I actually like the idea of failures as a general behind the scenes part. It should not apply to any build job in terms of collapse or loss of materials. Making the job move a little slower would be acceptable, but buildings just have to work.
My take on how to implement it goes along with something that has bugged many of us since the beginning. A 700L stone produces 5 mugs each with a 1L weight. Where did the other 695L go? I will get back to this at some point.
First, dwarves are gaining some levels of natural stats in the next version.
Skills are being more closely tied to specific stats. This is an excellent step. A concept used for skills in the GURPS roleplaying system is that skills have a default value. Driving for example is agility-0, this means that what you learn through basic observation in your earlier life is enough to get the default skill level and a decent agility is enough to make most rolls successful. Practice in the skill then increases it from that default value, and if you later gain more agility that will increase all skills that default on that stat. This type of skill system is the most accurate method I have ever seen, and explains why some people are just really good at whole classes of activities with no specific training. Hopefully DF heads in this direction.
Second, failures should give experience.
Since many people used the stone ring as an example I will address it in particular. Let's say a 700L stone is already at the workshop. The dwarf wants to make a ring, so the first thing they would do is cut a smaller chunk off the large stone. They aimed to get a 5L chunk, but failed horibbly. What they got was a 30L chunk and little bit of experience. Now they need to cut the 30L chunk down, repeating step 1 with a small bonus for easier working size; but fail again. This time they had the chisel at the wrong angle and only wasted time, but still gained some experience. Thier third attempt is a success and they have a 5L chunk. They can move to subsequent steps in creating the ring.
Third, gets back to that question of where the other 695L went.
It should heal and still be in the workshop for later use. If the dwarf making the ring fails later causing the destruction of thier work piece the destroyed part should be on the floor of the workshop as rubble and they would go back to cutting a new work piece off the stock, step 1 again. Rubble should work as clutter for the workshop and occasionally require a cleaning job. All materials currently have a significant disparity in weights between the finished good and the raw stuff. Raw materials need to have an actual weight assigned when they are created. Logs might range from 200-350L depending on the skill of the woddcutter; stone could be 250-700L depending on the skill of the miner and should leave a rubble in the spot that was mined; gems would always be mined with care even by the worst miner and ores are smeltable even in tiny chunks, both should have the amount of useable material set at creation of the world/region with only tiny amounts of loss based on skill.
Smelting:
The ores to bars should calculate a useable weight based on the useable weight of the ore chunk, losses from skill failures here would mostly be in fuel usage. It is reasonable to lose small amounts to slag which requires cleaning. When melting things down it should again be nearly lossless.
Smithing:
Would start with the smith selecting a bar out of the bar stock that is the right size. Failures here could include the smith selecting too small of a bar and having to start over once they realize it. Anything that makes a complete failure out of the material just means it needs to be melted back into a bar. These become scraps in the workshop of the appropiate metal type with the weight accumulating and are automatically marked for melting. Cleaning the forges would just require melt jobs to be set at smelters. When enough weight is built in the mystical smelter accumulators that already exist a new bars item would appear.
Brewing:
This already has a multiplier for how much to make from the raw plants. Losses of materials here should result in a reduction at the final end. Let's say we up the multiplier from 5 to 10. 3 sweetpods are being brewed and the brewer is quite skilled but clumsy. The final step of transferring the brew to barrels has a major failure. She tripped on her beard and hit the spigot before the finish barrel was in place, 3 units are spilled on the floor. This makes the final production 27 instead of 30. Small failures would result in the loss of 1 finished unit or the step taking longer.
Cooking:
It is a bit different since it currently adds all of its ingredients up. Until recipes exist and cooking is made more complex we can just use something like the brewing above. Loss of 1 from the finished product.
Gem cutting:
Is in a way generating a new raw material. The cutting of a large gem should be something that we specifically ask for, I have had a noble specifically mandate large gem creation. Normal cutting could be thought of as producing many small stones that are mystically accumulated by weight. Encrusting jobs use a certain amount of weight based on some size value for the item being encrusted. There should also be a job to remove gems from an item. Failures here could damage the item being (de)encrusted.
Decorating is basically the same as encrusting.
Engraving could produce marred images and bad likenesses.
Obsian shortswords:
These should be seperated into 2 tasks. Make stone blades done by a stone crafter would produce some weight of some quality blades. The wood crafter then makes the short sword and it gets its quality as a modified result of the blade quality. Poorly affixed masterpiece blades do not result in a masterpiece sword, but isn't a poor quality sword either.
Fourth, along with critical failures there needs to be critical successes.
These would always result in a higher skill gain on that step. They should often result in the step finishing faster. They can also result in a quality bump for the finished product. Going back to the ring example lets say it has 10 steps, and gives 3 exp per step. The dwarf failed on his first 2 attempts at step 1 and succeeded on the third try. He has 9 experience now. His first try at step 2 succeeds in a big way. It is a eureka moment, and he gains 30 experience finishing step 2 instantly.
Fifth, any system that realizes artificial losses from failures has to have different grades of failures.
Just like combat: you failed and missed, swing again; you critically failed, stumbling during your swing at the giant's head you trip and stab him in the stomach; you fail, instead of pulling the sword out you slice downward eviscerating and castrating him; the giant gives into pain. Having failures turn into successes in interesting ways doesn't generally work with an abstract crafting situation, but it is rather easy to make failures often not cause much damage. Failures can and should cause small injuries, and injuries should make it harder to do the task.
Sixth, the change ends the simple this job takes 100000 frames/agility/skill and that job takes 80000 frames/agility/skill.
Instead the dwarf prays to the RNG with its skill and materials are changed to be over abundant. That over abundance can be adjusted later as tiles gain a more fixed concept of size. If you didn't get it yet this really makes the RNG happy. It gets to be involved multiple times in the creation of each item. It can make jobs go faster or go slower. It receives sacrifices of raw materials at the altars of the workshop, and rewards Its faithful with higher quality goods. This kind of change requires a complete rewrite of the item creation system to be done correctly.
Seventh, the steps involved any job can be totally abstract.
They are just there to give the dwarf more chances at improvement and more worship to the RNG. I just made up descriptions for things to help demonstrate the concepts. The final steps in most craft things are polishing up. These later steps should have almost no chance to destroy the item. This helps to keep the massive item lost failures from really setting the dwarf back time wise. Sure they have to start over. Production rates for unskilled dwarves would be very low because of this, but they would tend to gain skills at the same rate.
Author
Topic: Dwarves can fail (Read 44052 times)