G.Skill announced its Trident Z5 Royal NeoX DDR5 series on September 9, 2026, completing G.Skill's rollout of AMD EXPO Ultra Low Latency support across every tier of its AM5 memory lineup — from the no-frills Flare X5X to the polished-mirror heatspreaders of the flagship Royal family. The new kits run at DDR5-6000 with EXPO ULL support and are offered in four timing configurations, with CL26 as the tightest — making the Royal NeoX the first G.Skill Royal memory to offer primary latency this aggressive inside an EXPO ULL profile. That combination raises a question any buyer choosing between the CL26 and CL36 configurations needs to answer: does tighter primary CAS latency still add meaningfully on top of the sub-timing gains EXPO ULL already delivers?

The short answer is yes, but only if your CPU is the right one — and the reason comes down to a specific architectural feature of AMD's AM5 memory subsystem that shapes every latency optimization on the platform.

What EXPO ULL Already Delivers (and What It Leaves on the Table)

AMD's EXPO Ultra Low Latency — part of the EXPO 1.2 specification introduced at Computex 2026 in Taipei — extends the standard EXPO profile by adding four sub-timing parameters to the data stored on the module's Serial Presence Detect chip: tREFI (the DRAM refresh interval), tRRDS (row-to-row delay), tWR (write recovery time), and a VDDP voltage lock. Of the four, tREFI produces the largest real-world effect. In standard DDR5 operation, the memory controller periodically suspends all CPU requests to force-refresh the capacitors inside each DRAM cell, which leak charge continuously. The JEDEC DDR5 baseline sets that refresh interval at roughly 7,800 nanoseconds. The standard EXPO profile runs it at approximately 3,892 nanoseconds. An EXPO ULL profile extends it to approximately 15,569 nanoseconds — about four times as long — meaning the controller spends far more of its time serving CPU requests and far less in mandatory housekeeping.

Independent testing by TechSpot, conducted in July 2026 across five games on both the Ryzen 7 9700X and Ryzen 7 9800X3D at 1080p, confirmed the practical result: non-X3D Ryzen users see 7 to 15 percent higher average frame rates with EXPO ULL compared to a standard EXPO kit at the same DDR5-6000 speed. AMD's internal benchmark data, covering 30 games on a Ryzen 7 9700X, claims approximately 4 percent higher average frame rates versus standard EXPO DDR5-6000 and 13 percent versus JEDEC-default DDR5-5600, with a 15 percent improvement in 1 percent low frame rates.

What the tREFI extension does not touch is primary CAS latency — the number of clock cycles the controller waits from the moment it requests a specific memory address to the moment the first data word is ready. That is where CL26 enters.

What CL26 Adds on Top

CAS latency translates directly into nanoseconds using a simple formula: (CL ÷ memory speed in MHz) × 2,000. At DDR5-6000, a CL36 configuration produces 12.0 nanoseconds of first-word latency. A CL26 configuration at the same speed produces 8.67 nanoseconds. The gap is 3.33 nanoseconds — the amount of time the memory controller spends waiting for the first byte of data that CL26 removes from every cache miss, regardless of what tREFI is doing.

Because tREFI optimization and primary CAS latency address different bottlenecks in the memory pipeline, they compound rather than cancel out. tREFI reduces how often the controller is locked out of serving requests during refresh cycles. CL reduction cuts how long the controller waits once it has started serving a request. A buyer choosing between CL26 and CL36 Royal NeoX kits — both carrying identical EXPO ULL sub-timing profiles — is choosing how much additional first-word latency reduction to layer on top of the refresh-interval gains.

Community testing data, encompassing multiple CAS latency configurations at DDR5-6000 across CPU-bound gaming workloads, places the practical benefit of CL26 over CL36 at approximately 1 to 4 percent FPS gain when the processor is the bottleneck. The upper end of that range appears in minimum frame rate metrics rather than averages — the same pattern as EXPO ULL's own gains. At 4K, or in any scenario where the GPU is the frame-rate ceiling, neither the sub-timing improvement nor the CAS latency difference produces a measurable result.

Why DDR5-6000 Is the Only Speed Where These Gains Fully Stack

Both EXPO ULL sub-timing gains and primary CAS latency reductions deliver their maximum benefit specifically at DDR5-6000 — and the reason is architectural. AMD's Zen 4 (Ryzen 7000) and Zen 5 (Ryzen 9000) processors contain three coupled clock domains: MCLK (the DRAM's actual operating frequency, 3,000 MHz at DDR5-6000), UCLK (the memory controller clock, which runs at either 1:1 or 1:2 relative to MCLK), and FCLK (the Infinity Fabric clock, which connects CPU cores to the memory subsystem, auto-set to approximately 2,000 MHz), as detailed in AMD Zen 5 UCLK architecture analysis.

