I don't like wall jumping - brainstorm
schkorpio
I can mspaint Join Date: 2003-05-23 Member: 16635Members
I can mspaint Join Date: 2003-05-23 Member: 16635Members
<div class="IPBDescription">as a way to gain speed</div>what other options do we have, apart from bunny-hopping and strafe-jumping?
Please post your suggestions.
(i'm wondering if simply having the skulk move faster when on walls and cielings would be easier, but then I like the skill factor of strafe jumping and bunnyhopping too).
Please post your suggestions.
(i'm wondering if simply having the skulk move faster when on walls and cielings would be easier, but then I like the skill factor of strafe jumping and bunnyhopping too).
Comments
Skill based movement would be soooo gooood
As for alternatives, I don't know. Only other movement based skill is skiiing (tribes) but that would not work in NS
<!--quoteo(post=1975421:date=Sep 10 2012, 07:08 PM:name=got_quail)--><div class='quotetop'>QUOTE (got_quail @ Sep 10 2012, 07:08 PM) <a href="index.php?act=findpost&pid=1975421"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->Why don't you like wall jumping?<!--QuoteEnd--></div><!--QuoteEEnd-->
The thing about wall jumping, is that it is pretty binary, you are either good at it, or terrible at it. Practicing will make you better at it, but there isn't a nice smooth skill curve to it. Secondly, you get a sort of fixed benefit from it, the difference between, a good wall jump and a bad one, is too drastic.
The other thing that urks me about wall jump, is that it emphasizes the most noob unfriendly part of bhop, the tight timing requirement. Timing requirements, don't really fit in FPS ( imo) and they tend to introduce binary type benefits. Either you've practiced enough to hit the timing consistently, or you haven't. Most competitive players advocated adding in 'quake jumping' as a way to remove the timing requirement of bhop. (quake jumping is where pressing and holding the jump button in air allows you to jump instantly when you hit the ground) The only reason people had bunny hoping scripts was to make this timing more lax. Added to this problem is there isn't really good feedback for when you need to hit the jump button, so learning and using wall jump is frustrating.
And finally, wall jump is also silly because it causes people to do a really gimpy version of bhop any way because you have to spam jump to keep your speed up after you do a few successful wall jumps. So you still end up with people mashing jump, getting frustrated about timings, and needing to grind to get good at something, but you don't get any of the fluid player expression you got with bhop.
In NS2 it appears you have mostly wet non-directional sound even for unoccluded sound sources; I cannot hear any hint of a HRTF, only the most basic directional audio. Occluded sound sources are just a muffled mess.
Bunnyhopping towards a decent player just doesn't work unless they are distracted; they won't just hear you a mile away, they'll hear that you are bunnyhopping from the rythmical *thwap, thwap, thwap...* noise and which way you are moving. The smooth arcs in which bunnyhoppers move are really easy to track. Bunnyhopping requires space; the faster you go, the larger the turning radius if you want to keep your speed; you need to plan your way around obstacles. Failures that cause you to lose speed tend to be soft failures; you make a mistake in timing or turning radius and you slow down, but you don't stop dead in your tracks.
The NS2 sound system doesn't allow good directional sound over long distances(<b>by choice!</b>) and wall jumping is not so simple and rythmical that you can tell if a player is wall jumping or just two skulks occassionally jumping. Wall jumping emphasises sliding along walls, where most of the mess is and it's very easy to get snagged by something; when you get snagged you tend to lose almost all of your speed. Wall jumping makes you move in a spastic, unpredictable and harder to hit fashion.
The funny thing is that I can't come up with that many. That's part of the reason why the whole "we're going to build a bhop replacement" was treated with such worry.
Unreal Tournament series has some jump trickery in it and Tribes has the skiing mechanics, but I don't think they really work in NS styled cramped environment and melee related gameplay. Tribes system requires large outdoor arenas and - as far as I can tell - the UT system is pretty strictly ranged combat in very arenalike surroundings. After that it's mostly Half Life and Quake related movement quirks.
I don't think it's impossible to build up a completely new system, but it's not something you paste on top of a otherwise complete game. Everything starts from the way the engine handles the player movement and so on.
