KobaBOS Spreadsheet

[kobayashi]

That is quite strange cause everything appears in the Main page called Master. As a test, if you enter 1,1,1,1,1,1. The Odds of A winning (light yellow) box should have either 0% or 1% in it. If not, there is definitely something wrong.

There should be a Master sheet and eight other sheets (named one,two....eight). These eight sheets also have an something like the input table but they are not highlighted in blue. Would you be entering you numbers in the wrong sheet?

Do you want me to send you another zip file in case the one you have was corrupted?

 

[Field Marshal Klesh]

Originally posted by kobayashi:
That is quite strange cause everything appears in the Main page called Master. As a test, if you enter 1,1,1,1,1,1. The Odds of A winning (light yellow) box should have either 0% or 1% in it. If not, there is definitely something wrong.

After entering A=1,1,1 B=1,1,1 nothing happens on the Master page. The odds of A winning (little yellow box) stays at'####' and doesn't change.

There should be a Master sheet and eight other sheets (named one,two....eight). These eight sheets also have an something like the input table but they are not highlighted in blue. Would you be entering you numbers in the wrong sheet?

I am entering the data on the Master page for sure.

Do you want me to send you another zip file in case the one you have was corrupted?

Sure. I don't know what would've corrupted it, but we can try again. Thanks alot.
FMK

 

[willemvanoranje]

He's right! Same thing is happening to me, now! And it doesn't work on other computers I have access to either! They ask for #NAAM? (NAAM means 'NAME'). And all the other pages are full of it. I don't undersand. I'm sure it worked the first time. Could it have something to do with 'translations'or so? What language did you use in programming (English, American English, something else)

FMK is right, it says ######. I don't think it has got anything to do with a corrupted zip, the extractor will warn you if it is corrupted.

------------------
Concordia res parvae cres****.

[This message has been edited by willemvanoranje (edited April 16, 2001).]

 

[Field Marshal Klesh]

You got it dude, thats exactly what is going on...

 

[willemvanoranje]

This is what is written in D11:
=C9/(C9+F9)

This is what is written in F13:
=(One!A7+Two!A7+Three!A7+Four!A7+Five!A7+Six!A7+Eeight!A7)/80/C7

This is what is written in F14:=(One!A8+Two!A8+Three!A8+Four!A8+Five!A8+Six!A8+Eeight!A8)/80/F7

Corect me if I'm wrong FMK. My guess there's something wrong with the eight!-things. I don't think it has something to do with Office 97 or 2000. I tested it on both, and both didn't work.

------------------
Concordia res parvae cres****.

 

[Field Marshal Klesh]

For me:

D11: ######

F13: #NAME?
F14: #NAME?

 

[kobayashi]

Folks, sorry for the problem. I have downloaded the zip from the download section and it works fine for me. Lets assume then that it is a version problem. You'll have to help me fix it as I don't have acess to Office 2000.

Willem, you seem to understand a bit more of what's going on so I'll work with you first. FMK can take a well deserved rest for the time being.

My guess is that the spelling of one of the functions has changed in the 2000 version and when it sees a function it doesn't recognise, it will assume that it is a label (which hasn't been created of course)

First I'll assume that the Eeight is a typo when you posted the message. Next, I'll need you to go to any one of the locations that is specified in F13:

(One!A7+Two!A7+Three!A7+Four!A7+Five!A7+Six!A7+Eight!A7)/80/C7

i.e. A7 of any other sheet other than Master. You can jump to any sheet by clicking on the tabs at the bottom. See if the A7 cell is also gone nuts and if it is, see where the cell reference leads to and continue to trace the path unitl the original fault.

Ultimately, this will lead to the two columns titled A dice and Bdice which have a function

=RANDBETWEEN(0,$B$5) and =RANDBETWEEN(0,$E$5).

 

[Field Marshal Klesh]

Originally posted by kobayashi:
FMK can take a well deserved rest for the time being.

*Kicks up feet and lights up a smoke*

 

[Field Marshal Klesh]

Any progress boys? I got an itchy mouse finger...

 

[kobayashi]

Yeah any progress Willem?

