Weaponsmith Archive

Hey D'Tox,

I think Scruftydog has a point here. Though both she and I are assuming that the same system for making powerups goes for making weapons.

I was crafting the other nite powerup after powerup for the hell of it. 2 Hrs later I saw some very big differences in the way the product turned out just by changing the way I experimented.

1. Each initial Assemble produced different stats even when modifiying the same stat ie. Speed or Max Dmg.


2. Using 4-5 experiment points at a time out of 10 almost never gave me more than 2 Stats being modified.


3. Using 1 stat at a time I was able to consistently make powerups that gave 3 & 4 Stat increases.

So far there are many instancesof crafted items showing a decimal like Crafting tools. Installations. blah blah. I think it would be a safe guess to say that all crafted items should have this decimal shown but just isn't for simplicity sake. What we should try and do is see over the next couple of days to make several weapons with using varying amounts of experiment points. Personally, I've always done 3-4 experiment points at a time. So I'll experiment when I get a chance.

got a couple questions for ya dtox

did you experiment the range or effectiveness lines?

If you did experiment them, was it before or after you experimented the damage/speed?

Something I have noticed with a lot of weapons, especially dx2's and other weapons that don't come anywhere near maximum stats, ie. you can only experiment 6 total gems, is that most of the time the effectiveness rating, ham experimentation, will always come out lower than the other three. If its a gun like the dx2 where i know i will have extra points after doing damage/durability i spend the first point into the subpar category.

Ok, what is a subpar category, notice that sometimes when you make the initial object that the bars show red lines on the edges of the ratings. This means that particular stat is not up to par for that weapon, which i believe tends to make chances of experimentaion failure greater. I have seen a bad experiment on all three of the other categories drop damage off a gun and speed off of gun. This leads me to believe that if a category is severly underpar, thus showing the red stripes then it effects overall experimentation for the whole gun.

I have yet to conclusively show that it increases damage/speed but if it can lower them then it would be most likely it can also raise them.

Not sure if this has anything to do with what your seeing but something to look into.

There are definite rounding errors which can make it so all the ticks become yellow, but there is still room for more experimenting. I've seen it happen before many times.

As to how to experiment, I have been experimenting differently with different schematics for about 3 months now. Depending on what you are making (don't know if complexity is involved) you HAVE to experiment differently. For me, laser weapons are all 3,3,3,1 anything else fails misserably. If i try that on a flamethrower, only 1 will be a great success. On flamethrowers I do 2/2/2/... Every once in a while I try different ways to use the experimentation points, and my methods have not changed in the last couple of months. Right now, on average, I got about 80-90% perfect builds when I hand craft, except for those special nights when the dark side of the force is running strong.

On one of the components 3/2/3/2 will get me perfect builds 90% of the time, any other method i tried got me perfect builds around 10-20% of the time.

I also know from previous testing that experimenting 3 points at a time, all greats, can get you more rewards then 1 point at a time.

My most recent example, advanced rifle stock. A friend was getting 32/32dmg out of them. I experiment 1 at a time like he did, same results. I mess around with different combinations, and got 39/39. He tried the same combination and also got 39/39. (in either case we didn't fill up all the yellow ticks) So this was not luck, as we can, and have repeated it.

Experimentation has worked like this for me as far as I can remember back, at least 3 months. For my 2 cents about effectiveness of stations and such, I think the effectivesness changes the optimal experimentation method. Maybe a 0 station gives optimal success to 1 at a time, while a 40 crafting station makes 3 at a time optimal. If they use a normal distribution to model success probabilities, then the 40 crafting station, with optimal at 3 points at a time, would have worse success at experimenting 1 at a time then the 0 effectiveness station. I think this is why we get conflicting reports on higher effectiveness stations. They experiment the same way on both stations, when I think the stations effect the optimal experimentation options.

Conclusion, from what I have seen, each schematic has different optimal experimentation policies. You just gotta find out what it is.

