Matrix Games Forums

To End All Wars: Mountain InfantryPandora: Eclipse of Nashira Announced! Deal of the Week: Command Ops goes half price!New Fronts are opening up for Commander: The Great WarCharacters of World War 1Sign of for the Pike and Shot Beta!More Games are Coming to Steam! Return to the Moon on October 31st! Commander: The Great War iPad Wallpapers Generals of the Great War
Forums  Register  Login  Photo Gallery  Member List  Search  Calendars  FAQ 

My Profile  Inbox  Address Book  My Subscription  My Forums  Log Out

Leo's ASW Test on 1.8.0.6 - NS vs ASW

 
View related threads: (in this forum | in all forums)

Logged in as: Guest
Users viewing this topic: none
  Printable Version
All Forums >> [Current Games From Matrix.] >> [World War II] >> War In The Pacific - Struggle Against Japan 1941 - 1945 >> Leo's ASW Test on 1.8.0.6 - NS vs ASW Page: [1]
Login
Message << Older Topic   Newer Topic >>
Leo's ASW Test on 1.8.0.6 - NS vs ASW - 10/11/2007 9:02:04 PM   
matchwood

 

Posts: 63
Joined: 9/9/2004
Status: offline
VSWG mentioned in his previous post about using NS instead of ASW because of an advantage in range. Doing the previous test I noticed an apprent advantage of ASW so I repeated the test to confirm spotting instead of sucessful hits. The table below shows spotted Subs by a squadron of H8K first on NS, then restarted on ASW. Fatigue, commander, location and number of subs to spot were equal.

The results show that ASW is almost twice as effective at spotting subs. I haven't tested the ratio of spotting to sucessful hits but I would expect that more spotting would translate into more hits (as long as the plane has a primary bomb load of non torps!




Attachment (1)
Post #: 1
RE: Leo's ASW Test on 1.8.0.6 - NS vs ASW - 10/11/2007 9:22:53 PM   
Procrustes

 

Posts: 633
Joined: 3/30/2003
From: Upstate
Status: offline
Interesting stats - thanks.  IIRC the effectivness of both is also correlated on distance from the base.  I thought that ASW was supposed to be most effective close to the airbase, while NS represents more wide-ranging searches and are more likely to discover ships and subs at range.  Do you have any thoughts on that?

Best,

(in reply to matchwood)
Post #: 2
RE: Leo's ASW Test on 1.8.0.6 - NS vs ASW - 10/11/2007 9:29:27 PM   
matchwood

 

Posts: 63
Joined: 9/9/2004
Status: offline
NS found a sub up 8 hexes away, ASW up to 7 - there didn't seem to be a huge difference.

NS doesn't seem to find anything more than about 8 away period, but that's an observation not from the test.

When I get a chance I can redo the test and check for distance.

I'm supposed to be working!

< Message edited by matchwood -- 10/11/2007 9:34:02 PM >

(in reply to Procrustes)
Post #: 3
RE: Leo's ASW Test on 1.8.0.6 - NS vs ASW - 10/11/2007 11:44:12 PM   
mantill


Posts: 1118
Joined: 3/2/2005
From: England
Status: offline
I don't want to get all mathematical here but if ASW searches at half range compared to NS you would expect spotting to be 4 times more efficient. As half the range equates to a quarter of the search area.



_____________________________


(in reply to matchwood)
Post #: 4
RE: Leo's ASW Test on 1.8.0.6 - NS vs ASW - 10/12/2007 1:54:01 AM   
VSWG


Posts: 3432
Joined: 5/31/2006
From: Germany
Status: offline

quote:

ORIGINAL: matchwood

NS found a sub up 8 hexes away, ASW up to 7 - there didn't seem to be a huge difference.

This is very interesting!

I've just asked this in the other thread, too: What are the experience levels of these air groups? IIRC forum wisdom says that units with low exp are very ineffective in spotting/hitting subs, can you confirm this?

