Page 1 of 1

LCU Data Query

Posted: Mon Nov 29, 2010 11:35 am
by herwin
I'm interested in writing an R script to allow me to assess the combat power of LCUs. Is there some way of dumping the LCU data as a summary .csv and tables of <LCU,Device,count> for the initial organisation and the TO&E organisation?

RE: LCU Data Query

Posted: Mon Nov 29, 2010 2:04 pm
by n01487477
ORIGINAL: herwin

I'm interested in writing an R script to allow me to assess the combat power of LCUs. Is there some way of dumping the LCU data as a summary .csv and tables of <LCU,Device,count> for the initial organisation and the TO&E organisation?
Herwin,
I'm surprised you haven't looked at witpload before... or is this not what you wanted ?

Open a dos command shell

In the SCEN folder

witploadae.exe /e /s01

where s30 is the scenario number.
Then look in witploc001.csv

All the info is there and will need to be linked with the devices in witpdev001.csv

RE: LCU Data Query

Posted: Mon Nov 29, 2010 2:17 pm
by herwin
ORIGINAL: n01487477

ORIGINAL: herwin

I'm interested in writing an R script to allow me to assess the combat power of LCUs. Is there some way of dumping the LCU data as a summary .csv and tables of <LCU,Device,count> for the initial organisation and the TO&E organisation?
Herwin,
I'm surprised you haven't looked at witpload before... or is this not what you wanted ?

Open a dos command shell

In the SCEN folder

witploadae.exe /e /s01

where s30 is the scenario number.
Then look in witploc001.csv

All the info is there and will need to be linked with the devices in witpdev001.csv

Looks good. It will take a bit of dancing through my a**e backwards, but I think I can do something with Java. Does anyone want a combat power calculator?

RE: LCU Data Query

Posted: Mon Nov 29, 2010 2:26 pm
by Tijanski
ORIGINAL: herwin
ORIGINAL: n01487477
ORIGINAL: herwin

I'm interested in writing an R script to allow me to assess the combat power of LCUs. Is there some way of dumping the LCU data as a summary .csv and tables of <LCU,Device,count> for the initial organisation and the TO&E organisation?
Herwin,
I'm surprised you haven't looked at witpload before... or is this not what you wanted ?

Open a dos command shell

In the SCEN folder

witploadae.exe /e /s01

where s30 is the scenario number.
Then look in witploc001.csv

All the info is there and will need to be linked with the devices in witpdev001.csv

Looks good. It will take a bit of dancing through my a**e backwards, but I think I can do something with Java. Does anyone want a combat power calculator?
That sounds interesting but it depends on what a combat power calculator calculates and how it calculates it.

RE: LCU Data Query

Posted: Mon Nov 29, 2010 4:21 pm
by herwin
ORIGINAL: Tijanski

ORIGINAL: herwin
ORIGINAL: n01487477

Herwin,
I'm surprised you haven't looked at witpload before... or is this not what you wanted ?

Open a dos command shell

In the SCEN folder

witploadae.exe /e /s01

where s30 is the scenario number.
Then look in witploc001.csv

All the info is there and will need to be linked with the devices in witpdev001.csv

Looks good. It will take a bit of dancing through my a**e backwards, but I think I can do something with Java. Does anyone want a combat power calculator?
That sounds interesting but it depends on what a combat power calculator calculates and how it calculates it.
It will summarise the unit attributes, calculate AV, and calculate both soft and hard firepower for both the current strength and the TO&E strength. It will also take into account such things as commander skills, unit experience, morale, disruption, preparation, and fatigue to generate various estimates of actual firepower and AV. The only thing it won't take into account is corps and theatre commander preparation.

RE: LCU Data Query

Posted: Mon Nov 29, 2010 6:32 pm
by davbaker
Sounds good to me, I'm pretty useless at estimating this stuff in WITPAE, getting better, but it's all rather bloody trial and error.
Anything that give some pointers would be useful!

RE: LCU Data Query

Posted: Mon Nov 29, 2010 6:40 pm
by herwin
OK, the tables load into R really smoothly. I can convert them into R data frames and save them as tab delimited easily. Then I can use a StringTokenizer to load them into Java. The one issue may be memory usage.

RE: LCU Data Query

Posted: Mon Nov 29, 2010 7:30 pm
by Tijanski
ORIGINAL: herwin
It will summarise the unit attributes, calculate AV, and calculate both soft and hard firepower for both the current strength and the TO&E strength. It will also take into account such things as commander skills, unit experience, morale, disruption, preparation, and fatigue to generate various estimates of actual firepower and AV. The only thing it won't take into account is corps and theatre commander preparation.
That sounds very good. Most of that is already in the downloadable csv files from witploadae, but a different perspecitve is always welcome. What do you mean by "take into account"? That is the code phrase the developers use to say "we can't tell you". Are you one of the developers? They seem to be secretive these days. Can you say what "take into account" means? If you are a developer and don't want to comment, just say so and I will back off.. Thanks in advance.

RE: LCU Data Query

Posted: Mon Nov 29, 2010 7:41 pm
by Pascal_slith
herwin isn't one of the developers.  He's one of the ones that have been asking quite a few of the 'hard questions' about some of the logic of the system over the years. To the benefit of the rest of us. A mine of info, too.

RE: LCU Data Query

Posted: Mon Nov 29, 2010 9:43 pm
by Tijanski
ORIGINAL: Pascal

herwin isn't one of the developers.  He's one of the ones that have been asking quite a few of the 'hard questions' about some of the logic of the system over the years. To the benefit of the rest of us. A mine of info, too.
So he is not a devloper? but he is a friend of yours and so that means what?

