How exactly does centerspeed work?

[arcanine2009]

Ok.. I just did some calculations...


Using this formula:
(5 * viewKick) / Centerspeed = Time in seconds to return to center

----------------------------------------------------------------------------------------
SMGs

MP5k:
w/out grip: 5 X 80 (view kick in right direction, which is highest recoil value for mp5k)/1700 = .235 seconds
w/ grip: 5 X 80/1700+100 = .222

AK74u:
w/out grip: 5 X 65/1600 = .203
w/ grip: 5 X65/1600+ 200 = .18

Uzi:
w/out grip: 5 X 80/1600 = .25
w/ grip: 5 X 80/1600+100 = .235

Skorpion
w/out grip: 5 X 50/1700 = .14
w/ grip: 5 X 50/1700+100 = .138


Ars

Enfield:
5 X 60/1500 = .20

Famas:
5 X 60/1600 = .1875

M14:
w/out grip: 5 X 80/1400 = .285
w/ grip: 5 X 80/1400+200 = .25

FN FAL:
5 X 70/1400 = .25


>Sniper Rifles

WA2000:
5X65/825 = .39

Dragunov:
5X65/500 = .65

L96A1
5X100/500 = 1 <-- also note that L96A1's RPM is 60 RPM also, or 1 round per second..

PSG1
5X95/100 = .475




Does this mean that.. M14 w/ grip = FN FAL in overall recoil (they have same RoF), and PSG1 has a better recoil than Dragunov(again same RoF also), and is better at leading second shots?

[arcanine2009]

Jan 4, 2011 17:52:36 GMT -5 Ishbane said:

Jan 4, 2011 2:20:43 GMT -5 toto said:

M14 - Heavy upward kick
FAL - Small, random kick in all directions

Pretty much, yes.

But since m14w/ grip takes the same time to recenter as FN Fal, does that mean they have the same exact recoil, minus the difference in direction kicks?

Feb 16, 2011 12:29:37 GMT -5 psijaka said:

@ mannon

Just plugged the numbers for the M14 and FN FAL into the theory part of my spreadsheet and I get the following percentage of second shots fired on target, IF the guns were fired at their maximum rate (a big if in reality).

M14 -------- 30.11%
FN FAL --- 23.34%

This is derived mathematically.
Reassuring that our numbers match to 4 significant figures.

Just for fun I popped the numbers into my brute force simulator, firing 50,000 shots, and I got:

M14 -------- 30.35%
FN FAL --- 23.08%

Not exactly the same, but close enough to give me confidence that the 3rd and 4th shot figures I derive from the simulator are reasonably accurate.

Nice. Why aren't these numbers added in the recoil thread?
Also.. What about M14 w/ grip?

And if shots were fired at half the speed of the 625 RPM allowed, does it mean accuracy would increase by 2X?

[mannon]

We understand all that. In fact I have a chart with all that information already calculated, because it looked interesting. But you need to understand what you are calculating.

If you leave FireTime out of the calculation all that you are showing is how long it takes to return to center, and the only real use for that info is it tells you how quickly you can spam the trigger to pace fire a weapon manually with 100% accuracy. Though you'd probably be better off firing a burst. Still it's handy as a reference to have some idea how long to wait between bursts. Some weapons require a longer pause between bursts than others, though this number alone doesn't tell you the answer for bursts because a 3 shot burst may, or may not, have put you significantly more off target than even the maximum kick.

BTW... this is also calculating the absolute maximum, which is kinda overkill. If you pace fire a weapon slow enough all of it's shots land dead center then go against someone who pace fires the same gun at a rate that only 80% of the second shots recenter completely they will win more than 80% of those fights. Keep in mind that just because 20% of their shots won't land exactly on target that doesn't mean they can't hit you. Many could still be close enough, though they might also get some compound recoil if a shot doesn't recenter and the next kicks the same direction to enough degree.

Once you get into the other stats and actually use the firetime that's when things get interesting.

Does M14 w/grip = FN FAL? Well... in terms of ROF and some other things sure. But in terms of recoil... Not even close.

First off the M14+Grip has a narrow horizontal kick that only ranges a total of 80 VKU's while the FAL's horizontal kick ranges 120 VKU's and isn't centered. The M14 has a taller vertical kick range of 120 VKU vs the FAL's 90.

