Droid Engineer Archive
Thread: It happened again :)
By the way, I've been keeping an eye on your Live successes on this. Well done, indeed.
MrWizzard wrote:
before I applied the experimentation, I had the full line of boxes to fill, which is what grabbed my eye this time... the bar went all the way to the end (it started at 29%, with a rating of 9 --8 is normal at 25% or so
Starting at 29% was what I saw most every time that I had hit the 20 Harv. mods when P8 was in testing. Sounds like this is pretty much the starting point to hit uber good Harvest modules. If you don't assemble to 29% or so...you're not gonna get there at all.
/bow2
Most Respectfully,
Btw, does anyone think it strange that the 2 people who got a 119 rating and a 117 rating have both had it happen to them twice, while it has only even happened a handful of times? Interesting to consider what this could indicate. No, it can't be just dumb luck (the odds are too great), so what is it?As for it being a bug, I have said elsewhere that I don't think it is. If it was a bug of some sort, you'd expect it to be a perfect rating, a zero rating, or some weird thing like that. I don't think a bug would produce just a very good droid within normal parameters.
Btw, both times that I got a 19.xx rated module (for a 117-rated droid), I wasn't using any skill modification of any sort. No toolbelt, no wine, no manufacturing city, no nothing. And I was standing in front of a decent, but not great crafting station (current one is 40+ rating, first one was 33). This seems to show to me that all that stuff doesn't have any bearing on this occurring.
Rippen
P.S. - I have 40 of the 117-rated harvest droids left to make, and then the schematic runs out.
- It is totally random. Every time somebody reports it, they give completely different explanations as to what we need to do to make it happen. Last time it happenned the person insisted we need to use the crafting foods. This time we're told it's not necessary and that maybe we need to be in some kind of city that gives us a crafting bonus.
- It goes against the rule that the stats of the resources determine the best you can do when you craft an item. If what people are reporting is not a bug, then it means there are two different contradictory rules for how crafting works. 99.99% of the time thenumber of experimentation boxes availablefor you to fillis fixed by the stats of your resources. But, if you are so lucky as to get one of these "super duper ultra amazing" assemblies, the rules completely change. When that happens, you get extra experimentation boxes to fillthat by the ordinary rulesare not there.
- finally, if they did this on purpose, the code to manage it looks pretty complicated because it is so excruciatingly non-linear. I personally don't think that the developers would have taken the time to implement such a change on purpose with so many other pressing issues.
In any case, what it looks like to me is that the code that sets up the experimentation boxes after the assembly is finished makes some assumptions about what can and cannot happen during the assembly. But some relatively recent change to the crafting systemcauses that assumption to be invalid in some rare instances if your assembly happens to hit "certain values". When that assumption becomes invalid, the numbers don't add up right and you end up getting extra experimentation options.
That's just my opinion anyway.
BTW: I have seen this happen before. I regularly experiment on everything, even things that don't matter, just to increase the probability that I will see this "feature". It happenned to me when I was crafting a droid brain (at least I think it was a droid brain. it might have been something else, but it was definitely some item for which experimentation is irrelevant). I used utter crap for resources; stuff that ordinarily only lets me get about 35% on the experimentation max. Yet I got some bizarre assembly results and all of a sudden I was able to experiment way past 50% on the item.
Message Edited by Malitevv on 09-01-2004 12:31 AM
It would be interesting to see if a stack of 6 of those 12's (all identical) and then another stack of 6 of the 13's would produce a different final rating. In other words is the hidden decimal rating on the two models different.
Straker_Atrella wrote:
Hmm ok, when I get home, I'll need to look at that 8/10s of a box thing. Never really looked at partial boxes, nor noticed them really.
How do you explain that if I fill all the boxes one time, I get a 18 rated module with 13 quality, then the next time I get one with 12 quality. Yet all the boxes are filled the same way. This leads me to believe that even in the experimentation stage things are not as simple as we think.
The type of module and the actual scores that are associated with experimentation are irrelevant so far as this point is concerned. It's just an issue of the experimentation percentage. Your experimentation success for each bar is always a number between 0% and 100%. And it is that percentage I am talking about. How that percentage translates to a stat is a secondary detail so far as I can tell. The crafted item stat in question will be equal to percent*(the maximum value for that stat). But that doesn't play a role in determining what percentage you get.
Each experimentation box equates to 10% of the experimentation. And the stats of the resources determine how many experimentable boxes will be there. So suppose the item requires one resource type and depends on100% overall quality and the overall quality of your resource is say 650. Then you will have 7 experimentation boxes available to you when you start experimenting. If you fill them all successfully, your final experimentation percentage will be 65%. If the OQ of the resource used is say 982 then you will have 10 experimentation boxes available to you when you start experimenting and if you successfully fill them all, your final experimentation percentage will be 98%. Now there is evidence that it is actually 98.2% in that case. Because you can see the result of that extra 0.2% when you combine things like detonation or combat modules together and watch how the hidden fractional percentages add up. In the case of detonation modules it is very easy to experiment and test that this is the case. For more complex items, the max experimentation that you can achieve is determined by the weighted sum of the stats of the resources, but the basic idea is the same.
In any case, that is the way it usually works.
In any case, when this "ultra-amazing assembly" happens these rules change. When the ultra amazing assembly happens people are observing that they are able to experiment to a percentage that is higher than what the resources should allow. If you got an "ultra-amazing" assembly when crafting a final droid while using 980 OQ chemical, you would find that you are able to experiment all the way to 100% even though the 980 on the resource's OQ would ordinarily set the experimentation percentage cap at 98%.
In terms of what you are asking, if you successfully completely fill all of the relevant experimentation boxes, the result should always be the same. Your not being clear about what you mean by "filled the same way". When you say that do you mean that the final experimentation percentages reported for each experimentation bar are identical?? Even if they appear to be, you can't know if they really are unless all relevant experimentation bars are completely and totally experimented. When you say "filled the same way" do you mean that all experimentation boxes are completely full and there is nothing left for you to experiment on? Or do you mean something else?
Straker_Atrella wrote:
Let's use your partial boxes point. Let's say it says you should get 8.75 boxes. What if with a decent assembly, it was modified by the assembly role and you actually got 8.80 boxes. Would you actually visually notice a difference? Probably not.
I'm not looking for visual differences. I'm looking at differences in the stats that come out at the end. I've seen no case where the stats you get for a given set of resources are not identical from one case to another if you use the same resources and successfully fill all the experimentation boxes that determine the value of that stat. If the resources suggest 8.75 boxes are what you should have available to you, then the final stats of your item are always 87.5% of the best possible for that item stat if you successfully experiment all the way to 87%. You will never be able to experiment to 88% in this case unless you get the freak "ultra-amazing assembly". In the absence of the ultra-amazing assembly, I have never seen any correllation between the max experimentable percentage and the assembly result.
If you have evidence of something different, I would be interested in seeing it. It sounds to me like you are talking about the case where you haven't completely filled all the experimentation boxes for the experimentation bar that sets the quality of the item. When that happens, the assembly result does matter of course. But that is not the case we are talking about. We are talking about the fact that when this freak "ultra-amazing" assembly occursthe best possible experimentation you can achieve for a given baris artificially raised.