Jump to content

- - - - -

The Gauss / Particle Projection Directive - Feedback


1263 replies to this topic

#921 Sandpit

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPip
  • Veteran Founder
  • Veteran Founder
  • 17,419 posts
  • Facebook: Link
  • Twitter: Link
  • LocationArkansas

Posted 04 August 2014 - 07:15 PM

View PostCimarb, on 04 August 2014 - 07:08 PM, said:

It is not an issue where they cannot do it. It is an issue where doing it causes severe hit registration issues. I agree with you that it should be done/figured out, but just stating WHY they are not doing it currently.


I don't see how slowing your reticle convergence speed down affects any of that. The shot still hits exactly where it aims, it jsut takes a split-second longer to get where you're wanting the shot to line up. It doesn't change the hit registration mechanic at all.

So I swing left to shoot
with an AC2 there's no delay
AC5 = .2 seconds before it moves all the way where you want it
AC 10 = .4 seconds
AC20 - .5 seconds
PPC = .3 seconds
Gauss = .5 seconds

All that means is when I swing my reticle around to the left, it takes just a split second longer for that AC20 to swing around and get a good shot. There's nothing that requres any additional mechanics, just an adjustment to the speed at which the reticle moves. If they can't code that, well then MWO is in much deeper problems than I ever suspected

#922 Aim64C

    Member

  • PipPipPipPipPipPipPip
  • 967 posts

Posted 04 August 2014 - 07:15 PM

View PostKageru Ikazuchi, on 04 August 2014 - 06:39 PM, said:

All this talk about convergence being the "solution" to the pin point damage "problem" makes me wonder ... has anyone, ever, gotten global (as in world-wide) server authenticated hit detection with a convergence mechanic "right" in a real-time multiplayer video game?


You just threw out a bunch of terms you didn't understand.

The short answer to this is that any game where a projectile 'spawns' from something other than the player camera can be coded to support what is known as "boresight." This means that the weapon fires in the direction the barrel is aligned.

Since we already have converging ballistic projectiles that can fail to properly converge - the answer is "yes."

Engines like RealVirtuality make the 'issues' PGI is having look like ... well... like a ship of fools.



#923 ReXspec

    Member

  • PipPipPipPipPipPipPip
  • Knight Errant
  • Knight Errant
  • 502 posts
  • LocationOrem, Utah

Posted 04 August 2014 - 07:20 PM

View PostSandpit, on 04 August 2014 - 07:10 PM, said:

there's already crosshair bob

with the implementation of clan ACs, IS ACs are in a good spot as FLD. It gives them an advantage over clans.

I like the beam durations where they're at now.

now sure what you mean by reticle bloom though?


Crosshair bob? In thirdperson, yes. In firstperson, no. Or, if there is, it does not effect where pilots place their shots.

Sure, you like it now, but if they increase convergence times, or introduce some sort of crosshair deviation while firing/moving they would need to shorten beam times.

Reticule bloom is basically an accuracy penalty when firing weapons with any sort of recoil or heat. The cone-of-fire expands, thus producing reticule "bloom."

For mechwarrior, I imagine if P.G.I. did implement a combination of systems like these, or reticule bloom, it would be barely noticeable, but would effect cone-of-fire to the point where shots wouldn't land in the same spot on a target if you managed to place the reticule in the same place.

Some players view this as "penalizing" players for good aim, but I humbly disagree. Knowing where shots are going to go, or slowing down movement and fire enough to reduce the cone as much as possible is another element players would need to adapt to, and learn how to work around. Not to mention, it would add more realism to the game that would serve to immerse the player as well as balance the problem of convergence we are facing now.

Edited by ReXspec, 04 August 2014 - 07:26 PM.


#924 IraqiWalker

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • The 1 Percent
  • The 1 Percent
  • 9,682 posts
  • LocationCalifornia

Posted 04 August 2014 - 07:29 PM

View PostSandpit, on 04 August 2014 - 07:08 PM, said:

you're talking about sway

I'm talking about convergence. Adjusting a few lines of code to slow down how fast that reticle moves according to the weapon size. Completely different issue.

Maybe that's where the miscommunication is coming in?

I'm not talking abotu anythign other than simply slowing down how fast the reticle moves when moving your mouse to aim. That's it. I'm not talking about revamping the whole targeting system. Simply adjusting the speed of the reticle. That doesn't require any extensive rework of any of the code or mechanics.

The ONLY thing it MIGHT require is adding in a targeting reticle for multiple weapons. Give each ballistic and the PPC its own unique reticle. AC20 gets a red one
AC10 gets a blue one
AC5 gets a green one
Gauss yellow
PPC purple

done, all PPD problems solved. Simple, elegant, and actually attacks the issue. This isn't a complex solution at all


Okay, now I understand what you're saying.

By the way, didn't we have gradual convergence back in Closed Beta?

View PostCimarb, on 04 August 2014 - 07:08 PM, said:

I never said they should be equal, even though they actually are quite identical in TT rules (where everything is random FLD).

Even in TT you could make called shots, and aim at specific parts, but they were harder, and most people never used them.


View PostReXspec, on 04 August 2014 - 07:10 PM, said:


The PP FLD idea was stupid in the first place except for weapon systems that were specifically PP FLD in both Battletech canon, and virtually every iteration of the Mechwarrior game.

And now we see WHY PP FLD autocannons are a bad idea.

I can't believe P.G.I. thought that would be a good idea.


Oh I don't like PP FLD ACs, however, right now, they provide an interesting distinction and a big role in balance between clan tech and IS tech.


View PostReXspec, on 04 August 2014 - 07:20 PM, said:


Crosshair bob? In thirdperson, yes. In firstperson, no. Or, if there is, it does not effect where pilots place their shots.

Sure, you like it now, but if they increase convergence times, or introduce some sort of crosshair deviation while firing/moving they would need to shorten beam times.

Reticule bloom is basically an accuracy penalty when firing weapons with any sort of recoil or heat. The cone-of-fire expands, thus producing reticule "bloom."

For mechwarrior, I imagine if P.G.I. did implement a combination of systems like these, or reticule bloom, it would be barely noticeable, but would effect cone-of-fire to the point where shots wouldn't land in the same spot on a target if you managed to place the reticule in the same place.

Some players view this as "penalizing" players for good aim, but I humbly disagree. Knowing where shots are going to go, or slowing down movement and fire enough to reduce the cone as much as possible is another element players would need to adapt to, and learn how to work around. Not to mention, it would add more realism to the game that would serve to immerse the player as well as balance the problem of convergence we are facing now.

No it penalizes players that don't use alphas 24/7. Reticule boom works with sustained fire. Like MGs in other FPS games.

However, you don't want that here.

A variable reticle that is impacted by velocity for example is a much more reasonable solution. You want to place an accurate shot? Drop your throttle to below 85%, and take aim.

#925 ReXspec

    Member

  • PipPipPipPipPipPipPip
  • Knight Errant
  • Knight Errant
  • 502 posts
  • LocationOrem, Utah

Posted 04 August 2014 - 07:32 PM

View PostIraqiWalker, on 04 August 2014 - 07:29 PM, said:



A variable reticle that is impacted by velocity for example is a much more reasonable solution. You want to place an accurate shot? Drop your throttle to below 85%, and take aim.


I like that idea.

#926 Kageru Ikazuchi

    Member

  • PipPipPipPipPipPipPipPip
  • The Determined
  • The Determined
  • 1,190 posts

Posted 04 August 2014 - 07:39 PM

View PostAim64C, on 04 August 2014 - 07:15 PM, said:

You just threw out a bunch of terms you didn't understand.

The short answer to this is that any game where a projectile 'spawns' from something other than the player camera can be coded to support what is known as "boresight." This means that the weapon fires in the direction the barrel is aligned.

Since we already have converging ballistic projectiles that can fail to properly converge - the answer is "yes."

Engines like RealVirtuality make the 'issues' PGI is having look like ... well... like a ship of fools.

No, I understand the terms ... but I'm also trying to understand what everyone else is proposing as the solution.

Let's see the terms I used ...
- global (world-wide) server authenticated hit detection ... that means when I shoot at something on my screen, it takes a fraction of a second for that information to be communicated back to the server to determine whether or not I hit ... and then takes another fraction of a second to notify me (and my target) of the effect my hit has. If I have 200+ ms ping, that means that the total time from pulling the trigger to seeing the effect can be as much as half a second.
- convergence ... the weapons point where your reticule is aimed, even if they are mounted on different sides of the 'mech

OK ... maybe I should have stated the question more clearly to the problem at hand ... has anyone gotten it right in cryengine? (which, as I understand, was delivered to PGI with client-side hit detection)

Now that I think about my previous post, though, I did make an assumption that was false ... hit detection calculations are already being done for each particle in a volley, whether that particle is an AC/20 shell or a "tick" of a laser.

Also, I agree that we do already have "convergence", it's just instantaneous, rather than delayed ... Sandpit's idea of different times to achieve convergence for different weapons kind of makes sense for torso mounted weapons, and maybe for arm-mounted weapons ... it is interesting but I'm not sure how much difference it's going to make when most of the aiming is being done by twisting your torso and moving your arms (and not moving individual weapons).

Edited by Kageru Ikazuchi, 04 August 2014 - 07:49 PM.


#927 Sandpit

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPip
  • Veteran Founder
  • Veteran Founder
  • 17,419 posts
  • Facebook: Link
  • Twitter: Link
  • LocationArkansas

Posted 04 August 2014 - 07:46 PM

View PostReXspec, on 04 August 2014 - 07:20 PM, said:


Crosshair bob? In thirdperson, yes. In firstperson, no. Or, if there is, it does not effect where pilots place their shots.

Sure, you like it now, but if they increase convergence times, or introduce some sort of crosshair deviation while firing/moving they would need to shorten beam times.

Reticule bloom is basically an accuracy penalty when firing weapons with any sort of recoil or heat. The cone-of-fire expands, thus producing reticule "bloom."

