102 replies to this topic

[Troutmonkey]

oldradagast, on 09 February 2016 - 04:00 PM, said:

Nothing shows a person's deep understanding of a topic more than watching them start their argument with groundless, sweeping accusations, such as what you've done. If you bothered to read the posts, plenty of elite players support removing instant, perfect convergence.

Convergence is not skill. If the only thing separating you from the unwashed masses of the "steering-wheel underhive" is that you know about boating lasers and clicking on a mech with them for instant damage, you were never skilled to begin with.

Instant, perfect convergence has been at the heart of nearly every balance problem in the game. Everything from ghost heat to jump-jet nerfs to Gauss nerfs, to PPC nerfs has been forced because of the catastrophic and unbalancing effect of being able to put a stupid amount of damage on a single point on a mech with no real effort - or skill -whatsoever. Hell, even the need to rescale mechs and evaluate them so heavily by their hitboxes has been forced on us because of how easy it is to carve off components at long ranges with little to no real effort.

What, exactly, is unreasonable about proposing a solution to fix one of the most obvious and serious problems in the game? Note that a cone of fire not only fixes this problem, but is also within the game engine's ability to handle AND makes the game behave somewhat more like Lore, where mechs were not perfectly accurate. I don't expect an answer since people who launch flame posts never have solutions, just whining.

QFT

[wanderer]

Quote

Except if Lights have a fast acquisition time and are slower to be acquired that fixes a lot. Also gives good balance for lights vs assaults

Which, as the actual good part of the pile of garbage that was "laser ghost damage on Test" , was the case. Lights had the most agile and far-reaching sensor suites.

Honestly, the info part was interesting and had potential, it was pairing it with Paul's ideas on laser damage that flushed the whole concept in the toilet.

[MischiefSC]

wanderer, on 09 February 2016 - 04:29 PM, said:

Which, as the actual good part of the pile of garbage that was "laser ghost damage on Test" , was the case. Lights had the most agile and far-reaching sensor suites.

Honestly, the info part was interesting and had potential, it was pairing it with Paul's ideas on laser damage that flushed the whole concept in the toilet.

Scale your expectations. The laser fall-out wasn't ideal but it was based on an excellent concept - an otherwise solid IW concept with role warfare tied to direct weapon performance. Sure, convergence is better but the underlying idea was solid. Best I've seen or of pgi even.

My fear is that the backlash got them to abandon the idea completely instead if just change the execution a bit.

[Troutmonkey]

wanderer, on 09 February 2016 - 04:29 PM, said:

Honestly, the info part was interesting and had potential, it was pairing it with Paul's ideas on laser damage that flushed the whole concept in the toilet.

Yeah I liked the idea of the Info Warfare part. It didn't need the ghost damage. All it should have been was lights get the longest range sensors and are the slowest to lock on to, while assault get shorter range sensor and could be locked quicker and just scale between them based on weight. It would have mechs in the lower tonnage categories a small edge over the heavier ones. There doesn't need to be a huge gap between each end of the spectrum, -+ 300m between the longest and shortest range would have been enough. Say 700m range and .5 second lock ons for 100t assaults, 1000m range for 20 and 2 second lock ons tonne lights. Wouldn't go any wider than that.

Adding different ranges for every different mech and all this other fluff about target sharing was too over the top for a first pass needlessly complicated it

Edited by Troutmonkey, 09 February 2016 - 04:46 PM.

[Solis Obscuri]

I find the biggest problem with the instant pinpoint convergence is that it really trivializes player accuracy to a large extent. Placing shots well, and repeatedly placing shots to the same component (particularly with dissimilar weapon types) becomes inherently less favorable than boating a lot of the same weapon and pounding the alpha-strike key. At 35-50 concentrated damage per shot, you don't really need consistent placement unless your target is titanically armored (which usually coincides with having titanically big hitboxes, anyway), anything that connects with even a graze has a good chance of crippling or destroying the target unless it's taken any damage already. The absence of any real "aiming" mechanic (breath hold, weapon sway, stabilizing, etc.) further diminishes any distinction between a good "called shot" and just plain "slop".

Which is all the frustrating because it's made mech selection and map tactics all fairly stagnant and kept optimal loadout choices unimaginative for a period of over three years, while at the same time making the game play less like an armored combat vehicle simulation and more like a stock FPS but with really big, slow, easy to hit targets.

MWO has always had potential to be a lot more than it is now, but these kinds of unresolved issues just keep holding it back.

[Kristian Radoulov]

oldradagast, on 09 February 2016 - 04:00 PM, said:

Convergence is not skill. If the only thing separating you from the unwashed masses of the "steering-wheel underhive" is that you know about boating lasers and clicking on a mech with them for instant damage, you were never skilled to begin with.

