An EEMBC Benchmark


This table lists all of the scores uploaded by members or licensees. It is not required that members or licensees upload scores, and in many cases the scores they create are for internal use only. However, EEMBC requires that before any license holder can use a score publicly (i.e., not under NDA) it must first be uploaded to the EEMBC website.

Scores that are certified have undergone a rigorous analysis by the EEMBC Certification Lab. Certification is a benefit only available to members, and guarantees the score adheres to the official run-rules for that benchmark.

Vendors who certify their scores may use the "EEMBC Certification Logo" () on their press release, product packaging, etc.


Each ULPMark-CoreMark entry contains three score tuples, one for performance, and two for energy-efficiency. The performance score indicates the best CoreMark score. The energy scores indicate the best ULPMark-CM scores, with one score at 3.0V and the other at a 'best' voltage. In order to facilitate energy comparisons between different products, all entries require one energy score at 3.0V.

The score tuple contains three components: the ULPMark-CM score (which is an energy metric), the voltage at which the ULPMark-CM score was collected at, and--importantly--the CoreMark performance score associated with the energy efficiency score. By showing all three scores together, the performance vs. energy tradeoffs are clearly illustrated: "ULPMark-CM Voltage [CoreMark]"

The ULPMark-CM score is unitless, but is derived from iterations per micro-Joule, divided by 1000. This represents the amount of energy required to perform one CoreMark iteration. The scale factor of 0.001 is simply for convenience. Scores are reported to three significant figures since there is a +/-3% run-to-run tolerance.

{{ error }}
{{ first }} − {{ last }} of {{ total }} platforms

Copyright © EEMBC