I have been reading various loot theories and am going to a conduct a series of experiments that collect data people can go crazy with. Going to start out by establishing baselines: S.I. Combat Knife First hunt: S.I. Combat Knife 161 Carabok Celeste Quarry ([30817, 9651, 109]) Avg. Cost per kill 4.267 pec Eco 2.344 dmg/pec Loot is in line with expectations. Eco is lower than I had anticipated; I will do several more baseline hunts to bring the total sample size to ~1000 for this setup (and confirm the eco), but that will be after Gold Rush event. S.I. Combat Knife 0 -6.87 PEDTotal expenses. -6.87 PED...Animal Hide +15 +0.15 PEDAnimal Muscle Oil +17 +0.51 PEDAnimal Oil Residue +60 +0.60 PEDBLP Pack +2521 +0.25 PEDBasic Cloth Extractor +1 +0.01 PEDBlazar Fragment +12 -0.00 PEDCarabok Hide +33 +0.66 PEDCarabok Leg Fur +3 +0.60 PEDDiluted Sweat +85 +0.85 PEDEnhanced Cloth Extractor +1 +0.01 PEDInferior Cloth Extractor +12 +0.12 PEDNova Fragment +141 -0.00 PEDShrapnel +20449 +2.04 PEDSocket I Component +1 +0.10 PEDWeapon Cells +4197 +0.42 PEDTotal returns. +6.32 PED...Result 91.99% -0.55 PEDEdit: filled in missing information about first hunt, edited commentary slightly in response to cost calculations.
Thanks, Starkiller. I appreciate the well-wishes. Gold rush hunts are not part of my actual experiments, but in the event the data is useful, here is the record of my next hunt. 2nd Hunt: Ozpyn 347 Magurg Breeder 1 Gold Rush Category I Instance 41644 damage inflicted. 120.01 dmg per creature implies 9.1% overkill (versus 110 hp per creature as per entropedia.) Kill cost/creature 43.9 pec Eco = 2.731 dmg/pec Heal points on self, 1294 Heal costs 0.8% Vivo S10 eco 9.585 hp/pec (Switched from Universal Ammo to BLP packs at 25 minute mark. Universal ammo consumption of 120.72 ped during first 95 minutes implies 31.77 ped BLP consumption in final 25 minutes, rather than 13.11 ped shown. Extra 18.66 ped BLP should be added to consumption & loot. Actual costs 167.39 v. loot of 161.95 (both increased by extra BLP consumed) mean 96.75% return.) BLP Pack -131081 -13.11 PEDBullseye I (L) 0 -1.38 PEDNeurobiotic Booster A1 1mg -2 -0.02 PEDOzpyn BP S1X1 0 -12.15 PEDUniversal Ammo -1207200 -120.72 PEDVivo S10 0 -1.35 PEDTotal expenses. -148.73 PED...Animal Hide +894 +8.94 PEDAnimal Muscle Oil +1374 +41.22 PEDAnimal Oil Residue +235 +2.35 PEDArkoin +24 +0.24 PEDBlazar Fragment +48 -0.00 PEDCat.1 Event Token 1 point +17 +0.17 PEDCat.1 Event Token 10 Points +2 +0.02 PEDCat.1 Event Token 5 points +1 +0.01 PEDNova Fragment +276 -0.00 PEDScott & Barlow BP-6 (L) +1 +12.18 PEDShrapnel +436716 +43.67 PEDSocket I Component +62 +6.20 PEDSoft Hide +4 +0.40 PEDThin Wool +19 +4.75 PEDTier I Component +70 +7.00 PEDWeapon Cells +145407 +14.54 PEDWool +8 +1.60 PEDTotal returns. +143.29 PED...Result 96.34% -5.44 PEDEdit: added heal eco. Does anybody know whether heal points on self from instance stats includes the heal over time effect of Vivo S10? My calculated heal eco is well below the level stated by entropedia, but it would be in line with entropedia numbers if heal over time is not included in instance stats.
