<!--quoteo(post=1821207:date=Jan 4 2011, 01:32 AM:name=Chris0132)--><div class='quotetop'>QUOTE (Chris0132 @ Jan 4 2011, 01:32 AM) <a href="index.php?act=findpost&pid=1821207"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->If marines are hard to hit you could just add a wider spread to melee attacks.
After all, it's kind of silly to have them work only at one point directly in front of you when you're swinging claws/chomping huge teeth at people.<!--QuoteEnd--></div><!--QuoteEEnd--> +1
Well since from what I gather by some of the latter tweets they are making the marines cone of fire virtually non existant, more akin to NS1. I'd say that if a marine is jumping up and down like crazy his cone of fire should grow a little, making it slightly harder to hit something.
<!--quoteo(post=1821688:date=Jan 5 2011, 03:04 PM:name=Cruor)--><div class='quotetop'>QUOTE (Cruor @ Jan 5 2011, 03:04 PM) <a href="index.php?act=findpost&pid=1821688"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->Well since from what I gather by some of the latter tweets they are making the marines cone of fire virtually non existant, more akin to NS1. I'd say that if a marine is jumping up and down like crazy his cone of fire should grow a little, making it slightly harder to hit something.<!--QuoteEnd--></div><!--QuoteEEnd-->
Spread/cone of fire is related to precision, while aim is related to accuracy.
With high accuracy, you'd also desire high precision. You would kill a skulk fastest, and conserve the most ammunition. With low accuracy - due to the quick motion of the target, the close quarters and the marine's aiming ability - some imprecision is helpful, because although it means not every one of your shots will hit, ideally enough of the shots will hit to kill the skulk - but you will waste ammunition. There has to be a balance: the average player's accuracy, versus the appropriate precision. Not every player has perfect accuracy, so you mustn't have perfect precision. Currently though, the NS2 rifle has too little precision, and isn't balanced well against the fire rate of the rifle / size of the clip (and of course, how those relate to the dps of the skulk).
I guess the best way to balance would be to consider a) probability distributions and cone of fire (perhaps a normal distribution with regards to radius from centre of the cone of fire would work best - i.e. most shots fire through the centre of where you aim, but less shots hit fire through a position away from the centre of where you aim), b) a "most likely distance" between rifle and skulk, c) the skulk's "shape" at its "most likely orientation" (facing you? facing away? side-on?) and "most likely position" (due to the movement of the skulk and the aim of an average player). d) the fire rate of the rifle and/or the size of a magazine, and how those relate to the "most likely" dps of a skulk At that "most likely" distance the probability distribution/density should overlap with the skulk shape almost perfectly, and there should be enough shots that actually hit to kill the skulk, before it kills you. There's a lot to consider. It's a science.
Another consideration you could have is to have the probability density/distribution (and cone of fire) tighter with less shots - so you could shoot a skulk at a distance if you did bursts of fire and they would land almost exactly through the centre of your crosshair; but the probability density/distribution (and cone of fire) flatter and wider with more shots - so you'd spray and pray (which is what tends to happen anyway) at close distances and they would land roughly equally around the cone of fire, or maybe even less towards the centre. Of course, if you have impeccable aim (accuracy), you'd still do bursts of fire at close distances, to conserve ammunition and maintain precision.
It'd be easier to show this all graphically, but I hope it gets through to you anyway. Edit: I tried to illustrate accuracy, precision, and different probability densities (in a very extreme sense): <img src="http://img411.imageshack.us/img411/6135/accandprec.png" border="0" class="linked-image" />
what is the circle on the bottom right meant to illustrate? if you maintain your point of aim and your cone of fire increases, your grouping is more spread out, but not in a donut shape.
See the graphs above each cone of fire - showing different probability density distributions for the same cone of fire. They <b>do not</b> represent increasing cones of fire, they are all <b>the same cone of fire</b> (I've said so), but with different probability distributions. The first from the left: normal distribution (bell curve) - more shots grouped towards the centre, less as you tend to the outside. Middle: flat or even distribution - shots even around the cone of fire.* Furthest on the right: inverse bell curve - shots tend away from the centre of the cone of fire. *The four accuracy/precision distributions assume flat or even distribution, but with changing cones of fire (precision). Pretty sure that was all illustrated and explained already. The reason I illustrated these is to show graphically (and thus more intuitively) what I've <b>discussed in my post</b>.
