Home » PLAYER'S HQ 1.13 » v1.13 Combat/Weapon Academy » Practice / Trainer / Student Effectivness
Practice / Trainer / Student Effectivness[message #328796] Wed, 04 December 2013 08:48 Go to next message

Registered:November 2007
Location: Croatia
I've always wondered how much time I needed to train using these assignments. Also, I never knew how much faster can I train my mercs by using the STUDENT-TRAINER assignment vs plan PRACTICE.
So I've made some tests to see how effective are the assignments.
I tested using the latest 1.13 r6671

PRACTICE assignment:
I tested using 2 IMPs:
IMP "All 50" has all stats set to 50 except for WISDOM 80.
He reperesents a potentially good merc with a low stat of 50 that you want to increase. For example Fox (WIS-76,STR-55) and you want to increase her STR.

IMP "All 80" has all stats set to 80 (including WIS)
He represents a good/excelent merc with a stat of 80 that you want to increase. For example Len (WIS-83,AGI-80) and you want to increase his AGI

Both of them are LVL 1 and have no Traits, Personalities, Disabilites or Backgrounds that could interfere with the calculation.

I tested the time needed for:
IMP "All 50" to gain +1 in a skill and +5 in a skill
IMP "All 80" to gain +1 in a skill
I didn't test +5 for the "All 80" IMP as this would take a lot of time (both in-game and real time)

Once the merc gained a point I'd stop time compression, record the time and load the game to repeat the test.

Each test was repeated 3 times and the final value is an avarage (hours are rounded down). Sleep time is included with one exception - If a merc went to sleep with just a few subpoints left for an increase, then I woke him up to finish the job in the next hour (otherwise there would be an error by ~8 hours needed for sleep).
Here's the results:

The COMPARISON row represnts the avarage time for each stat so you can see how fast a certain stat grows in comparison to others. AGILITY, as the slowest growing is the benchmark at 1.0X.

In the attributes department we can see a typical difficulty in rasing AGI. In order of need time, then come DEX, STR, HLT and LDR. A suprise for me was that STR training takes a bit longer then HLT. Also I didn't know that LDR training was the fastest of all.
As far as skills go, no supprises for me here. MRK takes the longest, then come MEC, MED and EXP as the fastest.

There you go, now you know how much time you'll approximately need to raise your stats by PRACTICE.
Please keep in mind that the higher the stat, the more it takes for it to raise (duh!)

STUDENT / TRAINER assignments

The point of this test was to determine how much faster is the STUDENT assignment vs the PRACTICE assignment.

For this test I used the "All 50" IMP as the universal STUDENT.
For TRAINER's I used 2 IMPs:
1. The "All 80" IMP from the above PRACTICE test as a genaraly good merc - he's the "Moderate trainer".
2. A new IMP called "Max trainer"
He has all stats 100 (including WISDOM), LVL 10 and the "Teaching trait". He Represents the best possible trainer of any stat.

The avarage result between the "All 80" and "TrainMAX" merc shuld represent a very good TRAINER you would typically use in-game.

I only tested the MRK, EXP & MED skills with +5 gain, to conserve on time. I assume that the advantage of STUDENT above PRACTICE should be universal along all stats, so there was no need to test all stats. More on this below (***)

Here's the results compared to the previous PRACTICE data:


The table here shows the times need for +5 gain with the Moderate and Max trainers. Below that is a comparison between PRACTICE and STUDENT (Moderate and Max).
As I mentioned earlier, the middle ground between Moderate and Max should represent an in-game scenario.

The Average of the 3 tested skills is what I have for now so...

For the time being I can conclude that the STUDENT assignment is on avarage ~43,3% more efficient then PRACTICE.

For more info on stats and leveling see Hedrocks how to series HERE. Hopefully someday we'll come up with a more detailed formula on stat gains from assignments, exactly from the code.

*** I was wrong, as it seems. There is no universal gain in STUDENT vs PRACTICE along the stats. There's a difference of ~25% between gains of +5 between MRK,EXP and MED when looking at the results from the table above.
When looking at Max trainer increases there is less deviation but still a ~5% difference.
There could be 2 reasons:
1. The game system has different calculations for STUDENT & TRAINER assigns for each stat.
2. Most likely, my test are a bit off. Maybe I have an error somewher or I should make more repetitions to get more even data.
I'll make a few more test and report back.

[Updated on: Thu, 05 December 2013 05:34] by Moderator

Re: Practice / Trainer / Student Effectivness[message #328800] Wed, 04 December 2013 16:42 Go to previous messageGo to next message
Sam Hotte

Registered:March 2009
Location: Middle of Germany
You most likely have an considerable influence of chance/luck in your tests since you'll need quite a lot of values to rule out chance by the "law of large numbers".
3 tests is far too low to allow for conclusions about expected values.

Another factor would be experience level as at least the teacher is getting better with higher level (not sure, though, about the student/practice and if difference of levels between student and teacher may also be important).
Re: Practice / Trainer / Student Effectivness[message #328854] Sun, 08 December 2013 16:53 Go to previous message

Registered:November 2007
Location: Croatia
Tnx for your input, Sam. Your probably right, I'd need a lot more test. I just got an idea. I'll use facilities to uber increase the performenca of task. This will allow me to make a lot of test in a short time.
Previous Topic: Use of 10x scope?
Next Topic: how do i move my mobile militia?
Goto Forum:

Current Time: Mon Apr 24 18:14:58 EEST 2017

Total time taken to generate the page: 0.00887 seconds