Then we get CenterSpeed involved. They may recover from a maximum kick in the same amount of time but the M14+Grip has a 200 CenterSpeed advantage over the FAL, which means a wider swath of kicks will recenter. In the vertical range this helps the M14 stay pretty close to the FAL on recenter percentages despite the larger kick area. M14+Grip recenters vertically on 51.2% of it's shots and the FAL 52.09%, so just over half. The M14 still kicks over a larger vertical range when it doesn't recenter, though.

But, it's the horizontal range where you really see a difference. The M14+Grip is nicely centered where the FAL will kick left... a lot left. In fact it is about twice as likely to kick left than right and can probably kick about twice as far. Then we look at the odds of recentering and the M14+Grip really shines. 76.8% of the time the M14+Grip will fire it's second shot horizontally dead center. That's HUGE! That is literally the best odds of recentering horizontally by any weapon that isn't a shotgun or sniper rifle, and those only do better due to slow firetimes. The FAL can only manage it 44.8% of the time and due to the off center kick it will kick it-self away from being in range of recentering horizontally, while the M14+Grip will stay on a narrow vertical kick pretty much strait up.

Sorry, but no, they are very different guns.


As for what we've been trying to figure out. We know how long it takes to recenter and how big a kick can be recentered in X amount of time by Y centerspeed... But what we don't yet know is exactly where the view is when a second shot is fired before it has recentered. And that's very important to know when you're trying to analyze 3rd shots and beyond.

If the view only traveled at a flat speed out from the center and back it would make our lives much simpler. But this isn't really the observed behavior as I recall it. Doing that would make the view jerk away from the center (which we do observe) but then suddenly jerk back at the apex of the kick.

This is why we've been discussing how to model the behavior, so that we can figure out the exact position of a kick at any point in time. Then we can simulate it and build mathematical probability models.

[arcanine2009]

Thanks for the info on M14/FN Fal! Which one do you think is better though?

Regardless, I've always been a huge fan of the M14.


Yes, the fire time is what i want to factor in especially, when it comes to centerspeed.


I know fire time is a huge factor, though when I'm comparing M14 w/ and w/out grip Vs FN FAL, and Dragunov Vs PSG1, I know they have the same RoF (625 for the 2 Ars, and 235 for the sniper rifles I'm talking about)


Is there anything you can say about Dragunov and PSG1? They fire up most of the time right? I know they have the same RoF, but Drag w/ less overall recoil, but less centerspeed as well.

I calculated that PSG1 recovers faster (vertically) than the Dragunov, so I'm wondering if this means PSG1>Drag in recovery center speed time.

[mannon]

This isn't nice and easy to read, but here is a big fat wall of numbers on every weapon/attachment combination of any relevance. (As near as I can tell.)

I've considered doing something easier to read, but I want to make it nicer looking and show more than just recenter percentages and we're still figuring that out, so I've not bothered. ;p

spreadsheets.google.com/pub?hl=en&hl=en&key=0AsohXFbJSr4ZdDk0Ml9jbHB1OWZhUW00TGVBZ0pyZVE&output=html

PS. I may have a few errors in the spreadsheet still. Most of the columns are just figures I thought might be interesting to compute. If so just let me know and I'll correct them. I might also have some rounding, but I attempted to avoid it. None of the numbers are rounded before being used in calculations, however. So there shouldn't be any rounding errors, it's just in the final formatting.

[psijaka]

Feb 19, 2011 19:52:41 GMT -5 arcanine2009 said:

Jan 4, 2011 17:52:36 GMT -5 Ishbane said:

Pretty much, yes.

But since m14w/ grip takes the same time to recenter as FN Fal, does that mean they have the same exact recoil, minus the difference in direction kicks?

Feb 16, 2011 12:29:37 GMT -5 psijaka said:

@ mannon

Just plugged the numbers for the M14 and FN FAL into the theory part of my spreadsheet and I get the following percentage of second shots fired on target, IF the guns were fired at their maximum rate (a big if in reality).

M14 -------- 30.11%
FN FAL --- 23.34%

This is derived mathematically.
Reassuring that our numbers match to 4 significant figures.

Just for fun I popped the numbers into my brute force simulator, firing 50,000 shots, and I got:

M14 -------- 30.35%
FN FAL --- 23.08%

Not exactly the same, but close enough to give me confidence that the 3rd and 4th shot figures I derive from the simulator are reasonably accurate.

