actually on the target, or you are firing weapons with various travel times at the same moment, and their impact times will stagger, thereby not necessarily converging at the same point.

[WrathOfDeadguy]

Get rid of convergence and, while you do increase TTK, you'll also render huge numbers of 'Mechs useless because the new meta will favor only those with a significant number of closely-grouped hardpoints.

Think about it: a 'Mech like the HBK-4P does not need to wait for a lock, because it has a cluster of six lasers that will already hit the same component when fired... and they're high-mounted, to boot. The Crab, on the other hand, will become instantly worthless, because it has no clusters greater than two, and it can only carry larger weapons on low, widely spaced mounts- and therefore always has to wait for convergence to fire. JVN-10P would be a great missile light, with two clusters of 3M close together on the chest, whereas the Oxide would be worse than useless with one cluster of two and two single launchers way out on the arms. The MAD-3R with its tight 3B group would be a great 'Mech for dakka, while the Jagermech, which was built around dakka, would be terrible at it. The Battlemaster and Banshee would remain relatively powerful, but let's not even talk about the poor Awesome.

So... right away, to avoid obsoleting huge numbers of 'Mechs (many of which are already only marginally viable), PGI has to implement a new set of TIG, lock, and convergence quirks to make up the difference between the new top performers and the bottom-of-the-barrel 'Mechs that have few or no clustered weapons.

Remind me again how this solves more problems than it causes? Reticle shake (but preferably CoF, with a dynamic circle instead of an annoying wobbly crosshair) at high heat levels or throttle settings I can totally get behind. Eliminating convergence without target lock... now that's just asking for trouble, without even taking into consideration the oceans of salt it would produce. We don't need any more factors complicating our already convoluted mess of a balance system.

[VanillaG]

WrathOfDeadguy, on 08 June 2017 - 08:55 AM, said:

Remind me again how this solves more problems than it causes? Reticle shake (but preferably CoF, with a dynamic circle instead of an annoying wobbly crosshair) at high heat levels or throttle settings I can totally get behind. Eliminating convergence without target lock... now that's just asking for trouble, without even taking into consideration the oceans of salt it would produce. We don't need any more factors complicating our already convoluted mess of a balance system.

This is partly my problem in that I am conflating accuracy with convergence. I would leave the existing convergence mechanic in place and just use the reticle shake to reduce the pinpoint accuracy without a weapons lock. I am not sure if reticle shake moves the reticle off the mech if it will effect the convergence but that is something that could be worked out later.

[Champion of Khorne Lord of Blood]

Khobai, on 08 June 2017 - 07:52 AM, said:

The easiest and next best thing would be not allowing weapons to do any damage beyond optimum range without a target lock.

That's almost exactly what PGI did, but they were even softer on it and just made it not do as much damage without a target lock and reduced the optimal range just a little in the case of not being locked to further show that damage is reduced if you aren't targeting them.

For whatever reason the community, after begging for info warfare and being mad at people who never lock targets, went and started a huge dumpster fire shouting about the new buzz words "Ghost Range" and "Ghost Damage" and shouting at PGI to undo it and never try this again.


Sometimes this community is one of the worst things holding the game back.

[CK16]

I think convergence definitely needs to be looked...one thing this game has over the past renditions and many other games is that pinpoint aim, and no, being able to put a dot on a screen on your target is not a high skillz, sorry not sorry. What takes more skill, target shooting with a zero'ed in scope or stock iron sights (assuming zeroed in as well.) . There needs to be some sort of convergence time/lock or a Cone of fire....I prefer the first option and this also opens the doors for targeting computers to have a use. But also this would help with time to kill. Cause well snap shots just as IRL become luck if you hit, but a well aimed shot takes time but ends up if good hitting your target where ypu want

[Pariah Devalis]

Dakota1000, on 08 June 2017 - 09:17 AM, said:

That's almost exactly what PGI did, but they were even softer on it and just made it not do as much damage without a target lock and reduced the optimal range just a little in the case of not being locked to further show that damage is reduced if you aren't targeting them.

For whatever reason the community, after begging for info warfare and being mad at people who never lock targets, went and started a huge dumpster fire shouting about the new buzz words "Ghost Range" and "Ghost Damage" and shouting at PGI to undo it and never try this again.


Sometimes this community is one of the worst things holding the game back.

