Intel Alder Lake DDR5 Memory Scaling Analysis with G.Skill Trident Z5

One of the most nerve-racking elements of Intel’s launch of its latest 12th generation Alder Lake desktop processors is its support for DDR5 and DDR4 memory. Motherboards are either one or the other, while waiting for DDR5 to hit the market. While DDR4 memory isn’t new to us, DDR5 memory is, and as a result, we’ve been reporting on the release of DDR5 since last year. Now that DDR5 is here, although difficult to obtain, We know from our review of the Core i9-12900K that DDR5 performs better at base settings compared to DDR4. To study the scalability of DDR5 on Alder Lake, we used a premium DDR5 memory kit from G.Skill, the Trident Z5 DDR5-6000. We are testing the G.Skill Trident Z5 kit from DDR5-4800 to DDR5-6400 to CL36 and DDR5-4800 with timings as tight as possible to see if latency plays a role in improving performance as well.

DDR5 memory: scalability, pricing, availability

During our launch day review and analysis of Intel’s latest Core i9-12900K, we tested many variables that could impact the performance of the new platform. This includes the variation in performance when using Windows 11 compared to Windows 10, performance with DDR5 and DDR4 at official speeds, and the impact of the new Performance and Efficient hybrid cores.

With all the different variables in this exam, the goal of this article is to assess and analyze the impact of DDR5 memory frequency on performance. While in our previous articles on memory scaling we have generally focused on the effects of frequency, but this time we wanted to see how tighter latencies can impact overall performance as well. .

ASUS ROG Maximus Z690 Hero motherboard with G.Skill Trident Z5 DDR5-6000 memory

Addressing the price and availability of DDR5 memory at the time of writing, TLDR is that it’s currently hard to find in stock, and when it’s in stock it gets expensive. With a massive global chip shortage that many attribute to the coronavirus pandemic, the drought has pushed prices above MSRP on many components. Interestingly enough, it’s not DDR5 itself that’s causing the shortage, but the power management controllers that DDR5 uses per module to achieve higher bandwidth are scarce. As a result, the increased costs can be compared to a kind of early adoption fee, where users who want the latest and greatest will have to pay to own it.

Another variable to consider with DDR5 memory is that a 32GB (2×16) kit from G. Skill Ripjaw DDR5-5200 can be found at retailer MemoryC for $ 390. In contrast, a more premium and faster kit such as G.Skill Trident Z5 DDR5-6000 is priced at $ 508, an increase of about 30%. One of the things to consider is that a price increase is not linear with the increase in performance, and that goes for just about every component in memory, graphics cards, and even processors. The more premium a product, the more it costs.

Activation of XMP 3.0: it’s technical overclocking

In March 2021, we reported that Intel had effectively terminated its “Performance Tuning Protection Plan”. It was essentially a warranty extension for users planning to overclock Intel processors, which could be purchased at an additional cost. One of the main benefits of this was that if users damaged silicon with higher than normal voltages (VCore CPU and memory related voltages), users could effectively return the processors to Intel on a replacement basis. similar. Intel said very few people took advantage of the plan to pursue it.

One of the variables to note when running Intel’s Xtreme Memory Profiles (XMP 3.0) on DDR5 memory is that Intel classifies this as overclocking. This means that when RMA of a faulty processor, running the processor with the original settings but activating it, an XMP 3.0 memory profile on DDR5-6000 CL36 is something they consider to be an overclock. This could inherently void the processor warranty. All processor manufacturers adhere to JEDEC specifications with their recommended memory settings for use with any given processor, such as DDR4-3200 for its 11th generation (Rocket Lake) and DDR5-4800 / DDR4-3200 for its processors. 12th generation (Alder Lake).

When it comes to overclocking DDR5 memory on the ASUS ROG Maximus Z690 Hero, we did all of our testing with Intel’s Memory Gear at 1: 2 ratio. We tested the 1: 1 and 1: 4 ratios but with little success. When enabling XMP on the G.Skill kit, it automatically sets the ratio 1: 2, with the memory controller running at half the speed of the memory kit.