For mechwarrior, I imagine if P.G.I. did implement a combination of systems like these, or reticule bloom, it would be barely noticeable, but would effect cone-of-fire to the point where shots wouldn't land in the same spot on a target if you managed to place the reticule in the same place.

Some players view this as "penalizing" players for good aim, but I humbly disagree. Knowing where shots are going to go, or slowing down movement and fire enough to reduce the cone as much as possible is another element players would need to adapt to, and learn how to work around. Not to mention, it would add more realism to the game that would serve to immerse the player as well as balance the problem of convergence we are facing now.

It's there you jsut don't see it. The same bob you get in 3pv is the what you get in first person. I thought the same thing the first time I tried out 3PV but after a few tests and a few people explaining it I can see how it works now.

That's why I suggested lasers remain pinpoint. That's their advantage since they're DoT

Ahhh ok, I think homeless bill suggested something similar (or someone else did or I might have lol) but when you fire off more than the equivalent of 1 AC20 then your corsshairs "jump" slightly ensuring that you can't just stand still (which is dumb anyhow in my opinion) and not continuously maintain perfect convergence. You fire off big weapons, the reticle "jumps" slightly so it has to come back down on target. Taking jsut a split second. Now you can fire again before it converges, it's just that your shot will hit wherever the reticles are when you click.

View PostIraqiWalker, on 04 August 2014 - 07:29 PM, said:


A variable reticle that is impacted by velocity for example is a much more reasonable solution. You want to place an accurate shot? Drop your throttle to below 85%, and take aim.

I don't remember if we did or didn't to be honest. It's been too long.

Variable reticle is a good "title" for what I'm talking about I guess. Same principle. Except it's affected by weapon size. Keep lasers pinpoint, adjust all ACs 5 and above and the PPC and I really believe you just solved most of the PPD and FLD and weapon combo issues.

#928 Col Jaime Wolf

    Member

  • PipPipPipPipPipPipPipPip
  • Bad Company
  • Bad Company
  • 1,214 posts

Posted 04 August 2014 - 08:09 PM

both are bad ideas.

why not buff lasers to compete?

like many have said if IS long range is nerfed then clan ER lasers will become top dog.

but really paul have you ever considered that you just dont know what your doing?

#929 Sandpit

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPip
  • Veteran Founder
  • Veteran Founder
  • 17,419 posts
  • Facebook: Link
  • Twitter: Link
  • LocationArkansas

Posted 04 August 2014 - 08:14 PM

View PostMellifluer, on 04 August 2014 - 08:09 PM, said:

both are bad ideas.

why not buff lasers to compete?

like many have said if IS long range is nerfed then clan ER lasers will become top dog.

but really paul have you ever considered that you just dont know what your doing?

I don't think it's not knowing what he's doing, I think he just feels he knows "better" and really thinks he's doing "what's best". Unfortunately none of his customers agree with him.

#930 Ultimax

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 6,979 posts

Posted 04 August 2014 - 08:42 PM

View PostCimarb, on 04 August 2014 - 07:08 PM, said:

I never said they should be equal, even though they actually are quite identical in TT rules (where everything is random FLD).


TT is kind of irrelevant here, since everything has been changed and in their own way PGI has tried to create differentiation among these weapon systems with varying degrees of success.


View PostCimarb, on 04 August 2014 - 07:08 PM, said:

I absolutely love how the Clan ERPPCs work, so I do not agree with you about the "non-viable" thing. They are very viable, which is actually the entire reason they are again on the chopping block in this thread.



Clan ER PPCs do 15 damage, not 10 like IS ER PPCs.

Clan ER PPC doesn't have a 90m min range like the IS PPC does.


Reduce IS ER PPC to spread, and you won't often see this weapon on the field outside of a few sniper lights...maybe.

The reason they are on the "chopping block" is because people are complaining about one, single, Dire Wolf build that is actually meta compliant never-mind it's terrible maneuverability and crappy top speed - and also because Paul refuses to touch other mechanics in favor of erroneously, and repeatedly, nerfing individual weapon systems.

Being on the chopping block is irrelevant, when you have almost 50 pages of players saying that these are bad ideas - this is because Paul has mistaken what the actual problem is. It is not Gauss, or PPCs, or being able to fire your 30+ tons of weapons together.

Edited by Ultimatum X, 04 August 2014 - 08:44 PM.


#931 Mazzyplz

    Member

  • PipPipPipPipPipPipPipPipPip
  • Ace Of Spades
  • 3,292 posts

Posted 04 August 2014 - 10:05 PM

http://mwomercs.com/...auss-ppc-issue/

this!
support this!

#932 Sandpit

    Member

  • PipPipPipPipPipPipPipPipPipPipPipPip
  • Veteran Founder
  • Veteran Founder
  • 17,419 posts
  • Facebook: Link
  • Twitter: Link
  • LocationArkansas

Posted 04 August 2014 - 11:41 PM

View PostMazzyplz, on 04 August 2014 - 10:05 PM, said:


no thanks, no less complicated and unnecessary than Paul's idea. There's no need at all for all these charge and trigger mechanics. That's where things start getting complicate and harder for people to use effectively, especially new players in this game.

