Leo's air ASW Search TEST WitP v1.795...

[Apollo11]

Hi all,

One month ago I posted the thread:

"Leo's ASW TEST (and scenario) - problems spotted!"

tm.asp?m=967245&mpage=1


The tests were done with v1.70 BETA and now I compared results with new v1.795 BETA that contains changes in air "ASW" routines...


Here is what I got in v1.70 BETA:




NOTE:
My original typo (last 0 was "lost") - last two numbers in table should be "10000ft" and "20000ft"!



Attached at the end of this message is what I get with v1.795 BETA.


Now... I know that this is just one test (relationship between number of submarines sighted and selected altitude) but it seems that number of sightings is still at least 10x too high... dunno...


Leo "Apollo11"

[Ron Saueracker]

The model assumes the sub is on the surface, that's the problem. As an aside, ever wonder why there is no submerged speed value?

[Bradley7735]

Hi Leo,

Have you counted the number of attacks and the number of successful attacks? My understanding of the code changes is that the number of attacks and successful attacks have been reduced. Not necessarily the number of sightings. Also, I think that a sighting report won't necessarily put an enemy sub on the map. (like you can see the sighting report, but looking at the map doesn't always show a sub.)

I haven't tested this, but I think it's the attacks that have been tweeked. Not the actual sightings.

bc

[pompack]

ORIGINAL: Bradley7735

Hi Leo,

Have you counted the number of attacks and the number of successful attacks? My understanding of the code changes is that the number of attacks and successful attacks have been reduced. Not necessarily the number of sightings. Also, I think that a sighting report won't necessarily put an enemy sub on the map. (like you can see the sighting report, but looking at the map doesn't always show a sub.)

I haven't tested this, but I think it's the attacks that have been tweeked. Not the actual sightings.

bc

If I understand correctly, the sightings probability is now a function of platform altiude. This is in addition to the reduction in hit probabilities as a function of commander rating.

[Ron Saueracker]

Have you tried the test vs just one sub?

[Apollo11]

Hi all,

ORIGINAL: Bradley7735

Have you counted the number of attacks and the number of successful attacks? My understanding of the code changes is that the number of attacks and successful attacks have been reduced. Not necessarily the number of sightings.

The number of attacks I did not count but it was much lower than with prevoous versions (i.e. WitP v1.795 lowered this)...

Also, I think that a sighting report won't necessarily put an enemy sub on the map. (like you can see the sighting report, but looking at the map doesn't always show a sub.)

I counted the subs placed on map thet were discoverd by enemy side.

Therefore sub sighting = sub on the map.

I haven't tested this, but I think it's the attacks that have been tweeked. Not the actual sightings.

Hmmm... might be...


Leo "Apollo11"

[Apollo11]

Hi all,

ORIGINAL: pompack

If I understand correctly, the sightings probability is now a function of platform altiude. This is in addition to the reduction in hit probabilities as a function of commander rating.

This is what I thought as well... but my test didn't show this as much as I was hoping... <SIGH>


Leo "Apollo11"

[Apollo11]

Hi all,

ORIGINAL: Ron Saueracker

Have you tried the test vs just one sub?

Nope... but it should not matter at all...


I used this (36 submarines in total - number of sighted I gave in my initial post here in thsi thread):




Leo "Apollo11"

[Ron Saueracker]

ORIGINAL: Ron Saueracker

Have you tried the test vs just one sub?

Be interesting to see how many of the aircraft on search spot the only sub in the vicinity. If more than one aircraft do spot the puppy, well then matey, we actually do have an obvious case of overspotting.[:)] Seems to be getting worse than better since I mentioned the sub spotting was too high during UV. Hmmm

[Ron Saueracker]

ORIGINAL: Apollo11

Hi all,

ORIGINAL: Ron Saueracker

Have you tried the test vs just one sub?

Nope... but it should not matter at all...


I used this (36 submarines in total - number of sighted I gave in my initial post here in thsi thread):




Leo "Apollo11"

A picture is worth a thousand words! Pfffft[X(] F--k that' ridiculously absurd. Man this is way off. Maybe two or three in a concentration like that may have been spotted on a perfect visibility day with flat sea.