_____________________________


(in reply to matchwood)
Post #: 5
RE: Leo's ASW Test on 1.8.0.6 - NS vs ASW - 10/12/2007 4:27:18 AM   
matchwood

 

Posts: 63
Joined: 9/9/2004
Status: offline
Average exp of the H8Ks when doing both NS and ASW tests was also 75.

I would expect though that more sightings = more exp gain although nothing significant over 12 turns. A couple of points maybe. Nothing to explain 2x better results.

(in reply to VSWG)
Post #: 6
RE: Leo's ASW Test on 1.8.0.6 - NS vs ASW - 10/12/2007 3:14:17 PM   
Apollo11


Posts: 22566
Joined: 6/7/2001
From: Zagreb, Croatia
Status: offline
Hi all,

Again - thanks for not forgetting my (old - I did it in v1.4 I think - I am at work now) test and testing with latest version!

The more tests the more things we can, hopefully and possibly, report as not OK and let our valiant programmers fix!


Leo "Apollo11"

_____________________________



Prior Preparation & Planning Prevents Pathetically Poor Performance!

A & B: WitW, WitE, WbtS, GGWaW, GGWaW2-AWD, HttR, CotA, BftB, CF
P: UV, WitP, WitP-AE

(in reply to matchwood)
Post #: 7
RE: Leo's ASW Test on 1.8.0.6 - NS vs ASW - 10/12/2007 4:18:50 PM   
trollelite

 

Posts: 444
Joined: 1/29/2006
Status: offline
Hey sir, can you please include dive bombers and torpedo bombers in your test? I guess D3A is more effective in NS than ASW.

And dare I say only "spot" is a bit useless, what more important is these SS are actually "attacked", can you include the number of actual attack in your report?

(in reply to Apollo11)
Post #: 8
RE: Leo's ASW Test on 1.8.0.6 - NS vs ASW - 10/15/2007 6:08:37 PM   
Procrustes

 

Posts: 633
Joined: 3/30/2003
From: Upstate
Status: offline

quote:

ORIGINAL: matchwood

NS found a sub up 8 hexes away, ASW up to 7 - there didn't seem to be a huge difference.

NS doesn't seem to find anything more than about 8 away period, but that's an observation not from the test.

When I get a chance I can redo the test and check for distance.

I'm supposed to be working!


Very interesting - thanks!

(in reply to matchwood)
Post #: 9
RE: Leo's ASW Test on 1.8.0.6 - NS vs ASW - 10/15/2007 10:43:08 PM   
Halsey


Posts: 4916
Joined: 2/7/2004
From: Indianapolis Indiana USA
Status: offline
Is what can really throw the results in these tests, is this.

There seems to be a TF density modifier for searches.

Put 20 seperate sub TF's out and there will be easy spottings and attacks.
Now put 20 subs in one TF, and the results will be way different.

My advice for gameplay is this.
One concentrated multi sub TF should go to a central location.
Then seperate sub TF's should can deployed from the combined TF.

Or, as I do in my games.
I stagger departures of my subs.
2-4 days per each sub, and vary their speed settings.







_____________________________

AE WITP Land Air Bitmap downloads

https://sites.google.com/site/aewitplandairartwork/home/ae-witp-bitmap-downloads

(in reply to Apollo11)
Post #: 10
RE: Leo's ASW Test on 1.8.0.6 - NS vs ASW - 10/16/2007 1:11:51 AM   
Pascal


Posts: 1637
Joined: 8/20/2003
From: in New England now after driving across US from CA
Status: offline
To do good sampling, you should repeat this sequence at least 20-30 times to see if you really get this difference in results. Remember, you are at the mercy of a random number generator, thus to eliminate 'good' and 'bad' sequences, you have to run your test many, many times.

If the programming is done right, the test of the ASW and NS subroutines should have been run a couple thousand times to see if results tallied with historical results...