<!--quoteo(post=1820305:date=Dec 30 2010, 07:22 PM:name=peregrinus)--><div class='quotetop'>QUOTE (peregrinus @ Dec 30 2010, 07:22 PM) <a href="index.php?act=findpost&pid=1820305"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->The devs could stop people firing while jumping, or create a jumping fatigue meter.<!--QuoteEnd--></div><!--QuoteEEnd-->
<!--quoteo(post=1822047:date=Jan 7 2011, 07:36 AM:name=Harimau)--><div class='quotetop'>QUOTE (Harimau @ Jan 7 2011, 07:36 AM) <a href="index.php?act=findpost&pid=1822047"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->See the graphs above each cone of fire - showing different probability density distributions for the same cone of fire. They <b>do not</b> represent increasing cones of fire, they are all <b>the same cone of fire</b> (I've said so), but with different probability distributions. The first from the left: normal distribution (bell curve) - more shots grouped towards the centre, less as you tend to the outside. Middle: flat or even distribution - shots even around the cone of fire.* Furthest on the right: inverse bell curve - shots tend away from the centre of the cone of fire. *The four accuracy/precision distributions assume flat or even distribution, but with changing cones of fire (precision). Pretty sure that was all illustrated and explained already. The reason I illustrated these is to show graphically (and thus more intuitively) what I've <b>discussed in my post</b>.<!--QuoteEnd--></div><!--QuoteEEnd--> yeah your explanation confused me. ive fired plenty of machine guns and the spread never ends up like circle 3. what situation are you suggesting causes a grouping like a donut? from my experience the the larger your cone, the more its spread out
or are we throwing away realism and basing this of a computer game cone of fire?
<!--quoteo(post=1822331:date=Jan 7 2011, 09:02 PM:name=Loey)--><div class='quotetop'>QUOTE (Loey @ Jan 7 2011, 09:02 PM) <a href="index.php?act=findpost&pid=1822331"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->or are we throwing away realism and basing this of a computer game cone of fire?<!--QuoteEnd--></div><!--QuoteEEnd--> Yes. I was exploring options: Different distributions of bullet spread.
in that case, just increase the cone significantly when jumping. the cone could also increase gradually the longer your burst of fire. for example after 10 rounds radius increases by +2, at 15 rounds +3 etc
Jusst wanted to bump this and through a question out there aswell. I see more and more bunny hopping from players of late, is there any planned form of anti bunny hopping going to be implemented into NS2? Some may argue it requires skill others not so much, but for me, bunny hopping in NS2 is extremely out of place. It break the immersion, gameplay and feel of the combat to me.
Would be really nice to see a simple cooldown between jumps or a move reduction after a jump or something, just to curb the amount of bunny hopping ingame that I see at the moment. I'm pretty sure it's not intended gameplay to see marines and skulks alike just bouncing up and down all over the place.
I would suggest implementing some form of non-visible fatigue meter - It seems too many people would complain otherwise.
You could base it off of a variable which increases at a moderate rate with each tick (so it may take 1 second or so to reach its clamped maximum) while not in the act of jumping. When one jumps, this variable drops by just over 1/4. For reasons of this suggestion, I'm going to say the clamped maximum is 2, hence when you jump 0.6 could beremoved.
The jump impulse/power would be clamped at a maximum of what it is currently, and is multiplied by this fatigue variable - meaning you can jump twice at normal height before you start suffering at all.
This is because when the variable is greater than one, the jump power would be clamped to its maximum, after one jump it goes down to 1.4, still clamped to maximum power, and then 0.8. Because 0.8 is less than one (full power of the jump), the jump power suffers by 0.2, and is only 80% as powerful. The next jump drops it to 0.2, resulting in a very small hop. Another jump and it's down to -0.4, preventing one from jumping at all until the variable increases to greater than 0 (about 210ms at proposed regeneration rate).
The above is an example of a perfect bunny hopper (absolutely negligible time spent on the ground), hence no variable increase is apparent until under 0.
This setup does not effect any other forms of motion, and only punishes bunny hoppers (which frankly breaks immersion and fun A LOT for me).
I dislike the idea to remove jump or limit jump and slow jog (waht I call sprint :P).
I think that if the aliens get a flashlight to highlight marines the frantic jumping wont really help them, as then you see if they are still infront of you or if it is a wall.
Limiting bunnyhopping could be ok, but seeing as going backwards means instant stopping I think it would be a bad idea. However, if you ask me, the best way to implement such a limit would not be fatigue or such, it should slow down the players, altough maybe (for a better flow in the game), only do it if the player already jumped once in the last second or similar. Or maybe make the slowdown depend on the difference in height of jump and land positions (the higher you land the less speed you lose, like irl).