Nice. Why aren't these numbers added in the recoil thread?
Also.. What about M14 w/ grip?

And if shots were fired at half the speed of the 625 RPM allowed, does it mean accuracy would increase by 2X?

I intend adding data on semi auto weapons but the % of second shots on target is less meaningful as, being semi autos, you will not be able to ge tthe second shot away at the minimum fore time (not without a modded controller, anyway). Of more relevance is the maximum time taken for the gun to recover from worst case kick. Am working on this.

Firing at half speed would result in a big increase in second shots on target; again I need to do some more sums and think about how to present this information.

[mannon]

Yeah I was thinking about semi auto weapons earlier and considered that it might be a good idea to use a more human capable firetime for the figures. Unfortunately it's difficult to say exactly what constitutes a good firetime for that, especially when talking PC vs Console as mouse clicking is far better for high paced firing than the trigger buttons on controllers. There's also a large diversity among players.

On top of that some weapons will perform better at different paces, and not merely spammed as fast as possible. Maybe we should list the time to recenter both horizontally and vertically and then maybe the time to reach a certain recenter % goal or some such. (For the % I'd say list the time for both X, Y, and both at once. You particularly want to see the directions split out for vertical recoiling guns since they are still very usable when you minimize the horizontal but simply accept the vertical kick.

I dunno. They are definitely something of a special case, though. For now I've just left everything in my table as is with the direct firetimes as I don't even really know for sure which weapons are semi-auto and which aren't. ;p

Hmm... Come to think of it, maybe they aren't such a special case. Perhaps I should just work out desired firetimes for certain odds of recentering and give those for all weapons. After all even a full auto weapon can be paced out for greater accuracy

If you tap fire a gun you can achieve anything between it's base recentering chances and 100% depending upon how quickly you tap out your shots.

[psijaka]

I think what I will display for semi autos in the table is % of second shots on target assuming fastest fire time, and then the time at which the gun will fully recover from worst case kick. Both can be accurately derived theoretically.

I will try to update the recoil table ASAP, but other things to do.....

If a gun is not displayed in the recoil table, then it is not a full auto. Not all the excluded guns are semi autos, but they will still have min fire time so the same principles will apply, even if recovery is 100% (bolt action)

[psijaka]

Feb 19, 2011 19:52:41 GMT -5 arcanine2009 said:

Feb 16, 2011 12:29:37 GMT -5 psijaka said:

@ mannon

Just plugged the numbers for the M14 and FN FAL into the theory part of my spreadsheet and I get the following percentage of second shots fired on target, IF the guns were fired at their maximum rate (a big if in reality).

M14 -------- 30.11%
FN FAL --- 23.34%

This is derived mathematically.
Reassuring that our numbers match to 4 significant figures.

Just for fun I popped the numbers into my brute force simulator, firing 50,000 shots, and I got:

M14 -------- 30.35%
FN FAL --- 23.08%

Not exactly the same, but close enough to give me confidence that the 3rd and 4th shot figures I derive from the simulator are reasonably accurate.

Nice. Why aren't these numbers added in the recoil thread?
Also.. What about M14 w/ grip?

And if shots were fired at half the speed of the 625 RPM allowed, does it mean accuracy would increase by 2X?

Have updated the recoil chart to include the M14 and FAL

Halving the RPM would give us:
M14 firetime 0.192 = 78% second shots on target
FAL firetime 0.192 = 71%

[psijaka]

Have tried to simulate a ten round burst assuming that Viewkick is a velocity and that centerspeed is an acceleration. Some initial results from a selected single burst are shown below:

Back to the drawing board I'm afraid; discovered a major flaw. In the red plots below, I incorrectly scaled the effects of centerspeed as an acceleration, so what we are seeing below is essentially just velocity * time. When I subtract 0.2*centerspeed*time squared the result are chaotic. I just don't think that this theory will work.



The green line is the method I have used previously, assuming that viewkick is displacement, the red line is the new method, treating Viewkick as a velocity. These are for Famas and Aug, no attachments.

The first images I posted contained a serious error, so they have been removed.

The "velocity" simulation has been arbitrarily scaled by a factor of 17, to allow us to compare the results from the 2 methods (it is the shape of the plots that we are interested in).

