Upgrading DDR3 RAM from 1600 to 1886 seems like a small tweak, but are you really getting much better results?
Upgrading DDR3 RAM from 1600 to 1886 seems like a small tweak, but are you really getting much better results?
Here are some system details: The Xeon E5 1650 V2 has 6 cores and 12 threads. It was pushed to a 4.2 gigahertz clock speed while running on air (the heat stays stable, and it worked well after testing with stress tests). A Chinese X79 motherboard is included, paired with 24 gigabytes of memory split into two parts: four gigabytes each on dual-channel slots. All the RAM runs at 1600 megahertz using a 120-gigabyte module. The computer uses Windows 10 and some games that run fine on both a 500-gigabyte SSD and a 2 terabyte hard drive. An MSI Nvidia GeForce GTX 1660 Gaming X with 6 gigabytes of graphics power is installed. Some BIOS settings are not flexible, but I am trying to set the memory timing instead. Currently, I have left the timings alone because I do not want to mess around too much with them. This computer is mostly used for gaming and runs all games smoothly at 1080p resolution with over 100 frames per second. I'm wondering if there would be any real benefits to overclocking the memory, or if just using it as-is is enough? Thanks a lot in advance.
Marginal, really. Back when ddr3 ram speed scaling was a thing, there used to be a cool article about it. It might work on your rig if you check the real-world tests instead of those fake synth ones. The difference is basically 1% or less anyway. I'd skip checking that out. Check this link for more info: https://www.anandtech.com/show/7364/memo...on-haswell
If it's going at 1600Mhz, I'm just going to leave it alone, though you could try making the timings tighter if that helps. Putting two different sized chips together is tough enough, and messing with the overclocking might cause more problems than what you'd actually want to do right now.
I think it's marginal. Back when there was a good article about DDR3 ram speed scaling, maybe that applies to your setup. Check the real-world tests instead of just the ones made in a lab. It seems like they won't make more than 1% difference. I wouldn't bother worrying about it.
Thank you everyone so much for your replies. I really appreciate them.