???????????????????????? lol

[Dino Might]

Mead, on 09 February 2016 - 10:58 AM, said:

I am honestly curious just exactly what part of that you think is a personal attack...

I'm curious as well. Challenging someone's unsupported claims is a personal attack nowadays. I think I violated a safe space or something.

[Koniving]

Mister Blastman, on 09 February 2016 - 10:40 AM, said:

Precisely to the variable hitboxes. Give Gargles and Summies some toughening™ if it is needed.

The legs are a good point. Legs could be divided in stripes...

Lights get one leg armor value (no stripes)
Mediums get two stripes
Heavies and assaults get four stripes (front, back, left, right)

Or hell, make mediums/heavies/assaults get the four stripes and leave lights at one or maybe give them left/right at most. Some folks have a hard enough time as it is with light 'mechs.

Or just triple leg armor values and leave them as is now. Same with arms if needed. It's save on computational and implementation costs.

I'm glad I'm not the only one who has thought of such an idea. Your Plan B has been thought of by me as of early 2015, although admittedly in a somewhat different form and with a different intended purpose in which something very similar to your idea was a bi-product of my attempt to remove "dice rolls" from through-armor criticals and 'skill-based' crit targeting.

Specifically, each mech would have crit slots assigned to certain parts of the body with both internal and external 'slots'... mind you this is for a Battletech Simulator pipe dream I'm doing 3D models and animations for rather than MWO. For example the Hunchback.
See the drum?

The ammunition by default is on the left torso, where the drum is.

On the 4SP in Battletech, the drum is on the middle of the back. Where the ammunition is! The HBK IIC, the drum is on the middle of the back but extrudes over both shoulders. The ammunition for the HBK IIC is in either side torso in a drum that overlaps all three torsos.

The Hunchback is a squat, thick medium mech of 50 tons that is not nearly as large as its equally armored, thinner, taller and lankier 50 ton mechs and as such... hasn't got the internal space to store its ammunition, thus the drum on the outside.

Reigning it into the topic, this drum would contain the HBK's ammunition as the designated "Ammo bin" slots. As such targeting this drum would work towards doing damage to those 'slots'. Now take the Hollander's Gauss Rifle, unlike most 35 ton light mechs the Hollander has a big rifle with a barrel nearly as long as its body is tall. Some slots interior, some exterior. Now if you switched that with an MG you may as well consider that tiny thing internally stored. Now expand. Look at the Atlas. Canonically there are only 3 variants with a full 20 missile rack, despite how almost every model carries an LRM 20 or LRM 15. Why is this? Because the Atlas didn't have the internal space, and so the LRM-20 is a 5 tube external hip-mounted launcher. On the Atlas C, K, and S2 it is internally mounted in the torso (on the K and C, it is split across left and right torso visually).




As you can picture this Cattlemaster-mounted LRM-20 would definitely be on the 'external' slots.

Now, imagine this expanding to everything from heatsinks to actuators. This would require a hitbox system that is far less generic, and so as a bi-product of this effort I found myself conceptualizing a far different system. This would take your 32 leg armor and divide it across several sections and accommodate for "Design Quirks." For example a mech with Exposed Actuators would find itself with less protection on the actuators themselves; this in turn however would free up armor points to go on other parts of the legs allowing them to be stronger. A mech with Protected Actuators would actually have more armor dedicated to their actuators to protect them, but as such other sections of the legs would have reduced armor potential.

The plan would automatically assign the generic "Leg armor 32 units" to each part in percentages. When accompanied with Koniving's Battletech lore-based weapon variants, there is no issue about a section having armor with decimals as the weapons themselves would have decimals. (i.e. the Austen-Armstrong Industries J11 AC/5 carried by the SHD-2K is a 90mm cannon that holds 8 shots per cassette and 20 cassettes per ammo bin [ton]. Each shot delivers 0.625 damage).

Leg sections included each actuator (yes, foot too), as well as front-thigh, rear-thigh, shin and calf. Body sections included separated plates akin to your Plan B but my design would assign them uniquely per chassis. The head would include "Visor," "Hatch," and several "Shielding" sections which can include parts of the cockpit (life support, sensors, etc). The arms would include several sections as well as their actuators.

[brroleg]

Mister Blastman, on 09 February 2016 - 09:36 AM, said:

Quote

ARMS ONLY

I add that ONLY arms with lower arm actuator (which are able to sway in X coordinate)

[Ramseti]

Mister Blastman, on 09 February 2016 - 10:05 AM, said:

Cone of fire reduces skill, though.

