Efficiency Vs. Particle System - A discussion on the merits of hitscan

[Dr. Flard]

As i play pyro from day to day, i find myself wondering, "Why are the flame mechanics so bad? I mean they artistically feel right, but why are they so ungodly inefficient at the task they perform.". (besides valve's lousy logic) I can't help but wonder why so many people want to hang on to such inefficient mechanics; is it because it feels right?, is it because it has existed since the game had launched and everyone is too used to it to comfort?

I believe most of the problems with the Flame part of the Flamethrower are directly related to the current particle system and that by replacing it with a hit-scan structure, it would correctly reward players for the skill involved. With the current system it is very difficult to predict where your damage is going due to particle physics and lag. This combination of the two creates an area of randomness, if you will, a sort of anomaly that prevents the user from attaining maximum dps let alone hitting with most of your damage. The fact that the damage causing particles are invisible and have a misleading representation makes using the weapon more prediction and leading than actual aiming.

The particle system seems rather, gimmicky, for lack of a better term.

Although a hit-scan model doesn't have that artistic feel to it, it has a more effective feeling to it, like your actions/skill have more impact on the outcome of a combat situation. It lacks the same style and feel that veterans of the class are used to, but it gives an immediate sense of control. Players would no longer have to worry about whether or not the particle system will screw them on a situation that they feel they should have otherwise won. (might not be the same for all people though) Obtaining maximum damage would no longer feel like a game of prediction and luck, rather a game of time and accuracy/skill. (like it should be)

I don't want to say I'd like to see the particles go, but it just isn't cutting it, the current system just causes problems now-a-days and I wouldn't mind seeing it updated to be more efficient at it's task.

What are others thoughts on this dilemma? Would you try to make the current system work, or would you prefer to adopt a more efficient structure for which the weapon should operate on?

[DRaGZ]

It might be pretty nice to just have hitscan fire with fire-like tracers that stop at a certain distance (basically how melee works). It would also need to be penetrating hitscan, since that is how the flamethrowers work.

[Dr. Flard]

It might be pretty nice to just have hitscan fire with fire-like tracers that stop at a certain distance (basically how melee works). It would also need to be penetrating hitscan, since that is how the flamethrowers work.

Oh no question, all the features that the old system had, the new system would have. (although some would have to be reconfigured) Afterburn would have to work on a system of, "Ignition is caused if exposed to flame for X seconds/X amount of damage", and the cone would have to be tweaked and tightened for skill purposes. But since it would take more aim/skill it could be granted with slightly higher dps or slightly increased range. (i'am sure the penetration would work seeing as the machina uses it)

[DRaGZ]

Naw, ignition should still be instant, DPS and range shouldn't be changed.

Actually, they should just use a cone similar to the airblast, just have it stay out there while you hold down M1.

Actually, it would also be difficult to figure out how the ramp-up/ramp-down works because the flamethrower has some funky characteristics when it comes to that.

[ungulateman]

For now, double the flame particle's speed and halve their lifetime. Easy change, would be a massive improvement with next to no work, etc.

Then, start Beta testing major overhauls to the Flame Thrower system.

Unfortunately, Valve can't sell a new Flame Thrower particle/hitscan system for 10 bucks in the Mann Co store, so we'll never see any results (beyond a boycott or something but lulz).

[Dr. Flard]

Naw, ignition should still be instant, DPS and range shouldn't be changed.

Actually, they should just use a cone similar to the airblast, just have it stay out there while you hold down M1.

Interesting, either idea works.

Actually, it would also be difficult to figure out how the ramp-up/ramp-down works because the flamethrower has some funky characteristics when it comes to that.

True.

The current system only has fall-off and it's based on particle lifetime. (never even thought about adding ramp-up to it, interesting indeed)

[DRaGZ]

I always thought the Pyro's flamethrower should reward keeping the enemy at the apex of its fire, kinda like a tipper from Marth's sword in Smash Bros. So, at near the very end of the fire's tip, it would do more damage than right up close. Not only would that promote spacing skills for the Pyro, it would also include a way of skill-indexing the act of burning things.