Problems in Windows 10: Priority and main schedule

As we pointed out in our review of the Intel Core i9-12900K processor, in some software environments there may be unexpected performance behavior. When a thread starts, the operating system (Windows 10) assigns a task to a specific kernel. As the P-Cores (performance) and E-Cores (efficiency) on the Alder Lake hybrid design have different performance and efficiencies, it is up to the planner to ensure that the right job is on the right kernel. Intel’s intended use case is that debugging software takes priority and everything else is moved to background tasks. However, in Windows 10 there is an additional caveat: Any software set to a lower priority than normal (or lower) will also be considered a background and placed on electronic cores, even if it is put in focus. point. Some high performance software sets itself as below normal priority in order to keep the system running it, so there is an ideological conflict between the two.

Various solutions to this exist. Intel told us that users can either run dual monitors or change the Windows power plan to high performance. To investigate the issue during testing, all of our testing in this article was done with the Windows power plan set to High performance (as I do for motherboard reviews) and running the tests with the Windows power plan. ‘high performance active power supply.

On top of that, I also used a third-party planner, Project Lasso software, to check for variations in performance. I can safely and confidently say that there was a margin of variance of about 0.5% between using the high performance power plan and setting affinities and priorities using the software. Project Lasso.

It should also be noted that users running Windows 11 should not experience any of these issues. When properly set, we didn’t see any difference between Windows 10 and Windows 11 in our original review of the Core i9-12900K, and so to keep things consistent with our previous tests for now, we’d like to stick to Windows 10 with our patch applied.

Test bench, configuration and hardware

As this article focuses on the evolution of DDR5 memory, we used a high-end Z690 motherboard, the ASUS ROG Maximus Z690 Hero, and a high-end ASUS ROG Ryujin II 360mm AIO CPU cooler. In terms of settings, we left the Intel Core i9-12900K at the default variables depending on the firmware, with only changes made to the memory settings.

Configuring DDR5 Memory Scaling Test (Alder Lake)
Processor Intel Core i9-12900K, 125W, $ 589
8 + 8 cores, 24 threads 3.2 GHz (5.2 GHz P-Core Turbo)
Motherboard ASUS ROG Maximus Z690 Hero (BIOS 0803)
Cooling ASUS ROG Ryujin II 360 360mm AIO
Power source Corsair HX850 80Plus Platinum 850 W
Memory G.Skill Trident Z5 2 x 16 GB
DDR5-6000 CL 36-36-36-76 (XMP)
Video card MSI GTX 1080 (1178/1279 Boost)
Hard disk Crucial MX300 1TB
Case Open Benchtable BC1.1 (Silver)
Operating system Windows 10 Professional 64-bit: Build 21H2

For the operating system, we used the most widely available and recent version of Windows 10 64-bit (21H2) with all current updates at the time of testing. (For those wondering about our GPU selection, the truth is that all of our publishers are in different locations around the world and we don’t have a single pool of resources. This is Gavin’s regular test GPU. until we can get a replacement; which in this current climate is unlikely. – Ian)

DDR5 memory frequencies / latencies tested
Memory Frequency / times Memory IC
G.Skill Trident Z5 (2 x 16 GB) DDR5-4800 CL 32-32-32-72
DDR5-4800 CL 36-36-36-76
DDR5-5000 CL 36-36-36-76
DDR5-5200 CL 36-36-36-76
DDR5-5400 CL 36-36-36-76
DDR5-5600 CL 36-36-36-76
DDR5-5800 CL 36-36-36-76
DDR5-6000 CL 36-36-36-76
DDR5-6200 CL 36-36-36-76
DDR5-6400 CL 36-36-36-76
Samsung

Above are all the frequencies and latencies that we tested in this article. For scaling, we selected the G.Skill Trident Z5 memory kit because it had the best overclocking capability of any DDR5 kit we received at launch. Right out of the box, it was rated the highest for frequency, and we’ve taken it one step further. G.Skill Trident Z5 memory has been tested from DDR5-4800 CL36 up to and including DDR5-6400 CL36, but also a special case of DDR5-4800 CL32 for lower CAS latencies. Details on our overclocking exploits are later in the review.

Read on for more information on G.Skill’s Trident Z5 DDR5-6000, as well as our review of DDR5 memory scalability on Intel’s Alder Lake. In this article, we cover the following:

  • 1. Test overview and configuration (this page)
  • 2. A closer look at the G.Skill Trident Z5 DDR5-6000 CL36
  • 3. Processor performance
  • 4. Game performance
  • 5. Conclusion


Source link

Steven L. Nielsen