DDR5 vs DDR4: Which RAM is best for gaming and content creation?

Before we get into content creation, let's start with a synthetic benchmark that measures RAM speed specifically. This'll allow us to see the theoretical maximum performance boost we'd see in a task that is limited only by memory speed, setting a kind of upper bound on the sort of gains we'd see from faster RAM in other tasks like content creation or gaming.

For this, we used Aida64's cache and memory benchmark, which includes four RAM-specific tests measuring read speeds, write speeds and copy speeds, plus latency. As well as our out-of-the-box settings for each of our memory kits, we also decreased the frequency towards the baseline in regular increments while keeping timings the same - and explored some simple overclocking by raising voltages and frequencies by a small amount.

You can see how, as memory frequencies ramp up, we get a corresponding increase in read, write and copy speeds. Our DDR5-6600 overclock allows us to breach the 100K barrier in read speed, with write and copy speed closer to 90000MB/s. Contrary to our expectations, latency at this 6600MT/s overclock is lower than at the JEDEC standard of 4800MT/s. Otherwise, our results are logical, with the fastest overclock resulting in a 33 percent boost to read memory read speeds from the slowest to fastest DDR5 frequencies. From our fastest DDR4 (4400MT/s) to our fastest DDR5 (6600MT/s), there's a 50 percent boost to read speeds which of course makes perfect sense. However, measured latency is relatively similar between the two results, at around 68ns despite being CL20 on DDR4 and CL40 on DDR5.

Content creation workloads from our CPU reviews return once again, as we see how faster RAM impacts real-world performance. In Handbrake, our H.264 encode average frame-rate is nearly identical between the fastest DDR4 and DDR5 options - 69.49 for DDR4-4000 and 70.66 for DDR5-6400. That's far less than we anticipated, and shows that there's not really a need for DDR5 RAM in this sort of workload at present. The H.265 (HEVC) results are similar, with all tests showing a frame-rate between 28.53 and 29.34fps (around three percent).

Similarly, we don't seem to be memory frequency limited in any meaningful sense in Cinebench, with slightly higher results with DDR4 (~10450) than DDR5 (~10200) suggesting latency may be more of a factor here.

It's all a bit of a nothing-burger so far in terms of the real-world testing, but games could prove a far more interesting springboard for discussion. Let's move into those now, as we test Ashes of the Singularity Escalation, Counter-Strike: Global Offensive, Far Cry 6 and Crysis 3 Remastered.

DDR5 vs DDR4: Which Z690 motherboard makes sense?

• Introduction, hardware breakdown, test system
• Content creation benchmarks: Cinebench, Handbrake, AIDA64 [This Page]
• Gaming benchmarks: Ashes, CS:GO, Far Cry 6, Crysis 3 Remastered
• DDR5 vs DDR4 on Z690: the Digital Foundry verdict