[Dr. Flard]

For now, double the flame particle's speed and halve their lifetime. Easy change, would be a massive improvement with next to no work, etc.

Then, start Beta testing major overhauls to the Flame Thrower system.

Unfortunately, Valve can't sell a new Flame Thrower particle/hitscan system for 10 bucks in the Mann Co store, so we'll never see any results (beyond a boycott or something but lulz).

Yeah i don't have much trust/faith in valve either, but i do believe in luck, no matter how small the chance is.

So anyway, you wouldn't mind the transition just so long as it's better than the current? That seems to be the common answer when i ask friends and acquaintances the same thing.

[Dr. Flard]

I always thought the Pyro's flamethrower should reward keeping the enemy at the apex of its fire, kinda like a tipper from Marth's sword in Smash Bros. So, at near the very end of the fire's tip, it would do more damage than right up close. Not only would that promote spacing skills for the Pyro, it would also include a way of skill-indexing the act of burning things.

<--- <3 playing as Marth

That is something i haven't heard before, and i've seen alot of strange ♥♥♥♥ from these forums. That sounds like an awesome flamethrower variant for sure, though i don't think that would be good on all flamethrowers. (I'm still thinking in the traditional sense, Closer = Hotter sorta thing)

Edit: I just don't think new players would easily get the hang of something like that, the more someone thinks they are getting a traditional flamethrower the more confused they will be when they find out it's not.

[Munkyman720]

I always thought the Pyro's flamethrower should reward keeping the enemy at the apex of its fire, kinda like a tipper from Marth's sword in Smash Bros. So, at near the very end of the fire's tip, it would do more damage than right up close. Not only would that promote spacing skills for the Pyro, it would also include a way of skill-indexing the act of burning things.

Roy is better, and I hate Nintendo for leaving him out of Brawl :\

But I agree with Flard that this could be an interesting mechanic for a new Flamethrower.

[DRaGZ]

Well, if you're dealing with real-life flamethrowers, the flames are actually much more potent near the end of the stream because it's been oxygenated enough for the fire to get really hot. In fact, in many WWII flamethrowers can't even ignite the entire stream, and the first half or so of fuel pouring out actually isn't ignited.

So that would be an excuse to justify the mechanics, if it even needs one.

[4812622]

Roy is better, and I hate Nintendo for leaving him out of Brawl :\

Ike fights for his friends. Roy got nothing on that.

[DRaGZ]

Roy is better, and I hate Nintendo for leaving him out of Brawl :\

But I agree with Flard that this could be an interesting mechanic for a new Flamethrower.

I actually mained Roy, but Marth was just so clearly better it was ridiculous.

[Munkyman720]

Ike fights for his friends. Roy got nothing on that.

Ike also fights for noobs who spam his special attacks.

I actually mained Roy, but Marth was just so clearly better it was ridiculous.

Well I main Mario and Pikachu, but between those two I always liked Roy better.

[Dr. Flard]

Well, if you're dealing with real-life flamethrowers, the flames are actually much more potent near the end of the stream because it's been oxygenated enough for the fire to get really hot. In fact, in many WWII flamethrowers can't even ignite the entire stream, and the first half or so of fuel pouring out actually isn't ignited.

So that would be an excuse to justify the mechanics, if it even needs one.

Traditional for tf2, i mean.

Ike also fights for noobs who spam his special attacks.

Ike was so damn slow you had to spam them to compete. (ooooh did i spam his up-B move)

"Efficiency Vs. Particle System - A discussion on the merits of hitscan and smash brothers"

[4812622]

Ike also fights for noobs who spam his special attacks.

Because they're ♥♥♥♥ing awesome. Also, Counter spam takes skill.

[DRaGZ]

Brawl is all spam anyway.

♥♥♥♥♥♥♥it Sakurai...

