Conclusion

To be, or not to be… no that’s now how it goes, to DDR4 or to DDR5 on Intel Alder Lake, that’s why we are here today. Intel has released its 12th Generation Intel Core Processors A.K.A. Alder Lake CPUs and Intel Z690 platform. With this platform is support for two RAM standards, DDR4, and brand new DDR5. The difference is in the motherboard, you will be able to buy either Z690 Socket LGA1700 DDR4 motherboards or Z690 Socket LGA1700 DDR5 motherboards.

The question is then asked, should I stick with DDR4, or go all out and upgrade my RAM also to DDR5? Does it really make a difference? In this review today that is exactly what we did, we took an Intel Core i5-12600K CPU and tested it on two different motherboards, one being a DDR4 motherboard, and the other a DDR5 motherboard. In this way, we can answer this question and see if it matters.

If you haven’t yet, read our full Intel Core i5-12600K DDR4 Alder Lake CPU Review first. In that review, we introduced the new CPU and platform and reviewed performance compared to the competition. However, we did it uniquely on a retail GIGABYTE Z690 GAMING X DDR4 motherboard with DDR4 memory. This represented a great value proposition, a nicely priced CPU, and a saving cost using DDR4, you can see how that turned out in the review.

Thanks to MSI, they sent us a brand new 32GB kit of Kingston FURY Beast DDR5-5200 memory and their flagship MSI MEG Z690 UNIFY motherboard which supports DDR5. We were now fully equipped to make DDR4 versus DDR5 comparisons and will be able to review more Alder Lake CPUs on this motherboard with DDR5 in the future. Stay tuned for a full review of this motherboard as well.

DDR4 versus DDR5 Performance Results Synthetic

Starting with 3DMark’s CPU Profile test, testing max threads and single-thread we didn’t seem to see any kind of performance differences. This test is very CPU-focused, with a specific type of workload, but shows that just from a CPU performance perspective, DDR type is not changing the performance of the CPU either way.

Geekbench 5’s Multi-Core benchmark out of all the other benchmarks showed the most difference in synthetic testing. It showed us an 8% performance advantage with DDR5 versus DDR4. This test works on multiple types of workloads at once, and with various types of workloads, DDR5 did help. This wasn’t replicated in single-core performance though, where the opposite was true, and the better latencies of DDR4 helped more when strictly bound by one core’s performance.

SiSoftware Sandra backed up the fact that there wasn’t any difference between DDR5 or DDR4 in integer or floating-point performance. In fact, floating-point multi-thread slightly favored DDR4, possibly due to better latencies. The real benefit was shown in AIDA64 memory bandwidth tests. Memory bandwidth is where DDR5 shines and is able to show off its advantage over DDR4. It allowed a 30-35% performance improvement in memory read, write and copy performance bandwidth. If bandwidth is something you need, DDR5 has it in gobs. We did though find that the memory latency was slightly increased with DDR5.

The rendering performance was pretty much a wash. With Cinebench, Blender, and V-Ray we found no differences going with DDR5 over DDR4. It just didn’t affect rendering performance to any real degree. The only exception, and it was small, was HandBrake transcoding video, DDR5 helped shave a little time off the total time, but it wasn’t major. I suppose if you have hours long of video, it could add up to a noticeable difference, but that’s an extreme case. It would benefit you more to use something like QuickSync to speed up exporting video.

DDR4 versus DDR5 Performance Results Games

When it comes to games, well much is the same there as well. In most of the games, DDR5 made no appreciable difference no matter the resolution. Generally, we found in every game 4K is completely GPU bound, even on an RTX 3080 Ti and DDR5 did nothing to help game performance in the most demanding games. 1440p also was mostly the same, only in a couple of games did we find a few percentages difference at 1440p, but nothing that you would notice during gameplay.

Only at 1080p were there some percentage differences around 8% in Far Cry 6. However, that seems to be an outlier, and compared to the other games, it was not the norm. Even at an 8% difference, it didn’t necessarily change the gameplay experience, just gave you more frames at an already well above 100FPS gameplay experience.

It seems to us that if you are gaming, the only way DDR5 would ever make a difference is if you have a very fast GPU like an RTX 3080 Ti or 3090, and you play games at 1080p. In that scenario, you might see some percentage differences right under 10%. However, if you are running a GPU under those, and playing at more reasonable resolutions like 1440p, you just aren’t going to get any benefit from DDR5 for gaming. The biggest thing that will make a difference for you is upgrading your GPU. Whether you have DDR4 or DDR5, you are going to have the same gameplay experience.

