If you’ve ever copied a huge folder to an SSD and watched the transfer speed start blazing fast, then suddenly drop off a cliff halfway through, you’ve met SLC caching in the wild. It’s one of the most misunderstood tricks in modern storage, and it explains why your drive’s advertised speed and its real-world speed sometimes disagree by a wide margin. The short version: your SSD is faking a faster type of memory for as long as it can. Here’s what that actually means, and why it matters when you shop for a drive. If you’re mid-research, our picks for the best NVMe SSDs for gaming and the best budget SSDs lean heavily on how well each drive handles this.
The short answer
SLC caching is when an SSD treats a portion of its slower, denser flash memory as if it were fast single-level-cell memory. Most consumer drives store three or four bits per memory cell (that’s TLC or QLC), which is cheap and roomy but slow to write. SLC mode stores just one bit per cell, which is much faster. So the controller carves out a chunk of the drive, runs it in fast single-bit mode as a write buffer, and dumps incoming data there first. You get SLC-class speed on writes without paying for a drive that’s all SLC. Once the cache fills, though, writes fall back to the drive’s native speed. That’s the cliff.
This isn’t a niche feature. It’s in nearly every SSD you can buy, from a $40 SATA drive to the 1TB PCIe Gen4 NVMe in a $3,970.99 ASUS Ascent GX10 AI machine. The difference between good and bad drives is largely how gracefully they manage this cache under sustained load.
Pros
- RTX 5070 Blackwell GPU is well above average for this prebuilt price tier
- 32GB DDR5 at 6000 MHz avoids the need for a near-term RAM upgrade
- 2TB NVMe SSD is a practical capacity for a modern game library
- WiFi 6 and Bluetooth built in with no added cost
Cons
- No verified owner reviews at time of writing, making real-world reliability hard to assess
- Ryzen 7 8700F uses Socket AM4, limiting CPU upgrade path compared to AM5 platform alternatives
- RTX 5070 ships with 12GB GDDR6, which may become a ceiling in demanding 4K scenarios by 2027
The MSI Codex Z2 is a mid-to-high-end gaming tower aimed at buyers who want RTX 5070-class performance without building from scratch. Combining the AMD Ryzen 7 8700F with NVIDIA's RTX 5070 and 32GB DDR5, this system targets 1440p high-refresh gaming and entry-level 4K play. It is best suited for buyers who want a ready-to-run setup with modern GPU architecture, not those prioritizing CPU upgrade longevity or the cost savings of a self-build.
The RTX 5070 is the headline component here. Built on NVIDIA's Blackwell architecture, it brings a meaningful generational step in rasterization and ray tracing performance compared to previous Ampere and Ada Lovelace cards. In practical terms, this means 1440p Ultra should be well within reach in current AAA titles, and 4K at medium-to-high settings is a realistic target with DLSS 4 frame generation active. Paired with the 8-core Ryzen 7 8700F boosting to 5.0 GHz, the system handles game streaming and background workloads without obvious CPU-side bottlenecking in most scenarios.
MSI has put some effort into the thermal design. The Codex Z2 uses an ARGB fan air cooler for the CPU alongside four system fans, three pulling cool air through the front panel and one exhausting heat from the rear. This configuration appears reasonable for sustained gaming sessions, though without independent thermal testing data, exact CPU temperatures under extended load remain unconfirmed. The compact tower footprint at 16 x 8.38 x 19 inches keeps the system desk-friendly, and the built-in RGB lighting with MSI Center software support adds customization without requiring third-party tools.
There are several considerations worth taking seriously before purchasing. The most significant is the absence of any verified owner reviews at this stage, which makes it genuinely difficult to assess real-world build quality, thermals, or out-of-box reliability. Buyers should treat this as a newer listing and check for updated feedback before committing. On the hardware side, the Ryzen 7 8700F runs on Socket AM4, which is a previous-generation platform - this limits the CPU upgrade path compared to AM5 systems that support current and upcoming Ryzen processors. Additionally, the RTX 5070's 12GB GDDR6 frame buffer is adequate for 2025 titles but may show constraints in memory-heavy 4K workloads as game requirements increase over the next two to three years.
