questions

I'm pretty confused on the exp formula. I was testing some fights and I can't get the outputted exp. Ever.

I've fought bots at 0.0 ratios and , by the formula, you should get 0 exp.

I fought a bot with a 0.0 ratio and I got 66 exp. I have 2 int. I plugged in the formula and I get only 1.12 exp.. No Exp buff.

I then fought a 0.01 bot with 200% exp buff. My formulated value was still 52% of the actual exp. What am I missing?

Effectively, I think you count the minimum ratio as 0.1 to make the formula still work, so the effective ratio range is somewhere between 0.1-3 where the exp is different.

Tryhard is right, but it's 0.01, not 0.1. If you give me a list of inputs (with wins and losses instead of ratio to avoid rounding problems) with the expected and actual output, I can take a look.

Screenshot of Excelsheet

The "variables" are 20 because I won each one. This is run in an excel sheet. The only result that is close is the second test. No idea what's happening. Also, none of these bots had 0.0 ratios.

I got the same EXP value (i.e. not EXP Bots4) from those numbers for test 1, so the most likely explanation is that one or more of the inputs is incorrect.

Shot in the dark: Are you including blocks and attempts from both the attacker and defender?

Ya, Total Blocks and Total Attempts.

Can you post a screenshot of the fight summary chart with a new calculation where the expected doesn't match the actual?

Fight -> http://i.imgur.com/DroLq0t.png

Excel -> http://i.imgur.com/vnmcTSi.png

http://bots4.net/forum/1/3162

I cracked the training part, but not fighting.

o/

Weird, I get the same non-matching results as you, both manually and with actual game code...

Woo! I'm not going crazy then!

Found the problem. In test 5, your spreadsheet lists Ag30 as having 1,908 wins. It currently has 1,806. Using the correct number produces the expected result.

So..That fixes every other one too? :P

Can't say without knowing what bots you were fighting. Can you post another fight and spreadsheet screenshot where the expected output does not match the actual?