I always thought q2 / q3 cpma / warsow had more skill based movement than vq3.
One way of designing a skilled based movement is to try to be smart and think really hard about mechanics. Another one is just to do a complex and open movement system and to let the players exploit it.
For example the game makes sure that you walk down slopes at the same speed than walking them up. You could let players gain speed on downward slopes and see if they can use it at their advantage. You could also replace the rigid collision with some muscle model (spring plus damper) and see what it gives.
I'd like to see somewhat looser timing requirements and somewhat better boost from what it is now.
IMO....
The easy way too fix this is to remove the "timing" system all together, this should cause the learning curve to be linear (the faster you can walljump = sooner you can wall jump again) and not so exponential.
The faster you chain jumps the faster you will accelerate so the "skill ceiling" should not change, but it will be quite a bit more accessible for new players.
Anywho- thats my 2 cents.
The easy way too fix this is to remove the "timing" system all together, this should cause the learning curve to be linear (the faster you can walljump = sooner you can wall jump again) and not so exponential.
The faster you chain jumps the faster you will accelerate so the "skill ceiling" should not change, but it will be quite a bit more accessible for new players.
Anywho- thats my 2 cents.<!--QuoteEnd--></div><!--QuoteEEnd-->
This actually sounds really good!
The reason I don't like the current wall jump, is that it just doesnt make any sense to me, and i think it is because of the binary nature of it.
one way to reintroduce the "skill" but keep the skill curve smooth might be to make the speed boost based off launch angle.
<img src="http://img35.imageshack.us/img35/8914/luanch.png" border="0" class="linked-image" />
Maybe have a sweet spot where jumping 120 degrees launch angle results in the highest speed boost, and any further away from that gives diminishing returns. So if you need to drastically change direction, you might loose your bonus speed. You would also loose a lot of speed if you hit a wall straight on, instead of obliquely because you wouldn't be able to get a launch angle bigger than 90 degrees.
Makes an interesting balance / judgement call between speed vs predictability because if you vary your trajectory you will slow down. Also makes it so just mashing jump won't end up increasing your speed too much.
Could potentially reduce base skulk movement a bit to compensate for wall jumping being much more accessible.
I think the current incarnation of wall jumping is the fairest (to both skulk and marine) and "best" that it has ever been throughout the NS2 developments.
Though I think the skulks sluggish feel is more because a base marine can out run you with sprint than him being actually slow.
-Remove the timing system.
-Speed gains should be on jump, not on landing.
And the big one
-add a jumping (push) on walls, like you would have when your jumping off the ground
Why do this, well basically wall-too the same wall jumping is pretty unintuitive, and a little too easy.
This would make it reallyREALLY hard, and incentives thinking about your jumps. You could do things where you uses walls to basically get free leaps at your pray, or just fly around there head. Overall I think this would add a lot of depth to wall jumping where it lacks.
One way of designing a skilled based movement is to try to be smart and think really hard about mechanics. Another one is just to do a complex and open movement system and to let the players exploit it.
For example the game makes sure that you walk down slopes at the same speed than walking them up. You could let players gain speed on downward slopes and see if they can use it at their advantage. You could also replace the rigid collision with some muscle model (spring plus damper) and see what it gives.<!--QuoteEnd--></div><!--QuoteEEnd-->
I really enjoyed Warsow's spin on movement mechanics. Made them integral, fluid, and a part of the tutorial was dedicated to it. So much fun to vault your way around maps.
I might be in the minority here, but maybe the cleanest solution is remove skill-based movement? Bhop has the problem of the timing requirements and it's unintuitive to learn about, requiring external sources. Wallhop maintains both of these problems. Trying to figure out a better system is difficult. So, perhaps stop investing in it and just remove it?
quoting this in the hope that someone "looks at warsow"
if you're interested in fps movement, that is one of the games you need to see in order to inform yourself
it does walljumping better than ns2, and it also has pretty much every other fps movement mechanic rolled into one
The main point to take from this thread is that the current skulk movement in NS2 is not as good as it should be and that there are ways to improve it that shouldn't require too much work.
EDIT: Give bunnyhop to marines too