Difference between North Carolina & Alabama Gunnery

[LittleWhiteMouse]

So I wanted to see just how much the 0.1 sigma difference between Alabama and North Carolina really mattered. For this, I put together the following limited test.

• Make sure replays are enabled.
• Load up the training room on Ocean Map with a 3-Cap Domination and a 20 minute setting.
• Bring USS Alabama with no mods. Do not move. Aim at the buoy at A cap, 16.53km out due west of the starting position.
• Begin firing AP shells using only A-turret. Repeat with every reload cycle.
• Repeat using North Carolina, following the same steps.
• Load the first replay. Align the camera overtop of the target buoy using freecam and zoom out. Take screencaps of each volley's shell fall.
• Load the second replay. Use the buoys around the target buoy to align the camera as closely as possible.
• Use image editor to collate all of the shell falls and ensure perspective and scale is the same.
• Collate all of the shell full data.

 

Here's the results:

 

LEGEND

• Yellow - Target Buoy
• Purple - Reference Buoys to align results.
• Red - North Carolina
• Blue - Alabama
• Buoys are approximately 350m apart.

 

Thoughts: There isn't enough data to draw an accurate conclusion, but the preliminary data is already indicating that North Carolina seems to have less 'extreme' dispersion than Alabama. The maximum observed dispersion with Alabama is half the distance between two buoys more than North Carolina. Both ships appear to have comparable groupings towards the center of the target. More testing is needed to draw accurate conclusions, especially to properly identify the extreme maximums of dispersion for each ship.

[TheDreadnought]

The problem is, the Alabama is a much better design than the North Carolina.  So they've had to nerf the hell out of it to get it down to tier 8.

[RivertheRoyal]

Damn, the work required to do this....

Hat's off to you, LWM. 

[TheZeke00]

Great stuff Mouse. 

 

So what's the final verdict; is the difference statistically significant? 

[Estimated_Prophet]

Dang Mouse; +1 just for dedication.

 

Want Alabama because I've been to the actual ship twice now; this test and your earlier review only reinforce that desire.

[Jakob_Knight]

What happens if both ships' shells impact the exact same place?  Does it count for one but not the other, are both eliminated from the picture, or does it become a hit for one of the ships based on random chance?  I ask this because that might skew the final statistic if a more accurate analysis is desired beyond eyeball evaluation.

 

Edit - And echoing the comments by others, very impressive work there, Mouse!

[LittleWhiteMouse]

Great stuff Mouse. 

 

So what's the final verdict; is the difference statistically significant? 

 

Keeping in mind that the 200 or so shots for each ship aren't statistically sufficient enough to draw an accurate conclusion (I'd need about 6 times as many shots), at their worst, Alabama's guns appear to deviate a shell's splash width further than North Carolina's.  However, both group most of their shots in and around the target point so frequently for that kind of thing to be very rare.

 

Here's the data, separated per ship.  Dots stack on one another.  So they get lost.

Alabama

North Carolina

[TheZeke00]

Also this would require a lot of testing, but I wonder if Alabama's slightly shorter stature has any effect/correlation with her shell placement in comparison to NorCal. 

 

(For full broadsides, mind you)

 

[RivertheRoyal]

 

Keeping in mind that the 200 or so shots for each ship aren't statistically sufficient enough to draw an accurate conclusion (I'd need about 6 times as many shots), at their worst, Alabama's guns appear to deviate a shell's splash width further than North Carolina's.  However, both group most of their shots in and around the target point so frequently for that kind of thing to be very rare.

 

So, in other words, 0.1 sigma difference doesn't amount to all that much? That's good to know for future reference. 

[LittleWhiteMouse]

It should be noted that I'm aiming just in front of the yellow target buoy -- at it's waterline (or as best can be estimated from 16.53km out).  It's fascinating to me that in the tests so far, more shells seem to land short than long.

[GrandAdmiral_2016]

 

Alabama and her sisters were designed for the 16-inch 45 caliber gun from the outset. North Carolina and Washington were designed for a 14-inch gun, and armored against them, originally to comply with the Treaty Limits, with an upgrade clause to be invoked if any other potential enemy went above 14 inches. Italy and France broke ranks and went for 15-inch guns out of fear of each other, as did Germany. Japan refused to confirm that they would stay at 14 inches, and informed that future construction would, in all probability, use 16 inches like Nagato (a bald-face lie By the Japanese Foreign Minister when the question was asked by the US Ambassador because the decision to go to Yamato's 18 inchers had already been made). Alabama and her sisters are probably better brawlers, but size and stability issues made them poorer gunnery platforms than North Carolina. Great AAA platforms, though! See Freidman's design history on US BBs.

