Search the Community
Showing results for tags 'ddr3'.
Found 2 results
DDR3 results suspiciously low
Bojanum posted a topic in Benchmarking, system performanceHi there, I have one problem and I think you guys/girls can help me out with it. My "Cache and memory benchmark" results are low, at least I think they are low,especially "write speed" and latency. I tried changing timings, voltage and nothing helped. Default timings are 13-11-10-30 CR1.With 10-10-10-28 CR1 I managed to lower the latency for a bit and that's it. Please, if you can, help. Thank you in advance. I forgot to write modules name: Kingston HyperX Fury 1866MHz 3x8Gb
From the JEDEC Standard 21-C, Annex K: Serial Presence Detect (SPD) for DDR3 SDRAM Modules, Document Release 6 the following definitions for "Byte 31: SDRAM Thermal and Refresh Options" bit 1. Extended Temperature Refresh Rate 1 = Extended operating temperature range from 85-95 ï‚°C supported with standard 1X refresh rate 0 = Use in extended operating temperature range from 85-95 ï‚°C requires 2X refresh rate The bit is to indicate if a 1X refresh can be used in an extended temperature range or if a 2X refresh rate must be used. The AIDA64 report reports for the "Extended Temperature Refresh Rate" item of the "Memory Module Features" list either "Supported" with the check box checked or "Unsupported" with the check box unchecked. These status verbiage does not reflect that nature of the bit being one of two choices. Per the JEDEC spec it would be more informative if the "Extended Temperature Refresh Rate" item showed "2X" or such and the check box unchecked and "1X" or such with the check box checked.