Okay, I will wrap my changes with the following labels:

;========= G A P S T A R T =========
;========== G A P E N D ==========

yet, you will need to change many counting numbers in UpgradePacks.upc. Since I have made already the work, are you sure you don't want to give up you custom EUF settings for the short time (so to speak ;)) required to test the guns?

P.S: I think I have just discovered how to change the standard ordinance loaded in each gun :03:

:up:

what for? I will just put changes to the end of the file..

not quite understood what did you meant :hmm2:

are you just sure that they don't have to be in numerical order? :hmm2:

Do you remember that when you had removed AP shells from some guns, they still had AP shells as their standard ordinance, and that you had to wait for HE shells to be loaded before the gun could start firing? :03:

I don't think they should be in numerical order.. will try..

ah.. now I see.. that's great if you could change this..

Okay, sorry for the delay... I was busy with some PM's :03:

well at the end I found a few evidences that 20mm AP shells were included in U-boat outfit, and no evidence that they weren't used, so I decided to keep them. Nonetheless, setting the most commonly/most abundant shell types as standard ammunition could be a cool feature :yep:

Returning to my tasks now... :up:

sorry about that :03:

not at all. My pleasure :D

Things are going to be funny :up:

I didn't have time to run systematic tests of the Flak guns, but I still wanted to get a sense of the changes. I ran four tests with our modifications and four tests without it. Here are the results.

Setup
U-boat: VIIC, novice crew, Turm IV, 2 x 20mm C/38 Zwilling, 37mm M42
Aircraft: Mosquito Mk. VI, 500 m altitude, 2000 m distance
Conditions: 12:00, calm seas, excellent visibility (default weather settings)

Reworked Guns Tests
Test 1: aircraft damaged on first pass, destroyed on second by C/38
Test 2: destroyed on second run [couldn't determine which gun delivered the fatal shot]
Test 3: damaged on second pass, destroyed on third by C/38
Test 4: undamaged on first pass, destroyed on second by C/38

Stock Tests
In all four tests, the Mosquito Mk. VI was destroyed on the first pass, before the C/38 Twins had to reload. As I'm sure both of you realized long ago, it's absurd how accurate the stock AA gunners are. It was news to me though :o

General Impressions
The M42 feels like the heavy AA gun it is. In stock testing, all the AA guns were trained on the plane by the time I managed to turn around after loading the mission. In the Reworked Guns version, I can see them working to track the target, especially the ponderous M42. I can't wait to see the SK C/30 struggle to keep up with fast-moving aircraft :03: The whole thing looks more natural to my admittedly untrained eye. I did not notice any obvious clipping problems with muzzle flashes or audio loops. The C/38 Zwillings seem much more effective than the M42, but that might just be due to the volume of fire they can produce. I didn't notice any horribly misplaced shots from any of the guns, but I might have missed something while being strafed by the British flyboy :O: Clearly, the changes have reduced the guns' accuracy, though they are still able to shoot down an aircraft with little problem.

Anyway, I know that's not the sort of comprehensive test you need to evaluate the mod, but it's all I had time for tonight. I'll try to do some more tomorrow after my first exam.

EDIT: I also clocked the C/38 Zwilling's reload time at about 14.8 seconds on average. The M42's was about 8 seconds. Keep in mind this is with a novice crew.

Thank you Keiser for your detailed reports :up:
please look below fo some remarks:

Seems that with our current settings the FlaK C/38 is much more effective than the M42 (by the way, I ma wondering if during your tests the M42 has ever managed to score a single hit). Incidentally, the C/38 is also the gun with the fastest elevation speed. This makes me to think that we need to reduce the difference between the fastest and the slowest gun, at least in therms of elevation rates.
Another aspect that we should take into account is that, even in real life, the bigger 37mm guns could have been more effective at medium/long range than at close range, in consideration of their worse handling and of their longer range. This leads to my next question: which firing order (close/medium/long range) had you issued during the tests?

Not all the guns were equally accurate in stock game. The single C/30, for instance, was horribly out of pace, no matter sea state or crew veterancy level set through single mission options. The worst part is that their error was not random; they were uncalibrated and they fired regularly on top or below the target. Now, thanks to Volodya/Wamphyri's tweaks, all the guns are aiming correctely, but we have the opposite problem: we need to destabilize them, or planes won't pose any significant threat :-?

I am glad to hear that. It means that we are on the right direction. But as I told you before, now we need to finetune our settings (the game realities that Volodya was talking about a while back) :03:

Okay, time to start replacing stock sounds with more varied sounds :up:

Okay, this another good news, because, as I had wrote in my PM of yesterday, I had slightly modified Volodya's settings, and I was worried I could have done some mistake leading again to wrong aiming :up:

Yes we have still much to do, for making them lesser accurate (see above).

data\UPCDataGE\UPCCrewData\SpecialAbilities.upc:

data\UPCDataGE\UPCUnitsData\UnitParts*.upc (one file for each U-boat model in game):

The answer to our doubts regarding gun reload times and their accuracy could be in one of the above files. I am still struggling to undertand how their settings are applied in game, and if veterancy levels in single missions are roughly equivalent to passive ability levels in campaign.