The question remains. What do you mean by "take into account"?

RE: LCU Data Query

Posted: Mon Nov 29, 2010 10:36 pm
by n01487477
ORIGINAL: herwin

ORIGINAL: Tijanski

ORIGINAL: herwin


Looks good. It will take a bit of dancing through my a**e backwards, but I think I can do something with Java. Does anyone want a combat power calculator?
That sounds interesting but it depends on what a combat power calculator calculates and how it calculates it.
It will summarise the unit attributes, calculate AV, and calculate both soft and hard firepower for both the current strength and the TO&E strength. It will also take into account such things as commander skills, unit experience, morale, disruption, preparation, and fatigue to generate various estimates of actual firepower and AV. The only thing it won't take into account is corps and theatre commander preparation.
The problem with what you're doing is that it's only a snapshot at the beginning of the war. The csv files are just the scenario files and as soon as turn 1 is resolved then you're working off inaccurate data.
morale, disruption, preparation, and fatigue
... will not be in the witpload csv files.

You'd be much better off connecting to the Tracker DB or generating csv files from that to get the data... and it does some of the things you're including already...

Having said that any user generated help / calculator is always welcome.

RE: LCU Data Query

Posted: Tue Nov 30, 2010 7:31 am
by herwin
ORIGINAL: Pascal

herwin isn't one of the developers.  He's one of the ones that have been asking quite a few of the 'hard questions' about some of the logic of the system over the years. To the benefit of the rest of us. A mine of info, too.

I was involved in witpstaff...

RE: LCU Data Query

Posted: Tue Nov 30, 2010 10:58 am
by floydg
ORIGINAL: herwin

ORIGINAL: Pascal

herwin isn't one of the developers.  He's one of the ones that have been asking quite a few of the 'hard questions' about some of the logic of the system over the years. To the benefit of the rest of us. A mine of info, too.

I was involved in witpstaff...

As Damian said, things like disruption and fatigue are only in the save files, so you have a few options to get it:
[*] extract from WitPTracker database
[*] extract from WitPStaff database
[*] write a plug-in for WitPStaff
[*] work with Matrix to get direct access to the save files

Floyd

RE: LCU Data Query

Posted: Tue Nov 30, 2010 1:49 pm
by herwin
ORIGINAL: floydg

ORIGINAL: herwin

ORIGINAL: Pascal

herwin isn't one of the developers.  He's one of the ones that have been asking quite a few of the 'hard questions' about some of the logic of the system over the years. To the benefit of the rest of us. A mine of info, too.

I was involved in witpstaff...

As Damian said, things like disruption and fatigue are only in the save files, so you have a few options to get it:
[*] extract from WitPTracker database
[*] extract from WitPStaff database
[*] write a plug-in for WitPStaff
[*] work with Matrix to get direct access to the save files

Floyd

I know a bit about writing plug-ins for WitPStaff. That might be the best way to handle things--it would save doing the rest. The problem is that Woos hasn't been visible for a year--he disappeared about the time we were working on the intel plug-in, and the program needs updating for the new Java version.

RE: LCU Data Query

Posted: Mon Dec 06, 2010 12:26 pm
by herwin
I've put together a R script to calculate combat power based on the various csv files that can be dumped from a scenario. I've noticed a few things:

1. soft firepower is a constant times the AV count. For most combined arms units, the constant is about 20.
2. Load cost for a combat unit is approximately the soft firepower.
3. Artillery and armour generate a bit more soft firepower per AV.
4. hard firepower is more complicated. On the other hand, it functions by reducing the AV and soft firepower that can be applied in the battle.

So you can continue to use AV as a proxy for combat power. Pay attention to where the opponent has armour, and make sure likely lines of advance are blocked, preferably by forces with antitank firepower.

The most important factors in determining who wins are the multipliers--terrain, fortifications, preparation, leadership, combat skill, morale, experience, disruption, supply, fatigue, and higher HQ support. Given a choice between adding AV to a hex and improving the multipliers in the hex, usually go with the latter. Any unit not immediately involved in combat should be constantly working on improving preparation, morale, experience, disruption, and fatigue. Any unit not needed immediately for combat in the hex occupied or moving should be in rest or reserve. If you have specialists that are not needed, remove them from the hex to rest or place them in reserve.

RE: LCU Data Query

Posted: Tue Dec 07, 2010 7:20 am
by herwin
I'd like to put together a detailed summary of the effects of non-material factors on LCU combat and the ways the non-material factors can be improved. This is a starting list with a few comments:

1. Terrain--up to 4x increase in AV for the side holding the hex. (x4--urban, x3--mountain & swamp, x2--jungle & forest) Can be improved by fortification (+33% per level). Going from 0 to 9 fortifications seems to reduce your casualties due to antisoft firepower by a factor of 3.
2. Preparation for target hex--if the hex is not being prepared for, AV halved. If it is being prepared for, AV multiplied by 1+preparation/100. Unknown effects on firepower.
3. Leader land skill--?
4. Leader leadership--?
5. Leader administrative skill--?
6. Leader inspiration--?
7. Morale--? Improved by rest/training once preparation reaches 100%.
8. Unit experience--I don't know the effect. Improved by being involved in combat (1-2 %).
9. Disruption--?
10. Fatigue--?
11. Supply--AV halved if insufficient, quartered if none. Unknown firepower effect.
12. Support--?
13. Corps preparation for target hex--?
14. Army preparation for target hex--?