[Munkyman720]

Ike was so damn slow you had to spam them to compete. (ooooh did i spam his up-B move)

Yeah, which is why he's crap.

Because they're ♥♥♥♥ing awesome. Also, Counter spam takes skill.

lol

Brawl is all spam anyway.

♥♥♥♥♥♥♥it Sakurai...

Yeah, pretty much. Melee ftw.

[DRaGZ]

The sad part is that Brawl isn't broken because of the roster. It's broken because of the ♥♥♥♥ty game mechanics Sakurai insisted on using. If the core mechanics were fixed, Metaknight would not be such an incredibly powerful force and the game would be just so much more fun to play.

[Munkyman720]

The sad part is that Brawl isn't broken because of the roster. It's broken because of the ♥♥♥♥ty game mechanics Sakurai insisted on using. If the core mechanics were fixed, Metaknight would not be such an incredibly powerful force and the game would be just so much more fun to play.

Very much agreed.

[Xzalander]

The big problem with making the Flamethrower hitscan is you no longer have the side effect of the flamethrower range being shorter while moving forward and longer when moving backward.

You end up with a fixed range cone, regardless of movement direction or speed.


Edit : Also Kirby says Hi.

Over

and

Over

and

Over

and

...

[DRaGZ]

The big problem with making the Flamethrower hitscan is you no longer have the side effect of the flamethrower range being shorter while moving forward and longer when moving backward.

You end up with a fixed range cone, regardless of movement direction or speed.


Edit : Also Kirby says Hi.

Over

and

Over

and

Over

and

...

I don't think that is necessarily a bad thing, to have a fixed distance.

Also, I actually preferred spinning over and over again, when characters used to have actual turning animations in the original Smash Bros. Jigglypuff in particular hops in a circle adorably.

[Dr. Flard]

The big problem with making the Flamethrower hitscan is you no longer have the side effect of the flamethrower range being shorter while moving forward and longer when moving backward.

You end up with a fixed range cone, regardless of movement direction or speed.


Edit : Also Kirby says Hi.

Over

and

Over

and

Over

and

...

Whats wrong with having a fixed range? I would think it would be beneficial to accurately know how far your main weapon reaches. (Considering the finite range and all)

Haven't you looked at the Phlog's visuals and wished They were accurate? (though the weapon is still garbage)

Speaking of Kirby, who here abused his dive-bomb/pile-driver throw? (it was the forward throw i believe) Though it always annoyed me when i went for a suicide edge throw and they broke out of it while just i plummeted full speed to my death.

[Xzalander]

Whats wrong with having a fixed range? I would think it would be beneficial to accurately know how far your main weapon reaches. (Considering the finite range and all)

Haven't you looked at the Phlog's visuals and wished They were accurate? (though the weapon is still garbage)

Speaking of Kirby, who here abused his dive-bomb/pile-driver throw? (it was the forward throw i believe) Though it always annoyed me when i went for a suicide edge throw and they broke out of it while just i plummeted full speed to my death.

I don't think it is wrong to have a fixed range on the flamethrower, it just means it would require some extra rebalancing, especially as it would make W+M1ing more popular.

As for Kirby my favorite was to abuse his Throw move by initiating it, breaking it myself (which pushes them out), Shapeshift Bombing them, breaking out after contact (pushes them out again) then B-Up sword slashing them down. Sure I go with them, but its just funny to hassle people with combos like that.

My favorite non suicidal, is to throw them off the arena and sit there suckin em up and spitting em back out.

[Dr. Flard]

I don't think it is wrong to have a fixed range on the flamethrower, it just means it would require some extra rebalancing, especially as it would make W+M1ing more popular.

Well it's not whether it would take more time and effort to balance, it's whether you would prefer the merits associated with hitscan or would you rather try and get the particle system to stop being so damn inefficient.