Final Points

After our review today, we think we hit the nail on the head in our Intel Core i5-12600K DDR4 Alder Lake CPU Review. The combination of the cheaper Intel Core i5-12600K at $289 and existing cheaper and more abundant DDR4 RAM is the best value for an Alder Lake build at the moment.

DDR5 is expensive, DDR5 is scarce. DDR5 while running at high frequencies, suffers from high latencies and timings at the moment. It is new, that is expected. It took time for DDR4 to mature. Eventually, DDR5 will improve over time, it will mature, there will be faster frequencies, way faster than DDR4 could ever go, and timings and latencies will improve as well. We just aren’t there yet.

Right now you can save some money and sanity by re-using your DDR4 if you want to upgrade to an Alder Lake CPU. Or, you can buy some fast DDR4 and get it cheaper and have more to pick from, compared to DDR5, again saving you money and allowing you to build a system right now instead of having to wait to find DDR5 in stock. In our testing today we showed that it just doesn’t make a difference. DDR5, right now, doesn’t offer any advantages. Will it eventually? Probably, especially once frequencies really ramp up to those 6000MHz+ ranges and latencies improve.

Intel made the smart decision to support both DDR4 and DDR5 on this platform and we are glad Intel did not force us into DDR5 only just yet. There is one other thing to consider though, and that is that the more premium motherboards being sold are based on DDR5, there are more premium DDR5 boards than DDR4 boards. It seems that manufacturers are focusing on the very high-end being DDR5, and not DDR4. This also means motherboard costs go up if you enter into a DDR5 build as well. Perhaps we will see more premium DDR4 Z690 boards pop up if DDR5 remains as expensive and scarce as it is, we certainly would like more options if the platform supports both DDR standards.

When DDR5 does ramp up and is actually at an affordable price, upgrading to DDR5 will be the future on Intel and AMD platforms. However, don’t expect miracles with DDR5, it is not going to change your gameplay experience drastically, especially GPU limited scenarios. DDR RAM speeds are not what is holding back your gaming performance.

Discussion

Don’t Miss Out on More FPS Review Content!

Our weekly newsletter includes a recap of our reviews and a run down of the most popular tech news that we published.

Brent Justice

Brent Justice has been reviewing computer components for 20+ years, educated in the art and method of the computer hardware review he brings experience, knowledge, and hands-on testing with a gamer oriented...

Join the Conversation

5 Comments

  1. How is that MSI cooler doing? CPU cooling for alder lake is my main big question atm and MSI liquid coolers seem to be one of the few that are compatible that I can get over here.

    Especially worried after reading that article where they show bent CPU heatsinks with dodgy cooler mounts.

  2. [QUOTE=”Denpepe, post: 45055, member: 284″]
    How is that MSI cooler doing? CPU cooling for alder lake is my main big question atm and MSI liquid coolers seem to be one of the few that are compatible that I can get over here.

    Especially worried after reading that article where they show bent CPU heatsinks with dodgy cooler mounts.
    [/QUOTE]

    It’s the best cooler that I’ve had on the test bench (on the 10980XE though). I’m sure Brent has better feedback on Alder Lake specifically…

  3. I’ll have more experience in the near future, have it installed on a 12700K right now, and then 12900K, so far as far as the 12600K goes it went real well, and that CPU stayed cool very well with it, however, I have been running it on 100% fan speeds for testing purposes, so it’s loud. I will be getting to temp testing on the 12700K soon, and then 12900K ,and we’ll really see how it holds up. As for bent heat spreaders, haven’t noticed anything odd. I will say though that with the screws screwed into the backplate it’s a lot looser than I would have thought it would be, or I should say, compared to other coolers, but it’s not a problem once you screw down the block on top of the screws with the thumbscrews, it tightens up nicely, and you really can’t over tighten it, it has a natural stop. Make sure to check in next week to see how the temps on the 12700K did, then in a couple more weeks the 12900K.

  4. I will leave a link to the article I read, it’s kind of though to read guess it did not translate well but found it interesting. [URL]https://www.igorslab.de/en/bad-cooling-at-alder-lake-problems-at-socket-lga-1700-on-the-lane-among-all-remedies/[/URL]

    Does look like the risks are bigger with cheaper boards.

  5. [QUOTE=”Denpepe, post: 45073, member: 284″]
    Does look like the risks are bigger with cheaper boards.
    [/QUOTE]
    That’s par for the course for a number of reasons. That being said, I did notice that the back plate EK offers for it’s LGA 1200 / 2066 socket waterblocks is much thicker and beefier than what I’ve seen on pretty much every generation going way back to the pre-Sandy Bridge era.

Leave a comment