<!--quoteo(post=1826399:date=Jan 23 2011, 04:42 AM:name=Frohman)--><div class='quotetop'>QUOTE (Frohman @ Jan 23 2011, 04:42 AM) <a href="index.php?act=findpost&pid=1826399"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->If you read through any of the later posts in this thread, you'd know that this is not what the majority is suggesting.
Please read before posting.<!--QuoteEnd--></div><!--QuoteEEnd--> Not his fault the title is misleading. In which case, he's on board with everyone 'later' in the thread.
Jump needs to be in the game for plain mobility reasons. However, I am also seeing a stupid and annoying amount of bunny hopping. If nothing is done, this will become a norm through out the game because currently there is no drawback to jumping while in the midst of a fight. It makes it harder to be hit by an alien while not causing you any negatives at all. By the way there is a reason almost all modern games except extremely arcady and cartoony shooters have completely removed bunny hopping mechanisms.
The solution I think could possibly work is to not allow people to jump while they are firing any two handed weapon, as this would be pretty dam hard anyways. Once they perform a jump, the weapon will be lowered and mimic the motion someone makes with their arms while jumping not allowing them to fire. Such animations will indicate to the new players whats happening and why they can't fire there weapon.
The one exception to this would be single handed weapons like the pistol and axe as they're light and only take one hand to fire making it fire easier to shoot and jump.
twilitebluebug stalkerJoin Date: 2003-02-04Member: 13116Members, NS2 Playtester, Squad Five Blue
edited January 2011
I guess they can make the marine's view bob noticeably upon landing. On the other hand, it will only be a client-side effect, which can easily be hacked and removed for competitive advantage, just like the view shake when firing weapons.
You guys have to be joking about remove jumpin for marines.. If marines are jumping all over the place then level up your skill and learn to kill them. Did any of you here actually play NS1?
Skill based movement adds depth to a game and is quite fun actually. You also have to take in consideration that it is hard to jump around and aim, so if people can do both why should we penalize them for being better? Immersion? You're playing a game with teleporting aliens in space so why should you expect it to be a "simulation". It seems to me you guys want to eliminate any sort of skill to NS2 so we can all win participation trophies.
Comments
After all, it's kind of silly to have them work only at one point directly in front of you when you're swinging claws/chomping huge teeth at people.<!--QuoteEnd--></div><!--QuoteEEnd-->
+1
inb4: competitive, elitist purists.
It works half-arsed to swing your view around swiftly when biting at marines, the zone hit isn't exactly dead center because of various lags.
Easier.
At close range spread is a benefit, not a burden.
Spread/cone of fire is related to precision, while aim is related to accuracy.
With high accuracy, you'd also desire high precision. You would kill a skulk fastest, and conserve the most ammunition.
With low accuracy - due to the quick motion of the target, the close quarters and the marine's aiming ability - some imprecision is helpful, because although it means not every one of your shots will hit, ideally enough of the shots will hit to kill the skulk - but you will waste ammunition.
There has to be a balance: the average player's accuracy, versus the appropriate precision. Not every player has perfect accuracy, so you mustn't have perfect precision. Currently though, the NS2 rifle has too little precision, and isn't balanced well against the fire rate of the rifle / size of the clip (and of course, how those relate to the dps of the skulk).
I guess the best way to balance would be to consider
a) probability distributions and cone of fire (perhaps a normal distribution with regards to radius from centre of the cone of fire would work best - i.e. most shots fire through the centre of where you aim, but less shots hit fire through a position away from the centre of where you aim),
b) a "most likely distance" between rifle and skulk,
c) the skulk's "shape" at its "most likely orientation" (facing you? facing away? side-on?) and "most likely position" (due to the movement of the skulk and the aim of an average player).
d) the fire rate of the rifle and/or the size of a magazine, and how those relate to the "most likely" dps of a skulk
At that "most likely" distance the probability distribution/density should overlap with the skulk shape almost perfectly, and there should be enough shots that actually hit to kill the skulk, before it kills you.
There's a lot to consider. It's a science.
Another consideration you could have is to have the probability density/distribution (and cone of fire) tighter with less shots - so you could shoot a skulk at a distance if you did bursts of fire and they would land almost exactly through the centre of your crosshair;
but the probability density/distribution (and cone of fire) flatter and wider with more shots - so you'd spray and pray (which is what tends to happen anyway) at close distances and they would land roughly equally around the cone of fire, or maybe even less towards the centre.
Of course, if you have impeccable aim (accuracy), you'd still do bursts of fire at close distances, to conserve ammunition and maintain precision.
It'd be easier to show this all graphically, but I hope it gets through to you anyway.