And how does requiring more aim encourage w+m1? Why do people consider on the whole that using the pyro's main weapon is somehow a noob tactic? Heavy's minigun is a hell of a lot easier to aim and obtain greater amounts of damage (that cone is massive) with than any incarnation of the flamethrower, and the only reason why most don't consider it as such is because the heavy is naturally slow. (which isn't a damn problem if you know how to position yourself)

[Incarnati0n]

I don´t think they´ll ever change the flamethrower particles.

I still have a tiny bit of hope for a lightninggun-esque unlock though

[Dr. Flard]

I don´t think they´ll ever change the flamethrower particles.

Neither do I, but thats not the point.
Well it's not whether it would take more time and effort to balance, it's whether would you prefer the merits associated with hitscan or would you rather try and get the particle system to stop being so damn inefficient.


I still have a tiny bit of hope for a lightninggun-esque unlock though

Which i would assume that it was based on a hitscan model. So would you prefer something like that, or increasing the particle reliability. (if that at all is even possible)

[Incarnati0n]

Which i would assume that it was based on a hitscan model. So would you prefer something like that, or increasing the particle reliability. (if that at all is even possible)

Particle reliability is more likely to happen (although still highly unlikely). Keeping the flamers as close to how they are now, but with improved hitreg would be optimal.

So improved particles for the current ones, with a lightninggun unlock as a side dish.

PS: Bring back melee samus, brawl's samus was horrid

[DRaGZ]

Particle reliability is more likely to happen (although still highly unlikely). Keeping the flamers as close to how they are now, but with improved hitreg would be optimal.

So improved particles for the current ones, with a lightninggun unlock as a side dish.

PS: Bring back melee samus, brawl's samus was horrid

Brawl's version of everyone was horrid. The only characters that really improved was Kirby and Pikachu.

[Incarnati0n]

Brawl's version of everyone was horrid. The only characters that really improved was Kirby and Pikachu.

True, true. Although I always felt like melee was often too hard to get into, especially with the L-cancelling stuff. You had to either stay awful forever or break through this granite skill-barrier, which simply was too much for me when wanting to play casual-pro

I mean look at this (http://www.youtube.com/watch?v=vXgpGBbh5r8)

Brawl on the other hand took it 7 steps too far, cutting out every single advanced technique and making it possible to reach the skill ceiling of any class in a matter of weeks

[Ls777]

ITT: people who don't know bout my brawl minus

[Incarnati0n]

ITT: people who don't know bout my brawl minus

I've seen it but never really bothered with it. Do you play that on an emulator or something?

[DRaGZ]

True, true. Although I always felt like melee was often too hard to get into, especially with the L-cancelling stuff. You had to either stay awful forever or break through this granite skill-barrier, which simply was too much for me when wanting to play casual-pro

I mean look at this (http://www.youtube.com/watch?v=vXgpGBbh5r8)

Brawl on the other hand took it 7 steps too far, cutting out every single advanced technique and making it possible to reach the skill ceiling of any class in a matter of weeks

You can master Brawl in a few days. Seriously, the apex of technical skill can be achieved in about a few days, the rest comes to pure fundamentals and playing absurdly defensively.

Melee actually isn't that fast compared to, say MvC2 or ST, where some of the most important combos in the game can come down to single-frame links or really strange setups that involve manipulating how the screen moves. If you see how fast the button inputs in those games are, it's ridiculous.

I'd say the original had the best mix. The lack of an air-dodge made it so getting an opponent in the air was a big deal and made juggles a huge part of the game. L-cancelling was butt-easy, so it wasn't something you needed to master as much as it was something to be aware of. There was no wavedashing, but it wasn't really needed as movement options were actually pretty good in that game. The mechanics favored the approacher, which made stupid all-defense shenanigans impossible. And then combos were amazing in that game, most characters could 0-death on nearly any stage.