BTW FMK, if you post the stats that you want to experiment with here, I will process the results for you as an interim measure. Just put them in a colums

for example

A H F D H F Awin Bwin
12 2 2 8 4 2 26(31) 74(35)

A wins 26% of the time and has 31% hit points left
B wins 74% of the time and has 35% hit points left

 

[willemvanoranje]

Sorry guys. I'm at school now. I can't do anything.

------------------
Concordia res parvae cres****.

 

[willemvanoranje]

I think I found the problem and have the ability to fix it. Everyone who has problems with it, should install the Analysis Tool Pack. You can check how to do this in the help-files.
The problem is the RANDBETWEEN. You need the Tool Pack to be able to use it.

------------------
Concordia res parvae cres****.

 

[kobayashi]

Here is a very helpful message I received from Richard Arbuckle III:

*****
I was looking at your Civ II Battle simulator and had the same problem that a few people have posted about, i.e. the majority of cells across the 9 sheets have the #NAME? error.

After a bit of investigation I found the cause of the error was the RANDBETWEEN function. Checking the help I found that to use that function you must have installed the Data Analysis add-in, which I hadn't done.

Once I added the Data Analysis add-in everything works fine. As I am not a registered member of the site, I haven't bothered to post this there, but thought you may be interested to know. Since you obviously already have the Data Analysis add-in installed it must be confusing as to why your spreadsheet wasn't working for anyone else.
*****

To make a long story short, you must choose Tools/Add Ins/Analysis Tool pack. Seems I did this a long time ago and forgot about it. It doesn't require any diskette or anything and the only difference is that your Excel will take slightly longer to launch.

Many thanks to Richard. I have made a new zip which includes instructions regarding this on the front page and sent it to thunderfall. If you have already downloaded the previous version, you do not need to download the new one as it is exactly the same.

Added later:
...Hey WillemO (clap clap)discovered the same thing and posted a message while I was typing this one..


[This message has been edited by kobayashi (edited April 19, 2001).]

 

[willemvanoranje]

The problem for me now is that even now I installed it, the program doesn't translate. SO I'll have to do it all manually. Lucky for me it are only eight tabs.

------------------
Concordia res parvae cres****.

 

[Field Marshal Klesh]

BOOYAH!!!

That did it boys. I had to find my Microsoft Office2000 disk, but it installed no problem and ran smooth after install. This program is gonna be great!
Thanks again for all the help fellas.

FMK

 

[deity]

There's a huge difference between the BOS and the Probability Computer .exe file.
The latter seems to be more accurate.
------------------
deity of THE DEITIANS

[This message has been edited by deity (edited April 25, 2001).]

[This message has been edited by deity (edited April 25, 2001).]

 

[kobayashi]

KobaBOS is based directly on what is given in the manual. As I mentioned earlier in the thread, there is a more accurate method proposed based on binomial distribution - which I don't fully understand myself. There may also be other factors (like maybe there are different formulas for different cases???) which I did not take into account. Seems to be that KobaBOS is reasonably accurate if the numbers are not too small, but if anyone can point out what is missing or wrong I can try to modify what I have now.

FMK, if you have been trying it out, did you find any major discrepencies between the spreadsheet and actual testing?

 

[Field Marshal Klesh]

Originally posted by kobayashi:


FMK, if you have been trying it out, did you find any major discrepencies between the spreadsheet and actual testing?

Ah, koba... I actually have been heavy into that yet. Still running into other issues to deal with...

One thing that this won't help for is defensive bonuses from terrain. A lot of the fighting will be taking place on heavy defensive terrain which is sure to skew the results in favor of the defender. I'm moronic when it comes to mathematics, so I don't know how to add the terrain bonuses to the defender's stats.

If a defender has a 5 for def. and is on a mountain (standerd Mtn. defense) what would I enter for the defenders #.... 8, 9, 10? Where are you math gurus?

FMK

 

[kobayashi]

I don't know whether you know this, you can control the defense bonus of the terrains. In default, the Def Number of mountains is 6, which you have to divide by 2 = 3.

Therefore just multiply the defender's Defense factor by 3. 1 becomes 3, 2 becomes 6 and so on.

 

[Field Marshal Klesh]

I can dig that kob.

I tested for a bit today. Just a few units, but the simulator worked spot on for me. And that little equation you just posted is exactly what I needed. Thanks.

FMK

[This message has been edited by Field Marshal Klesh (edited April 28, 2001).]