https://www.sisoftware.co.uk/download-lite/
Just get the lite version and run the processor multi-core efficiency benchmark. Run it in both best-case and worst-case mode.
Will do.
I'm surprised your latency was that high.
It's about where it should be. 3200CL14 here: https://www.guru3d.com/articles_pages/amd_ryzen_threadripper_2950x_review,21.html
My numbers were "distributed mode". You're probably expecting local numbers? I won't get mid 70s (3200/3600*85.6) as I'm not running CL14, but I don't wind up in the mid 80s either because some of that is IF latency, not memory latency. I expect local latency to be almost exactly the same as the 3200CL14 numbers.
Current newegg pricing is $245 for the 3600CL15 pair vs $213 for the 3200CL14. 15% higher for ~13% better throughput and ~6% "gain" in latency. It's a pretty fair trade for bandwidth. If Ryzen ever gets the ability to use odd CLs, it would be a somewhat better trade in the latency department. I got the CL15 sticks over the 3600CL16 sticks mainly because I figured I'd increase my chances of being able to use these at 3200CL14 if I needed to. But I can't boot in 3200, so, joke's on me The 3600CL16 sticks are only 4% more ($223) than the 3200CL14, which seems pretty fair. What surprised me more was the increased throttling from running at higher memory speeds, which in my mind significantly lowers the value of the higher speed RAM, increases the value of running 3200CL14, and makes me even more unhappy that my board doesn't want to accept 3200 speed :sigh:. I'm as interested as you are in seeing the sisoft results.