At DDR5-6000, the memory controller (UCLK) runs at exactly 3,000 MHz — a 1:1 ratio with the DRAM (MCLK). Every command from the CPU crosses the DDR5-6000 UCLK sweet spot without a clock-domain translation penalty. At DDR5-6400 or higher, AMD's platform defaults to a 1:2 ratio — UCLK runs at half the DRAM speed — which introduces a translation latency that partially offsets the bandwidth gains from the higher clock. This is why memory tuning guides for AM5 consistently recommend leaving FCLK on Auto and running UCLK at Zen 5 DDR5-6000 sweet spot — and why DDR5-6000 is the highest speed at which that configuration is achievable without expert silicon selection.

EXPO ULL's tREFI optimization assumes the memory controller is operating efficiently. At DDR5-6000 with the 1:1 UCLK ratio, the tREFI extended window means the controller spends more of an already-efficient interface on real CPU requests. At DDR5-6400 or DDR5-7200 running in 1:2 mode, the tREFI gain is partially absorbed by the 1:2 translation overhead. DDR5-6000 is not merely a price-performance sweet spot — it is the specific speed at which AMD's memory subsystem is architecturally aligned to extract the most from both primary and sub-timing latency optimizations simultaneously.

Which CPU Determines Whether CL26 Is Worth the Premium

The X3D caveat documented in prior EXPO ULL coverage applies equally here. AMD's Ryzen X3D processors — including the Ryzen 7 9800X3D and the recently launched Ryzen 7 7700X3D — stack up to 96 megabytes of additional L3 cache on top of the standard Zen 5 die, as covered in the Ryzen 7 7700X3D cache review. That cache absorbs most of the memory requests the CPU would otherwise send to DRAM during typical gaming workloads. In TechSpot's independent July 2026 testing, EXPO ULL delivered near-zero gains on the Ryzen 7 9800X3D in the majority of tested titles — 1 percent in Cyberpunk 2077 and 2 percent in Marvel Rivals. If tREFI optimization cannot overcome the X3D cache's filtering effect, the additional 3.33 nanoseconds removed by upgrading from CL36 to CL26 will make even less difference.

For builders using standard Zen 5 processors — the Ryzen 7 9700X, Ryzen 9 9900X, or similar non-X3D chips — the compounding effect is real. EXPO ULL's 7 to 15 percent gains on these processors reflect a memory bottleneck the cache cannot fully absorb. Adding CL26 on top of those sub-timing gains targets the same bottleneck from a different direction: both the frequency at which the controller is locked out and the duration of each access are reduced simultaneously. Whether 1 to 4 percent additional improvement justifies a price premium over the CL36 Royal NeoX configuration depends entirely on what that premium turns out to be.

Pricing Remains the Open Question

G.Skill has not disclosed pricing for any Royal NeoX configuration. The G.Skill Royal NeoX announcement confirms worldwide phased availability through local distributors and retail partners, with no specific figures attached. The earlier Trident Z5 Royal Neo — the standard-EXPO predecessor to the Royal NeoX — had retail listings for the 32GB (2×16GB) CL26 configuration at prices starting above $699 at some retailers, with prior Royal Neo pricing history suggesting higher-end options above $899. The Royal NeoX's additional EXPO ULL certification and tighter binning will likely command a premium above that baseline, though no specific figure has been announced.

For context, the Flare X5X — the budget-tier EXPO ULL kit announced August 31, 2026 — is estimated by Tom's Hardware at approximately $509.99 for the CL36 configuration, placing it at roughly a 6 percent premium above a comparable standard EXPO kit at normalized July 2026 pricing. The Royal NeoX's premium aesthetics — the mirrored silver or gold aluminum heatspreaders and crystalline RGB light bar — have historically commanded an additional $100 to $200 over G.Skill's non-Royal tiers. A realistic estimate for the Royal NeoX CL36 configuration would be in the $600 to $700 range, with CL26 above that, though this is extrapolation until G.Skill publishes actual figures.

The broader DDR5 market context makes this decision more urgent than it appears. AI data center demand has diverted significant DRAM production capacity toward high-bandwidth and server memory, constraining consumer DDR5 supply throughout 2026. IDC projected consumer DRAM supply growth at 16 percent year-over-year for 2026 — below historical norms — and analysts across multiple research firms do not forecast meaningful price relief before mid-2027. TechSpot's July 2026 review of EXPO ULL memory opened with an explicit caution against DDR5 purchases at current prices unless necessary. That advice applies equally to the Royal NeoX: if your build can wait, the value equation improves significantly once pricing normalizes.

Specifications and Requirements

All Royal NeoX configurations ship in a single capacity: 32GB as a dual-channel 2×16GB kit. G.Skill has not announced higher-capacity options. The four timing profiles are:

The notably tighter tRCD and tRP values in the CL26, CL28, and CL30 configurations — 32 in place of the CL36's 76 for the fourth timing — reflect the additional binning required to validate modules at these latency levels across operating temperature. These specifications are confirmed in the Royal NeoX specification table.

