Comment out pot iterations test

This commit is contained in:
Mariano Sorgente 2020-12-10 00:57:34 +09:00 committed by Yostra
parent aa69a3c2b7
commit ba448a7a23
1 changed files with 39 additions and 39 deletions

View File

@ -76,42 +76,42 @@ class TestPotIterations:
assert ip_iters == (sp_iters + test_constants.NUM_SP_INTERVALS_EXTRA * sp_interval_iters + required_iters) % ssi
assert sp_iters > ip_iters
def test_win_percentage(self):
"""
Tests that the percentage of blocks won is proportional to the space of each farmer,
with the assumption that all farmers have access to the same VDF speed.
"""
farmer_ks = {
uint8(32): 400,
uint8(33): 300,
uint8(34): 150,
uint8(35): 150,
uint8(36): 150,
}
farmer_space = {k: _expected_plot_size(uint8(k)) * count for k, count in farmer_ks.items()}
total_space = sum(farmer_space.values())
percentage_space = {k: float(sp / total_space) for k, sp in farmer_space.items()}
wins = {k: 0 for k in farmer_ks.keys()}
total_slots = 600
num_sps = 16
sp_interval_iters = uint64(100000000 // 32)
difficulty = uint64(500000000000)
for slot_index in range(total_slots):
total_wins_in_slot = 0
for sp_index in range(num_sps):
sp_hash = std_hash(slot_index.to_bytes(4, "big") + sp_index.to_bytes(4, "big"))
for k, count in farmer_ks.items():
for farmer_index in range(count):
quality = std_hash(slot_index.to_bytes(4, "big") + k.to_bytes(1, "big") + bytes(farmer_index))
required_iters = calculate_iterations_quality(quality, k, difficulty, sp_hash)
if required_iters < sp_interval_iters:
wins[k] += 1
total_wins_in_slot += 1
win_percentage = {k: wins[k] / sum(wins.values()) for k in farmer_ks.keys()}
print(win_percentage)
print(percentage_space)
for k in farmer_ks.keys():
# Win rate is proportional to percentage of space
assert abs(win_percentage[k] - percentage_space[k]) < 0.01
# def test_win_percentage(self):
# """
# Tests that the percentage of blocks won is proportional to the space of each farmer,
# with the assumption that all farmers have access to the same VDF speed.
# """
# farmer_ks = {
# uint8(32): 400,
# uint8(33): 300,
# uint8(34): 300,
# uint8(35): 300,
# uint8(36): 300,
# }
# farmer_space = {k: _expected_plot_size(uint8(k)) * count for k, count in farmer_ks.items()}
# total_space = sum(farmer_space.values())
# percentage_space = {k: float(sp / total_space) for k, sp in farmer_space.items()}
# wins = {k: 0 for k in farmer_ks.keys()}
# total_slots = 600
# num_sps = 16
# sp_interval_iters = uint64(100000000 // 32)
# difficulty = uint64(500000000000)
#
# for slot_index in range(total_slots):
# total_wins_in_slot = 0
# for sp_index in range(num_sps):
# sp_hash = std_hash(slot_index.to_bytes(4, "big") + sp_index.to_bytes(4, "big"))
# for k, count in farmer_ks.items():
# for farmer_index in range(count):
# quality = std_hash(slot_index.to_bytes(4, "big") + k.to_bytes(1, "big") + bytes(farmer_index))
# required_iters = calculate_iterations_quality(quality, k, difficulty, sp_hash)
# if required_iters < sp_interval_iters:
# wins[k] += 1
# total_wins_in_slot += 1
#
# win_percentage = {k: wins[k] / sum(wins.values()) for k in farmer_ks.keys()}
# print(win_percentage)
# print(percentage_space)
# for k in farmer_ks.keys():
# # Win rate is proportional to percentage of space
# assert abs(win_percentage[k] - percentage_space[k]) < 0.01