#933 NoTime

    Rookie

  • Ace Of Spades
  • Ace Of Spades
  • 4 posts

Posted 05 August 2014 - 01:13 AM

About nerfing PPCs. Basically, if you make PPC projectile travel much slower it will only hit long range builds, since in close combat speed has no relevance. If you do so, IS will have even less response to Clan long range builds. Plus, I don't see why other PPC builds have to suffer because of a single slightly imbalanced build.

So, it is a good idea to target only Gauss+PPC builds but as mentioned many times before, not with such unclear methods.

IMHO, both Ideas are just equally bad.

View Poststjobe, on 29 July 2014 - 04:24 PM, said:

This is a MechWarrior game that uses the armour system from BattleTech tabletop, a system that isn't designed for pin-point accurate damage; it is designed for spread damage.

Where TT uses dice rolls to spread the damage we need something not-so-random, like movement-based reticule shake (a.k.a. head bob in other FPS games), or movement-based Cone of Fire, or non-instant convergence, or any of the multitude of ideas that have been presented over the last two years or so.

There's no shortage of ideas on how to fix the problem of instant convergence (I listed four in a previous post), which is the REAL problem these convoluted and twisted ideas in the OP are trying to address.


That's not a good idea. Implementing movement-based cone of fire or reticule shake will make even more players to stay put while firing. So we will have even more sitting ducks complaining of being targeted by superior skill players who don't forget to move, to jump, to torsotwist or even simply to take cover.

It is just about making maneuvers which lessen your exposure time.

Edited by NoTime, 05 August 2014 - 01:14 AM.


#934 Reno Blade

    Member

  • PipPipPipPipPipPipPipPipPip
  • The Blade
  • The Blade
  • 3,462 posts
  • LocationGermany

Posted 05 August 2014 - 01:42 AM

It's not only the "one" Direwolf build with 2GR/2PPC, but also 1GR/2PPC, 2AC5/2PPC and to a lesser degree AC10/2PPC and ac20/2PPC builds that are stronger than any kind of energy-only builds.
This is even stronger for IS PP FLD autocannons that don't have a burst.

All these alphas can delive a damage over 30 into one spot with very little problems and no beamtime = minimum exposure.
Clan ERPPC and UACs with splash/burst reduced the Clan builds to Gauss/cERPPC combos and these are the biggest problem.
As long as PPC, Gauss and ACs (for the IS) are one big bullet that does only need point&click easy aiming, they will always be better than DoT weapons (lasers) or spread (missiles and LBX).

Some people ask why the "other" weapons are not just getting buffed?
More SRM and LRM damage, shorter laser beams?
These weapons are in a way better balance state than the PP FLD weapons.
BUT, you can look at it from both sides, right? either of these groups are stronger or weaker compared to where the balance goal would be, right?
So does it matter which group to change?
Oh it does if the deviation to the goal is higher for one group.
And here comes one of the problems.
Perception of balance of the Community vs Paul and his crew.
There are so many people who think their precious weapons will be useless after a nerf that it doesn't even matter where the balance is currently. They just see their favourite weapon on the edge and start screaming.


The balance goal is simple.
- Give players the chance to react to enemy fire and spread incomming damage.
- Give players the chance to shoot what they want, but make it harder for strong combos than for balanced weapons and weapon numbers (see ghost heat).

Some people suggest slower aiming and it would also be part of a TT heatscale side-effect with high heat.
The issue is, that slower aiming only helps to reduce the accuracy of the whole volley, but don't give the target the chance to spread incomming damage.
Take a direwolf with 2Gauss and 3ERPPCs, stand 700m away from the target and use armlock, if you don't have all weapons in the arms.
Even with a slower reticule, you only have to twitch to the side by 5 degrees or so at that range. Hardly a problem for aiming.

With that said, we also have the double-basics of the skill tree that gives a lot of speed for aiming and turning as well as accelerating and decelerating.
Without these, heavy and assault mechs would feel very sluggish and aiming would be harder.
But at the same time, twisting to spread damage would be a LOT harder for slower mechs.

This would give medium and light mechs the edge at med/short ranges over slower mechs.
But it would mean the slower mechs would need to be sniping or LRMing to prevent their weekness to affect them too much.
Brawling would be harder if you wanted to use twisting to spread damage (see victor nerfs).

So in the end, we could gain some changes with this, but with side-effects.

Going back to the weapons itself.
The easiest way to reduce the strenght of these combos is to reduce the pinpoint or the FLD part of some or all of them.
Thats why some suggest more slash for all PPCs.
Remember, you still hit with FULL damage with splash CERPPCs. (minus the chance to lose one arc if you hit arm or leg)
Compared to the burst UAC20 where you need to land the shots to do all damage = need way more skill and luck.