[ValkyrWarframe]

Maybe locking onto a ship could yield different results?  That dispersion looks awfully big.

[LittleWhiteMouse]

Maybe locking onto a ship could yield different results?  That dispersion looks awfully big.

 

Locking on does tighten dispersion but hits make it difficult to judge high dispersion.  It's also more difficult to get a ship to move to the exact same spot for repeat performance.

[Dr_Richtofen]

 

Locking on does tighten dispersion but hits make it difficult to judge high dispersion.  It's also more difficult to get a ship to move to the exact same spot for repeat performance.

 

If only we had a Buoy or something small like it that can't take damage to lock onto in training rooms to test this.

 

Seems interesting though. Would like to see more data to find out if the Alabama outlier shells continue to follow the tend they showed here.

[Kitsunelegend]

 

Locking on does tighten dispersion but hits make it difficult to judge high dispersion.  It's also more difficult to get a ship to move to the exact same spot for repeat performance.

 

Maybe get a really small ship and park it right on top of a selected buoy? Obviously would need help from someone else of course, and maybe a back up ship in case the other one gets killed. Just an idea.

[Axiomaticranger]

Very interesting post, LWM! While I'd agree that 200 (or so) shots are a comparatively small sample size, it does begin to suggest a trend.

I'm also glad to see (not really) that the majority of the shots that miss seem to be falling short, as this lines up with what I see when driving my BBs.

When I miss I'd always thought "nah, just confirmation bias that my shots drop short more often than any other type of deviation", but your test clearly indicates otherwise. 

Again, nice work!

[Vanessaira]

Science!  World of Warship style! <3 <3 <3

[amade]

There is another way of getting lock on dispersion debuff while shooting at a fixed target.

 

Get a ship to park way to the side of the target buoy so you don't hit it (but try to make sure the ship is at roughly the same range as the buoy). Lock on the ship, aim for the buoy. That way you won't hit the ship but still be able to shoot at the buoy without the dispersion penalty.

[Battlecruiser_Repulse]

I can't tell the difference between blue and purple, so I guess I'm SOL.

[Lert]

Maybe get a really small ship and park it right on top of a selected buoy?

 

1) A few hits would destroy that ship, removing the locked on target

2) Locking to a target narrows dispersion, this would be counterproductive. When judging relative gunnery, the narrower the dispersion, the smaller the measurable difference. If we are to see a difference, it needs to be as apparent as possible.

[Sir_Davos_Seaworth]

Mouse, first of all thanks for all you do. This is all very interesting, but in battle I don't think any of us would be able to tell the difference. Don't get me wrong, I like many of you geek out on this stuff, but we seem to be splitting some pretty fine hairs...

[Lert]

This is all very interesting, but in battle I don't think any of us would be able to tell the difference.

 

I think that's exactly the point Mouse is making. People are claiming that Alabama is less accurate than North Carlina based on 0.1 less sigma, but the difference is provably so minor that the vast majority of people won't notice. Strictly speaking yeah, Alabama is less accurate, but then, is that difference really enough to worry about?

[LittleWhiteMouse]

Maybe I'll do New Mexico (1.5 sigma) versus Arizona (1.8 sigma) to show a more dramatic difference.  Or maybe Arkansas Beta (1.5) versus Colorado (2.0).

[Submarine_Wahoo]

Maybe I'll do New Mexico (1.5 sigma) versus Arizona (1.8 sigma) to show a more dramatic difference.  Or maybe Arkansas Beta (1.5) versus Colorado (2.0).

 

Based on your results, I think a +/-0.5 sigma would be a more telling difference than comparing a +/-0.3 sigma.

[LittleWhiteMouse]

 

Based on your results, I think a +/-0.5 sigma would be a more telling difference than comparing a +/-0.3 sigma.

 

Yes, but Arizona and New Mexico are close competitors while the other two ships more certainly are not.  Alternatively I could do Fuso (1.5) vs Nagato (2.0).