Bio Engineer Archive

I mostly sampled the Bolle Bols in endor from another BE's advice. Found it worked perfectly considering they were a yellow mob and all. What i did was i already had masked scent, popped me can of jawa beer, and did the sampling. Not sure if mask scent did anything but i guess i followed someonez advice. And if the mob decides to aggro on me i just hopped on my bike and speeded outta there Did this all the way to master. You fail less also higher u get in DNA sampling.

---

Squygxicus wrote:

But the thing that keeps me pulling my hair out is trying to figure out the samples and how they fit etc. Theres usually too much information for me to "remember" when I put the combine together so when I get the final "result" its like well how did I do that. Also at the low levels Im using crappy DNA and I get nothing over a one digit percentage point except the last stat.

Just my personal experience. Im sure things will get easier when I move up but right now its almost too frustrating to figure out.

---

For me it's this uncertainty and complexity that was the most attractive. Until I get deeper into the data collection and analysis I won't be able to tell if my goal will be met or not. But sometimes it's the pursuit of a goal that is the reward.

There is a lot of good information out regarding forumula and such and I am studying everything I can find so as to not have to reinvent the wheel.

What I want to be able to do is have a customer coem into the shop or send me an email requesting a certain type of pet with specific stats or stat ranges. I want to be able to plug these design criteria in and have my tool first tell me if the combination is possible. If it isn;t then it will tell me some variations that are possible (less ham, higher CL, etc).

Then after selecting a design template, it will give me a "shopping list with various types of DNA to go hunting for each of the slots. I will not begin with a design for multi-generational pets quite yet until a single-generation pet tool is working.

This will require me to sample from a wide variety of creatures in order to develop a nice database. The database will hold a statistical model for each stat instead of specific values.. i.e, it will hold the Mean and Standard Deviation for the data distribution. In this manner, I will be able to study 2 or 3 types of creatures thouroughly (50 samples) and then model distributions for any creature given a small 3-5 piece sample. Over time, the database becomes more precise as more sample values are entered.

This sounds like work.. but for me, data analysis like this and then applying the data to making pre-designed pets is quite fun.