Would it fix the problem builds or would that break the whole weapon?
I say the change to all PPCs towards more splash is what is needed for now and better than any other "more complicated" fix. (it can easily be changed if we have better ideas in the future)
You will spread some damage, still kinda concentrated as the dmg is only to adjacent sections.
I think a distribution like this would give enough reason to field other weapons without removing the advantage of the PPCs (they are still FLD):
CERPPC: 7/3.5 +3.5
IS PPC: 6/2+2
IS ERPPC: 7/1.5+1.5
Edit: fixed numbers
It would still be more than your "average" beam damage (because you rarely have 100% uptime with lasers).

We could keep IS ACs to have a small edge over the Clans, but I rather have these as burst (with one less shell vs clans) too.
I'd also increase heat and or beamtime of Clan weapons a bit to balance it vs IS weapons, but thats just a sidenote.

Edited by Reno Blade, 05 August 2014 - 01:54 AM.


#935 P e n u m b r a

    Member

  • PipPipPipPipPipPip
  • The Nocturnal
  • The Nocturnal
  • 273 posts
  • LocationUnited Kingdom

Posted 05 August 2014 - 02:22 AM

Rexspec are we even playing the same game look there is nothing wrong with pin point damage what's wrong with it? It's good enough for SO many top games that are popular and no weapon recoil is different when you fire a fully automatic weapon with 20x the time to kill potential as mwo and even then the first few shots will land center crosshairs anyway, you do not have that here you have single shot cooldown weapons with low time to kill.
Brawling would need to be over powered and ranged redundant in this if you could roll about in the open and still go and smash those snipers in a open que with no coordination. Have you seen the competitve scene go look at that and tell me brawling is not strong and a prefered means to fight by many teams problem is in this game is that it does have old men and table toppers playing they in my opinion ruin the game with complaints founded in inexperience with the game we have the closest balance we have had in a while and it's about to be ruined because of people like you most the weapons are situational and it's just that gauss Ppc is situationally just the best thing for solo que. The game needs new players fresh ones pin point works it works for many titles it's not a problem here everyone had the same shooting mechanic it's true it's fair it's tried it's tested it works you guys just want to dilute the skill cap and make it all even there is no nice way of saying it but that is exactly what's going on.

Edited by L e 0, 05 August 2014 - 02:23 AM.


#936 n r g

    Member

  • PipPipPipPipPipPipPip
  • The Universe
  • The Universe
  • 816 posts

Posted 05 August 2014 - 03:01 AM

View PostReXspec, on 04 August 2014 - 06:08 PM, said:


Okay... I've held my tongue long enough, but I'm tired of comments like this.

One, if there IS talk about what strategy/gameplay option is OP/not OP, it's not in regards to what new players can and cannot do. It is a matter of what looks better or "simpler" to a new player.

Two, Mechwarrior IS NOT CS:GO.

The cancerous attitude that we need to eliminate is the tendency for both new and "competitive" players to choose low-risk, high-reward builds/strategies.

When comparing Brawling to Sniping, Brawling is quickly DYING in this game (if it is not dead already) because it is more difficult to Brawl then it is to park your ass in a nice position and pick off targets with 30 - 60 pinpoint alphas.

Yes. Sniping take skill in gunnery (aiming) and positioning, but Brawling takes skill in gunnery (yes, you do actually need to aim in ranges from 0 - 300 meters), piloting (you have to know how to torso-twist to spread damage, AND you have to know how to move, in what direction to move, and where to position yourself).

That stated, do I agree with this nerf? No. Absolutely not. If anything, reducing the speed of the PPC just makes it better able to sync with the AC-5/PPC combo, thus tunneling players into that abominable "meta." Not to mention it is outright punishing players for running a potent combination such as the PPC/Gauss combo.

I've said this before, and I'll say it again PGI NEEDS to draw inspiration from threads like this to fix issues with pinpoint damage.

They don't need to outright nerf weapons combos, they don't need to nerf weapons, they don't need to add more convoluted mechanics to solve an issue like this. Weapon locking is awkward and it always will be awkward.

I've mentioned this already in the link I provided, but, if they want to re-work pinpoint damage, they should start by making the first-person crosshair work like it does in third-person view. That could be phase one in testing some sort of semi-realistic gunnery when it comes to weapons/crosshair reaction.

Personally, I think they should have ballistics, weapons firing, and aiming work the same way it does in MW3, but I guess asking for that is too ambitious.

So translating crosshair behavior from third-person view, to first-person view would at least be a step in the right direction, I think.


CS:GO is a model for all games. It's the pinnacle of competition and one of the only games that takes its input from the competitive base NOT the joe-schmo pub scrubs

And how is PPC/guass low reward? Have you seen the heat on PPCs, or on jump jets, or charged a gauss while an atlas is bearing down on you?

Most of the players who initiate these ingame nerfs to this game are the mainstream "casual" base as I 've stated....

It's more than possible to brawl. In fact, I'd wager 80% of matches turn into brawls in this game, even on Alpine, just due to the high heat of PPCs, the new jump jeat heat, and the awkward charge mech on gauss...

How is sniping "low risk/high reward"? The "casuals" all focus on the "pinpoint alpha" aspect, but do you know what happens if you miss that alpha? that's 30points of damage down the drain...... and time for brawlers to close the gap and spam missles or srms...........