I know others have already commented on this, but this mindset just boggles my mind. If I know that firing 2 lasers will be pinpoint, 3 introduces some inaccuracy, and 4 lasers is just a disco party, then if I want to hit the same location with 4 shots I have to hit it twice (2 and 2). You, meanwhile, only have to hit it once using your 4 laser alpha. So if I can hit the same spot twice, how is my skill lower? Besides, how is accounting for movement, heat, distance, etc. (you know, like if this were a military-themed game) mean less skill? "You have to account for more, therefore you suck!"

[Mister Blastman]

Ramseti, on 18 February 2016 - 07:34 PM, said:

I know others have already commented on this, but this mindset just boggles my mind. If I know that firing 2 lasers will be pinpoint, 3 introduces some inaccuracy, and 4 lasers is just a disco party, then if I want to hit the same location with 4 shots I have to hit it twice (2 and 2). You, meanwhile, only have to hit it once using your 4 laser alpha. So if I can hit the same spot twice, how is my skill lower? Besides, how is accounting for movement, heat, distance, etc. (you know, like if this were a military-themed game) mean less skill? "You have to account for more, therefore you suck!"

Because cone of fire is random. You can't hit the same spot consistently with it. Thus it takes away skill. It leaves it to dice rolls.

The revised armor system will increase ttk and increase the skill required, simultaneously. If your skill is high enough, ttk will be the same or only slightly increased.

[Ramseti]

Mister Blastman, on 18 February 2016 - 09:21 PM, said:

Because cone of fire is random. You can't hit the same spot consistently with it. Thus it takes away skill. It leaves it to dice rolls.

The cylinder of fire implementation that went on for tons of pages doesn't, yet people still argue the same thing. No matter what, it doesn't reduce skill. That's like saying you're better at shooting guns on a rifle range because your laser pointer always hits the bulls-eye while my weapon has to deal with tons of variables.

[L A V A]

oldradagast, on 09 February 2016 - 04:00 PM, said:

Convergence is not skill.

Instant, perfect convergence has been at the heart of nearly every balance problem in the game.

Thank you.

Most accurate and pertinent observations made so far in any thread concerning this subject.

[oldradagast]

Mister Blastman, on 18 February 2016 - 09:21 PM, said:

Because cone of fire is random. You can't hit the same spot consistently with it. Thus it takes away skill. It leaves it to dice rolls.

The revised armor system will increase ttk and increase the skill required, simultaneously. If your skill is high enough, ttk will be the same or only slightly increased.

Disagree with the claim that unpredictability removes skill requirements. Imagine if real-world marksman fired guns - even more than one at a time - with perfect, pinpoint convergence. Suddenly, good aim and shooting well would no longer be an impressive sight and marksmenship would no longer be a skill. That is exactly what we have going on this game right now; also, note that nobody is proposing a huge cone of fire or rolling a 20-sided die to see if you hit.

As for increasing TTK, yes, that's part of the point, but the bigger issue is trying reduce the crutch of boating insanely powerful long-range alphas and just carving mechs up with ease at the click of a button. That is a zero-skill game - "huhuhuh, derp, mah lasers blew off your torso because I have great skills thanks to hitscan weapons and perfect instant convergence" - and makes the game less fun for players, particularly new ones, who, rightfully, think that a mech with multiple hitboxes should, you know, actually take damage across them vs. being cored out all the time.

Convergence is NOT skill.

A good player with good aim still lands a lot more damage on an enemy mech than a bad one with poor aim even with a cone of fire - and that doesn't even count all the skill in the game other than aiming. If the only thing separating a "skilled" player from a "steering-wheel using bad" is that his mech has more lasers or more Gauss rifles, than he was never skilled to begin with.

Edited by oldradagast, 19 February 2016 - 04:13 AM.

[Greyhart]

How about cone of fire if you are not locked on target and pinpoint if you are.

Because people need to learn the R key is over powered

[H I A S]

When it is so easy to blew up torsos, why dont die all robots with 80% hitpoints?

[CDLord HHGD]

I don't like instant convergence, just shouldn't work that way. I do think there should be time to converge (I am thinking a modified version of what we see on WoWS). This way, convergence skill and possibly quirks can be real things.

Cone of Fire is not ideal either, but in my Heat Scale post, I do offer a reticle shake based on heat levels that can have a similar effect.

Delayed convergence and a heat based reticle shake can have the desired effects.

[Mister Blastman]

oldradagast, on 19 February 2016 - 04:11 AM, said:

Disagree with the claim that unpredictability removes skill requirements. Imagine if real-world marksman fired guns - even more than one at a time - with perfect, pinpoint convergence. Suddenly, good aim and shooting well would no longer be an impressive sight and marksmenship would no longer be a skill. That is exactly what we have going on this game right now; also, note that nobody is proposing a huge cone of fire or rolling a 20-sided die to see if you hit.