Early days yet, but the results from the 2 methods are very similar.

I expect that the recoil from guns having a longer period between rounds will have their cumulative recoil increased if we treat kick as a velocity, as the time between rounds is squared when we are dealing with the effect of centerspeed as an acceleration.

More work needed, and I haven't even started thinking about "residual" velocity!

[mannon]

Feb 23, 2011 3:36:26 GMT -5 psijaka said:

The green line is the method I have used previously, assuming that viewkick is displacement, the green line is the new method.

Sorry, you seem to have a typo here. Which is which? Nice to see they work out pretty similarly, that's about what I expected and why I encouraged you to do the charts anyway, regardless of method.

[psijaka]

Feb 23, 2011 11:20:29 GMT -5 mannon said:

Feb 23, 2011 3:36:26 GMT -5 psijaka said:

The green line is the method I have used previously, assuming that viewkick is displacement, the green line is the new method.

Sorry, you seem to have a typo here. Which is which? Nice to see they work out pretty similarly, that's about what I expected and why I encouraged you to do the charts anyway, regardless of method.

Oops thanks for pointing this out. Have corrected.

But I discovered a bigger error - see above

[mannon]

I'm looking into it. I haven't really attempted simulations before I did those graphs. It definitely complicates the formula having to deal with both sides of 0 and reversing the gravity effect of centerspeed.

If it is a displacement/velocity model instead of velocity/acceleration then perhaps the velocity is applied more like an average velocity and the acceleration and deceleration are merely simulated. In such a system the viewkick would actually designate the apex of each kick. Centerspeed simply determines the average velocity and thus the time it takes to complete the entire kick. The actual initial velocity and acceleration would be scaled to hit the apex and recenter in the right amount of time.

It would work out a bit differently than the velocity/acceleration model and wouldn't squash things down quite so much numerically. But it should produce similar visual results and could match the observed behavior. To me it just seems a little backwards to specify the apex of a kick rather than the initial velocity, but then maybe it's more direct that way and thus easier to tweak the recoil if you at least know how far the peak of each kick is based on that. Then again I'm not too sure how well that works out when you figure in actual firetimes and centerspeeds.

Just need to do a lot more math... and some testing would help. Dammot... heh Tempted to find a torrent of CoD4 just to try and test some things, but meh... It's not like I'd be able to even play it on my old computer, and I'd probably get a virus, aside from feeling a bit guilty about it too. ;p I do try to not download things I actually have access to, even if I can't afford them. I feel oddly less guilty about things I just don't have any other access to, though.(eg. I don't have cable, but we do have netflix. But if I can't get it legally online or on Netflix, well... *shrug*) But that's a whole other issue...

[mannon]

I'm leaning more towards the displacement/averagespeed model... I still think a velocity/acceleration model could potentially work, but it has some quirks that might eliminate it with testing. (Or prove that it is in practice.)

The main quirk is that since velocity is applied at the moment the shot is fired regardless of where the view is at the moment, any kicks that would cross the X or Y axis get killed and simply resettle at zero. In practice this would make it impossible for a weapon to kick on one direction and then kick in the other direction without recentering first. Then again considering how often weapons actually do recenter, (a lot more than I thought previously) that still doesn't quite disqualify the model. But the good news is that should be pretty easy to test. Just give a semi auto weapon with a wide viewkick a low centerspeed so you can watch it. Fire off a shot so that it kicks to one side then let it drift back. Before it actually recenters fire it again. Keep doing that until it kicks hard in the direction of 0 and note whether it stops at 0 or kicks to the other side. I'm betting it kicks to the other side... which implies a viewkick=displacement(apex?) model. If on the other hand the shot kicks back towards zero but sticks there without crossing it then that implies a viewkick=velocity model.

[psijaka]

I have had a look back at Den's comments before his weapon damage chart -

Recoil is an area within four points and the random chance of the weapon kicking anywhere within that area.
The recoil causes the player's view to travel in that direction, a higher number means a faster camera jump and ultimately more recoil.
The kick focuses on the average between the points, so if a weapon kicks left 10 and right 60, the kick will almost always be to the right.
If a weapon kicks 60 both left and right, the sights have equal chance of jumping either way.
If both numbers are on one side of the arrows, then it'll always kick in that direction.

