ddr4 memory

[AlreadyOver]

Hey there, so im almost done with my build except for the memory and i dont know what to choose. I watched Linus video about ddr3 show not much difference betwen 800 mhz and 2400 but does this apply for ddr4?

So is there any difference betwen cl12 or cl16 and 2133 and 2666?

[Boink]

You won't notice a difference in almost all workloads and even then the performance difference is quite small. Don't concern yourself with latencies or clockspeeds of RAM.

[Allshevski]

3 minutes ago, AlreadyOver said:

Hey there, so im almost done with my build except for the memory and i dont know what to choose. I watched Linus video about ddr3 show not much difference betwen 800 mhz and 2400 but does this apply for ddr4?

So is there any difference betwen cl12 or cl16 and 2133 and 2666?

Yes it kinda does.. Unless you're doing CPU heavy tasks like processing data or APU gaming. CL marks are not always equal. For example low CL with low frequency will be equal to high CL with high frequency. 

[airdeano]

nope, not really. the latency is usually the hidden issue with higher speeds.

[AlreadyOver]

Gonna use the system for gaming, so is it ok if i go with  2x HyperX Fury Black 8GB DDR4 2133MHz CL14 ?

[MageTank]

For a moment, I thought I had actually convinced this forum that faster memory does in fact aid in gaming performance. Guess I was mistaken. Oh well, back to repeating myself.

 

Ram speed impacts gaming performance in situations with high CPU overhead. Plain and simple, cannot be refuted by any source. The difference depends on your hardware, with stronger GPU(s) showing a bigger difference depending on the titles. I have personally noticed a 10% difference in minimum framerates when going from DDR4 2133 CL15 to 2800mhz CL15. Going from 2800mhz CL15 ro 3200mhz CL15 did not make much of a difference at all with my Pentium G4400 and GTX 770. If i had a stronger GPU, it might have. 

 

Click my signature to learn more, or check out the following links from other sources if you find my personal tests insufficient:

 

http://www.overclock.net/t/1487162/an-independent-study-does-the-speed-of-ram-directly-affect-fps-during-high-cpu-overhead-scenarios (Old post on OCN covering the impact on X79 platform with R9 290X in crossfire, with several other posts having similar results)

 

 

 

If you refuse to believe the 3 sources i have provided, test it yourselves. Load your default JEDEC speeds, then test your favorite games. Then load your higher speed XMP profiles, and run the test again. It's really not that hard to test.