Maybe it's because the approach they took was probably the worst way to handle it, and the community called them out on it?

Ghost range made very little sense. Ghost damage, just as little. It also added an extra level of complexity that makes the game's already difficult learning curve harder for brand new players.

There were better ways to handle "infotech," and they chose not to go those routes. Their entire methodology seemed outright backwards. See: radar cross section vs what they actually chose to do.

[Deathlike]

This would simply be a massive buff to brawling, and would just invalidate the purpose of a lot of long range weaponry...

If you like your SRMs and Streaks, you're going to love it.

If you like your Gauss (charge), PPC, and ERPPCs (particularly IS), and ERLL you're going to hate yourself.

Edited by Deathlike, 08 June 2017 - 06:34 PM.

[Khobai]

Quote

they tried something like that in the Infowar PTS,
it was a Laser Lock Mechanic, where if you didnt have the Mech Targeted your lasers only had 60% range,
it didnt test well, and back lash was so bad that they dropped all of Infowar even though most of it was good,

Except the idea wasnt tried properly. It was set up for failure by PGI's incompetence.

ECM still granted stealth. Which made ECM anti-laser since it blocked sensor locks. Which is why people hated it so much.

ECM stealth needed to be removed to make the concept viable.

Quote

That's almost exactly what PGI did, but they were even softer on it and just made it not do as much damage without a target lock and reduced the optimal range just a little in the case of not being locked to further show that damage is reduced if you aren't targeting them.

For whatever reason the community, after begging for info warfare and being mad at people who never lock targets, went and started a huge dumpster fire shouting about the new buzz words "Ghost Range" and "Ghost Damage" and shouting at PGI to undo it and never try this again.

Sometimes this community is one of the worst things holding the game back.

Because they didnt get rid of friggin ECM stealth. So ECM was reducing laser damage. That stupidity is the main reason why the community was against it.

They need to get rid of ECM stealth then reboot the idea of damage being reduced without sensor locks. Having it directly affect damage is the only way to make sensors and holding sensor locks actually matter.

Quote

Ghost range made very little sense. Ghost damage, just as little. It also added an extra level of complexity that makes the game's already difficult learning curve harder for brand new players.

Weapons should do less damage at longer ranges. Lasers defract in the atmosphere. And projectiles shed momentum due to air resistance. And it makes sense having a sensor lock could account for some of those variables when firing the weapon.

Edited by Khobai, 08 June 2017 - 09:45 AM.

[kapusta11]

Complaining about pinpoint accuracy when average player's hit rate is 50%.

Edited by kapusta11, 08 June 2017 - 09:37 AM.

[Scyther]

I tend to favor game mechanics that are already in place and working, since it is somewhat harder to screw those up. Convergence, cone of fire, reticle shake, etc, all boil down to "your damage doesn't all always go to the exact point you are aiming at".

PGI / MWO already has several 'spread damage' mechanics. Just use one of those (eg., C-ERPPC damage to sections adjacent to the aim point). If you have a radar lock and are fairly stable (moving at less than 50% max speed, say), all your damage gets directed at aim point (for weapons that work that way, eg. not SRMs).

If you have radar lock and are moving fast, 66% of damage goes to target point, 34% goes to an adjacent section.

If you are moving fast and do not have radar lock, 50% of damage goes to aim point, two packets of 25% each go to random adjacent locations (including a potential miss for one packet).

Somewhere along the way, firing past optimal range should probably be figured in as well. Although since the damage is already reduced somewhat, and aim is bit more iffy at range, it wouldn't need to be too high.

This will encourage radar locks, somewhat slow down the pace of engagement, and increase TTK.

Lights would probably be impacted the most, although having Target Retention would help. Lights should probably get to fire as 'stable' platforms up to about 75% speed, since they would hopefully be designed with that in mind.

[Mystere]

VanillaG, on 08 June 2017 - 08:09 AM, said:

I am not trying to implement delayed convergence or cone of fire, just introducing the random reticle shake that you currently get when jumping. The main difference is that you would always have a reticle shake unless your crosshairs were over the targeted mech. I think that the reticle shake combined with your no damage past optimum would go a long way to making this more of a simulator than a FPS.

Frankly, I think reticle shake was/is a silly solution, and a random one is even worse. Reticle bloom practically does the same thing and is much more visually appealing. We are talking about HUDs and not iron sights after all.