HWMonitor Pro 1.53 download the new1/14/2024 Overall, your CPU might made training easier, but it didn't change the outcome of your used setĮxcept that the memory controller was never changed ~ only voltages around the IODie and some removed intentional limits Memory lottery also exists, but let's hope that you can run tRCD_RD 14 on 3800MT/s Sometimes up to set, it can need 1.53-1.55v for 3800, some only 1.5 I suggest you rework your 3734MT/s CL14 set again, and see where the issue is, preventing you from getting it stable on 3800MT/s Same would go for a 2700X with probably the same good or same not so good set of timings Your 3734MT/s CL14 will be equally stable and equally unstable, between both CPUs Overall CL14 most of the times, requires between 1.475-1.55v on higher than 3600MT/s freqĪt least that was the case with older memory PCBs and didn't change much recently Well, some hynix kits seem to scale linear in req voltage beyond 1.54v, just with smaller freq jumps ~ still requiring the same voltage jumps (logarithmic in the sense that, the further you go, the higher voltage jumps would be needed) It's a logarithmic scaling,and only b-dies behave nearly linear +4 steps up = ~200Mhz more requires +1 CL to be stable or +0.5-0.75mV more (signal integrity could be better ~ considering i got 4400 out of my A0 Vipers) Click to expand.Except that the memory controller was never changed ~ only voltages around the IODie and some removed intentional limits
0 Comments
Leave a Reply.AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |