Searching for Hynix MFR 8Gb devices
Searching for Hynix MFR 8Gb devices
Hello everyone, I own a 4x8Gb Corsair Vengance LPX 2400C16 (PN: CMK8GX4M1A2400C16). I’m not sure if others know but Corsair releases different models for the same part number, with varying chips and layouts. My current setup is a Ver5.30 single rank 25nm chip, while my previous one was Ver5.20 using 29nm dual rank. I’ve also noticed earlier Micron versions with 30nm dual rank, but I’m not sure about the latest ones. It feels like buying them separately is quite a gamble. I decided to test it mainly because of poor AC Odyssey performance, hoping for improvements. I managed to hit 3000Mhz at 16-17-17-38 CR1 tRFC 486 with 1.23v. What’s interesting is how it reacts to temperature and high voltage—previously, even a 1.35v setting caused crashes in Aida64/memtest, but lowering it to 1.23v gave me over 400% stability on both tests. It seems Hynix MFR chips can be quite sensitive to these factors, especially since the slots are packed closely together.
My setup is generally well-ventilated. The MB PCB sensor shows a maximum internal temperature of 48°C after an extended game session during a 25°C day. Right now I don’t have a spare fan to install, but I might consider getting one for testing purposes. The interesting aspect is the consistent timing at 3000Mhz. Using 1.23V performed well in my tests so far; however, increasing it to 1.35V causes immediate errors during memtest and failure after about six minutes of stress testing. I checked it manually and felt slightly warmer, though not excessively so. It was cooler than my GPU heatsink at idle (reporting 55°C), likely around 45°C. Here are the results I’ve gathered: 3000Mhz @ 1.23V = 400%+ memtest success, no errors; 2-hour stress test @ 1.30V = 10% failure; 14-minute test @ 1.35V = 2% failure; 6-minute test back to 2666Mhz @ 1.30V = 200%+ memtest success.
Wouldn’t that make more sense the other way around? If a 3000Mhz board runs at 1.23v, does that really indicate a solid 3000Mhz setup? What indicates unstable undervoltage RAM? Could it cause crashes or BSODs? Might it hang? I’m not sure what you’re seeing. Everyone seems to jump straight up to 1.35v, just like before testing at normal voltages. Also, what do your earlier test results show? Does it look like temperature effects or something else I’m overlooking?
But it failed the binning test at 1.35V. Remember, 3200MHz CL22 1.2V is extremely uncommon—only Crucial offers them for desktops today. Regardless of the cause, it crashes, triggers BSODs, and behaves erratically. MFR isn’t known to be very sensitive to temperature or degrade so rapidly at such low voltages. This isn’t an old Micron die that stops improving after 1.3 to 1.4V; MFR is unusual.
Haha, that does sum up my experience with it. Thanks, I am leaving like this for now; [email protected] 16 17 17 38 CR1 486tRFC, everything else is on auto if someone with those wanna try or help improving. Just to remind this is a single rank 25nm version 2400 16 16 16 39 CR2 462tRFC XMP stock. Results are 9000MBs average improvement on R/W/C and 10.0ns lower latency (Aida test) comparing to stock 2400.