[DFalcon]

ORIGINAL: Ron Saueracker


A picture is worth a thousand words! Pfffft[X(] F--k that' ridiculously absurd. Man this is way off. Maybe two or three in a concentration like that may have been spotted on a perfect visibility day with flat sea.

Ron, that is the allied player turn so he can see all the subs.

[Apollo11]

Hi all,

ORIGINAL: Ron Saueracker

A picture is worth a thousand words! Pfffft[X(] F--k that' ridiculously absurd. Man this is way off. Maybe two or three in a concentration like that may have been spotted on a perfect visibility day with flat sea.

Yes I know... and it's just 9 aircraft on search... 10x too many sightings IMHO...


BTW, please read in detail my tests (I did all other relationship tests: range, altitude, percentage):

"Leo's ASW TEST (and scenario) - problems spotted!"

http://www.matrixgames.com/forums/tm.asp?m=967245&mpage=1


Leo "Apollo11"

[Apollo11]

Hi all,

ORIGINAL: DFalcon

Ron, that is the allied player turn so he can see all the subs.

I did my tests with FoW = ON and number of sightings varied - please look at the numbers below...


Here is what Allied player sees:




Here is what Japanese player sees (after his best search result):




Here is what I got in v1.70 BETA:




NOTE:
My original typo (last 0 was "lost") - last two numbers in table should be "10000ft" and "20000ft"!



Here is what I got in new v1.795 BETA:




Leo "Apollo11"


P.S. [Edit]
Clarification added.

[Ron Saueracker]

ORIGINAL: DFalcon

ORIGINAL: Ron Saueracker


A picture is worth a thousand words! Pfffft[X(] F--k that' ridiculously absurd. Man this is way off. Maybe two or three in a concentration like that may have been spotted on a perfect visibility day with flat sea.

Ron, that is the allied player turn so he can see all the subs.

Really!? [:D] LOL Yeah, guess you are right . Maybe the right picture is worth something.[:)] Still, the chance is way too high according to everything from gut feelings to test results.

[Erik Rutins]

Any results or comments on the difference in attacks or the effect of the messages since the previous beta?

Regards,

- Erik

[Apollo11]

Hi all,

The TEST ASW scenario is here (it is here for past 30+ days when I first made it) and all can test - please do download it and do test it!


Download Link:

Leo ASW TEST Scenario #120.zip (10-16-2005)


Leo "Apollo11"

[Apollo11]

Hi all,

ORIGINAL: Erik Rutins

Any results or comments on the difference in attacks or the effect of the messages since the previous beta?

This test of mine was sightings only (i.e. to show how many submarines were discovered in most favored situation).

But what I can tell is that number of actual attacks that happen after the submarines were sighted was much smaller (I didn't count it though) - that's good!

What is still not good IMHO, is that way too many submarines are sighted even when search ASW aircraft is flying very high (10x too many IMHO)...


Leo "Apollo11"

[Nikademus]

Leo,

try it with not quite so many subs saturating the area around the base. (example, try 5-10 subs located at various distances from the base)

[Apollo11]

Hi all,

ORIGINAL: Nikademus

try it with not quite so many subs saturating the area around the base. (example, try 5-10 subs located at various distances from the base)

But what would that accomplish?


Leo "Apollo11"


P.S.
If you run my scenario you can see how search is done in WitP - clockwise concentric circles (from indide towards out)!

[Ron Saueracker]

ORIGINAL: Apollo11

Hi all,

ORIGINAL: Nikademus

try it with not quite so many subs saturating the area around the base. (example, try 5-10 subs located at various distances from the base)

But what would that accomplish?


Leo "Apollo11"


P.S.
If you run my scenario you can see how search is done in WitP - clockwise concentric circles (from indide towards out)!

Good question. I was thinking just use one sub as a target and see how many planes of the search pattern spot it but this is actually impossible to do. My bad there. If we could determine how many of the planes spot the one sub it would have been immensely useful in illustrating the relative ease of spotting a sub.