This summary of the video was created by an AI. It might contain some inaccuracies.
00:00:00 – 00:17:06
The video explores DDR5 memory scaling and tuning performance, focusing on compatibility with Intel's Core i9-13900K and other Raptor Lake processors. Collaborating with Buildzoid from the Actually Hardcore Overclocking channel, the presenter tests various DDR5 memory kits, such as G.Skill Trident Z5 and Corsair Dominator, across different speeds like 7200MHz and 6000MHz. Testing reveals that higher DDR5 speeds, while providing some performance boosts, offer modest gains and can lead to stability issues.
In games like Marvel's Spider-Man Remastered, Callisto Protocol, A Plague Tale: Requiem, and others, tuned configurations show slight performance improvements, but results vary, with some games showing minimal benefits. The DDR5-5600 CL28 kit is identified as the optimal balance of performance and cost for Raptor Lake processors. The speaker advises against sacrificing system stability for minor gains and notes potential variability depending on motherboard brands like MSI, Asus, or ASRock. The video concludes with a cautionary tale about the pitfalls of overclocking and emphasizes that minor performance differences do not justify the risk of instability.
00:00:00
In this part of the video, the focus is on examining DDR5 memory scaling and tuning performance with Intel’s Core i9-13900K desktop processor. The goal is to determine the optimal memory configuration for Raptor Lake processors and whether memory tuning is worthwhile. The segment acknowledges the collaboration with Buildzoid from the Actually Hardcore Overclocking YouTube channel for providing tuned timings. The presenter will evaluate the i9-13900K using various DDR5 memory speeds and plans to share comprehensive timing details. This data is intended to guide users on memory performance and tuning for 13th Gen Intel processors, specifically excluding comparisons with AMD processors.
00:03:00
In this part of the video, it is explained that not all CPUs will handle 7200 MHz with specific timings, as buildzoid and G.Skill note. For weaker CPUs, adjusting the frequency might be necessary. Testing aimed to provide broadly achievable data, similar to the approach for AMD’s Zen 4 processors. Different memory kits were tested, including G.Skill Trident Z5 RGB DDR5 7200, 6400, 6000, and Corsair Dominator DDR5 kits at various speeds. Two memory tuning configurations are mentioned, with the fastest using DDR5 7200 memory at 1.4 volts. The video also notes the challenges of pushing to 8000 MHz, requiring high-quality CPU and overclocking-focused motherboards, which may not be practical for most users.
00:06:00
In this segment of the video, the focus is on configuring DDR5 memory for Raptor Lake CPUs using Buildzoid’s tuned settings. The recommended modules are Hynix-based with ACL timings of 30 to 32. The video details testing on a Gigabyte Z790 RS Elite AX motherboard with an i9-1300k and GeForce RTX 4090. Initially, Marvel’s Spider-Man Remastered shows a five percent performance increase using tuned DDR5 7200 memory over 6400 stock. Further tuning yields a nine percent boost for DDR5 6000 CL30 memory. Comparatively, dual-rank DDR5 5600 CL28 achieves an 11 percent improvement over 7200 XMP, suggesting limited performance gain for higher-cost memory. In Callisto Protocol, memory tuning results show minimal performance increase, indicating negligible differences in this less memory-sensitive title.
00:09:00
In this part of the video, the speaker examines the performance differences between various DDR5 memory configurations in different games. They note that in “A Plague Tale: Requiem,” a tuned 7200 configuration is 7% faster than using XMP, while dual rank 5600 memory outperforms single rank versions by 5%. Despite modest gains, some peculiar results show little variance among single rank XMP configurations. In “Horizon Zero Dawn” and “The Riftbreaker,” tuned profiles offer minor benefits, with 7200 memory showing only slight performance boosts. “Shadow of the Tomb Raider” reflects a 7% improvement with tuned 7200 memory, showing better gains compared to other games. Finally, in “Watchdogs Legion,” the tuned 7200 configuration achieves a 9% performance increase, marking the highest gain observed. The analysis suggests that while faster memory can provide boosts, the differences are often minor.
00:12:00
In this part of the video, the speaker discusses tuning DDR5 memory kits to achieve various performance boosts, highlighting that a DDR5-6000 CL30 kit resulted in roughly a 9% performance increase, reaching around 200 FPS on average. The speaker provides an average frame rate performance comparison for a small seven-game sample using an RTX 4090 at 1080p, showing a 5% boost from DDR5-7200 and a 6% boost from DDR5-6000. Despite these gains, the speaker questions their worth due to potential stability risks in gaming.
The speaker identifies the DDR5-5600 CL28 kit as the “sweet spot” memory for Raptor Lake processors, offering significant performance at a lower price. However, they note that high-end users might prefer the more expensive DDR5-7200 for potentially better returns.
The speaker criticizes the hype around benchmarking with tuned memory timings, finding the actual performance gains modest. They acknowledge more extreme memory configurations could yield better results but depend on specific hardware and favorable conditions.
Additionally, the speaker mentions differences in performance between motherboards from different vendors, suggesting further testing is needed. They hint that using boards from MSI, Asus, or ASRock might yield more substantial gains from tuning memory, possibly even double those shown.
Finally, the speaker suggests that enthusiasts seeking top performance and willing to invest heavily will find the 1300k processor effective, though they express interest in comparing it to the 7950x 3D for a more comprehensive analysis.
00:15:00
In this part of the video, the speaker concludes by emphasizing that while flagship processors perform similarly straight out of the box, gamers should avoid compromising stability for a marginal performance gain. The speaker shares a personal anecdote about overclocking their Intel Core i9-12900KS rig which initially seemed stable but eventually led to crashes during crucial gaming moments, prompting a return to default settings. The speaker notes that the difference in performance was negligible, and they have since experienced no issues. The video wraps up with acknowledgments, including a shout-out to Buildzoid of Actually Hardcore Overclocking, and mentions additional content and exclusive perks available for Patreon supporters.