With all these nerfs or weapon complaints, the biggest overlooked factor is player skill...........

#937 Reno Blade

    Member

  • PipPipPipPipPipPipPipPipPip
  • The Blade
  • The Blade
  • 3,462 posts
  • LocationGermany

Posted 05 August 2014 - 03:23 AM

View PostE N E R G Y, on 05 August 2014 - 03:01 AM, said:


CS:GO is a model for all games. It's the pinnacle of competition and one of the only games that takes its input from the competitive base NOT the joe-schmo pub scrubs

And how is PPC/guass low reward? Have you seen the heat on PPCs, or on jump jets, or charged a gauss while an atlas is bearing down on you?

Most of the players who initiate these ingame nerfs to this game are the mainstream "casual" base as I 've stated....

It's more than possible to brawl. In fact, I'd wager 80% of matches turn into brawls in this game, even on Alpine, just due to the high heat of PPCs, the new jump jeat heat, and the awkward charge mech on gauss...

How is sniping "low risk/high reward"? The "casuals" all focus on the "pinpoint alpha" aspect, but do you know what happens if you miss that alpha? that's 30points of damage down the drain...... and time for brawlers to close the gap and spam missles or srms...........

With all these nerfs or weapon complaints, the biggest overlooked factor is player skill...........

1. games like CS GO use recoil because they have automatic weapons. This is where the comparison lacks, not the competitive part or the part where the community is involved.

2. I think you answered part of your last question yourself.
low risk/high reward because in case of high player skill.
You won't see many misses either with high skill.
You can brawl, but it's saver to snipe because of how cover works and gives high skilled snipers very low exposure while they can land their skilled shots exactly where they want.

#938 P e n u m b r a

    Member

  • PipPipPipPipPipPip
  • The Nocturnal
  • The Nocturnal
  • 273 posts
  • LocationUnited Kingdom

Posted 05 August 2014 - 03:46 AM

Go watch competitive matches see how many turn out to be a brawl and if it's sniping it will be lasers more so

#939 n r g

    Member

  • PipPipPipPipPipPipPip
  • The Universe
  • The Universe
  • 816 posts

Posted 05 August 2014 - 04:19 AM

View PostReno Blade, on 05 August 2014 - 03:23 AM, said:

1. games like CS GO use recoil because they have automatic weapons. This is where the comparison lacks, not the competitive part or the part where the community is involved.

2. I think you answered part of your last question yourself.
low risk/high reward because in case of high player skill.
You won't see many misses either with high skill.
You can brawl, but it's saver to snipe because of how cover works and gives high skilled snipers very low exposure while they can land their skilled shots exactly where they want.


So you want highly skilled players to suffer then? That's the worst response I've ever heard. You self admit to sniping taking more skill, and then imply that is somehow part of the problem and then that the higher skilled player should suffer.

"You won't see many misses either with high skill" so the "high skilled" player should therefor suffer for that? because all the new/novice/casual players will complain on the forums and convince the DEVs to go on a nerf banwave? WHAT???? So due to the "high skill" and all the hours put into practice, now it becomes "low risk".... logic 0/10? the same could be applied to any other concept i.e. the expert brawler (and there are some amazing brawlers in this game don't get me started) who spends hours perfecting his craft, but after countless hours now his "style" becomes low risk/high reward, so let's nerf SRMS or AC20's! How does that even make sense to you? Do you understand you are killing the game with that logic?

So players that spend 1000/2000 hours on CSGO playing AWP and one-shotting newer players, that's unfair too right? low risk/high reward because of their skill right? So we should all go on the CSGO official forums and complain about it and maybe the devs will make the AWP harder to shoot, or have a max range on it, so it has to be used up close... THEN, all the newer players who complain can much EASILY use p90's and run around and spray bullets and kill things, so they don't have to worry about those "OP AWPERS" anymore *cough scarcasm cough*. Anyway, notice the parallels here? This is what the mainstream community is doing MechWarrior Online.

pfffttt... And the CSGO comparison is valid. Of course it has assault rifles and this is a sci-fi robot shooter - BUT, in terms of gameplay, EVERY game company strives to make a popular, competitive game, that is well-balanced that will last for many years to come. CounterStrike has done this. It doesn't matter if its a MOBA, FPS or RPG, any company could imitate some of the models that CS has laid out, and they could only wish to attain a fraction of the success VALVE has established.

CounterStrike has done this and is why other companies can only dream of acheiving the feats that Valve did. It's all universal mate, the answers are right before your eyes, and much of this has been argued before, years before.

Alas, sadly, as I stated about 15 times earlier, it's only a matter of time before PGI follows in the footsteps of Call of Duty or Battlefield, turning a once great franchise into a diluted, mundane, low-skill cap pub game for casual players.......

..No competition scene, gimicky sales ($500 gold mechs), and total and utter dilution of the gameplay mechanics to make it "easier" for new players, who are essentially going to pump $$$ into the game, to play. The signs are all there, and many that I didn't even bother to list. Either this is your first PC game, or you are blinded by your love of the MechWarrior series and don't want to believe it (believe me, I love MechWarrior too, but the signs are here).