[Nirruden]

This subject comes up fairly regularly. I will summarize my thoughts:

1. A hitscan "laser gun" with more, and fixed, range would be interesting to test. If the additional range causes an issue, it might be interesting to test a flamethrower that has to score several hits to ignite (gotta shock 'em more to get 'em to burn) but I think losing ignition, even marginally, is probably more bad for the pyro than it first appears.

2. Doubling speed and halving lifetime might reduce the pyro's ability to influence the motion of his own projectiles (frex, sweeping a flamer to push particles over a wall or across an elevated ledge to ignite targets otherwise unreachable) in cases where the motion of the pyro causes the particles to travel more laterally when aimed in a different direction. I think I'd have to play with this one. I'm not sure how it would affect the situations where the scout is dancing in the flame graphic but never gets ignited.

If it wouldn't be entirely server crushing* you could go on to alter the geometry of the fireboxes (long and narrow, like rockets) and/or also alter the number of them upwards (frex, double number per second, halve damage dealt per hit) to better model the "stream of fire" effect, but it's more computationally expensive to create extra projectiles - more so than hitscan which are simple ray traces. Something like this would have a net damage change of zero , but increased particle count should help with laggy situations where quick rotation of the flamer does not produce a smooth sweep. (*:What I mean is the additional overhead of the additional objects. @34 pyros, double particle count = 22.5 * 34 more particles per second added to the game world and tracked fro collision checking and LOS traces on impacts.)

3. I think what causes most of the "swept flamer over target and didn't ignite" issues I see is the addition of effectively infinite player rotation speed to latency. The flame graphic may be continuous, but the particles do not behave as a truly continuous stream when spread out over a surface of rotation. Combine this with the fact that particles are not latency-corrected in any way and leads to a faulty expectation of the flame effects when the aim vector is quickly altered. This in no way impacts my desire to see the effects of this sort of change (or unlock) in play.

[Incarnati0n]

You can master Brawl in a few days. Seriously, the apex of technical skill can be achieved in about a few days, the rest comes to pure fundamentals and playing absurdly defensively.

Melee actually isn't that fast compared to, say MvC2 or ST, where some of the most important combos in the game can come down to single-frame links or really strange setups that involve manipulating how the screen moves. If you see how fast the button inputs in those games are, it's ridiculous.

I'd say the original had the best mix. The lack of an air-dodge made it so getting an opponent in the air was a big deal and made juggles a huge part of the game. L-cancelling was butt-easy, so it wasn't something you needed to master as much as it was something to be aware of. There was no wavedashing, but it wasn't really needed as movement options were actually pretty good in that game. The mechanics favored the approacher, which made stupid all-defense shenanigans impossible. And then combos were amazing in that game, most characters could 0-death on nearly any stage.

Sadly I was too young when SSB came out to play it on a serious level. The only thing I remember about it is that the throws were awesome.

On this subject, a friend and I have been searching for a fighting game we could play semi-serious without having to get too technical, would you have any suggestions? More technical than brawl, but less than melee

[DRaGZ]

Sadly I was too young when SSB came out to play it on a serious level. The only thing I remember about it is that the throws were awesome.

On this subject, a friend and I have been searching for a fighting game we could play semi-serious without having to get too technical, would you have any suggestions? More technical than brawl, but less than melee

Well, if you want to stay in the realm of Smash Bros.-style games...you'd probably want to go to the original Smash Bros. >_>

Seriously, though, it's out on the Wii Virtual Console, can use the Gamecube controllers (which are a less of a pain on your thumbs than the original N64 control sticks), and there's really no other game that's been able to really match that original style of gameplay.