Enabling EXPO ULL requires three conditions: a compatible AMD processor (Ryzen 7000 or 9000 on socket AM5), a 600-series or 800-series motherboard whose memory QVL includes EXPO ULL support, and a BIOS incorporating AGESA 1.3.0.1b or later. ASUS, MSI, Gigabyte, and ASRock have released AGESA 1.3.0.1b builds for both 800-series (X870, B850) and most 600-series (X670, B650) AM5 boards, though coverage varies by specific model. Buyers should confirm their board has received the update before purchasing.

The Royal NeoX supports ASUS Aura Sync, MSI Mystic Light, Gigabyte RGB Fusion, and comparable motherboard RGB control software for lighting customization.

Is It Worth It? A Buying Framework

The Royal NeoX is the right choice if all three conditions are true: you are building or upgrading an AM5 system around a standard non-X3D Ryzen processor, you are gaming primarily at 1080p under CPU-bound conditions, and you are comfortable paying the Royal premium for the showcase aesthetics of the mirrored heatspreader and RGB light bar.

The compounding benefit from pairing CL26 with EXPO ULL is real and directionally confirmed by the latency mathematics: EXPO ULL targets refresh-interval latency, CL26 targets first-access latency, and both operate on the same memory pipeline at DDR5-6000's uniquely efficient 1:1 controller clock ratio. For a non-X3D Ryzen builder who would buy an EXPO ULL kit regardless, the incremental cost of CL26 over CL36 — whatever G.Skill ultimately discloses — buys an additional 3.33 nanoseconds of first-word latency reduction on top of the sub-timing gains already delivered. Whether that is worth the premium depends on the actual price gap between tiers once retail listings appear.

If your CPU is an X3D model, save the money. If you are gaming at 4K, save the money. If your build timeline is flexible, waiting for announced pricing avoids committing to an unknown premium in an already-expensive DDR5 market.

Frequently Asked Questions

Does CL26 add any benefit on top of EXPO ULL, or does the sub-timing optimization make primary CAS latency irrelevant?

The two optimizations address different stages of the memory access pipeline and compound rather than cancel out. EXPO ULL's most significant parameter — tREFI — extends the interval between mandatory DRAM refresh cycles, reducing how often the memory controller is locked out of serving CPU requests. CAS latency reduction, by contrast, shortens how long the controller waits once it has begun a read access. At DDR5-6000, upgrading from CL36 to CL26 removes 3.33 nanoseconds of first-word latency from every cache miss — a gain that sits on top of EXPO ULL's existing tREFI improvement. Community benchmark data places the practical additional benefit at roughly 1 to 4 percent additional FPS in CPU-bound gaming workloads, with the more consistent wins appearing in minimum frame rate metrics.

Is DDR5-6000 still the recommended speed for Ryzen 9000 processors, or has a faster option become more competitive?

DDR5-6000 remains the architectural sweet spot for Zen 4 and Zen 5 processors specifically because it is the highest DRAM speed at which the AMD memory controller (UCLK) runs in a 1:1 ratio with the DRAM clock (MCLK) without switching to a performance-penalizing 1:2 half-rate mode, as confirmed by Zen 5 DDR5-6000 sweet spot analysis. Beyond DDR5-6000, the controller defaults to operating at half the DRAM frequency, introducing a translation overhead that partially negates bandwidth gains. Multiple independent scaling tests confirm that performance returns diminish sharply past DDR5-6000 for most Ryzen workloads. The Ryzen 9000 memory controller can technically sustain DDR5-6400 at 1:1 on some samples, but DDR5-6000 remains the guaranteed sweet spot accessible to all AM5 builders.

What does the Royal NeoX cost, and how does it compare to the non-Royal EXPO ULL kits?

G.Skill has not disclosed pricing for the Royal NeoX as of this writing. The preceding Royal Neo (standard EXPO, no ULL) had CL26 listings starting above $699 at some retailers, with higher-end retail configurations above $899, as referenced in Royal NeoX pricing context. For comparison, the Flare X5X — the budget-tier EXPO ULL kit announced August 31, 2026 — is estimated to land near $509.99 for the CL36 32GB kit. The Royal design premium has historically added $100 to $200 over G.Skill's equivalent non-Royal tiers. Buyers should treat any pricing figure as speculative until G.Skill publishes official retail numbers through its distributor network.

Should Ryzen X3D owners consider the Royal NeoX?

In most gaming scenarios, no. Ryzen X3D processors — the 9800X3D, 9700X3D, and related models — use a stacked L3 cache of up to 96 megabytes that absorbs the majority of memory requests during typical gaming workloads, making both EXPO ULL's sub-timing gains and CL26's primary latency improvement largely invisible to the processor. TechSpot's independent testing found near-zero performance differences between standard EXPO and EXPO ULL on the 9800X3D in most titles. Exceptions exist in workloads that overwhelm even the large cache — certain heavily threaded titles at 1080p — but these represent a minority of gaming use cases. An X3D builder's budget is better directed toward GPU, storage, or cooling improvements.