Let's all remember how much focus is put on making sales for trivial hero mechs, ingame cockpit items, 500$ gold mechs, skins for your mech, premium time (you can't even do a 1v1 without both players paying $$$, are you joking?), MC for mech bays......

Oh, and were is community warfare? Clearly, $$$$$$ is the big sign here and a total disregard for ACTUAL GAMEPLAY is obvious for an experienced PC gamer.

tl:dr

Edited by E N E R G Y, 05 August 2014 - 04:32 AM.


#940 ReXspec

    Member

  • PipPipPipPipPipPipPip
  • Knight Errant
  • Knight Errant
  • 502 posts
  • LocationOrem, Utah

Posted 05 August 2014 - 05:41 AM

View PostE N E R G Y, on 05 August 2014 - 04:19 AM, said:

So you want highly skilled players to suffer then? That's the worst response I've ever heard. You self admit to sniping taking more skill, and then imply that is somehow part of the problem and then that the higher skilled player should suffer.


Sniping is a low risk, high reward strategy that will always look more appealing to new players because they need to exhibit very little risk to themselves in order to put the most damage on a target. That is already an established fact.

It is also an established fact that brawling is much more dangerous and takes more skill because, you not only need to aim well enough keep your opponents head down while you're closing the distance, but you need to know how to spread damaged over the rest of your body once the distance is closed, while keeping fire on your opponent.

An archer's job is not as dangerous as a knight's, because the archer can sit behind the front lines and pick targets off with impunity. If the archer faces a threat of someone closing the distance, he can simply re-position, and keep himself away from a potential threat. It's really not that hard.

Again, sniping takes skill in two categories of piloting in mechwarrior: Aiming and positioning. Brawling takes skill in four categories: Aiming, positioning, boxing, and movement. Those are the facts. The opinion you are bringing to the table is the opinion of someone who hugs the "meta" for dear life, and then seizures when he realizes that a single playstyle is not viable in EVERY scenario anymore.

View PostE N E R G Y, on 05 August 2014 - 04:19 AM, said:

"You won't see many misses either with high skill" so the "high skilled" player should therefor suffer for that? because all the new/novice/casual players will complain on the forums and convince the DEVs to go on a nerf banwave? WHAT???? So due to the "high skill" and all the hours put into practice, now it becomes "low risk".... logic 0/10? the same could be applied to any other concept i.e. the expert brawler (and there are some amazing brawlers in this game don't get me started) who spends hours perfecting his craft, but after countless hours now his "style" becomes low risk/high reward, so let's nerf SRMS or AC20's! How does that even make sense to you? Do you understand you are killing the game with that logic?


We don't agree with this nerf proposal as much as you do, dude. I don't want to see a powerful combination such as a dual Gauss/ER PPC go down the drain because of a SEVERELY nerfed projectile speed, and a clunky weapons lock. There are ways to introduce balance to that combo without outright nerfing it like Paul is proposing to do. You've seen hundreds of reasonable suggestions to do so on this thread alone. Let's keep the suggestions flowing rather then simply white-knighting that particular playstyle with nothing but bluster and a rant against the players/developers, eh?

View PostE N E R G Y, on 05 August 2014 - 04:19 AM, said:

So players that spend 1000/2000 hours on CSGO playing AWP and one-shotting newer players, that's unfair too right? low risk/high reward because of their skill right? So we should all go on the CSGO official forums and complain about it and maybe the devs will make the AWP harder to shoot, or have a max range on it, so it has to be used up close... THEN, all the newer players who complain can much EASILY use p90's and run around and spray bullets and kill things, so they don't have to worry about those "OP AWPERS" anymore *cough scarcasm cough*. Anyway, notice the parallels here? This is what the mainstream community is doing MechWarrior Online.


Again, Mechwarrior is not CS:GO, nor is it an FPS. The day Mechwarrior becomes a first-person, twitch shooter like Counter-strike is the day Hell freezes over.

Plus, you do know all the controversy and design mentality/balance considerations that went into the transition from the weapons in CS:S to the weapons in CS:GO, right?

In CS:S, snipers were king in that game because you just needed to be good in two aspects of gameplay (much like snipers in Mechwarrior Online right now): Aiming and positioning. To make matters worse, the devs at Valve made snipers utterly lack much of the skill that real world snipers have to incorporate to do their job well. In reality, this is what snipers have to account for when picking a target (and this is coming from an Infantryman from the 173D ABCT, 1-503D, so I'd like to think I'm at least somewhat knowledgeable on this subject): Knowing the eye-relief on a scope and being able to account for that eye-relief in as little time as possible (realistically, finding the correct eye-relief on a 10 - 20x scope takes about two seconds), finding the correct sighting appropriate to the range of the target and gaining a proper sight picture, compensating for the bullets travel path/time to the target, and compensating for the targets movement.