Personally, I'm really interested in Skullgirls right now:

http://skullgirls.com/

It's a spiritual successor to MvC2 (because MvC3 really wasn't) and it's being made by an indie company and headed by Mike Z, a guy who's been in the competitive fighting game community for nearly two decades. He's intentionally making inputs and whatnot easy to do. For instance, there is nothing more complicated than a 360 motion, which the game recognizes when you try it so you can't accidentally jump when doing it. There's also no double motion inputs for supers, just double button presses, which makes pulling off supers very easy. While each character has a specialty, they all have a versatile-enough toolbox to have an answer for any situation (albeit, not necessarily always the most feasible or safe answer). There's also a lot of awesome little touches that, for some reason, haven't been thought of by other fighting game developers, like requiring 15 frames of holding down Start to pause the game (so you can't accidentally pause the game), counter supers that don't flash so the opponent can't react to it, an infinite prevention system, the ability to block simultaneous high and low blocks at the same time, optimization for pad players, etc. It's even going to come with a comprehensive tutorial that teaches you the fundamentals of 2D fighting games, like how to block properly, how to set up combos with chains and links, etc.

It's not out yet, though, apparently the release date is April 4. It'll be available on PSN and XBLA, and perhaps PC in the future.

EDIT: Man, I miss Captain Falcon's chokeslam. That was the move that made me want to main Captain Falcon forever, that chokeslam was such a baller setup move. And though it went away in Melee, he got so many other sick setups into knee in that game that it didn't really matter. And then Brawl made him so incredibly bad. Loss of aerial momentum when jumping is the worst thing ever.

[Ls777]

I've seen it but never really bothered with it. Do you play that on an emulator or something?

you throw it on an sd card and load it through the stagebuilder, works on any wii
(or use homebrew which loads faster if you have it)

[Kurohagane]

I actually suggested something like this:

http://forums.steampowered.com/forums/showthread.php?t=2608724

[shinquickman]

What is it about the Pyro that always sparks fighting game conversations around here? :confused:

Anyway, I don't particularly agree with a LG type flamethrower. It worked in Quake Live because everyone basically ran in scout speed, but that's not the case in TF2. Doing this would just turn the Pyro into a mini-heavy. Same issue I have with speeding up the flame particles.

I do agree with a tightening of the flame particle spread. And I also liked the idea Dragz brought up about giving the flamethrower a "sweet spot" at the tip of the flames. Combine the sweet spot dealy with a longer ranged flamethrower, and I actually think the Pyro would be suitably combat-capable without being brain-dead easy to use.

[DRaGZ]

What is it about the Pyro that always sparks fighting game conversations around here? :confused:

'Cause the Pyro actually makes the game feel like a first-person fighting game in three dimensions.

[Limeaide]

A way could use a hitscan flamethrower system without it being too conspicuous when your flames hit the target the moment you click is to have the range start at 0 and increase at the speed of the particles, if anyone understands what I'm trying to say here.

[Archmage MC]

I've seen "Double particle speed, half particle life" thrown out a lot, but how well would that actually work? sure it would be more consistant, but wouldn't the damage be lower due to the way the flamethrower works? And what about server load? Though I dunno if the particle system uses more resources than pipes or needles or not though.


If you guys wanna play a good SSB game, Project M (Brawl mod) should be coming out later this month. Give it a look. Though getting the mod on your Wii will take an SD card and some work.

And all you guys playing Marth, Roy, and Ike? Psh. Wario is where its at. Rides a motorcycle, about as fat as the heavy, and with aerial maneuverability that would make the scout green with envy. (though release grabs SUCKED, but hey, played well, he was the best matchup vs metaknight besides a ditto match.)


Or Captain Falcon, but thats Captain Falcon, and hes in a league all his own.

[DRaGZ]

I've seen "Double particle speed, half particle life" thrown out a lot, but how well would that actually work? sure it would be more consistant, but wouldn't the damage be lower due to the way the flamethrower works? And what about server load? Though I dunno if the particle system uses more resources than pipes or needles or not though.


If you guys wanna play a good SSB game, Project M (Brawl mod) should be coming out later this month. Give it a look. Though getting the mod on your Wii will take an SD card and some work.

And all you guys playing Marth, Roy, and Ike? Psh. Wario is where its at. Rides a motorcycle, about as fat as the heavy, and with aerial maneuverability that would make the scout green with envy. (though release grabs SUCKED, but hey, played well, he was the best matchup vs metaknight besides a ditto match.)


