Search Results
6/28/2025, 5:08:57 AM
>>105729806
a wise choice
>>105729816
>am i not doing 30/20 at the moment?
picrel
>well if I have to increase primaries by one to have the same performance compared to sticking to a 200 mhz lower speed what's the point tho
that's not how timings work, 3000C14 ~= 3200C15 ~= 3600C17
https://www.techpowerup.com/dram-latency
that's why i used 3200 12-16-12 and 3733 14-18-14, similar absolute latency (which means i didn't have to mess with VDIMM either)
>entire mobo has an aneurysm and will require a cmos reset if it's any lower yeah
then don't touch it
i do tRFC*(new_speed/old_speed)
so if i'm going from 3733 to 4133 i do 540*(4133/3733) ~= 600
>I'd be perfectly happy with 3200 already
start by seeing if you can boot 3200 with loose timings, 3200 22-22-22 with auto subtimings
mess with the drive strengths until you find the most stable setup
a wise choice
>>105729816
>am i not doing 30/20 at the moment?
picrel
>well if I have to increase primaries by one to have the same performance compared to sticking to a 200 mhz lower speed what's the point tho
that's not how timings work, 3000C14 ~= 3200C15 ~= 3600C17
https://www.techpowerup.com/dram-latency
that's why i used 3200 12-16-12 and 3733 14-18-14, similar absolute latency (which means i didn't have to mess with VDIMM either)
>entire mobo has an aneurysm and will require a cmos reset if it's any lower yeah
then don't touch it
i do tRFC*(new_speed/old_speed)
so if i'm going from 3733 to 4133 i do 540*(4133/3733) ~= 600
>I'd be perfectly happy with 3200 already
start by seeing if you can boot 3200 with loose timings, 3200 22-22-22 with auto subtimings
mess with the drive strengths until you find the most stable setup
Page 1