Skipped a few logs, but here is my 10th Gold Rush Instance: Ozpyn BP S1X1 w/ Bullseye I (L), no enhancers 328 Magurg Breeder 1 Gold Rush Category I Instance 39251 damage inflicted. 119.67 dmg per creature implies 8.8% overkill (versus 110 hp per creature as per entropedia.) Kill cost/creature 44.49 pec Eco = 2.690 dmg/pec Heal points on self, 1186 Heal costs 0.85% Vivo S10 eco 9.565 hp/pec A little slower because I was watching a movie. This was by far my worst, both on loot return and on points. Overkill was about the same as last time. Last time I had accuracy enhancers, so I think that explains some of the eco difference (since none broke it exaggerated the eco of my previous hunt). I still think that the instance doesn't count the heal over time, so actual healing eco is higher than shown. I may do one more GR instance to see if I can get a better normalized score, since this run was so poor. Either way, I will plan on continuing my hunting experiments soon. Bullseye I (L)0-1.32 PEDNeurobiotic Booster A10 1mg-1-0.01 PEDOzpyn BP S1X10-11.56 PEDPiron PBP-0 (L)0-0.06 PEDUniversal Ammo-1317316-131.73 PEDVivo S100-1.24 PEDTotal expenses.-145.92 PED...Animal Hide+550+5.50 PEDAnimal Muscle Oil+956+28.68 PEDAnimal Oil Residue+195+1.95 PEDArkoin+115+1.15 PEDBLP Pack+37131+3.71 PEDBlazar Fragment+270.00 PEDBullseye II (L)+1+7.34 PEDCat.1 Event Token 1 point+10+0.10 PEDCat.1 Event Token 10 Points+3+0.03 PEDCat.1 Event Token 20 points+1+0.01 PEDCat.1 Event Token 5 points+8+0.08 PEDFine Hide+1+0.25 PEDMagurg Male Pincer+1+0.20 PEDNova Fragment+252+0.00 PEDShrapnel+315086+31.50 PEDSocket I Component+103+10.30 PEDSoft Hide+6+0.60 PEDThin Wool+3+0.75 PEDTier I Component+35+3.50 PEDWeapon Cells+92282+9.22 PEDWool+5+1.00 PEDTotal returns.+105.87 PED...Result72.55%-40.05 PED
Back to establishing baseline for SI Combat Knife: 31 Monura Female 22 Monura Male 46 Carabok 20 Gallard 15 Arkadian Hornet Arkadia Event Area 2 [16978, 25667, 175] Dmg/pec 2.34 Eco is in line with last baseline test on SI Combat Knife. (Dmg used for calc is total hp of creatures killed, so eco does not compensate for overkill.) As can be seen, I was swunting a bit (51.9% of potential bottles gathered) so sample is small compared to how long it took me. Will continue to do that for the next few tests, since the sweat gained makes up for the TT losses on the hunt, and the defensive skill gain is higher when swunting. S.I. Combat Knife 0 -6.69 PEDUniversal Ammo -400 -0.04 PEDTotal expenses. -6.73 PED...Advanced Leather Extractor +2 +0.02 PEDAdvanced Stone Extractor +1 +0.01 PEDAnimal Hide +12 +0.12 PEDAnimal Muscle Oil +5 +0.15 PEDAnimal Oil Residue +72 +0.72 PEDArkadian Hornet Hide +2 +0.04 PEDBLP Pack +2235 +0.22 PEDBasic Wood Extractor +2 +0.02 PEDCarabok Hide +8 +0.16 PEDDiluted Sweat +42 +0.42 PEDGallard Hide +11 +0.22 PEDMonura Female Hide +22 +0.66 PEDMonura Male Hide +12 +0.36 PEDNova Fragment +123 +0.00 PEDPiron PLP-0 (L) +1 +0.59 PEDShrapnel +16933 +1.70 PEDVibrant Sweat +819 0.00 PEDWeapon Cells +3532 +0.35 PEDTotal returns. +5.76 PED...Result 85.59% -0.97 PED