Or Captain Falcon, but thats Captain Falcon, and hes in a league all his own.

I really don't like what Project M's goal is, that is to make Brawl like Melee. Brawl has its own feel and its own mechanics. They're broken, but only because they're poorly implemented and some of the choices were just downright terrible.

If I were to suggest anything mod, it would be the Balanced Brawl mod. It keeps the game as pretty much Brawl and doesn't drastically change too much about it, just the stuff that's broken. It's also already been finished for a long time. Project M is like...giving different characters different moves and stuff, at that point you might as well just build a new game.

[Archmage MC]

I really don't like what Project M's goal is, that is to make Brawl like Melee. Brawl has its own feel and its own mechanics. They're broken, but only because they're poorly implemented and some of the choices were just downright terrible.

If I were to suggest anything mod, it would be the Balanced Brawl mod. It keeps the game as pretty much Brawl and doesn't drastically change too much about it, just the stuff that's broken. It's also already been finished for a long time. Project M is like...giving different characters different moves and stuff, at that point you might as well just build a new game.

It is a new game honestly, but hey, they can do that when they've almost turned brawl into the next MUGIN, seriously you can do anything you want nowadays with the brawl engine it seems. Don't think you can have custom sounds longer or bigger than current sound files though.

And at least their goal of making a game with no tier list and trying to stick with that, and having some actual passion as they're getting squat for this can turn out a good thing.

[DRaGZ]

It is a new game honestly, but hey, they can do that when they've almost turned brawl into the next MUGIN, seriously you can do anything you want nowadays with the brawl engine it seems. Don't think you can have custom sounds longer or bigger than current sound files though.

And at least their goal of making a game with no tier list and trying to stick with that, and having some actual passion as they're getting squat for this can turn out a good thing.

That's the folly of their goal, though. You cannot have a fighting game that never develops a tier list. Even the most well-balanced fighting games in the entire world have tier lists, you're always going to have tier lists as long as you have characters that are just fundamentally different from each other.

If it ever gets a large enough community for there to be a competitive scene, and that's a really big if, you're going to get a tier list no matter what.

[Dr. Flard]

I've been ungodly busy this week.

Anyway...

This subject comes up fairly regularly. I will summarize my thoughts:

1. A hitscan "laser gun" with more, and fixed, range would be interesting to test. If the additional range causes an issue, it might be interesting to test a flamethrower that has to score several hits to ignite (gotta shock 'em more to get 'em to burn) but I think losing ignition, even marginally, is probably more bad for the pyro than it first appears.

While it does seem like losing instant ignition is a penalty of sorts, it really doesn't affect things that much. You would either catch them in your flames for such a time and cause x amount of damage or you wouldn't. Since the flamethrower would be hitting instantly the moment you slide the mouse over anyone, i think at that point ignition would be almost too ridiculously easy to obtain. (Spies would rage pretty hard, and unfortunately it would be fairly justifiable)

Now I am not saying it has be a lot, this time/damage could be very small, it just needs to be enough to verify whether you got a solid hit/s or you just barely glanced them with a single iota of damage.

2. Doubling speed and halving lifetime might reduce the pyro's ability to influence the motion of his own projectiles (frex, sweeping a flamer to push particles over a wall or across an elevated ledge to ignite targets otherwise unreachable) in cases where the motion of the pyro causes the particles to travel more laterally when aimed in a different direction. I think I'd have to play with this one. I'm not sure how it would affect the situations where the scout is dancing in the flame graphic but never gets ignited.