Working against the recoil is Center Speed (number in the bottom right of the Recoil box), a mostly arbitrary number.
Higher Center Speed means faster recovery as well as dampening the recoil, as CenterSpeed is always active from the instant you shoot.

Need to think some more....

[mannon]

I've been thinking about it and actually if the view can jump from one side of zero to the other... how exactly do you handle that? If the velocity were flat it would be easy, but that doesn't seem to mesh with my observations. Then again... I'm not really sure about that anymore. This whole thing is making me question my perception of recoil. I should really do some testing, though I really really wish I could alter the stats instead of just observing the game as is.

[psijaka]

@ mannon
I come back to my "fantasy recoil" model.

The Viewkick, chosen randomly, is where the gun "wants to go", or it's intended "destination", but recentering is applied to pull the destination back towards the axis. This would allow for the gun to cross the axis in response to the kick, but then be pulled back towards it.

Will do some more modelling on this.

Any thoughts on this, Den?

[][nquisitor Mateo]

So a gun with a macro for a 5 shot burst set to its number up there will have perfect accuracy ADS?

[psijaka]

Any gun will always recover fully in 5 x maximum kick / centerspeed seconds

[asasa]

Feb 25, 2011 8:35:06 GMT -5 ][nquisitor Mateo said:

So a gun with a macro for a 5 shot burst set to its number up there will have perfect accuracy ADS?

Yes, but its going to be a very low firerate. Manual single firing will yield better results as the numbers given are worst case. Obviously, thats rare.

[mannon]

Yeah it's not really necessary to tap fire at a rate that produces 100% recentering in all cases. In fact you're pretty much shooting yourself in the foot, if you'll pardon the phrase, to do so, unless your target is stationary and unaware of your presence.

Generally speaking TTK is going to be a factor as well as accuracy so you want to fire at a fast rate as well as an accurate one. It's really a judgement call, however. For example if you fire at 80% recenter rate then you have really good odds of each shot being on target, but also the ones that aren't should be fairly close, maybe even close enough to recenter to still hit your target, but you also keep your fire rate up. If your target is really far away maybe 90% would be better. For closer range even 50% or lower would be just fine. But the % you aim for is totally arbitrary and up to the user.

Try as we might to produce good statistical data on this to help inform users it still really comes down to hands on experience.

[mannon]

Feb 25, 2011 7:22:46 GMT -5 psijaka said:

@ mannon
I come back to my "fantasy recoil" model.

The Viewkick, chosen randomly, is where the gun "wants to go", or it's intended "destination", but recentering is applied to pull the destination back towards the axis. This would allow for the gun to cross the axis in response to the kick, but then be pulled back towards it.

Will do some more modelling on this.

Any thoughts on this, Den?

I dunno, I'm still iffy on exactly what happens during a cross axis kick. Here's my thing. I'm more or less sold on modeling viewkick on displacement now, though I still want to work on my velocity/acceleration model a little more. Might as well pursue all avenues... This definitely makes it possible to kick across the axis. But this also introduces a new wrinkle to the simulation which did not exist before. In other words we know that centerspeed defines the speed at which the view will recenter once kicked. However, descriptions of centerspeed say that it is always working anytime the view is not centered, from the moment the viewkick starts to when it ends. This along with the fact that the exact time of the entire kick is proportional to vkick and inversely proportional to centerspeed without any other variables needed, makes me think that centerspeed not only dominates the motion of the view once the kick has reached it's destination, but also the trip to the destination.

I can still get a nice motion out of it if I apply centerspeed as an average speed such that the initial speed is higher then drops along the path to the destination. It reaches actual centerspeed at the halfway point. At the destination it comes to a rest then heads back to center with the same exact acceleration curve. I just don't see any reason for the acceleration curve before the destination and after it to be different, nor do I recall that being the case in any of my observations. It always looks to me like it's the same going out as coming back in. But then... that's merely anecdotal evidence. More testing would be appropriate... preferably in slow motion with good video capture and accurate timestamps, or maybe even better still with modded gun stats.

