Performance improved noticeably after upgrading to new RAM.
Performance improved noticeably after upgrading to new RAM.
I think there are several memory controllers I’m considering. My Ryzen 7 1700 runs at 1GHz over spec for the CPU memory controller, and I also received a silicon lottery version on that chip. Another 1700 is running at 3200MHz RAM, with a 2666MHz memory controller on the first one. It took a while to fine-tune the timing, and a lot of trial and error (it works very well). I’m not sure the best way to measure it accurately. I’d appreciate seeing more posts from other I7 6700K owners who have experience with this chip.
It's more than just the CPU and IMC specs. We have no clue what board he's running on. No details on the exact RAM type... No screenshots or part numbers here. We're comparing low-end memory performance against 4000MHz achievements, but the voltage settings are uncertain without experience. 2400MHz likely offered better speed due to tighter timing, and Jedec + XMP probably work closely together. The 3200MHz board in this thread probably needs manual timing tweaks, while XMP seems to push CL17, which is likely where performance drops. Probably around 1.35V for the memory—if the board supports Vcore and Vcc, those can be adjusted up to about 1.15V depending on cooling and temperature. If Cache Optimization is an option, it could boost performance too. I'm not familiar with a 6700K, but if it's similar to an 8700K, a solid memory setup with a good board can last a long time.
Additional Tip: For measuring memory bandwidth and latency, AIDA64 is useful. At the top, click tools and choose cache and memory benchmarks to gauge performance direction.
I thought the board would work well with the overclocked K series chip, but I agree
The board runs an Asus Strix Z270H with Corsair RGB PRO 3200Mhz CL16 RAM. The XMP profile is now active, but there are problems with performance settings. I'm worried about how the system behaves if the memory controller struggles to support 3200Mhz. Specifically, I want to know if there will be minor lag or reduced frame rates at normal high settings—like noticeable stuttering or lower lows. I expect a significant drop, possibly around 50% or more in FPS, and I don’t want to accept any uncertainty.
The interesting aspect is that memory performance can vary by about 10% in actual gaming results, depending on the game. Some titles are more CPU-heavy like CSGO, while others rely heavily on graphics cards, such as the latest triple-A games. It's difficult to say whether memory changes affected your performance. Edit, reviewing your updated memory JEDEC profiles with CL over 20 shows you're much better suited to run XMP. Don't disable it.
Which monitor are you using? What refresh rate does it have? You're not experiencing crashes, just occasional lows—likely due to slower RAM. Which titles are you playing? How do you describe stuttering? Frame drops, temporary lock-ups, or mismatched frames?
BenQ 144hz runs smoothly without crashes, though frame rates drop significantly affecting gameplay. I compete at Apex level. Regarding the memory controller, pushing it too hard with RAM can cause similar issues if the RAM isn't built for it. The problem arose after installing new RAM because the BIOS retained old settings, causing a flashing error between CPU and DRAM indicators. I'm concerned this might damage the CPU controller.
I previously discussed a situation where posting wouldn't work with the new RAM. I was using a non-XMP profile at 2400Mhz, and there were red and yellow indicators (for my MOBO, these signify CPU and DRAM errors). Might this affect the memory controller?
Based on what I've learned, higher voltage with DDR4 can potentially harm the CPU memory controller. Intel likely won't confirm going beyond their specifications, but it's advised to keep it under 1.2V and use XMP at 1.35V out of the box. Running a memtest64 on the RAM at 1.35V and 3200MHz could reveal hidden issues. Regarding Apex, it's stored on a HDD or SSD?