This is a mixed bag. I have played around with the game and modded the flamethrower to try and test a few things. I have loosely tested double particle speed and half particle time. It's better than current, but if the server decides that you hit nothing with your particles, you still hit nothing. (Still has annoying issues, but better)

If it wouldn't be entirely server crushing* you could go on to alter the geometry of the fireboxes (long and narrow, like rockets) and/or also alter the number of them upwards (frex, double number per second, halve damage dealt per hit) to better model the "stream of fire" effect, but it's more computationally expensive to create extra projectiles - more so than hitscan which are simple ray traces. Something like this would have a net damage change of zero , but increased particle count should help with laggy situations where quick rotation of the flamer does not produce a smooth sweep. (*:What I mean is the additional overhead of the additional objects. @34 pyros, double particle count = 22.5 * 34 more particles per second added to the game world and tracked fro collision checking and LOS traces on impacts.)

While i do like where you are going with that, i'd have to say simpler is better in this case. (But by no means would i not like to see something like that in action)

3. I think what causes most of the "swept flamer over target and didn't ignite" issues I see is the addition of effectively infinite player rotation speed to latency. The flame graphic may be continuous, but the particles do not behave as a truly continuous stream when spread out over a surface of rotation. Combine this with the fact that particles are not latency-corrected in any way and leads to a faulty expectation of the flame effects when the aim vector is quickly altered. This in no way impacts my desire to see the effects of this sort of change (or unlock) in play.

All i can say is that i hate a large portion of the factors that go into the particle hit detection. Even when understanding the innermost workings of the weapon, i still shake my head in frustration when that point blank scout dances in front of me not on fire!

What is it about the Pyro that always sparks fighting game conversations around here? :confused:

I bet it's the Fire.

Anyway, I don't particularly agree with a LG type flamethrower. It worked in Quake Live because everyone basically ran in scout speed, but that's not the case in TF2. Doing this would just turn the Pyro into a mini-heavy. Same issue I have with speeding up the flame particles.

It doesn't have to be the same, it wouldn't be a lightning gun, at least not in the quake sense.

Technically, on a fundamental level the pyro is what one would call a mini-heavy. Since the two classes are obviously different i just don't understand why having a similar firing mechanic would suddenly cause the two to blur the lines between each other. (even though on a theoretical level they have been doing it for years)

I do agree with a tightening of the flame particle spread. And I also liked the idea Dragz brought up about giving the flamethrower a "sweet spot" at the tip of the flames. Combine the sweet spot dealy with a longer ranged flamethrower, and I actually think the Pyro would be suitably combat-capable without being brain-dead easy to use.

Tightening the cone just by itself probably wouldn't do much, the most i could see it doing would be increasing the particle hit percentage at the end of the flamethrowers range, but all the other problems would still put a damper on the overall effectiveness of it.

Dragz's idea was pretty interesting.

I've seen "Double particle speed, half particle life" thrown out a lot, but how well would that actually work? sure it would be more consistant, but wouldn't the damage be lower due to the way the flamethrower works? And what about server load? Though I dunno if the particle system uses more resources than pipes or needles or not though.

Well the damage wouldn't be lower, because since the damage is based on particle lifetime the particles would fizzle out faster, but they would move approximately the same distance because of the doubled speed. (i've modded the flamethrower to work that way in the past, but i didn't really check to see if it was 100% accurate)

As for server load, i don't see why it wouldn't be able to handle it. (servers can take a lot of abuse)

[TheDL]

flame particles are fine

[Dr. Flard]

flame particles are fine

Fine for what it is, Yes.

Efficient at what they do, not really.

[Strings]

What? No Gannondorf mains?

[TheDL]

They're efficient once you get used to it.

[DRaGZ]

What? No Gannondorf mains?

I mained Ganondorf in both Melee and Brawl.

Basically:

SSB: Captain Falcon
Melee: Captain Falcon, Roy, Young Link, Ganondorf, Mewtwo
Brawl: R.O.B., Ganondorf

[Dr. Flard]

They're efficient once you get used to it.

Getting used to an inefficient system doesn't make said system anymore efficient. (You still lose damage to lag, particle physics and zero visual aid no matter how "used" to the system you may be)

You can minimize how much the above three affect the outcome, but they still affect every outcome to a noticeable degree.