Doctor Archive

---

ShadovvIVIaster wrote:

My theory on this is that the values of the BECs are shown as integers whereas on the server they are stored as their accurate value.

---

Yes, I also believe that to be the case. It seems that rounding errors occur the most of small values, such as number of charges. If I would expect to see discrepencies in values, I would expect it to appear on things like charges.

Awesome post. I am trying to work out the math behind the entire process, and while I haven't done extensive searches of the forums yet, this post and Jaato's armoursmith one have given me many pointers in the right direction. I'm not that great at spotting patterns, but I am a programmer, and when it comes to making systems like this, I use the most basic formula I can and let the data do the work. Makes things easy to debug. I imagine the devs are probably working on a similar approach, but then again, they may not be.

So, incase anyone is interested, I did some of my own research and tried to blend it with mcglonec's data so that we can just have more data... because people like me just... like data...

Test 1
12.93 tool, 12.5 crafting station

Organic: Ewriob, Endorian Insect Meat
OQ: 983
PE: 993
UT: -
Inorganic: Carbewia, Duranium Steel
OQ: 820
PE: -
UT: 740

Results
power: 81 @ 27%
charges: 12 @ 23%

Experimentation start:
9 slots for charges / 2 already used
10 slots for power / 2 already used

Experiments, using 1 point at a time in power only

Exp1: great success
power: 83 @ 34%

Exp2: great success
power: 85 @ 41%

Exp3: great success
power: 87 @ 48%

Exp4: great success
power: 88 @ 55%

Exp5: great success
power: 90 @ 62%

Exp6: amazing success
power: 92 @ 70%

Exp7: amazing success
power: 94 @ 78%

Exp8: great success
power: 96 @ 85%

Exp9: great success
power: 98 @ 92%

Exp10: critical failure
power: 96 @ 85%

Test 2:
12.93 tool, 38.5 crafting station
Same resources, same initial stats as Test1

Exp1 using 8 points and recieving 9 worth: amazing success
power: 97 @ 91%

Exp2 using 1 point: amazing success
power: 98 @ 93%

Test 3:
Ran indentical to test 2 with oddly enough, identical results.

Some results andthoughts...

DELTA = (98 - 81) / (93 - 27)
= 17 / 66
= 0.25757
BASE = POWER - (EFF * DELTA)
= 81 - (27 * 0.25757)
= 81 - 6.95
= 74.05

Some other data...

OQ1 + OQ2 = 983 + 820 = (1803 / 2) * 0.66 = 594.99
PE1 + PE2 = 993 + N/A = 993 * 0.33 = 327.69
UT1 + UT2 = N/A + 740 = 740 * 0.33 = 244.2

OQ + PE = 594.99 + 327.69 = 922.68 or 92.268% (funny how my cap was 93% for experimentation)
** edit **
OQ + UT = 594.99 + 244.2 = 839.19 or 83.919%

This leads me to believe that you get n + 1 experimentation slot for every 10% of resource quality. ie.
(OQ + PE) / 100= 922.68 / 100 = ~92%
(OQ + UT) / 100 = 839.19 / 100 = ~83%

Power Exp Point Slots = 92 / 10 = ~9; n + 1 = 9 + 1 = 10;
ChargeExp Point Slots =83 / 10 = ~8; n + 1 =8 + 1 = 9;
** edit **

This is where I keep coming unstuck. What I want to know is:

• What determines the starting experimentation % and what the formula for that is


• What determines the cap of experimentation % and what the formula is....

I feel that it is directly related to the resource stats somehow, but I'm not sure exactly what it is. I hope someone out there can use my data, or at least direct me on how to perform a better test, or point me to some existing explanation of the system. I currently have a head cold and I've lost track of my thoughts....

Message Edited by MasterNerfSlayer on 05-05-2004 09:59 PM

Read next post, Lithium's code generator screwed this one up

Message Edited by LordOfFatness on 05-05-2004 10:10 AM

Message Edited by LordOfFatness on 05-05-2004 10:13 AM

---

MasterNerfSlayer wrote:

Organic: Ewriob, Endorian Insect Meat OQ: 983 PE: 993 UT: - Inorganic: Carbewia, Duranium Steel OQ: 820 PE: - UT: 740

Results power: 81 @ 27% charges: 12 @ 23%

Exp1 using 8 points and recieving 9 worth: amazing success power: 97 @ 91%

Exp2 using 1 point: amazing success power: 98 @ 93%

This is where I keep coming unstuck. What I want to know is:

• What determines the starting experimentation % and what the formula for that is


• What determines the cap of experimentation % and what the formula is....

Message Edited by MasterNerfSlayer on 05-05-2004 09:59 PM

---

((983+820)/2) * .66 + (993) * .33 = 922.98

(993 + 983 + 820) / 3 = 932

Message Edited by LordOfFatness on 05-05-2004 10:16 AM

---

LordOfFatness wrote:

---

MasterNerfSlayer wrote:

This is where I keep coming unstuck. What I want to know is:

• What determines the starting experimentation % and what the formula for that is


• What determines the cap of experimentation % and what the formula is....

Message Edited by MasterNerfSlayer on 05-05-2004 09:59 PM

---

For the cap, it just uses our standard formulas listed in the schematic. For example, for power on a Stimpack B, we have OQ * .66 + PE * .33. With your resources, we get:

((983+820)/2) * .66 + (993) * .33 = 922.98

I'm really not sure how their formula is actually coded, they could also use an equation like (OQ1 + OQ2 + PE) /3. This is essentially the same thing, but yields a slightly different result:

(993 + 983 + 820) / 3 = 932

Either one of these formulas can account for your 93% experimentation cap that you hit. The 922 can be rounded up to 930, or the 932 can be truncated to 930, since they don't account for the decimal when displaying the experimental percentage, it's difficult to say exactly what is done concerning this.

I'm going to play around with the initial experimentation levels to see what that formula is. I thought I had an idea, but it isn't panning out.

Message Edited by LordOfFatness on 05-05-2004 10:16 AM

---

I have been playing around with these formulas for a while now and there is one thing that has bothered me the whole time. The max power formula P=((OQ1+OQ2)/2000)*.66 + (PE1/1000)*.33 will only have a maximum value of 99%. This immediately sends a red flag up for me and leads me to believe that it is not entirely correct. I have tried different variations of it, but none of those seem to work any better. With that being said, I also have a formula that i think is the starting experimentation percentage which is: (((P*.15)+.15)*P)

Give it a try a let me know how it goes.

Stven of Sunrunner