Bio Engineer Archive
Thread: Additives more info needed.
Bracci
Thu Jul 08, 2004 11:54 pm
#27
Seiryuu I'm taking a numerical methods class in college this summer, and damnit, you blew my mind with that post. Yet one more reason I say that BE is not only the most difficult profession to master (not the title, but be the actual master of), but is also the profession with the most intelligent people in it. (of course, being an engineer I equate intelligence with math/science skill typically - biased, I know, but 
). Well done! I only hope we can either a) figure out w t f is going on, or b) get it fixed so that it *is* reproducible. I figure our pet line is so random, that maybe we could at least catch a break and have one line that is easy to comprehend/master 
.
). Well done! I only hope we can either a) figure out w t f is going on, or b) get it fixed so that it *is* reproducible. I figure our pet line is so random, that maybe we could at least catch a break and have one line that is easy to comprehend/master .
Seiryuu
Fri Jul 09, 2004 12:08 am
#28
For everyone's information, the range on all nutrients is 30.
MN = 30-60
BSN= 60-90
INN= 90-120
Please note your initial assembly success. This affects the final product percentage. An amazing lets you make a better product than a moderate success.
I will try to remember to do a few when I get home and post the results, Nancy.
MN = 30-60
BSN= 60-90
INN= 90-120
Please note your initial assembly success. This affects the final product percentage. An amazing lets you make a better product than a moderate success.
I will try to remember to do a few when I get home and post the results, Nancy.
Joilhath
Fri Jul 09, 2004 4:08 am
#29
Well I'm going to stick with my theory that OQ determines the initial reuslt of the tissue without experimentation and PE and to a lesser degree FL determine how high you cn get it after that point. I would have never agued the 33/33/33 untill I got these results.
using 658/993/999(fl/pe/oq) 874 weighted avg organic and 986/854/1000(fl/pe/oq) 937 weighted avg meat I get an x8 ie 58/88/118 nutrient, but if I use 970/968/841(fl/pe/oq) 917 weighted avg organic with the same meat I get x7 nutrients.
Now the 58's started as 38's before experimentation and the 57's started as 37's and iff I ass some 995 OQ wood to the higher avg but lower OQ floral I can get 58's that start at 38 before experimentation.
Now this isn't enough for me to know what the actually percentages are on the final outcome but it is enough for me to know that it's not 33/33/33 all the time at least.
Bracci
Fri Jul 09, 2004 5:28 am
#30
JayceMilam wrote:as we've have suspected it is not 33/33/33, but i also have been unable to establish a reliable formula. I don't have the math skills that some others have on here, but I also have gotten wildly varying numbers when tryign to weigh avg's. I tried to figure if the meat was calculated differently than the floral, then the two were avg'ed together but the numbers were so far off I gave up. Maybe we should make this our 19 answers question.
And the response would be "please provide more data". I'd rather Nancy just ask in the correspondent forums and give em some time to dig around in the code than ask TH for another "evasion" rather than an "answer".
Denizen_Nightstalker
Fri Jul 09, 2004 5:52 am
#31
I have heard different theories on one topic of this thread and that is.
When a person uses wood where flora can be used
It doesnt have Flavor or Potential Energy ...I have heard that that calculates as a 0.
I have also heard that it removes itself from the formula ..therefore requiring a less demanding resource.
Can a few people please clarify this for me ...
If somethin has 33/33/33
and you use something that only has one of the 3 ..
..does that mean that it calculates the missing numbers with a 0 or a N/A ?
D
Biytor
Fri Jul 09, 2004 11:58 am
#32
This is they way I've had it explained and it works well enough.
Resource OQ / PE / FL
Flora933 857 864
Meat 966 773 617
Wood 983xxx xxx
-------------------------------------
2882
Grozurr
Fri Jul 09, 2004 2:24 pm
#33
just wondering,,, what would happen to your sample if you modified it to have either 3 different resources or used the flora in 2 and meat in 1? as i was looking over this all i saw almost all 3 ways but i was wondering how much a difference would have, because using more meat tends to raise prices sharply.
as for the wood theory, i believe that it just leaves the missing values out instead of weighting them as 0. using some 900+OQ and 800+PE oats (no fl) i managed to get 84bsns (this is all from memory so sadly can't give specific #s), putting the oats in both flora food and organic spots, and i'm pretty sure if it had calculated it as 0 i couldn't have gotten 84's
Grozzer
CresenKitty
Fri Jul 09, 2004 7:47 pm
#34
I was attempting to make high lvl BSNs tonight and figured I would add to this information.