As for increasing TTK, yes, that's part of the point, but the bigger issue is trying reduce the crutch of boating insanely powerful long-range alphas and just carving mechs up with ease at the click of a button. That is a zero-skill game - "huhuhuh, derp, mah lasers blew off your torso because I have great skills thanks to hitscan weapons and perfect instant convergence" - and makes the game less fun for players, particularly new ones, who, rightfully, think that a mech with multiple hitboxes should, you know, actually take damage across them vs. being cored out all the time.

Convergence is NOT skill.

A good player with good aim still lands a lot more damage on an enemy mech than a bad one with poor aim even with a cone of fire - and that doesn't even count all the skill in the game other than aiming. If the only thing separating a "skilled" player from a "steering-wheel using bad" is that his mech has more lasers or more Gauss rifles, than he was never skilled to begin with.

I've done a lot of shooting, fyi. I'm a proud gun and carry owner, so I have a bit of experience here. Yes, we have this term called groupings and MOA and even when bench shooting there's still a very minor spread of fire that depends on the shooter, their trigger pull, wind and any bench devices used such as a bipod or more.

But the weapons in a Battlemech are fixed. What does influence spread is terrain differences and 'mech movement--we have that now. The problem is... the torso sections are SO BIG that we don't get much spread from it.

Now... if we change these torso sections into more sections... you'll start seeing more of that natural spread from movement.

What we both want is the same thing--to spread damage out. One does it with a cone of fire which will most likely irritate players and the other ups the hitboxes making it harder to hit that same spot over and over again. They both solve the problem where one is less random by algorithm.

My natural hitbox system more accurately models what armor is and it does--plus accounts for small 'mechs and large 'mechs to prevent the smalls from becoming tank monsters.

[Ramseti]

Mister Blastman, on 19 February 2016 - 05:58 AM, said:

But the weapons in a Battlemech are fixed.

They're fixed, but that doesn't automagically negate effects of recoil, computer inaccuracies, incoming fire, damage sustained (i.e., laser lens cracked), etc. I mean, the coaxial and pintle mounts in tanks are fixed, but they still suffer decreased accuracy when the main gun fires.

Quote

My natural hitbox system more accurately models what armor is and it does--plus accounts for small 'mechs and large 'mechs to prevent the smalls from becoming tank monsters.

I think this would be effective as well, certainly leaps and bounds above anything in-game now.

[Luminis]

What gets me with these threads isn't how skilled the person posting them is. What gets me is that they're pushing for a hugely complex fix to an easily fixed problem.

Okay, so, problem in question (at least from what I can tell):

TTK is too low because point-and-click, insta-hit weapons can shred components with pinpoint accuracy.

How does one fix that?

Decrease the heat threshold for all 'Mechs? No. First off, not all energy weapons are problematic and second, people will just sit in cover for a slightly longer period of time to cool down.

Increase the cycle time for lasers? No. People will just sit in cover longer to wait for the next alpha.

Increase the burn times for lasers? Getting closer, makes it easier to spread the damage when getting hit but also makes lasers more and more awkward to use - plus, a lot of the incoming damage can still be pinpointed.

Now, the "solutions" brought up in these thread: Cone of fire systems, removal of convergence and changes to the hit boxes. Let's have a short look at them, shall we?

Both a CoF system and a complete overhaul to the hit boxes would probably take PGI months to develop, yet a few more months to implement and test internally and roll out and yet another few more months to iron out most of the bugs (while the forums are up in arms about them breaking their game, I suppose). Same with hit boxes, really. Getting the CT and ST and overall geometry sorted seems to be a task in and of itself, so reworking all the hit boxes on everything? Months to get that done? Nah. Removal of convergence? Yeah, probably relatively easy to implement, or so I'd assume, but as has been mentioned earlier, that brings with it a huge slew of new balance issues, such as clustered hardpoints becoming a focal point.

So, what else is there?

Reduce the damage of ALL lasers.

By how much? Dunno. 10%, 15%, 20%, whatever. This, basically, solves a lot of things. First, it encourages binging weapons other than lasers. Yes, a laser will still hit exactly where you point your cross hair, it'll still have a burn time that'll allow you to hit-scan with it, it still won't require you to compensate for bullet drop off nor will it require you to lead your target. However, with lasers doing (considerably) less damage, there's a bigger incentive to bring weapons that require actual aiming. Without the need for ballistic quirks or buffing weapon systems and thereby shortening the TTK. At the same time, the TTK will be directly increased by the lower damage of laser weapons and indirectly by ballistics becoming more popular (due to them being harder to place with pin-point accuracy).

This might also drive people away from the peek-a-boo playstyle.

Basically: Why not go with the most feasible, because least convoluted fix that gets the job done?