Review on Timetec 64GB DDR4 3600MHz PC4-28800 CL18 Non-ECC Unbuffered Gaming Memory Module Upgrade by John Swain

Note: This is an addendum to my previous positive review. ON THE SURFACE IT FEEL GOOD, bought this 64GB "ryzen compatible" Timetec kit sold at 3600MHz for my new 3950x and I was able to get the timings to 16-18-18-36 at " Standard” voltage reduce. I even ran memtest86 on the thing and it was stable the whole hour I tested it. THIS IS HERE THINGS GOED ODD ODD So the RAM was set up and seemed to be working fine. The obvious next step was to test the new system I had just built. Every test I've run has shown my GPU, CPU, and NVMe SSD performance to be well below average (I mean the bottom 3%). I was very upset to see such poor results with my new system and tried everything in the book to improve them. I made sure power settings were open, checked bios settings and made changes, upgraded cooling to the point where my system was idling in my upper 20s and benchmark mid 50s, replaced my graphics card - even got a mobo a higher class. Even with all of those boxes checked, I could barely squeeze out an average bench with maxed overclocks across the board. MY SUPER LUCKY ACCIDENTAL DISCOVERY After customizing the BIOS settings, I had to reset the BIOS settings, and I was [unknowingly] lucky enough to do so. Forget about re-enabling the XMP profile or playing through the timings manually. During this time, I made another attempt to improve my test scores. With the motherboard set to default to 2400MHz, all of my hardware just magically seemed to perform above average for the first time since I assembled the system, but RAM, which used to be the only thing above average, suddenly became a lot more below that Average. When I looked closely, I noticed that it operates at a frequency of 2400MHz. Thinking the performance was so good thanks to my settings, I rebooted the BIOS and set the XMP profile to active, logged into Windows again and boom! Same crappy benchmarks for the CPU, GPU (which turned out to be a 2080ti) and even my PCIe 4 NVMe SSD. FIRING POINT At that moment it suddenly dawned on me that the problem must be in the RAM. this kit by visiting the Timetec website but it's not there at all. I tried using different software to get more information about this RAM and its specs, but the weird thing was that the manufacturer information always returned as "unknown". I finally Googled the model number printed on the sticker on the front of the RAM. It turns out that the exact model number (which is insanely long and unique) only appears on another manufacturer's website. Very oddly, this other manufacturer listed an identical 1.2 volt kit clocked at 2666MHz. However, I could not find a single trace of this manufacturer ever sold online. Guessing it was in fact an overclocked and heavily overclocked kit that had been rebranded, I tried the 2666MHz clock with shorter timings and for the first time since building my system I was able to get my entire Hardware (yes, including RAM) received. for optimal performance. !CURRENT LITMUS TEST Bought a more expensive 64GB 3600MHz kit from a reputable manufacturer, plugged it into my phone, enabled XMP and with the BIOS settings prepared, my system ran much better than average. In fact, I scored nearly 15 points better on default settings with legitimate 3600MHz memory installed than with overclocking galore and the Timetec kit. The problem is solved! WHAT HAPPENED TO ME I think this memory should be clocked at 2666MHz with 1.2V power settings and probably somewhere around 16-18-18-36 to sync. At this frequency it works exactly as expected without affecting the overall performance of the system. When running a 3rd gen Ryzen system at a much higher frequency and higher power settings, I believe the Zen 2 infinite structure should increase dramatically to prevent system instability (which works fantastically well, by the way). . All that extra stress on the system to maintain stability at 3600MHz severely reduces the performance of everything else. FINAL THOUGHTS If my suspicion is correct, and this is indeed a kit designed to operate at 2666MHz, then the value proposition of this product is significantly less. There is a lot of 2666MHz RAM in this price range that can be overclocked and overclocked and if I wanted to buy 2666MHz RAM and overclock I would buy memory listed at that speed and price. While this kit definitely achieved its advertised speed, it also degraded the performance of my entire system and cost me many hours trying to overcome what I believed to be the worst dice roll in silicon lottery history. For all of these reasons, plus what I consider almost certain dishonesty, I'm downgrading my previous 5 star rating to one star.