[regawdless]

It's time we take a fresh look at DDR4 vs. DDR5 memory, using the "new" Intel Core i9-14900K. The last time we compared DDR4 and DDR5 memory was nearly two years ago using the i9-12900K in 41 games. At the time, we were using DDR5-6000 and DDR4-3600 for testing but today we're upgrading that to DDR5-7200 and DDR4-4000 memory.

Interestingly, in a role reversal, DDR4 memory now costs quite a bit more than DDR5 (!). The G.Skill Ripjaws V Series 32GB DDR4-4000 CL16 kit is priced at $170, while the G.Skill Trident Z5 RGB Series 32GB DDR5-7200 CL34 kit is available for $130.

Mostly interesting for players using resolutions lower than 4k, because at 4k we get GPU limited fast. Very good results for 1440p players.
At 1080p: 7% faster average fps, 1% lows got faster by an average of 10%.

Some games showing no benefit at all, others give you 20-30fps more at 1440p.

Here's the 1080p overview:

Here's what that means in fps terms, some positive examples, showing off different resolutions because nobody with this kind of system plays at 1080p. 20fps more in BG3 at 1440p, Spider-Man also showing huge 30fps increases even at 4k.

DDR5 vs. DDR4 Gaming Performance

It's time we take a fresh look at DDR4 vs DDR5 memory performance using the Core i9-14900K. We're benchmarking with fast DDR5-7200 and DDR4-4000 memory, noting that...

www.techspot.com

At 1440p, it can often mean 10 - 30fps more, which is surprising to me and it seems like newer games will continue to show good gains with fast memory.

 

[Sean Mirrsen]

Now if only DDR5 motherboards were cheaper than DDR4 ones...

Also, what's so surprising? The PC is a whole of many parts. Any one part lagging behind others will pull the whole thing down. It's not that DDR5 offers a performance benefit over DDR4, it's that the high-end processors have interface rates so high that DDR4's maximum available frequency presents a bottleneck.

Right now I have an Athlon 200GE and 16 gigs of DDR4-2133. Even with the APU being the weakest of the Ryzen line (it's even called an Athlon), I'm sure I could get a decent performance boost just by upclocking the RAM to 3200.

 

[regawdless]

Also, what's so surprising? The PC is a whole of many parts. Any one part lagging behind others will pull the whole thing down. It's not that DDR5 offers a performance benefit over DDR4, it's that the high-end processors have interface rates so high that DDR4's maximum available frequency presents a bottleneck.

RAM speed never made a big difference before. The gains in gaming were small to non existing, it wasn't an important factor. Then DDR5 arrived in 2021, and overall, the gains weren't great, between 1-3% on averge. TechSpot benchmarked it two years ago. It was a bit over the place, with some games having big gains, others ran slower, most games had barely any gains.

With new CPUs and the newer games, that changed and fast DDR5 RAM now has a significant impact on performance in more games.

 

[SirKicksalot]

I recently bought 32 GB Kingston Fury Beast DDR5-6000 CL40. The CL number made by head spin while I was researching what to buy. Here's a good explanation: https://www.teamgroupinc.com/community/en/blog-detail/memory-latency/
I bought this memory because it has no RGB. If you're not using a stock cooler, this is important. I can't say how much better it is because I also replaced the CPU and motherboard.
NVME prices have also gone down.

 

[regawdless]

I recently bought 32 GB Kingston Fury Beast DDR5-6000 CL40. The CL number made by head spin while I was researching what to buy. Here's a good explanation: https://www.teamgroupinc.com/community/en/blog-detail/memory-latency/
I bought this memory because it has no RGB. If you're not using a stock cooler, this is important. I can't say how much better it is because I also replaced the CPU and motherboard.
NVME prices have also gone down.

Yeah I also updated my CPU and RAM lately, with huge gains. Didn't expect that upgrade to be that significant.