HIP17: Hex Density Based Transmit Reward Scaling
jamiew opened this issue · 2 comments
Author(s): @Carniverous19
Initial PR: #59
Start Date: 2020-10-20
Category: Technical
Rendered view:
https://github.com/helium/HIP/blob/master/0017-hex-density-based-transmit-reward-scaling.md
Summary:
Very little additional coverage is demonstrated by being able to witness multiple hotspots that are co-located or in close proximity to each other. On the other hand, being able to witness many hotspots that are in distinct locations demonstrates a hotspot is providing a large area of coverage. This is in contrast to the current PoC reward structure, where transmitters and high density areas see a significant portion of PoC rewards.
This HIP suggests a change to PoC rewarding to better rewards areas of coverage. Rewards are reduced for transmitting or witnessing hotspots in close proximity to each other.
To demonstrate how the density limit in a certain area can be elevated based on neighboring hexs at each resolution. Click on a hex to see the limit on number of hexs (1-4 at res8, 5-20 at res7) vs actual number of hotspots in the hex.
For resolution 8: https://carniverous19.github.io/hex_res8_limits.html
For resolution 7: https://carniverous19.github.io/hex_res7_limits.html
The key take away is cities see an increase in the allowed hex density addressing the concern that cities need more hotspots than rural areas. This uses neighboring hex densities as a proxy for population density / areas where coverage as needed as these are at least very loosely correlated (people buy and deploy hotspots, people also use products and services that benefit from Helium Network).
Of course you could still see hotspots in cities with low reward scaling if they are still way over these max densities (like 10 hotspots in a res8 hex with max limit of 4). But NYC / SF can have 4 hotpsots per res8 hex without any density limiting where small towns / central Alaska can only have 1 hotpsot per res8 hex without seeing reward scaling go down.
Finally, these density limits are all very tunable with chain variables so if you say "the mechanics look ok but the max densities are still too low in cities" that's fine, its a single (or maybe 2-3) chain variable changes to adjust the specific targets and limits, so I would say you should still approve the HIP with a suggestion to adjust chain variables.
Rough consensus among the community was discussed and ratified at the October 2020 community call: https://docs.google.com/document/d/1bMm2alBigBj3detA775Dn0Gz9UM5XczAeK9vnjBB3l0/edit#bookmark=id.1isrxrlxktk9
Next steps are writing an actual implementation, reviewing, testing and deploying