Hello All,
Question: which of the two equations should we use in the process of the selection of cables for FDR receiver tolerance testing? Mellanox has a question regarding the FDR receiver tolerance tests with linear cables (described in section 6.7.8.1 of the InfiniBand Architecture Specification Volume 2), specifically regarding the selection of cables for the test. Table 55 of the InfiniBand Architecture Specification Volume 2 defines the fitted insertion loss coefficients of the cables that should be used in the tests, with a reference to clause 85.8.4.2 of the IEEE 802.3.2012 document. However, there is a discrepancy between the ways the IEEE 802.3.2012 document and the InfiniBand Architecture Specification Volume 2 define the relation between the fitted insertion loss and the mentioned coefficients - the equation used in the IEEE 802.3.2012 document is , The InfiniBand Architecture Specification Volume 2 defines the fitted insertion loss as , where is the signaling frequency (14.0625 GHz). Thanks Rupert
image007.png
Description: Binary data
image008.png
Description: Binary data
image009.png
Description: Binary data
_______________________________________________ ewg mailing list ewg@lists.openfabrics.org http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg