Droid Engineer Archive
Thread: Assembly as important as Experimentation?
Page 1 of 2
Straker_Atrella
Mon Jun 14, 2004 4:02 pm
#1
Now, I havn't been a DE as long as many of you, only about 3 months. Yet in that time, I have crafted a lot of Droids and learned a lot. Yet recently a couple of posts have made me wonder about what I know.
It was my belief that when experimenting, if you filled all the boxes, that was as good as you were going to get with your current resources.
Yet now in other posts people are saying that a good or perfect "Assembly" will actually make something better, even if all the boxes are filled.
What experiances do others have with this? If this is true, I may need to craft a lot more modules by hand before I do a factory run.
Malitevv
Mon Jun 14, 2004 4:58 pm
#2
as far as I know "assembly" has been said to alter your chance of critical failure. but i've seen not definitive statement to that effect and something like your chance of critical failure is very hard to test empiricly so I can't say if that is the effect it has.
As far as it "enhancing other things"I don't see how it could. The max possible percentage that you can get from experimentation seems to be pretty clearly set by a weighted sum of the values for the statsof the resources used. The weights appear to be determined by the product of the percentages listed in the schematic and thenumber of resources of each type required,with the caveat that for resources that have a cap for a stat of a particular type the actual weights used are skewed internally. What I mean by that last point is this: if the things that matter are 50% cond and 50% oq, and the resource in question is inert gas (which never has a conductivity stat), then the 50% conductivity will be ignored and it will be treated as though the stats that mattered are 100% oq for the inert gas part only. This is all inference from the way in which i've observed things to work. But it seems to me that this is pretty close to the truth. For example, with simple modules that have simple requirements and not too many resources, it's possible to use this assumption and accurately predict what the max percentage is going to be before you've put in the resources. When resources like aluminum or steel are involved in something where conductivity matters, itsnot so simplebecause the cap on conductivity for steel and aluminum appears to be something less than 1000 (which I think skews the weights in an unknown way [if you knew what the game "assumes" the cap on steel and/or aluminum conductivity is you could probably predict for those ones too]).
But like I said, I may be wrong. But that is my best hypothesis that matches the facts.
Either way, if you assume this hypothesis is thinking in the right direction, it is possible that assembly bonuses factor in their too. But it should in theory be possible to test thatpossiblity that assemply bonuses matteron one of the simple modules where you are able to accurately calculate the max experimentation percentage from the resource stats (the final droid build where there is only one resource used and the max percentage effectiveness is clearly (OQ of the resource used)/1000 would be a good test case). You'd need an assembly bonus of at least +10 to test it though because for most skill tapes, the bonuses have an in game effect that kicks in only at +10. But if assembly plays a role you should observe that the max percentage effectiveness on the experimentation for the final droid build is NOT always (OQ of the resource used)/1000 but instead varies with the magnitude of the assembly bonus.
Gribnitz
Mon Jun 14, 2004 5:34 pm
#3
I know which post you refer too. I have +121 Experimentation base and +129 Assembly base. With the foods, that boosts me to +130 Experimentation and +143 Assembly. I get amazing assemblies perhaps 6 or 7 out of 10 times and I can't duplicate what has been done. Everytime I try, I always end with 88% for the final assembly whether I get an amazing success or a great success. I tried perhaps 50 assemblies using my very best resources just trying to get over that 88% without success. I may play around with it again, but eating up my resources just to try and get something I haven't experienced before made me cringe. All my resources are the best that have ever spawned on Kauri and are now extinct, so I hate burning it on something that may not really work. I compared the resources of Naturius and Kauri and Kauri's resources are actually quite a bit better. The steel alone was131 points higher average OQ/Cond and I can't get that lucky assembly.
I would really love to hear if anyone else can duplicate this.
Malitevv
Mon Jun 14, 2004 5:52 pm
#4
if there is any truth to this, there is no need to use quality resources to test it. use grinding resources where the best you can get is something rather low and see if your assembly bonuses play any roll in setting what the maximum experimentation percentage is.
craft MSE droids with no modules in them. use any metal and chemical that you have on hand in large quantities. But intentionally use low quality materials so that it is very easy for you to hit the maximum experimentation bonus with your existing experimentation points. :-)
If your max experimentation percentage possible on effectivenessis always equal to:
(53*(OQ of your metal) +13* (OQ of your chemical))/(53000+13000)
and this is independent of your assembly bonus. Then the post you are referring to is very likelyfalse.
Message Edited by Malitevv on 06-14-2004 05:56 PM
Message Edited by Malitevv on 06-14-2004 05:57 PM
Malitevv
Mon Jun 14, 2004 9:48 pm
#5
ok. i did the weighted sum in my head incorrectly in the MSE droid example. I'm at home with pencil and paper now though and have the correct weighted sum worked out correctly. The maximum experimentation for an MSE droid should be this:
[ 53/68 * (OQ metal) + 15/68 * (OQ chemical) ] / 1000
You should find this to be the truth. I've run some tests and no matter what materials I use, and what assembly bonuses I have in play, the maximum possible experimentation points for the MSE droid is ALWAYS exactly what the above formula calculates. If there was any truth to the idea that assembly bonuses affect experimentation, this would not be the case.
Straker_Atrella
Mon Jun 14, 2004 9:53 pm
#6
Well that's not exactly what was being said. They are saying that you will just as many experimentation boxes filled up, but you will have a better product. For example in the thread where this came from, 17 rated Harvestor modules, and 18 rated ones.
One person is claiming that with the exact same resources and same experimentation, they can get 18 rated ones instead of 17's if they get a Perfect assembly.
This isn't how I understood things to work, but I don't know for sure, that is why I am asking.
TheRealTK421
Mon Jun 14, 2004 10:33 pm
#7
Straker_Atrella wrote:
Well that's not exactly what was being said. They are saying that you will just as many experimentation boxes filled up, but you will have a better product. For example in the thread where this came from, 17 rated Harvestor modules, and 18 rated ones.
One person is claiming that with the exact same resources and same experimentation, they can get 18 rated ones instead of 17's if they get a Perfect assembly.
A more perfect assembly, yes.
Specific to the Harvest modules, I do think I've seen this to be the case but I've seen it elsewhere as well. It has to do, I think with how the numbers add up for successive great/amazing rolls. We've all seen that there are fractions in the background that can end us up with a slightly higher outcome.
I often had to shoot for Amazing rolls on TC when testing P8. If I didn't, more often than not, I couldn't get to 20.
/bow
Respectfully,
Straker_Atrella
Mon Jun 14, 2004 11:01 pm
#8
Wow, I guess I better get me some assembly tapes then, I didn't think that mattered.
Thx.
R0ZM4N
Tue Jun 15, 2004 6:26 am
#9
OK this is starting to look more and more like a crafting bug and a crippler at that.
I was staying well out of this conversation as it appeared to have too many unknowns and require a massive amount of experimentation, anything else would just be speculation….
Starting on a completely unrelated topic, I may have come across some more evidence.
So I’m crafting Stim dispensers. I figure having 12 experimentation points I *can’t* fill all the boxes but the current debate had me thinking if I spent a little time, I might get a few ‘Amazing’ results and get a better product. I had built 20 (I was counting, allowing myself 30 max so as not to waste resources); they had been a mix of Amazing’s and Great’s but I always ended up with the same result (barring any misshaps) of 9 speed/51 uses. Then, on number 21, I got an amazing success initial build, all of my experimentation’s were only Great but the module ended up 9 speed /55 uses.
I wasn’t using food or drink and I don’t live in a City. This means that the initial build is a new and important factor and as we’d all have to spend all day long churning out test after test after test in order to compete with the next guy down the road, that says BUG.
Somewhat concerned…
Cass
MachineZed
Tue Jun 15, 2004 6:59 am
#10
Cass I would not consider it a bug. It would make sense that a "Great" assembly would be better than a "Good" assembly, like wise the "Amazing" the best we can hope for.
I was under the impression thata Successful Assembly would increase the Starting %. It is at that point that where you start the experimentation. Making it easier to Fill up the Exeriment bar.
Either that or the assembly bonus is going into your count as to how high your "10" points of experimention can take you. I am taking the 10 to mean the one box. IE (just pulling numbers out of my rear) say you have 100 Assembly points and 100 Experimentation. Moving one box at a time, Maginal=5% higher,Good = 6% higher, Great =7% higher, Amazing=9%.