Has anyone messed with those parameters before? you, Volodya? :hmm2:

Yes, I think you're probably right that the M42 didn't score any hits. I was trying to pay careful attention to the 37mm tracers and did not notice any impacts, but my testing was not thorough enough to state this with certainty.

Is changing elevation/training speed the only way to effectively control accuracy? You mentioned in your PM that there is a chance recoil_dist might affect accuracy. For flak guns, do you think this refers to the distance the bolt travels after firing one round and before another is chambered? I found this recoil distance for the C/30, as cited by our old friend Mirsolaw Skwiot:

average recoil distance: 33mm
max. recoil distance: 44mm

These numbers seem quite small so I really don't know what to make of them. I have unfortunately not been able to find information for the other guns yet.

The guns were set to fire at medium range for all tests. I agree that the M42 should have a range advantage over the 20mm guns. However, I have a suspicion that our SK C/30 range settings might be too high... :hmmm:

It's an interesting dilemma. Do you an objective standard by which you are planning to measure the success of our testing? That is, how will we determine whether our guns are too accurate or not accurate enough?

You might want to double-check my report before starting collecting sound clips in earnest, as I have been known to be wrong :O: In any case, the 37mm is far too quiet, as this youtube clip of the SK C/30 indicates:

http://www.youtube.com/watch?v=xJUuwhx4Xr4

Notice also how slowly the gun seems to track around 0:41 :03:



I certainly haven't. I'm was also using the Reworked Abilities mod and the Equipment mod for the tests, in case they affect these parameters.

I suspect that 37mm gun's laying/elevation speed need to be raised a bit, nonetheless can you please give a "fire at long range" order, the next time you test our settings?

currently, yes.
In past, I and Volodya have been testing tolerance fator's effect, but without significant success. There are two of them, one for elevation and one for traverse. According to this old post in SHIII subforum, tolerance factors should be sort of maximum gyrostabilization angles: the bigger their value, the more stabilized the gun, the more accurate its aim.

I think so, and it should work the same way for cannons (except for the reloading mechanism, which indeed is not automatic). But I haven't ever messed with that setting; all I know about it, is what I (or anyone else for that matter) can deduce from its name. Its effect on gun accuracy is only my (vague) supposition. :yep:

The stock recoil distance setting for 20mm guns is 10 cm. I will make a test with the numbers provided by you, and see if anything changes. Anyhow, if you find other statistics regarding gun recoil distances, please post them here.

Why do you think so?

These are the theoretical ranges, according to the sources I have consulted (including the information provided by you):

navweaps: 8,500 m @ 45 deg
wikipedia: 8,500 m @ 37.5°
navypedia: 8,500 m
your document: 8,500m @ 35.7°

They all agree on 8,500 m, despite some discrepancies on the elevation angle required for attaining the aforementioned range. Also notice that this is not the AA ceiling range, which was way shorter, nor the effective range in AA firing, that was even shorter. SH5 gun controllers allow just one setting, which is the maximum theoretical range. I ignore how accurately the game deduces the ceiling range from the latter, but in any case, there is a setting for limiting FlaK guns long, medium and close firing ranges. IIRC, the stock maximum range setting is 2,000 m, which is pretty close to the historical 37mm FlaK's maximum effective range :yep:

During our past tests, I an Volodya were comparing different results by reporting the numebr of rounds required to shoot down a plane, or the number of dive attacks the plane had performed before being downed. This was using the same test mission with the same weather/crew veterancy settings.

yes, I had seen that video before. Unfortunately I could not capture its audio due to the musical comment on the background (never before I had thought that I could have hated Mozart :shifty:), but IIRC I had also managed to find another video whose audio is "cleaner". :up:

:yep: :up:

good choice. EUF was anyway required, as I have included the settings for testing the M42 twin FlaK in our testing version of the mod, and the files I had modified were from TheBeast's mod. I was going to inform you on it, but at the last moment I have forgot to do it. :oops:

Answering to your question, yes both the mods mentioned by you contain the settings I was talking about in my previous post. I think we should try and play with them with the aim to reduce FlaK's accuracy and to reduced the excessive difference in reload times among trained and untrained crews (your estimated times seem to me abit excessive even for a rookie). :salute:

Yes, I will test with the "long range" order too.


Yes you are correct :oops: I was rechecking some numbers and found a comment about the Kriegmarine's disappointment with the SK C/30's very poor AA ceiling (only 2000m practically, versus 6800m theoretically). I confused ceiling with range and therefore thought I might have provided you with incorrect information. My mistake :oops:

Yes, it's not often that Mozart ruins a video :O:

Agreed. I will test other settings.

When its time for historical depth charges :up:
http://traktoria.org/files/underwate...structures.pdf
plenty of graphs

:up:

first rate stuff sober, :yep:
Depth charges damage in in our todo list. I will save that essay in a safe place, and I will have a closer look into it at the right moment.

Thank you for sharing :salute:

U-boat gun malfunctions
 

Hi guys,

A thought that jumped to my mind in the last days is that our guns are lacking zone information.

I am wondering if any of you has ever suffered gun malfunctions: Flaks jamming, getting damage from enemy fire and letting you defenseless in the middle of a surface engagemnt, and requiring long minutes if not hours before they are repaired.

During my limited gaming experience I haven't ever had any of the above...