ALL OF THOSE FACTORS are what are supposed to go into sniping. In CS:S Valve chopped down those factors into nothing by making the eye-relief time practically nill, by making bullet travel-time instant, and making the bullet itself hitscan with nothing but a simple, barely randomized cone-of-fire manipulating the shots.

CS:GO slightly alleviated this problem by introducing "crosshair blur" by moving, but still, sniping is stupid easy in that game because you're still firing a weapon with no eye-relief compensation, no bullet-travel time, no sighting or ranging a target. In essence, the AWP is a one-shot laser with a slightly blurry and shaky scope when moving.

Even the devs at Valve conceded that the sniper had WAYYYYYYYYYY too much pull in both public and competitive matches. Which is why comptetitive matches only allow one AWP per game now. Not to mention, it is part of the reason why Minh "Gooseman" Le LEFT the developers of CS at Valve because he wanted to create a balanced Tac-ops shooter on his own.

WE DON'T WANT MECHWARRIOR TO TURN INTO CS:GO, E N E R G Y.

If you're expecting Mechwarrior to, one day, be CS:GO then you may want to quit now, because it will NEVER be that way. Not because P.G.I. doesn't want this game to be competitive, but because at least most of the players and P.G.I. know the design differences between a 31st century, vehicular combat simulator, and a twitch FPS shooter.

View PostE N E R G Y, on 05 August 2014 - 04:19 AM, said:

pfffttt... And the CSGO comparison is valid. Of course it has assault rifles and this is a sci-fi robot shooter - BUT, in terms of gameplay, EVERY game company strives to make a popular, competitive game, that is well-balanced that will last for many years to come. CounterStrike has done this. It doesn't matter if its a MOBA, FPS or RPG, any company could imitate some of the models that CS has laid out, and they could only wish to attain a fraction of the success VALVE has established.

CounterStrike has done this and is why other companies can only dream of acheiving the feats that Valve did. It's all universal mate, the answers are right before your eyes, and much of this has been argued before, years before.

Alas, sadly, as I stated about 15 times earlier, it's only a matter of time before PGI follows in the footsteps of Call of Duty or Battlefield, turning a once great franchise into a diluted, mundane, low-skill cap pub game for casual players.......


Oh my dear Christ... are you screwing with us right now? You've GOT to be screwing with us.

I'll say this again, and it will be the LAST time I'll say it: MECHWARRIOR IS NOT AN FPS.

This franchise has always been advertised as a "31st Century battlefield, vehicular-combat simulator."

That does not translate into: "tac-ops, first-person-shooter."

Srsly. GET OFF the notion that Mechwarrior is an FPS, and that it will EVER be equivalent to the design flop that is CS:GO.

Put that thought as far away as possible from your mind and at least PRETEND we're piloting giant, stompy robots with a lot of armor and huge guns. We're NOT playing human Spec-ops soldiers who can take out an opponent with a few good "plinks" to the chest.


View PostE N E R G Y, on 05 August 2014 - 04:19 AM, said:

..No competition scene, gimicky sales ($500 gold mechs), and total and utter dilution of the gameplay mechanics to make it "easier" for new players, who are essentially going to pump $$$ into the game, to play. The signs are all there, and many that I didn't even bother to list. Either this is your first PC game, or you are blinded by your love of the MechWarrior series and don't want to believe it (believe me, I love MechWarrior too, but the signs are here).

Let's all remember how much focus is put on making sales for trivial hero mechs, ingame cockpit items, 500$ gold mechs, skins for your mech, premium time (you can't even do a 1v1 without both players paying $$$, are you joking?), MC for mech bays......

Oh, and were is community warfare? Clearly, $$$$$$ is the big sign here and a total disregard for ACTUAL GAMEPLAY is obvious for an experienced PC gamer.


You're wanting to turn this game into CS:GO. That's the bottom line of your tirade, right? But you can't recognize that there are fundamental design differences between this game, and CS:GO. CS:GO IS NOT the end-all, be all, of competitive games. I cannot believe that you are implying that CS:GO is the end-all, be-all of competitive games.

Again, we don't agree with Paul's design proposals or decisions as much as you do, but at least we're trying to introduce some sort of balance into that equation by introducing very simple mechanics which will serve to further immerse the player in the game, while simultaneously balancing the gameplay styles and load-outs that are available to players.

As I said before, we don't WANT to eliminate sniping from this game, but we don't WANT to make it THE DOMINANT style of gameplay. In fact, ideally Mechwarrior is supposed to be a game a rock, paper, scissors. There is not supposed to be any one dominant style of play, there are supposed to be a variety of rediculous gameplay styles and load-outs--each with there own unique challenges/upsides.

If you want to take that away, and turn Mechwarrior into a unilateral, repetitive, "meta-war," then we don't want you here. I will not subscribe to P.G.I.'s "meta." I will not subscribe to yours. I will choose a variety of styles that plays best to my strengths rather then hugging a dominant playstyle for dear life and continuously defend it with bluster and tirades. All for the sake of modeling one completely different game into another.

Edited by ReXspec, 05 August 2014 - 06:23 AM.






7 user(s) are reading this topic

0 members, 7 guests, 0 anonymous users