Development Cycle Archive
Thread: Crafting Critical Failure Rates
- Bloodfin
Chrysalide wrote:
Hello everyone,
These numbers do fall within the expected parameters of the crafting system. Regardless, the significant number of people that have raised this issue does still lend this topic an air of legitimacy. I am interested in hearing any suggestions, comments, and/or concerns that the crafting community (or anyone at all) might have about critical failures as they relate to the crafting process.
I'll admit - as a DE I forget about crit fails sometimes. Just doesn't happen all that often, especially when using factory parts.
I think the changes for Architects with regards to crafting and structure building were quite kind. There's nothing like scraping together resources to build a structure that takes so many resources only to have them disappear due to a critical failure. I'm not sure the solution to just remove failures for structures was the best solution, but kudos to the Devs for erring on the side of players!
With experimentation, critical failures are a killer for Architects on harvesters. With the costs involved, 80% or more people will always want the best BER harvesters as anything substandard isn't worth their money. I've used a select few expletives during final experimentation of a harvester only to get a critical failire. This still results in a harvester, but one that generally will not sell and be a waste of input resources. I don't have any good suggestions to offer, however, as it wouldn't be fair to just remove any chance of critical failure and always get a great BER harvester.
Furniture is where I've had my biggest issues. Furniture was my main business as an Architect (oh how I hated structures), but even as a Master Architect I'd sometimes get critical failures on things like the rusty looking bookcase -- sometimes even 2-3 times in a row! It doesn't seem right that a Master would fail on such trivial things. As a crafter progresses, things under a certain level of complexity should never fail I'd think. Yes I had 15 functionality crafting tools and 30+ crafting stations, but I still got failures on seemingly trivial things.
I'm really glad this is being looked into. Even though I'm not crafting anymore I'm still scarred with the memory of it all.
Xanobia wrote:If you want to produce accurate results, try doing it on a love server
where IS this "love server"? i'd like to be playing in it instead, i'm pretty sure....
i ran a few numbers given you numbers, and i want to point out that, while four and a half percent doesn't sound that bad when you run together the experimentation fails and the crit fails this means that you lose (not quite) every tenth item. crafting ten items and getting nine isn't that great for master.
my main suggestion would be to make it so that experimentation will UNDER NO CIRCUMSTANCES make an item worse than it was to start out with. you can lose items to crit crafting fails, and you can waste experimentation points, but i wouldn't EVER punish the player for experimenting...
that's my input
I am somewhat suprised at the low number of crits you had. In medical experimentation, using tools rated over +14 and station rated over +43, I get crit failures around 1 in 10 attempts on average and have had up to 3 in a row and 4 of 5. Of course, after that 4 of 5 debacle, I logged off and found a bottle with the word "proof" on it and putthe whole encounter behind me. The crits outnumber the amazings by five fold to my experience.
I would really like to se the crit failures on final combines done away with. It is simply unacceptable to belive that a master archetect would fail so horribly in putting his town hall together that the whole thing was destroyed. The same can be said of all of us masters--artisans, docs, tailors, smiths--should not be punished to this degree by having so much wasted on 1 roll of the virtual dice.
Thanks for looking into this Chrysalide,
Vinaddar
Chrysalide,
Just for the sake of being complete. Did your tests all involve Crafting Stations and tools of the same operational effectiveness. What effects are these supposed to have oncritical failures? For example, a 40 operational crafting station and a 14 tool is the same/has half the critical failures of a 20 crafting station, 7 tool combo? Also, what about the effects of moderate successes. I have a 40.8 crafting station and a 14 tool and get a tremendous amount of moderate successes (which often only decrease stats) and great successes (which always increasethe stats as one might expect), but virtually no other types of success or failure at all.
I realize I could just be hitting outliers since were talking maybe 100 experiments in the past weekversus several thousands. I'd really like to see a fully documented,scientificstatistical test of crafting. We would need a minimum of 1,000 tests each with at least three diferent tools and crafting stations at different levels each tested in every combination. Of course, 10,000 tests would be more accurate. Its a lot of work, but much more statistically sound especially if done together by people of at least the same level.
Thanks,
This is my first post to the forums, however I have been viewing them since inception. I think one key issue your not looking at is a difference in criticals from one zone/planet to another. Let me give you an example. I have been attempting to craft sub-components for crafting stations for the last 3 days with better than a 60% critical fail rate. I finally gave up and made up some factory runs of the subs once I managed to get a good schematic for each. Now keep in mind I am a Master Artisan and Master Architect, my tools and stations are top of the line, I now am seeing the same problem with assembly on the station as I was seeing with the sub's. I have better than a 70% crit fail rate. I have tried changing my crafting stations, my tools, the sequence I use my experimentation points, nothing helps. So I inquire to a fellow crafter friend of mine thats making them on Talus(my shops on Tatooine btw) and not only is he not having the same problem but he is experiencing 0 failures of any kind using pretty much the same resources. When I was finally able to craft a station without a crit I managed 99% experimentation with using all optional sub components as well, however, the ratings would never break 30. This is using the same resources I was making 44's with 3 weeks ago. I dont know now if its the city my shops in, the zone, the planet, the moon, the stars. So I post a customer service ticket and you guys send me an electronic form letter telling me to post in the forums so here I am.
Well, those failure rates are in the neighborhood of a failure every 20-25 items you craft, and the same for the number of times you press that Experiment button. Now once every 20-25 items... well that is a tad annoying, but generally we can probably live with that. Experimentation, though, it's not like you are seeing that rate once per item like you do on assembly. You see that every time you hit that Experiment buttton, which could mean that on a given item you could concievably Crit Fail in experimentation on every other item you craft if you are being cautious and using only 1 EP every time.
I know that of the folks crafting there were more than a few Masters, but I know I was crafting at Master Medic which is only +50 in Med Crafting. I varied the points I used in experimenting as well, so sometimes I "played it safe" with just 1 EP at a time, other times I'd do 2 or 3, and sometimes I'd just go for it all with all 5 at once. Mostly I did it one by one, though. I used various levels of crafting tools and machines rating from +0 to... actually I don't remember what the other one is, but it's an OK station. +20-something, maybe? Nothing spectacular, but a positive value at least.
It should be noted that I don't know how the scripts handle it, but there may be some slight scewing with success rates on experimentation if any of us continued to experiment after a crit fail that brought us to 0%. Since all experimenations from that point on would be Amazing Successes, but would make no difference on the item.
I think the biggest issue is looking at the "Risk" scale in experimentation, though. If we were really looking at, say a 4% failure rate, then it would be nice to at least have that Risk scale tell us 4%, instead of 0%.