A quick query - is there any information on how HWiNFO64 is gathering the DRAM Stepping value for DDR5 memory?
I assume it is simply pulling the 2-byte HEX value from the SPD table and converting to ASCII, however I've seen strange values across a couple devices and was hoping to confirm whether it is an application issue or (multiple instances) of incorrect data in the SPD.
First instance is with Crucial CT32G56C46S5.M16D1 32GB SO-DIMM modules - these according to my understanding of the Crucial numbering should be Micron Die Rev D DRAM (DRAM Stepping: 0x44), however HWiNFO64 reports "G-Die (4.7)"
Second instance is Corsair CMK64GX5M2B5600C40 32GB UDIMM module gives a SDRAM Manufacturer of Micron and DRAM Stepping of B-Die (4.2).
Again this appears incorrect as Micron Die Revision B is a 24Gb DDR5 SDRAM (https://media-www.micron.com/-/medi...a-sheet/dram/ddr5/24gb_ddr5_sdram_dierevb.pdf) and therefore should only be present in 24/48GB modules.
I also tested with Passmark RAMMon and received the same results on the Corsair equipped system.
Is this an issue in how both applications are reading the SPD data, or are DRAM manufacturers not being diligent in how they populate the DRAM Stepping SPD field?
I assume it is simply pulling the 2-byte HEX value from the SPD table and converting to ASCII, however I've seen strange values across a couple devices and was hoping to confirm whether it is an application issue or (multiple instances) of incorrect data in the SPD.
First instance is with Crucial CT32G56C46S5.M16D1 32GB SO-DIMM modules - these according to my understanding of the Crucial numbering should be Micron Die Rev D DRAM (DRAM Stepping: 0x44), however HWiNFO64 reports "G-Die (4.7)"
Second instance is Corsair CMK64GX5M2B5600C40 32GB UDIMM module gives a SDRAM Manufacturer of Micron and DRAM Stepping of B-Die (4.2).
Again this appears incorrect as Micron Die Revision B is a 24Gb DDR5 SDRAM (https://media-www.micron.com/-/medi...a-sheet/dram/ddr5/24gb_ddr5_sdram_dierevb.pdf) and therefore should only be present in 24/48GB modules.
I also tested with Passmark RAMMon and received the same results on the Corsair equipped system.
Is this an issue in how both applications are reading the SPD data, or are DRAM manufacturers not being diligent in how they populate the DRAM Stepping SPD field?
