{
{0.0001, {1.0000, 1.0602, 1.3492, 1.3524, 1.5645, 1.9595, 3.4143, 5.4849, 6.8115, 10.9489, 19.8932}},
{0.0005, {1.0000, 1.0534, 1.1068, 1.2821, 1.5151, 1.7112, 3.6442, 4.7700, 6.9978, 12.2086, 22.5374}},
{0.0025, {1.0000, 1.0015, 1.0031, 1.0876, 1.4027, 1.6920, 3.3014, 4.8019, 7.6273, 13.5664, 24.1108}},
{0.0125, {1.0000, 1.0071, 1.1713, 1.3430, 1.8335, 2.6955, 5.3925, 8.6168, 15.0510, 28.3340, 54.4134}},
{0.0500, {1.0000, 1.2241, 1.3336, 1.6827, 2.4608, 3.7554, 7.3573, 12.4689, 23.2699, 45.0874, 86.5339}},
{0.0625, {1.0000, 1.1011, 1.2670, 1.5964, 2.4030, 3.6996, 7.1772, 12.4852, 23.3882, 44.7427, 87.8259}},
{0.3125, {1.0000, 1.2818, 1.5493, 2.2546, 3.7804, 6.5428, 12.9410, 24.4539, 47.6262, 93.4733, 185.1019}}}
The cost factor to penalize a search in an IBF with more then 64 bins.
The table contains experimentally derived cost factors that were collected when measuring the runtime of an (vanilla) IBF with 64/128/... technical bins given a specific FPR. Each measurement was normalized by the runtime of an 64 bin IBF.
Each run was conducted 5 times and the mean was taken over the measurements. Low FPR rates were observed to have a rather high variance on the runs.