SI Combat Knife 16 Monura Female 16 Monura Male 30 Carabok 16 Gallard 20 Arkadian Hornet Arkadia Event Area 2 [16978, 25667, 175] Dmg/pec 2.45 Better eco than the last couple of baseline tests, but 4% difference could be explained by less overkill. I do have a couple of shots on Arkadian Hornets with my Z12 Barbarella when I couldn't reach with my knife, but not enough to swing the eco as much as we're seeing. Too small a sample to draw any conclusions. After finishing out 1000 creatures with the SI Combat Knife and determining average eco for the whole series, I will baseline with some guns with a smaller damage range (PBP-0 series). More shots, less overkill, more precise calculation. Then I will have two separate baselines to work with for the next set of experiments. Depending on how those go, I may need to do a set on creatures with 20-25 hp with the SI Combat Knife to determine actual eco without so much overkill. Overall it was a good hunt: S.I. Combat Knife0-4.59 PEDUniversal Ammo-615-0.06 PEDTotal expenses.-4.65 PED...Animal Hide+12+0.12 PEDAnimal Muscle Oil+21+0.63 PEDAnimal Oil Residue+55+0.55 PEDArkadian Hornet Hide+5+0.10 PEDBLP Pack+2068+0.21 PEDBasic Wood Extractor+2+0.02 PEDBlazar Fragment+120.00 PEDCarabok Hide+8+0.16 PEDCarabok Leg Fur+5+1.00 PEDDiluted Sweat+21+0.21 PEDGallard Hide+4+0.08 PEDMonura Female Hide+2+0.06 PEDMonura Male Hide+2+0.06 PEDNova Fragment+930.00 PEDShrapnel+18295+1.82 PEDVibrant Sweat+610-0.00 PEDWeapon Cells+4661+0.46 PEDTotal returns.+5.48 PED...Result117.85%+0.83 PED
SI Combat Knife; Z12 Barbarella only as necessary when Hornets were out of reach. 42 Monura Female 39 Monura Male 54 Carabok 51 Gallard 22 Arkadian Hornet Arkadia Event Area 2 [16978, 25667, 175] Dmg/pec 2.47 S.I. Combat Knife0-9.71 PEDUniversal Ammo-9225-0.92 PEDZ12 Barbarella0-0.06 PEDTotal expenses.-10.69 PED...Advanced Stone Extractor+2+0.02 PEDAnimal Hide+18+0.18 PEDAnimal Muscle Oil+17+0.51 PEDAnimal Oil Residue+115+1.15 PEDArkadian Hornet Hide+14+0.28 PEDBLP Pack+3152+0.32 PEDBlazar Fragment+190.00 PEDBody Fat+2+0.16 PEDCarabok Hide+13+0.26 PEDDiluted Sweat+10+0.10 PEDGallard Hide+67+1.34 PEDHair Gel+6+0.78 PEDInferior Cloth Extractor+1+0.01 PEDMonura Female Hide+14+0.42 PEDMonura Male Hide+5+0.15 PEDNova Fragment+190+0.00 PEDPiron PBP-0 (L)+1+0.46 PEDShrapnel+30888+3.08 PEDSocket I Component+1+0.10 PEDVibrant Sweat+76-0.00 PEDWeak Cloth Extractor+3+0.03 PEDWeapon Cells+5050+0.51 PEDTotal returns.+9.86 PED...Result92.24%-0.83 PED
SI Combat Knife. 41 Monura Female 40 Monura Male 37 Carabok 47 Gallard 20 Arkadian Hornet Arkadia Event Area 2 [16978, 25667, 175] Dmg/pec 2.45 S.I. Combat Knife 0 -9.71 PEDTotal expenses. -9.71 PED...Animal Hide +41 +0.41 PEDAnimal Muscle Oil +29 +0.87 PEDAnimal Oil Residue +135 +1.35 PEDArkadian Hornet Hide +2 +0.04 PEDBLP Pack +3480 +0.35 PEDBlazar Fragment +11 0.00 PEDCarabok Leg Fur +1 +0.20 PEDCorria Timber +3 +0.60 PEDGallard Hide +38 +0.76 PEDMonura Female Hide +9 +0.27 PEDNova Fragment +141 0.00 PEDPiron PBR-0 (L) +1 +1.46 PEDShrapnel +26063 +2.60 PEDWeapon Cells +6458 +0.64 PEDTotal returns. +9.55 PED...Result 98.35% -0.16 PED
SI Combat Knife 46 Monura Female 36 Monura Male 34 Carabok 43 Gallard 32 Arkadian Hornet Arkadia Event Area 2 [16978, 25667, 175] Dmg/pec 2.49 S.I. Combat Knife0-9.71 PEDUniversal Ammo-205-0.02 PEDZ12 Barbarella0-0.01 PEDTotal expenses.-9.74 PED...Advanced Leather Extractor+1+0.01 PEDAdvanced Stone Extractor+2+0.02 PEDAnimal Hide+41+0.41 PEDAnimal Muscle Oil+28+0.84 PEDAnimal Oil Residue+96+0.96 PEDArkadian Hornet Hide+17+0.34 PEDBLP Pack+2893+0.29 PEDBlazar Fragment+860.00 PEDCarabok Hide+10+0.20 PEDCarabok Leg Fur+1+0.20 PEDDiluted Sweat+13+0.13 PEDGallard Hide+13+0.26 PEDMonura Female Hide+4+0.12 PEDMonura Male Hide+20+0.60 PEDNova Fragment+169+0.00 PEDShrapnel+20400+2.05 PEDWeapon Cells+3907+0.40 PEDTotal returns.+6.83 PED...Result70.12%-2.91 PED
