(
Jump to the original submission)
Mon 11 Apr 2011 10:06:23 PM UTC, comment #7:
Re-did the second spreadsheet with the fixed probabilities. Looks great.
Attaching fixed spreadsheet.
(file #23205)
|
Mon 11 Apr 2011 09:58:27 PM UTC, comment #6:
Re-did the first spreadsheet with the fixed probabilities. It looks perfect.
Attaching the table generator, and the fixed spreadsheet.
(file #23203, file #23204)
|
Mon 11 Apr 2011 08:17:16 PM UTC, comment #5:
The discrepancy between warbot and lordsawar illustrated in target2.gnumeric has been resolved. The charts are very close now.
The fix is committed in svn revision 2652.
Unfortunately the fix doesn't make a whole lot of sense to me, but:
1. it's a symmetrical fix,
2. it results in fewer lines of code, and
3. the charts line up. :-)
Attached target4.gnumeric.
(file #23202)
|
Mon 11 Apr 2011 07:35:22 PM UTC, comment #4:
I noticed that the sheet labeled "warbot" was mislabelled. The sheet actually comes from a table on heeter.net.
I made a new sheet called warbot that has values generated by warbot. I could not cajole it to give 15 strength points to the defender, so the final row of the table isn't filled.
The warbot values look pretty similar to the heeter.net values.
Attached target3.gnumeric.
We can still see a noticeable bulge on the topmost line in the lordsawar chart.
(file #23201)
|
Mon 11 Apr 2011 06:07:50 PM UTC, comment #3:
I think the error might be illustrated in this spreadsheet of 1 on 1 battles of varying strengths.
Attached.
Even in the 1 on 1 battles, the dragons were always over-performing.
(file #23200)
|
Sun 10 Apr 2011 10:04:48 PM UTC, comment #2:
Here are 3 more charts.
1. 2 Elephants vs N Pegasi
2. 3 Heavy Cavalry vs N Pikemen
3. 2 Archons vs N Spiders
These charts look a little more accurate.
(file #23198)
|
Sun 10 Apr 2011 04:57:50 PM UTC, comment #1:
Whoops, I made a mistake in the dragon vs heavy infantry slide.
Re-attaching gnumeric file.
Is a 3 to 4 percent difference okay?
(file #23193)
|
Sun 10 Apr 2011 04:39:48 PM UTC, original submission:
Here are 3 charts showing a series of battles between multi-army stacks.
1. a unit of dragons vs 1 to 32 scouts.
2. a unit of devils vs 1 to 32 scouts.
3. a unit of dragons vs 1 to 8 heavy infantry.
Sometimes the error can be equal to a whole army unit. I think that is too much error.
|
Depends on the following items: None found
Items that depend on this one: None found
Follow 9 latest changes.
| Date |
Changed By |
Updated Field |
Previous Value |
=> |
Replaced By |
| Mon 11 Apr 2011 10:06:23 PM UTC | benasselstine | Attached File | - |  | Added multi-army-stacks-2-fixed.gnumeric, #23205 |
| Mon 11 Apr 2011 09:58:27 PM UTC | benasselstine | Attached File | - |  | Added multi-army-stacks-fixed.gnumeric, #23203 |
| | | Attached File | - |  | Added prob.cpp, #23204 |
| Mon 11 Apr 2011 08:17:16 PM UTC | benasselstine | Attached File | - |  | Added target4.gnumeric, #23202 |
| Mon 11 Apr 2011 07:35:22 PM UTC | benasselstine | Attached File | - |  | Added target3.gnumeric, #23201 |
| Mon 11 Apr 2011 06:07:50 PM UTC | benasselstine | Attached File | - |  | Added target2.gnumeric, #23200 |
| Sun 10 Apr 2011 10:04:48 PM UTC | benasselstine | Attached File | - |  | Added multi-army-stacks-2.gnumeric, #23198 |
| Sun 10 Apr 2011 04:57:50 PM UTC | benasselstine | Attached File | - |  | Added multi-army-stacks.gnumeric, #23193 |
| Sun 10 Apr 2011 04:39:48 PM UTC | benasselstine | Attached File | - |  | Added multi-army-stacks.gnumeric, #23192 |