If that's true then centerspeed is actually more of a universal speed. But then, what happens when a shot is fired that interrupts things and kicks us off in another direction again? I'm fairly certain you can easily apply centerspeed from the point of the kick's destination to center and come up with the right answer. If speed is constant then the view travels in only 8 primary directions and makes only sharp 45 degree turns when returning to center... which doesn't really mesh with what I think I've seen. If there is an acceleration and deceleration then you get more like the curves I charted describing the motion, even if centerspeed is used as the average speed. So where A describes a nonzero location where a second shot is fired and B describes the vkick destination that the second kick will go to before returning to Zero, we know how to go from point Zero to point A, and from point B to point Zero, but how exactly do we go from point A to point B? Is it a strait line? That seems unlikely to me as that would require synchronization between X and Y and except for exactly 45 degree kicks would require a different velocity in each axis. So do we use centerspeed as an average velocity again? Maybe. It also occurs to me that motion could get rather complicated in this.

Let me explain. With X and Y being independent, that necessarily means that the viewkick could hit the apex in one axis before the other. In fact in all but 45 degree kicks I would expect this to happen. But it also would apply to kicks from point A to point B which means that try as it might the view will never actually kick to exactly x, y two dimensionally unless it kicks in one of the 8 principle directions. Otherwise it's going to arrive at the apex in one axis before the other and start recentering.

What's confusing to me is in going from point A to point B do we just ignore the actual position from center, or are we still somehow tied to it? In the velocity model we are always always always tied to center, with gravity pushing down towards it. But in a displacement model I don't see how we really can be until we've hit the peak of that particular kick, rounded third base, and headed home.

Or rather, maybe we don't completely ignore center, but rather just pay attention to directionality and use centerspeed as an average speed. IF that is true then I think I can model a cross axis kick as it just changes the end behavior we're looking for. In other words if A is really far off to the right and B is closer to center, but on the same side of center, then we have to apply centerspeed in such a way that we just go from A to B then B to Zero but don't stop at B, because that would be retarded and contrary to observations. ;p Now if the positions are reversed and A is closer but B is further out then we DO stop at B as an apex and fall back in towards center. Now if A is to the right and B is across the axis to the left it's exactly the same as the second case where we do stop at B and then fall back towards center. During the kick from A to B we can otherwise more or less ignore the position relative to Zero... theoretically.

But... if that's not the case... if centerspeed IS still dragging at the view between the kick from A to B then crossing the axis becomes MUCH more complicated... but I also can't really see how to handle that without going to a velocity model in which case it should actually stop when it hits center.

Sorry this is so long. Just me musing on the potential complexities of it. It's really nice that we can figure out the recentering, but man trying to actually simulate the motion it-self is definitely way more complicated. I think we have enough open questions left unanswered to build several possible competing models. Maybe we should actually do that and then run simulations to see which seem to most closely match observed behavior, similar to weather forecasting, though that will be a lot of effort.

It'd be nice if we could do some direct testing... mebbe I will dig up a torrent after all for testing purposes alone. I'd rather not I'm sure it would take forever to download. If I had any money I'd just buy a cheap copy of CoD4 if I could find it.

[psijaka]

All this is beginning to "do my head in", just a bit!

One thing I tested recently was to see if a gun returns to it's starting point after a long burst. I tested with a Famas and loosed of a full clip (30) several times to see what happens, and the gun always seems to return to the position that it was when the first round was fired. No surprises there, I know, but I wanted to check.

Have done some more work on simulating the Velocity/Acceleration recoil model, but the results just don't add up. Subsequent shots are much more heavily biased towards the axes, and the differences between a low recoil gun (Galil) and a high recoil gun (Uzi ACOG) are just way too extreme. I'll not post pictures.

More thinking required, but I'm really not sure how much further I want to take this in the absence of some serious testing, which I cannot do on the Wii.

[Den]

Okay. So I modified the Super Duper Pretty Cool Guy so that it kicked horizontal five degrees per second with zero centerspeed and up five degrees with five centerspeed.

Any speed from viewkick or CenterSpeed is completely neutralized on the next shot.
Any straight upwards/downwards movement will be nullified by a straight left viewkick.

Centerspeed independently handles the X and Y axis.
If it were to kick ten up and one hundred left, the path taken would not be a linear route to the center. It would hit the zero on the pitch axis while still having a way to go on the yaw.


The maximum possible Viewkick is limited to 100 (or ten degrees).
Once it hits the edge (and almost no stock weapon can actually do that), any speed going that way immediately goes to zero, centerspeed will begin pulling back immediately.