Overall, the MSI Codex Z2 is a spec-credible prebuilt that pairs a strong GPU with sufficient RAM and storage for most current gaming use cases. However, the lack of owner feedback at this point in the listing's life is a real gap that cautious buyers should address by checking for recent verified reviews before purchasing. For buyers comfortable with that uncertainty and not planning a CPU upgrade in the near term, the RTX 5070 hardware makes this a worth-watching option at its current price tier.
The longer explanation
To get why this works, you need to picture how flash stores data. A NAND cell holds electrons, and the drive reads the voltage level to figure out what bits are stored. Single-level cell keeps it binary: charged or not, one bit. That’s easy to read and quick to write. TLC crams eight distinct voltage levels into one cell to encode three bits, and QLC pushes that to sixteen levels for four bits. More levels mean more precision required, which means slower, more careful writes and more error correction.
SLC caching sidesteps the slowdown by temporarily running TLC or QLC cells in one-bit mode. A cell that could hold three bits just holds one for now, trading capacity for speed. The controller writes fast into this pseudo-SLC region, tells your computer the write is done, and later, during idle moments, quietly re-folds that data back into dense TLC storage. You never see the second step. From your side, the drive just felt fast. That background migration is why leaving a PC on but idle for a few minutes often restores full burst speed for the next big copy.
How it works
There are two flavors of cache, and drives usually blend them. A static cache is a fixed slice of the NAND permanently reserved for SLC mode. It’s small but always available. A dynamic cache borrows empty space from the rest of the drive and runs it in SLC mode on demand, which is why an empty 2TB drive can absorb a much larger burst than the same drive at 90 percent full. The msi Codex Z2 desktop, for instance, ships a 2TB M.2 NVMe drive, and that headroom means its dynamic cache stays generous even after you’ve installed a pile of games.
The catch is arithmetic. Writing one bit where three could live means your usable cache is roughly a third of the free space it’s borrowing. Fill a 100GB dynamic region in SLC mode and you’ve spent about 300GB of native TLC capacity to do it. That’s why cache size shrinks as a drive fills up, and why a nearly full budget drive can feel dramatically slower than the same drive fresh out of the box.
Why it works this way
Manufacturers do this because it’s the cheapest way to make a slow, high-capacity drive feel fast for the workloads most people actually run. And most real usage is bursty. You save a document, install a game, copy some photos, then the drive sits idle. Those bursts almost always fit inside the SLC cache, so the drive feels quick nearly all the time while still using cheap, dense NAND underneath. It’s a clever bit of sleight of hand that keeps prices down.
The design only shows its seams under sustained heavy writes, like editing 4K video straight to the drive or moving a 200GB game library in one shot. That’s when you blow past the cache and hit native TLC or QLC speed, which on a QLC drive can crater from 5,000 MB/s to under 100 MB/s. For gaming and everyday work you’ll rarely notice. For a video editor hammering the drive for an hour, it’s the whole story. There’s also a thermal angle worth knowing. Sustained writes heat the controller, and a hot drive throttles on top of the cache running dry, so a cheap drive without a heatsink can feel doubly slow during a long copy. Good airflow and a simple heatspreader keep that second penalty off your back.
When you’d want this
Truthfully, you don’t choose SLC caching, it’s baked into every consumer SSD. What you choose is how much cache headroom you need, and that comes down to workload. If you game, browse, and run office apps, almost any drive’s cache covers you, and you can shop on price and capacity without overthinking it. Owner reports consistently show that light users never touch the cache limit in normal use.