_____________________________

So much WitP and so little time to play.... :-(


(in reply to matchwood)
Post #: 11
RE: Leo's ASW Test on 1.8.0.6 - NS vs ASW - 7/1/2008 9:59:10 PM   
vonSchnitter


Posts: 310
Joined: 7/2/2004
From: Germany - still
Status: offline
Hi Gents,

even though I am late to the party, some events in my current PBM as a JAP player (1.806, ABs extended map, latest CHS 2.08 scen with Nik Mod - 157 (?)) got me into a "test and re-evaluation" situation.

In the game it is late March 42 and my worthy opponent managed to thread a few of his subs through Malacca Strait and past Singapore. I got miffed to say the least.

So I remembered the various threads on Jap ASW - I had taken for gospel - downloaded Leos test scenario, learned a few things about the editor and got at it.

Even though my test - or sampling - base may be slim, there is a clear trend. As far as I can see.

One hint for further testing: Saved games with further tests - even with modifications - may result in false or misleading stats. I used an ASW TF in several tests. Guess what. The game played out identical every time - same ships attaking etc. Even when I changed the destination hex of the TF. Same outcome, with a saved game. Strange.

If you care to test, reload the scenario fresh - or use the editor to make the process less tedious.

Here is part of an e-mail I send my opponent:

I used "Leos" test scenario for a start and added any type of Jap plane type to the testing:

In general: All Airgroups have 27 planes, 80 experience, 90 morale and zero fatigue.
ASW/Search are performed at 100%, 1000 ft alt.

If it comes to ASW there are three issues: Find the subs, keep them hiding (to hinder them to attack), attack them and kill them.

AGs on Naval Search do detect Subs. However, the detection level is about less than half as effective as for those on ASW.
Even though Nells/Bettys detect more subs on search as Helens/Sallys/Lillies the relation is thus:
Nells/Bettys on search find about more than 1/3 more subs on search as Army level bombers.
Nells/Bettys on ASW find 1/2 more subs as Army level bombers.
Nells/Bettys on search very rarely attack subs - and virtually neve score.

ASW

Army level bombers on search find less than half the number of subs on search (compared to ASW), do attack occasionally and score a chance hit than and now.
Nells/Betties on ASW do attack Subs at half the rate of Army Groups on ASW, with half the chance of hitting per attack.
Army bombers on ASW have a ratio of detection/attack/hit of about 7/2/1
Lillies on ASW have a chance of multiple attacks/hits on a single sub at 2:1 compared to helens/ sallies.

Army DBs are quite useless at both Search and ASW.
On Search they just find about 1/2 to 1/3 of the Army LBs - and in both cases almost never attack ( I employed them at different alts with no discernable difference)

IJN Float planes (Alf, Jake, Pete) are pretty good at detecting Subs on ASW, but do not better on search as Army Babs, Dinah etc. And they almost never attack.

Scaling:

Naval search for sub detection with more than one unit is for the most part ineffective. Under good conditions a second unit may find some 25% more.
On ASW scaling is nearing 2/3ds.

Training and Commanders:

I do suspect, Naval ability of commanders have a say. However, since there are very few Army leaders with naval skills beyond 25, the issue is moot.
Search and ASW are very nice for training - having the proper leader in place, at the right time.

end of quote,

Anyway, Did some tests with ASW Hunter/Killer groups and was rather surprised with my findings,

Will report later


_____________________________



Remember that the first law of motion is to look where you're going. A man with a stiff neck has no place in an airplane.
Technical Manual No. 1-210, Elementary Flying, War Department, Washington,

(in reply to Pascal)
Post #: 12
Page:   [1]
All Forums >> [Current Games From Matrix.] >> [World War II] >> War In The Pacific - Struggle Against Japan 1941 - 1945 >> Leo's ASW Test on 1.8.0.6 - NS vs ASW Page: [1]
Jump to:





New Messages No New Messages
Hot Topic w/ New Messages Hot Topic w/o New Messages
Locked w/ New Messages Locked w/o New Messages
 Post New Thread
 Reply to Message
 Post New Poll
 Submit Vote
 Delete My Own Post
 Delete My Own Thread
 Rate Posts


Forum Software © ASPPlayground.NET Advanced Edition 2.4.5 ANSI

0.082