(yet) another AH fighter firepower comparison

[Angus]

Bullets vs objects is nowhere near the same as bullets vs aircraft. Aircraft take into account penetrating power, explosive power, many other aspects including internal damage systems of the target itself.

Objects are very flawed, and IMO I think they negate a lot of the round power of MGs, and I think they soak up too much "splash damage" from cannons, but this is only personal guesswork.

Since we don't have HTC's exact numbers (other than saying 1 50cal is worth 3 30cal, 1 hispano worth 3 50cal, etc), and we know that certain relationships seem to match up to the real things, I tend to lean more towards Tony Williams than the flawed "rounds to kill a hangar".

IMHO objects like field acks should not suffer so much from cannons vs being hosed by machine guns, - the simulation would be killing the crew and not the gun, or?

[moot]

Yes, I've read a lot of his articles, and I like the work that he's done.  But a good reputation and frequent citation is not evidence that his figures match Aces High.  Looking to his work to figure out lethality is comparable to asking how fast the P-47 rolls in the game, and then reading a book about it instead of testing the roll rate yourself in game.  You'll probably learn a lot of interesting things and enrich your historical understanding of the P-47, but come away with a misunderstanding.

I understand that damage against objects takes no account of kinetic energy loss, which is accounted for in damage versus planes.  However, it seems that the most strident debate is generated over auto-cannon lethality, which is mostly chemical energy, kinetic energy only to a lesser extent.  This is where discrepancies exist between the two methods of ranking, i.e. Ki-84 vs Spit8.

If by this you simply mean that I should compare how well Williams matches the hangar destruction data, it begs the question because that's what we're arguing about in the first place.  Did you mean something else?

Did you try to compare them or not?  

How many different ways would it be modeled in AH?  The factors in Williams' formula are what it will always boil down to when you compare the guns in the game.  Kinetic power and chemical charge.  You're doing the same thing as in the ENY arguments.  Arguing the theory to no end instead of assessing what concrete evidence there is and going from there.  Back then you were saying the ENY values were wrong and that there needed to be an exact math formulation of the ENY values based on paper figures, and never (that I saw) admitted that an intermediate step (plain human quantitative/qualitative guesstimate; HTC arbitrary) is good enough for the time being (however long the ENY formulation debate would take - forever), nevermind that it might just be good enough. There's only so many integers to set the ENY to, and that HTC guesstimate certainly looks like it's beyond the depreciating returns of the chase for "perfect values".

The only way that taking Williams' values is a mistake is if the data points/curves don't even look familiar.  Which you could rule out for good by actually crunching the numbers right off the bat.

IMHO objects like field acks should not suffer so much from cannons vs being hosed by machine guns, - the simulation would be killing the crew and not the gun, or?

You wanna stand in that ack hole while someone shots a burst of 20mm within it, without any body armor?

[Lusche]

For your convenience... I have converted both charts in this thread into Z-scores:

[Anaxogoras]

You're doing the same thing as in the ENY arguments.  Arguing the theory to no end instead of assessing what concrete evidence there is and going from there.

This time you are the one doing this.  Williams' account is the the theoretical account, hangar test data is practical and concrete.

[moot]

No.  The topic is AH fighter firepower comparison - that's level 0 of abstraction. If Williams' formulation matches the AH hangar data to negligible difference, it's no more excessively theoretical than the very topic itself is. It's as good a baseline as any to build on for any more practical, concrete extrapolations on how AH fighters compare in firepower. In fact, it's better, because it gives us a clue (a confirmation) as to how things are modeled in AH, as opposed to extrapolating firepower against planes from firepower against buildings.  We see that the Williams figures are likely "good enough" to build a z-score from.  You wanted to reject it apparently without even looking at the data. Purely on principle.

As far as I can tell, we can now build a couple of more specific charts, e.g. z-score against a target flying at zero relative velocity and short, medium, and long range. Maybe make it a 3D bar chart with Z-Score, Range (three sets), and ambient speed, as dimensions.

[morfiend]

For your convenience... I have converted both charts in this thread into Z-scores:

(Image removed from quote.)

 WoW lusche they are remarkably close!!!

 I do have 1 question tho,it seems the tempest has more firepower than the mossie,what loadout did you have in mossie??

The 100 hispano rounds that the tempest has over mossie couldnt better the 3100 303 rounds??

  I'm only curious and applaud your work.

 

[Lusche]

I do have 1 question tho,it seems the tempest has more firepower than the mossie,what loadout did you have in mossie??

The 100 hispano rounds that the tempest has over mossie couldnt better the 3100 303 rounds??

Ammo capacity is not figured in, it's a pure 1-second-burst comparison.

The Tempest is carrying the Hispano V, the Mossie the Hispano II. The V has about 25% higher rate of fire, which more than equalizes the additional 4 rifle-caliber Brownings the Mossie is carrying.

[morfiend]

ahhhh thx.
   

[bustr]

Can't you use the offline mission planner to create a one on one scenario to test lethality? Make the con fly straight and level at a fixed speed while you start from a fixed distance and speed behind it? I thought the offline mission planner had that much granularity. Being on the training staff you should have the settings to duplicate the environment in the MA. But then it would be testing offline with no internet connection issues to interfere with the interaction.