But with higher Assembly bonus we will say 110 Assembly points. The Marginal Experimentation now = 6% higher, Good=7%, Great=8%, and Amazing=10%
These are just two of my theories of how Assembly works in the game enviroment
Erillion
Tue Jun 15, 2004 7:17 am
#11
Repeated assembly without crafting food/drink (but still in Research Center).
Again got harvest module rating 18 instead of 17 (my record: 19 with crafting food). Difference : better assembly result with the rating 18.
As usual experimentation was maxxed out with 9 ex points (I usually use 7 points, then the rest - usually 2 ... then I reach max).
Seems like my humble rating 114 harvest droid crafting (see thread that started this all) suggests a new trend in crafting in general. But only for people doing a LOT of hand crafting or being extremly lucky. I suggest you spread the word so people test it with other professions. I wont call it a bug - but its definitely a different behaviour than I (and others) were used to in crafting.
Have fun
Delos Stardust
R0ZM4N
Tue Jun 15, 2004 7:18 am
#12
MachineZed wrote:
Cass I would not consider it a bug. It would make sense that a "Great" assembly would be better than a "Good" assembly, like wise the "Amazing" the best we can hope for.
I was under the impression thata Successful Assembly would increase the Starting %. It is at that point that where you start the experimentation. Making it easier to Fill up the Exeriment bar.
Either that or the assembly bonus is going into your count as to how high your "10" points of experimention can take you. I am taking the 10 to mean the one box. IE (just pulling numbers out of my rear) say you have 100 Assembly points and 100 Experimentation. Moving one box at a time, Maginal=5% higher,Good = 6% higher, Great =7% higher, Amazing=9%.
But with higher Assembly bonus we will say 110 Assembly points. The Marginal Experimentation now = 6% higher, Good=7%, Great=8%, and Amazing=10%
These are just two of my theories of how Assembly works in the game enviroment
Lol 
You see this is why I wanted to stay out of this one, you CAN’T play with numbers because the DEV’s will never let us know what those numbers are.The FACT is, if my one off experience marries with the details coming in from other DE’s on other servers and indeed other crafters, then forget the numbers, the facts of life will become thus.We will all sit around for hour upon hour trying to get an initial roll that is completely random and outwith our control. That is either very poor design or, more likely, a bug.
Do you see why I’m concerned now flower? I think TK needs to hold this one up to the DEV light to see if there are any cracks in it.
Message Edited by R0ZM4N on 06-15-2004 03:20 PM
Malitevv
Tue Jun 15, 2004 9:33 am
#13
MachineZed wrote:
I was under the impression thata Successful Assembly would increase the Starting %. It is at that point that where you start the experimentation. Making it easier to Fill up the Exeriment bar.
This makes perfect sense to me. But in that case,it means thatthe benefit of the assemblybonuses above and beyond masteris being grossly over stated here. And that Gribnitz is being misled into thinking there is some magical relationship between assemblypoints and an items stats which doesn't exist, and is wasting his resources trying to do something that can't happen.
IfMachineZed is right aboutwhat assembly does then assembly bonuses will never "make the difference" between being able to get a all the experimentation boxes filled up and not being able to do so, but only increase the probability that you are going to do so. Saying it makes the difference means that it couldn't be done without the assembly bonuses from food and clothing, which isn't true. You can fill all the boxes without any assembly bonuses beyond those that a Master DE has intrinsicly most of the time anyway. You can do it without any experimentation bonuses beyond a master de's experimentation bonuses. I often craft in my armor with all my clothing put away because it just doesn't make much difference at all either way.
I amusually able to experiment all the boxes on an item where the maximum is 100% when crafting in the woods ofNaboo with my crafting droid, wearing no clothes, and eating no food. Sometimes it takes two or three tries, but since I do everything in factories, the fact that it takes an extra try is irrelevant.
In any case, I understand now. This shows that assembly is of similar importance to experimentation in terms of the value of the bonuses. But any claims that it's going to make all the difference between getting to 19 and not getting to 19 on a harvest module are false. Any master DE can, with quality resources, experiment all the way to 100% effectiveness on a harvest module with no assembly or experimentation bonuses whatsoever with a small amount of luck.
Page 1 of 2