Edit: I tried to illustrate accuracy, precision, and different probability densities (in a very extreme sense):
<img src="http://img411.imageshack.us/img411/6135/accandprec.png" border="0" class="linked-image" />
They <b>do not</b> represent increasing cones of fire, they are all <b>the same cone of fire</b> (I've said so), but with different probability distributions. The first from the left: normal distribution (bell curve) - more shots grouped towards the centre, less as you tend to the outside. Middle: flat or even distribution - shots even around the cone of fire.* Furthest on the right: inverse bell curve - shots tend away from the centre of the cone of fire.
*The four accuracy/precision distributions assume flat or even distribution, but with changing cones of fire (precision).
Pretty sure that was all illustrated and explained already. The reason I illustrated these is to show graphically (and thus more intuitively) what I've <b>discussed in my post</b>.
This this and more this :-)
They <b>do not</b> represent increasing cones of fire, they are all <b>the same cone of fire</b> (I've said so), but with different probability distributions. The first from the left: normal distribution (bell curve) - more shots grouped towards the centre, less as you tend to the outside. Middle: flat or even distribution - shots even around the cone of fire.* Furthest on the right: inverse bell curve - shots tend away from the centre of the cone of fire.
*The four accuracy/precision distributions assume flat or even distribution, but with changing cones of fire (precision).
Pretty sure that was all illustrated and explained already. The reason I illustrated these is to show graphically (and thus more intuitively) what I've <b>discussed in my post</b>.<!--QuoteEnd--></div><!--QuoteEEnd-->
yeah your explanation confused me. ive fired plenty of machine guns and the spread never ends up like circle 3. what situation are you suggesting causes a grouping like a donut? from my experience the the larger your cone, the more its spread out
or are we throwing away realism and basing this of a computer game cone of fire?
Yes. I was exploring options: Different distributions of bullet spread.
Would be really nice to see a simple cooldown between jumps or a move reduction after a jump or something, just to curb the amount of bunny hopping ingame that I see at the moment. I'm pretty sure it's not intended gameplay to see marines and skulks alike just bouncing up and down all over the place.
You could base it off of a variable which increases at a moderate rate with each tick (so it may take 1 second or so to reach its clamped maximum) while not in the act of jumping. When one jumps, this variable drops by just over 1/4. For reasons of this suggestion, I'm going to say the clamped maximum is 2, hence when you jump 0.6 could beremoved.
The jump impulse/power would be clamped at a maximum of what it is currently, and is multiplied by this fatigue variable - meaning you can jump twice at normal height before you start suffering at all.
This is because when the variable is greater than one, the jump power would be clamped to its maximum, after one jump it goes down to 1.4, still clamped to maximum power, and then 0.8. Because 0.8 is less than one (full power of the jump), the jump power suffers by 0.2, and is only 80% as powerful. The next jump drops it to 0.2, resulting in a very small hop. Another jump and it's down to -0.4, preventing one from jumping at all until the variable increases to greater than 0 (about 210ms at proposed regeneration rate).
The above is an example of a perfect bunny hopper (absolutely negligible time spent on the ground), hence no variable increase is apparent until under 0.
This setup does not effect any other forms of motion, and only punishes bunny hoppers (which frankly breaks immersion and fun A LOT for me).
Cheers,
Frohman
Please read before posting.
I think that if the aliens get a flashlight to highlight marines the frantic jumping wont really help them, as then you see if they are still infront of you or if it is a wall.
Limiting bunnyhopping could be ok, but seeing as going backwards means instant stopping I think it would be a bad idea. However, if you ask me, the best way to implement such a limit would not be fatigue or such, it should slow down the players, altough maybe (for a better flow in the game), only do it if the player already jumped once in the last second or similar.
Or maybe make the slowdown depend on the difference in height of jump and land positions (the higher you land the less speed you lose, like irl).
Please read before posting.<!--QuoteEnd--></div><!--QuoteEEnd-->
Not his fault the title is misleading. In which case, he's on board with everyone 'later' in the thread.
The solution I think could possibly work is to not allow people to jump while they are firing any two handed weapon, as this would be pretty dam hard anyways. Once they perform a jump, the weapon will be lowered and mimic the motion someone makes with their arms while jumping not allowing them to fire. Such animations will indicate to the new players whats happening and why they can't fire there weapon.
The one exception to this would be single handed weapons like the pistol and axe as they're light and only take one hand to fire making it fire easier to shoot and jump.
Skill based movement adds depth to a game and is quite fun actually. You also have to take in consideration that it is hard to jump around and aim, so if people can do both why should we penalize them for being better? Immersion? You're playing a game with teleporting aliens in space so why should you expect it to be a "simulation". It seems to me you guys want to eliminate any sort of skill to NS2 so we can all win participation trophies.