Theros Ironfist BK
Master Weaponsmith, Master Typist

Hmmm...somewhere in the middle of this post, you guys forgot that I got a BETTER DX2 by using ALL of my Exp. Points at once, as opposed to 1 or 2 at a time. Meanning, what I observed was that the 2 at a time method was not as good as doing all at once.

Oh, and an interesting point, I actually spent more points by using all of them at once. I spent 6 on the first experiment, 3 on the 2nd, and 1 on the 3rd. Doing 2 at a time, i only did 4 experiments.

Is it possible that using more Experimentation points at once will give you the best stats, plus allow for more experimentation?

Hmmmm...I'm gonna have to find a Weapon experimentation skill tape and pick a weapon I can max out fully (maybe a Lance)

Well, okay, here's what I've noticed on artisan crafting, whereI actually get amazing successes and whatnot.

1. Plunk down 6 experiment points at once, get a regular 'great success'. Total experimentation will raise by 40% (I'm making up that number for comparison). It would theoretically take 2 more points to max out the item (again, made up).


2. Plunk down 6 experiment points one or two at a time, get 'great successes' on each. total experimentation goodness will raise somewhere around 40% (maybe more or less). It would still take 2 more points to max out the item.


3. Plunk down 6 experiment points at once, this time get an 'amazing success!!!'. Total experimentation goodness raises by 50%. it only takes 1 more point to max out the item.

However, in all of these cases if the experimentation goodness was equal (say, 60%), you'd expect basically the same stats.

Except of course when the game starts doing its special space-math.

I guess I went a bit quick on one point I made. I can't remember the item, but there was one where if I put 1 point in at a time, 10 times in a row, and got all greats, I did not get as good results as duing 3/3/3/1, all greats. In this case it was by about 4-5% more.

i do believe that the more you experiment at once, the more you get per point that is spent. How much more? thats a good question. Since you'll still hit the max % barrier, I think it will only matter in some of the cases where there is rounding error.

Theros Ironfist BK
Master Weaponsmith, Master Typist

I have observed this behavior on some weapons and many components. I always try the 1 or 2 point experimentation method and the all points experimentation method for any weapon or component schematic that I make. On some items 1 method willgive better speed results while the other gives better damage results. Sometimes using a combination of the 2 methods yields the best results (this happened when making adv stocks).

For power handlers and projectile mechanisms (both regular and advanced) using all the points at once will eventually always produce the best results.I may get equal the results using the 1 or 2 point method but I've never gotten better that way.

For weapons I have found this behavior to be uncommon (maybe 1 in50 weapon schematics) and dependent of the resources used. It could just be a result of the benefical 'moderate success' behavior that you've observed - when using multiple experimentation points these may be occuring during some iterations but not reported in the final result message.

I'm not sure if any of my data is statistically significant but if definitely effected how I crafted weapons.

Isis

---

DToxWeaponsInc wrote:

Hmmm...somewhere in the middle of this post, you guys forgot that I got a BETTER DX2 by using ALL of my Exp. Points at once, as opposed to 1 or 2 at a time. Meanning, what I observed was that the 2 at a time method was not as good as doing all at once.

---

But you also said you just made one gun two-at-a-time, and it turned out to be 2.5. Two-at-a-timestill has a possibilityto land a hidden moderate success. Which is why I thought, from reading your post, that your first gun has the possibility of being only 2.5 as a result of having a hidden moderate success, not because of your experimentation method.

Did you try to make a DX2 using one-point-at-a-time (to ensure you had no hidden moderate successes)? If after several runs, you find it's not possible to achieve 2.4 with all great successes, then yeah, I'd say you have something here.

if they would let me transfer all my resources and my character to test center i would gladly test this kind of stuff. There is no way I am going to waste thousands of good resources though on a live server just to grind through 50 of the same item to test theories.

I know the test center people probably don't want a copy over but could you imagine how fast we could test stuff like this if just 10 or 20 of us that post and keep up with the bugs had our supplies on test to work with.