Is DDR5-8000 Worth It? The Ryzen AM5 Test

I watched the video that this article is based upon and while watching it, it took me back to when I building my Ryzen 7700X-based system, specifically the thought of what RAM I was going to use. I said... Self? If Steve recommends a specific memory kit, far be it from me to not use that same memory kit. And I did.

I went with the the G.Skill Trident Z5 Neo F5-6000J3038F16GX2-TZ5N (I opted to go with no RGB, hence the lack of an R). Everything about the memory kit is the same in all the reasons that matter.

I just enabled EXPO and Memory Context Restore and instant success. No issues. Rock solid stable. OK, so I could tweak the timings a bit using that Zen Timings program, but my system is stable and that's all I could ask for.

I just wanted to say that your recommendation on memory kits helped a lot when building my system.
 
Wtf is this article literally just copied hardware unboxed's video stole there pictures. Not even a signal mention of hardware unboxed in this article. This by my understanding copy right infringement

Hardware Unboxed is the name of the author's youtube channel. He started out as a writer on TechSpot before he started making YouTube videos.
 
Well not very related to the topic but anyone has some experiences with the kit Ram TeamGroup T-Create 6000 CL30 ? I'm planning on buying them instead of GSkill
 
This is a horrible test! Do this with an 8700G, then run gaming benchmarks where the RAM speed does matter and LLM testing as well. Considering the 8000 series desktop CPUs are the only models in the AMD lineups with an NPU and giving that access to a large pool of high speed RAM will most certainly allow for a boost in performance across the board.
 
This is a horrible test! Do this with an 8700G, then run gaming benchmarks where the RAM speed does matter and LLM testing as well. Considering the 8000 series desktop CPUs are the only models in the AMD lineups with an NPU and giving that access to a large pool of high speed RAM will most certainly allow for a boost in performance across the board.

No, we all know RAM is more important on iGPU, we want to see non-iGPU benchmarks. This isn’t that kind of review, if you want one ask nicely, don’t be a ****.
 
No, we all know RAM is more important on iGPU, we want to see non-iGPU benchmarks. This isn’t that kind of review, if you want one ask nicely, don’t be a ****.
Yes, but it is also more important on an APU like the 8700G, because it allows the NPU to access the full amount of RAM for neural processing on Large Language Models. Right now your options are limited to things like an M3 Ultra. It isn’t the best solution, but it is the cheapest.
 
Last edited:
God, seeing how cheap non-ecc DDR5 RAM is kind of makes me sick... just dropped $1600 US on 256GB of V-Color RAM (6000) for my Threadripper... $110 for 32GB - I dream :(
 
Given the BIOS nightmare I went through getting my AM5 to run DDR-6000 with my 7700X, I wouldn't even think to push the platform higher. I even used the memory kit that AMD shipped to reviewers.

AMD did a bait and switch on smooth DDR-6000 support. I still slow boot to this day.
 
This article was useful, but I would really have wanted to see 2 additions: Testing productivity outside of gaming. How does video processing go? How about some CAD work? Even something as basic as browser speed.

The other addition would be another speed in between 6000 and 8000 Mhz. There are a lot of 6400 kits out there, for example.
 
This is a horrible test! Do this with an 8700G, then run gaming benchmarks where the RAM speed does matter and LLM testing as well. Considering the 8000 series desktop CPUs are the only models in the AMD lineups with an NPU and giving that access to a large pool of high speed RAM will most certainly allow for a boost in performance across the board.
I'd also like to see a review of the 8700G with higher Ram speed but I don't think your introduction is appropriate.
 
Can you try it next with intel?

(I'm not sure if there was a test conducted already--I didn't check)
 
VERY informative! I am a 4K gamer, so I'm perfectly fine with my 6000/CL30 RAM. thank you!
 
Great test, as usual, not much to be gained but the 12% difference between the 5600 & 8000 was surprising. As long as my PC's memory doesn't cause a crash or some weird issue, I don't care about it much. For the past 10 years stability and size are more important to me than all-out performance. (that's what she said, lol)
 
What voltage does it run at? Ryzen 9x series is rumored to be sensitive to ram running over the recommended 1.2v. I see some kits are listed as high as 1.4v which is a huge "no-no" for the longevity of your cpu
 
Last edited:
It would be nice to know how they run at 1.25v, 1.30v, 1.35v not just at headline spec (1.20v & 1.40v). If I want to run at 1.30v, what's the best MT/s & timings I can use?
 
Can you try it next with intel?

(I'm not sure if there was a test conducted already--I didn't check)
Intel would be worse as the AMD memory controller is actually better, but these chips appear to be the Royal Neo line and they’re not believed to be CUDIMMs, where as the Intel compatible DDR5 8000 chips are CUDIMMs and thus have their own menory controller on the stick.
 
Based on these results it seems foolish to spend ~$100 more on a RAM kit for gaming. At that point just buy the X3D CPU instead.
 
Back