Architect Archive
Thread: Thoughts on Critical Failures, etc.
After failing on two heavy harvesters in a row over the past couple of days, I began to get a little frustrated. Something seemed different after the patch (which I'll discuss below) and I figured I'd just use the time to set up some factory runs of relatively simple items. I then failed twice in a row trying to make wall module schematics, which seemed highly improbable for a master architect.
One of the things I don't understand is why a critical failure (on my PC anyway) doesn't clear the crafting draft schematic (i.e., I fail ona combine, and all of the materials still appear to be there, and I need to reboot the tool to fix). One hypothesis, and it is certainly no more than that, is something has changed server-side that is sending an event to the client which isn't being properly interpreted. This event, right or wrong, is being interpreted as a critical failure but not in the same way the 'old' ones were.
For the moment I'm not bothering to build anything. I can't afford the catastrophic losses, and while my results are far from statistically significant they have a definite impact on my ability to function from a pragmatic point of view. I'm guessing that this failure rate is artificially high now, and needs to be addressed as soon as possible.
That being said, I think there are some refinements that could reduce the frustration that comes from failure on large items and reduce the randomness without recognizing that there should probably be some risk. I generally support the idea of 'partial critical' failures, but there are some other things I'd like to see too.
1. We should know our probability of success when crafting. In my mind, this would involve seeing the effect of crafting tool, crafting station, complexity, and materials used, to generate a clearly transparent probability of success (ideally breaking down into Critical Fail, Marginal, Good, etc). This needn't be a number, but a rating from "Very Easy" to "Extremely Difficult" would help, if we knew roughly what the labels meant.
2. I believe we should be given the opportunity to 'Insure' a crafting attempt.This could take various forms, the simplest of which would be where the crafter had the option to 'pay insurance' and guarantee that the combine would not critically fail. The cost for insurance could be computed in a similar manner to the experience value of an item, using the amount of raw resource as a driver, and further modified by the failure percent computed by the system. Without trying to figure out an optimal formula for interrelating these two, I think that a cost of 0.5-1.0cr per resource would be reasonable for a Master constructing Master level items (such as the PA hall).
A slightly more advanced and 'realistic' option would be that the insurance cost doesn't prevent a critical failure, but rather prevents the resources from being destroyed if one occurs (or, alternatively, creates the resources at some in-town location where they can be retrieved). The key difference here is that with every attempt, insurance must be purchased, so the cost (in credits, at least) can climb if more than one failure occurs, without being totally debilitating.
3. Architects shouldn't be the de facto miners in the game simply because they use the most resources. I'm not sure what the Miner did in Beta, but having a profession that is more efficient at gathering resources makes alot of sense. I'd love to see miner-specific harvesting units, perhaps ones that are incredibly efficient at extracting one specific type of resource (such as the Fusion Generator is today). If Miners could efficiently gather resources, refine them (to change characteristics), and homogenize them (to combine smaller specific stacks into larger more general ones), perhaps the market prices of finished goods and raw materials will relate in a more realistic fashion.
4. Plenty of other ideas, but I will save for a later time. 
I'm writing this mostly out of frustration, so I know these aren't the only solutions (or the only problems!) but I wanted to get some thoughts, particularly on the insurance. Does it sound feasible, and would it be desireable from a balance perspective? Whether it would be desireable from a practical perspective would likely depend on how much insurance cost, relative to the probability of failure, so I'm just throwing out the general concept for the moment.
Rearden Roark, Master Architect, Gorath Server
I have alsonoticed that my failure rate seems to be really high lately. Here is a list of what I have failed on and what I have built in the last 24 hours.
Built:
5 Meduim Min.
1 Solar
1 Small House
Total Items Built = 7
Failed:
3 Medium Min.
1 Solar
1 Small house.
Total items Failed = 5
As you can see I have failed almost as much as I have succeed. This represents a 41% failure rate on final assemblies. I find this to be unacceptable.
Recent successes
3 Small houses
1 heavy mineral
1 heavy gas
Recent Failures
2 PAs in a row (I will throw myself off a bridge)
1 Food Factory
3 Small houses