That brings my tracked total with SI Combat Knife to 977, which I will consider to be close enough to 1000 to call it a decent baseline. 176 Monura Female 153 Monura Male 362 Carabok 177 Gallard 109 Arkadian Hornet Total Cost 48.39 Dmg/pec 2.45 Total Return 43.8 TT 90.51% For my first experiment, I want to separate dmg done from creature hp. To do this, I will knife each creature once, then sweat while it recovers, then kill it once it's dry. I will continue to calculate dmg/pec using the base hp of creatures killed. Ideally it should decrease to below 2.00 over sample size of 1000 punies, which would represent at least an 18.37% decrease in 'eco' but driven by a corresponding increase in actual damage done. If loot multipliers are based on actual damage or decay, hunt results should fall within normal variance (85-95%). If loot multipliers are actually based on creature hp, results would be expected to fall to below 78%. Any outcome between 78% and 85% would be considered inconclusive and I will expand the sample size for baseline and this experiment to ~2000 punies.
It's an interesting experiment you're doing (if I understand it correctly). However, did you also consider the increased defense-cost while you're sweating it for the mob to regain health? I've done my small share of experimenting, but I found that they are all inconclusive. The loot swings from one sample to another is just too large. You might also, from my own experience, get a result well between 85-95%, yet with a standard deviation of 10-15%. Good luck with the punies, and remember that Entropia is about having fun!
If you want to know why this might not be a good idea, do it in a team of two: A does some damage to the mob; A & B sweat it until it fully recovers; B kills it; How much loot does A get? Spoiler (in my experience, A doesn't get anything. Been some time since the last experiment though, so things may have changed)
Since I'm sweating punies, I don't use armor or fap, and heal up naturally when necessary. This is an interesting point, and highlights a potential issue with the test design. If the system is "losing track" of initial damage once creature heals to full, as the example illustrates, then the test might not conclusively separate damage done from the health of the creature. It would expect a lower rate of return even if loot multiplier is driven by a single-variable (damage) model. A few hundred creatures into the experiment, I am seeing lower returns, but the sample isn't large enough for it to be outside of normal variance yet. These wide standard deviations are why I'm trying to keep a sample size around 1000 for control group and test group. From what I've read, it looks like the sample size to narrow the return into a 2-3% band for any particular setup is about 5000 loot events. I think that indicates 1000 loot events would be enough to narrow the expected band to 10-15% (so a result more than 15% outside of expected values can be considered conclusive, more than 10% indicative, and less than 10% inconclusive). I could perform larger tests, but when I'm testing a setup or technique that is indicative of lower return after the first 1000 loot events, larger tests have significantly greater costs.
