Compare implementations
Two numbers may be put side by side when they came from the same point on the same machine. Everything on this page follows from that.
IMPLEMENTATIONS4POINTS MEASURED20MACHINES3COMPARABLE PAIRS0
WHERE NUMBERS EXIST20 of 240 cells measured
fairmath-fpga
fairmath-general-runner
fherma-general-purpose
Naive element-wise
test with random sleep
test with sleep
testl2n3
every seed passedsome seeds failedclaimed, not measurednot claimed
N2203 on fairmath-general-runner
One number is not a comparison. It is here because it is the only measurement at this point on this machine.
POINT
MACHINE
IMPLEMENTATIONMEDIANACCURACYSECURITYRINGCAPACITYKEYSMEMORYSEEDS
01test with random sleepg-arakelovtest with random sleep3.63 sfastest hereexactbyte for byte—not declared———10 MBpeak10 of 10
NOT IN THIS COMPARISON
Naive element-wiseClaimed and not yet run here.NOT MEASURED
test with sleepOutside the region its author says it handles, so it was never offered this.NOT CLAIMED
testl2n3Outside the region its author says it handles, so it was never offered this.NOT CLAIMED
HOW ACCURACY IS JUDGED
Every accuracy figure here is the output of the verifier in this specification’s testing bundle — one function, run against what each seed left behind. It travels with the specification, so every implementation is judged by the same code.THE ONE RULE
A time means nothing without the point it was measured at and the machine it ran on. Fixing both gives a comparison; releasing either gives a reading of something else — how an implementation scales, or where it runs. Nothing on this page averages across machines, and no implementation is judged on a point it never claimed.