Difference between revisions of "Talk:Quadrant Delta-Delta-Delta Force"

Latest revision as of 01:11, 11 January 2010

Some raw data to maybe give someone better with numbers a chance at figuring out how the XP ties into stats.

Level 87 Gadgeteer
Str: 983 (808)
Int: 1180 (937)
Ref: 2073 (1520)

All +XP effects have been factored out for the numbers below.

Jellyfish Bot: 1469, 1469, 1723, 1836, 1751, 1667

Delta Queen: 1949, 1695, 1780, 1638, 1638

Squid Bot: 1723, 1780, 1921, 1723

Rho Master: 1695, 1780, 1582, 1751, 1836, 1695, 1723

Aquatic Landshark Bot: 1610, 1921, 1864, 1582, 1893

Bass Bot: 1780, 1667, 1949, 1780, 1582, 1780

Upsilon-Iota Bot: 1723, 1497, 1780, 1751

Prowler: 1723, 1638, 1780

Cristiona 20:48, 29 April 2008 (MST)

From yamum: i took a look at your post on Quadrant Delta-Delta-Delta Force. i believe it is poosible that the XP range could be avg stats to highest/main stat based on your results. i believe the range also to be sort of a personal XP cap.

This seems to jive with my gains, more or less. All of the numbers fit that range, but I think there might be some kind of variable, as I haven't broken 2000 yet. Cristiona

For what it is worth, I did about 50 turns in Delta-Delta-Delta and gained an average of about 550 XP. My stats are:

Lvl 66
Str 458 (416)
Int 682 (505)
Rfl 404 (384)

I will collect 40 observations in this same format when I get some time over the weekend and then we can try to build a regression model to tease out the logic. I would be helpful to have observations from more people though so if others could post 40 observations like Cristiona did that would be very helpful. Hodag

I asked Ryme, and all the bots should be the same, so we'll only need to record XP gains, not which robot gives which. Cristiona

I did a comparison between Delta-Delta and the Fiendish Pit. I started with the following stats:

Lvl 66
Str 458 (416)
Int 581 (505)
Rfl 404 (384)

and burned 50 turns off in each zone. What I found was:

Delta 30,537 points / 50 turns = 610.74 points average Pit 28,042 points / 50 turns = 560.84 points average

That is somewhat different from my earlier test where I only got 550 points in Delta. The only noticable difference was that my intelligence was lower (682 vs 581) but that would suggest the lower your stats, the higher your points! I will continue to collect some more data points and see what happens. I am not happy with the number of observations just yet.

In theory, the pit is running off level, not stats. The difference comes out with heavily skewed stats. Anyway, I've hit a new high: 2189 XP, with my highest stat being 2284. Highest -95... Also, re-reading, it's entirely possible that you were just within RNG swinging... especially since your pit gains were off from the expected average.

Anyway, I may have found the lower bound for the Delta. I managed to pull in 1469 XP when my average was also 1469. I haven't managed lower than that, so our range may be (average buffed stats) - ((highest buffed stat)-95(±?))

Of course, that doesn't jive with your 610 average for the Delta... did you have any +XP effects running? Cristiona

XP gain of 2247. Buffed Ref of 2304. Difference of 57. Getting closer. Cap may be (max buffed) - 50. Cristiona

New Theory

Having pulled in gains higher than my highest buffed stat is a little confusing. I find myself wondering if the cap might be highest buffed stat + level. Don't have hard numbers to back this up right now though. Also, I wonder if lower bound might not be an average of all buffed stats, but your highest unbuffed stat. This is a little difficult for me to test due to my stat distribution, though. Cristiona

A theory that fits

The formula itself isn't my own suggestion, but here are my observations on how it works.

XP for the other zones always involves N*randint(0,10), i.e. a uniform distribution between 0 and 10, summed N times. Taking this as a hint, the XP on Delta^3 fits ceiling((50+2*randint(0,10))*SumOfBuffedStats/150).

As noted above the lower bound is very close to the average of stats, in fact it is 0.3*SumOfBuffedStats, where the average is 0.33*SumOfBuffedStats. The cap is 0.47*SumOfBuffedStats. Considering that the distribution of stats is for most classes 0.25/0.25/0.5 (see here, the cap would appear as if it is the main stat.

So far this fits the data points I observed. Let me know if you see a data point that doesn't fit. --Muhandes 09:11, 11 January 2010 (UTC)

Data

aquatic landshark bot (442) - 201 enerbun / 23 cybertronium / 24 mona lisa drivetrain / 20 plasteel plating

bass bot (412) - 205 enerbun / 25 mona lisa drivetrain / 32 Aquasson logo

Delta Queen (461) - 240 enerbun / 23 cybertronium / 52 strut bar

jellyfish bot (424) - 228 enerbun / 50 laser macrodiode

prowler (453) - 218 enerbun / 21 laser macrodiode / 41 mona lisa drivetrain

Rho Master (426) - 210 enerbun / 44 titanium plating

squid bot (457) - 237 enerbun / 24 cybertronium / 21 strut bar

Upsilon-Iota Bot (426) - 193 enerbun / 20 cybertronium / 46 strut bar

3500 total turns

Encounter rate seems to be more or less evenly divided (do we have a way of saying that on the wiki, or is it just generally assumed that all areas are like that?). Adventuring with +item gear to spade the drop rate of enerbun, it seems to be 50% for all enemies. For other items, I separated like and unlike numbers to get the percentages (that is, aquatic landsharks and bass bots clearly drop drivetrains at a similar rate so I combined the results from either to get the percentage for both... while the prowler clearly has a different rate, so it was determined separately). --Jesus 06:15, 25 June 2009 (UTC)