From Discworld MUD Wiki
Revision as of 07:51, 13 May 2020 by Thalic (Talk | contribs) (Update regarding ranges.)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search


The command has one failure and four success messages: the failure message doesn't provide any information about the item value while the success messages report a random value that lies within a tolerance band around the real item value. The size of the tolerance band depends on the grade of success:

  • You're not very confident in your estimation. (+/- 20%)
  • You feel pretty sure that your estimation is in the general region of the real value. (+/- 8%)
  • You feel pretty confident that your estimation is close to the real value. (+/- 4%)
  • You feel confident that your estimation is on the spot. (exact value)

The probability to get a specific grade of success depends on the skillcheck (see below).

Update: Ranges might be +/-10% for pretty sure and +/-5% for pretty confident. Not sure if this has changed or is the result of faulty experimentation; working on it. Thalic (talk) 07:51, 13 May 2020 (EDT)


The command performs a skillcheck of the item value against the relevant skill in the tree.

20190104 Valueing Success Rate Graph.png

The skillcheck seems have six segments with different probabilities for the five results:

  • segment 1 - always succeed
  • segment 2 - minimal chance of failing, different ratio of intermediate successes
  • segment 3 - small chance of failing, different ratio of successes, failure eating into "on-spot" success rate
  • segment 4 - moderate chance of failing, minimal chance of "on-spot" success, small chance of lowest level success, different ratio of intermediate successes
  • segment 5 - high chance of failing, only lowest success message
  • segment 6 - always failing

Note: The transitions between segments are possibly step responses, the gradients in the diagram are caused by the granularity of jewellery values available for experiments. For the same reason the width of each segment is only approximate. The command results have a high variability and even with almost 100k samples the results are rather noisy. The diagram shows the approximate results for valueing jewellery with a bonus of 561 Thalic (talk) 17:19, 4 January 2019 (EST)