35 Berries: 919/839/1000
25 Berries: 919/839/1000
20 Meat: 971/986/999
Used all but 1 of my experimentation points (maxed out)and ended up with 94% on experimentation. However it only had +88 for the additive (are we positive that perfect is a 90?)
The ave of mine were 926.97/ 879.12/ 989.67 which averaged out to 931.92 or 93%. Not quite the same as the 94% I achieved.
aswex
Fri Jul 09, 2004 8:22 pm
#35
I realize I'm going to frustrate a lot of people with this post. I just hope those people realize that by posting some, but not all of what I know, I will completely confirm acouple suspicions. In RL I have adegree in Statistics, and I've spent the last few months as a BE collecting data and performing statistical analyses (designing multivariate regression models to be exact). I can predict the max-experimented percentage on any additive tissue within 1%. I will confirm the following: 1) The FL/PE/OQ ratio of 33/33/33 is incorrect. It is somewhat close, but inaccurate enough to provide the occasional surprising result. 2) The weight of each resource used (organic, flora food, creature food) is very closely related to the number of units of each resource used in the schematic. Please forgive me for not posting THE FORMULA, but I would hate to bring a complete end to discussions such as these and encourage the devs to nerf the formula (effectively flushing all of my work down the toilet). I realize that I've gained enough valuable insights from these boards tooffer something back. I also thank those of you discussing the use of wood as an organic. I have not tried that, and I look forward to experimenting in my lab, especially since someone claimed a +86 BSN using wood- that certainly raised my eyebrows. One final note- on a few occasions my final product has ended up with a percentage well-above expected- enough for at least an additional +1 bonus. Subsequent experiments with the same ingredients yielded the expected percentage. I may have selected different ingredients than I wrote down, but I am suspicious. This has only happened 2 or 3 times.
Respectfully yours,
Isscossk Osli, owner
BE Tissues by Isscossk (Naritus)
Seiryuu
Fri Jul 09, 2004 8:28 pm
#36
To figure out the value of the BSN from the percentage, multiple it by 30 and add 60.
Using your 94% example:
(0.94 * 30) + 60 = 28.2 + 60 = 88
To get 90, you will need absolutely perfect 1000 stats in every field. (Assuming there is not some hidden factor.) Because of this, the obtainable maximum will be 89, which requires 96.67% once experimentation is done.
I don't think I said this in my analysis above, however the percentage shown (plus the hidden decimal value) does follow this without fail. The only thing in question is how it reaches this percentage because given the 33/33/33 formula, the end result should usually be different.
Using your 94% example:
(0.94 * 30) + 60 = 28.2 + 60 = 88
To get 90, you will need absolutely perfect 1000 stats in every field. (Assuming there is not some hidden factor.) Because of this, the obtainable maximum will be 89, which requires 96.67% once experimentation is done.
I don't think I said this in my analysis above, however the percentage shown (plus the hidden decimal value) does follow this without fail. The only thing in question is how it reaches this percentage because given the 33/33/33 formula, the end result should usually be different.
aswex
Fri Jul 09, 2004 8:54 pm
#37
OK. After thinking it over, I realized that I only confirmed things that people were pretty convinced of. So, to actually be of benefit to the discussion, I'll give you the following spoilers: 1) OQ is the most important spec for your final product and the PE is the least important, but 33/33/33 isn't abad estimate. 2) The percentage displayed is the actual percentage rounded down. 3) When the percentage is used to calculate thebonus of the tissue, the bonus is rounded down.
By the way, after my last post a few minutes ago, I had an epiphany, altered my existingformula, and I cannow predict percentage and bonus with 100% accuracy. It's a proud moment for me,considering it's the first time I actually used what I studied in college 7 years ago.
Respectfully yours,
Isscossk Osli, owner
BE Tissues by Isscossk (Naritus)
aswex
Fri Jul 09, 2004 8:58 pm
#38
Quick addendum to my above spoilers: Use the actual percentage to calculate the bonus, not the displayed percentage. This is why you can have 73% BSNs with either a +81 or +82 bonus.
Respectfully yours,
Isscossk Osli, owner
BE Tissues by Isscossk (Naritus)
Bracci
Fri Jul 09, 2004 9:08 pm
#39
Now if I could just get someone to explain Taylor's Expansions to me...
GJ to those of you who most likely (do not take this as an insult as I typically classify myself in this category) take the typical insult of "math geek" as a compliment.
GJ to those of you who most likely (do not take this as an insult as I typically classify myself in this category) take the typical insult of "math geek" as a compliment.