Image at 16:9 resolution, default 65° FOV

Note that in this image, the box is a perfect 260x260 square.

Well, horizontal viewkick causes the camera to tilt.
The further along it goes, the more the camera tilts.
It actually stops at that point 130 pixels to the left/right of the point of origin and despite stopping in place, continues to twist... to a max of ten degrees.

Course, even though the camera tilts, the path taken by the x-only viewkick is still a straight line.

Using this perfectly vertical door jamb as reference...


The horizontal viewkick hits the edge in two seconds (since the viewkick speed is 5), but it continues to turn for exactly another two seconds.

This tilt is also a part of what the CenterSpeed needs to counteract - the camera will stay on the edge until the centerspeed takes care of the part of the tilt.

Essentially, the x axis has twice the "area" it can travel than the vertical. While a few weapons (like a rapid fire Kiparis) can reach the horizontal edge, none will reach full tilt. So... none of that is really important.

Though it does tell that horizontal viewkick causes the camera to tilt. 5° at the edge, 10° possible maximum.

[arcanine2009]

Mannon.. Considering all factors.. Do you think the M14 is better than the FN Fal or vice versa?

Don't forget that FN FAL has AR hipspread...

[Contrary]

M14 with grip, FAL without grip/for dual mags.

[psijaka]

Mar 2, 2011 16:53:12 GMT -5 arcanine2009 said:

Mannon.. Considering all factors.. Do you think the M14 is better than the FN Fal or vice versa?

Don't forget that FN FAL has AR hipspread...

Sharp eyed asasa pointed out that the FAL does not have AR hipspread.

It is 5-9 standing instead of the usual 3-7.

Both these guns have a firetime of 0.096 s

If by some miracle you were able to get the second shot off exactly on time, 30% of M14 second shots would take place with the gun recentered (39% with grip). The figure for the FAL is only 23%.

The FAL's worst case recenter time is 0.25s, same for M14 grip. The M14's worst case recenter time without grip is 0.286s.

So comparing the two guns:

M14 has a significantly higher chance that a fast second shot will be on target, even without grip.
M14 has better hip fire spread.
M14 has 1.5 head damage multiplier, making it a long range 2SK if you get a single headshot.

FAL has slightly faster worst case recenter time than the M14 without grip (same with grip)

NO CONTEST

[asasa]

Ugh... but it means you must use Scavenger for any high killstreak... while with the M60 you get the best of everything... except ADS + hipspread... and the slight 2HK range reduction... but still.. :/

I cant decide... M14 grip W/ Scav
M60 grip w/ Hardline

EDIT: M14 math is definetely wrong. Like I said earlier, it never kicks down - even with Acog where it couldnt possibly recover before the next shot.

I'm firing it @ 7 shots per second w/ a mod and not once did it fully recover.

-- And thats a big drop from 10+SPS in optimal conditions.

Can you try the math assuming it was.. 0-40 and 0-80 Y kick? [2x chance of getting 0-40]

[Contrary]

Mar 3, 2011 4:00:37 GMT -5 psijaka said:

NO CONTEST

but but but *snivel* I loooove my fal.

Actually that is a fairly convincing argument. Damnit I would try stealth M14 if it weren't for the fact M14 doesn't have Dual Mags.

[psijaka]

Mar 3, 2011 18:38:36 GMT -5 asasa said:

Ugh... but it means you must use Scavenger for any high killstreak... while with the M60 you get the best of everything... except ADS + hipspread... and the slight 2HK range reduction... but still.. :/

I cant decide... M14 grip W/ Scav
M60 grip w/ Hardline

EDIT: M14 math is definetely wrong. Like I said earlier, it never kicks down - even with Acog where it couldnt possibly recover before the next shot.

I'm firing it @ 7 shots per second w/ a mod and not once did it fully recover.

-- And thats a big drop from 10+SPS in optimal conditions.

Can you try the math assuming it was.. 0-40 and 0-80 Y kick? [2x chance of getting 0-40]

Yes, something troubles me about the M14. Really not sure what to make of it.

As for comparing M14 and M60; I wouldn't even try. I would usually go for the M60, but occasionally just fancy a semi auto, just for the challenge. Usually don't get a high score but rarely do I "go under" badly either. I use extended mags or grip; with Ghost/Steady Aim/Hacker. M60 is more Flak Jacket/Hardened territory.