[Test Puzzle] Molecular Weight Objective Update

Closed since over 2 years ago

Summary

• Created: November 08, 2023
• Expires: November 10, 2023 at 19:00 UTC
• Max points: 0

Description

We have implemented a new constraint for the Molecular Weight Objective. Now puzzle creators can set a lower and upper bound for molecular weight. In this test puzzle we will be testing the addition of this lower bound setting. When testing, it would be good to remove atoms until the small molecule is below the lower bound, and add atoms until the small molecule exceeds the upper bound. The addition can be easily done with the fragment panel in the Small Molecule Builder.

Top groups

• 1. Anthropic Dreams 1 pt. 17,950
• 2. Gargleblasters 1 pt. 17,807
• 3. Foldit Staff 1 pt. 15,177

Top soloists

• 1. LociOiling 1 pt. 17,950
• 2. HuubR 1 pt. 17,807
• 3. blazegeek 1 pt. 16,800
• 4. akaaka 1 pt. 16,530
• 5. beta_helix 1 pt. 16,530
• 6. Sciren 1 pt. 15,177

Comments

[HuubR]

Tried a wide range of weights (not hampered by any knowledge of biochemistry :-)
ranging from 16.043 (methane) to 199,947 (closest I could get to 200),
and from 600.025 to well over 900.

What I see is three ranges with linear behaviour:

weight < 200   ⇒   bonus = 5 * weight
weight > 600   ⇒   bonus = 5 * (800 - weight)
other values     ⇒   bonus = 1000

For weights over 800, the bonus is indeed negative.

Is it possible (and could it be useful to the puzzle setter) to use different slopes on the "small weight" side and on the "large weight" side?
In this example, the coefficient is 5 for both cases (or, actually, on one side it is -5).

[Sciren]

Thank you for the suggestion HuubR. I will look into the scoring for this objective more. For now it makes sense to keep the current adjustment as it inhibits the score falling into a negative range for the lower bound. It is indeed true that in the future we may have the bonus to provide a negative impact in the future though.