If you move large files constantly, you want a drive with a big cache and, ideally, more free space to feed the dynamic pool. Prebuilt machines lean on this too. A gaming desktop like the msi Codex Z2, priced around $2,067.34 with an RTX 5070 and 2TB of NVMe storage, carries enough capacity that its cache rarely runs dry during a game install. Bigger drives simply give the caching system more room to work, which is one more reason capacity buys you speed indirectly.
Pros
- GB10 Superchip targets 1 petaFLOP AI throughput suitable for on-device LM prototyping
- 128GB unified memory pool avoids CPU-GPU copy overhead common on discrete GPU rigs
- NVLink-C2C fabric gives higher effective bandwidth than PCIe Gen5 x16 for AI workloads
- ConnectX-7 enables two-node stacking, scaling to roughly 400B parameter model fine-tuning
Cons
- Limited independent owner fedback and third-party benchmarks available at time of writing
- 1TB NVMe fills quickly with modern checkpoints, external NVMe or NAS staging likely need
- Closed NVIDIA ARM-based stack, not a general-purpose x86 workstation for gaming or standard DC aps
The ASUS Ascent GX10 is a flagship developer-class AI appliance built on the NVIDIA DGX Spark reference design, pairing the GB10 Grace Blackwell Superchip with128GB LPDDR5x and a 1TB PCIe Gen4 NVMe SSD. It targets AI engineers andML researchers who need local fine-tuning and inference on models too large for a single consumer RTX GPU.
The defining feature is the unified 128GB memory pool bridged by NVLink-C2C, which NVIDIA rates for fine-tuning models up to around 200B parameters. Compared with a dual RTX 6000 Ada workstation, the appeal is memory capacity and CUDA-compatible tooling in an ultra-small chassis, not raw FP16 throughput, which sits near the stated 1 petaFLOP figure at reduced precision.
Trade-offs are typical of this tier and platform. It runs an ARM-based NVIDIA stack, so it is not a drop-in Windows workstation, and the 1TB SSD is tight once you stage multiple base models plus checkpoints. Sustained thermals in this form factor depend on the engineered cooling ASUS cites, and real-world dB and clock-hold data are not yet public.
Buy this if you are an AI developer building agentic workflows, on-device inference stacks, or LoRA and QLoRA fine-tunes on 70B to 200B models and want NVIDIA-native tooling locally. Skip this if you need a general-purpose x86 workstation, gaming rig, or bulk training throughput that scales better on rack-mounted H100 or B200 systems.
Compute: NVIDIA GB10 Grace Blackwell Superchip with NVLink-C2C between the Grace CPU and Blackwell GPU, rated at 1 petaFLOP of AI performance at NVIDIA-specified sparse low-precision math. Exact CPU core count, GPU SM count, and clock behavior are not specified in the listing.
Memory and storage: 128GB LPDDR5x unified memory shared between CPU and GPU, sized for fine-tuning models up to 200B parameters per NVIDIA. Storage is a single 1TB PCIe Gen4 NVMe SSD. Additional M.2 slots, DRAM channel count, and TBW endurance are not specified.
Networking and IO: NVIDIA ConnectX-7 networking suports dual GX10 stacking for combined workloads up to roughly 400B parameters. Wireless is WiFi 7 and Bluetooth 5.4. Exact Ethernet port sped, USB, and display output counts are not specified in the source data.
Chassis and software: Stackable ultra-small form factor with engineered cooling for sustained AI loads. Ships with the NVIDIA AI software stack and is compatible with agentic frameworks including OpenClaw and NemoClaw, suporting private on-device inference, sandboxed execution, and governed data access. PSU wattage, dimensions, and weight are not specified.
What to look for in an SSD
Read the sustained write spec, not just the peak. The big number on the box is the cached burst speed. The number that matters for heavy work is what the drive does after the cache empties, and good manufacturers publish both. Capacity helps too, since a 2TB drive keeps a healthier dynamic cache than a 500GB one at the same fill level. TLC drives recover their cache and hold native speed far better than QLC drives, so if you write a lot, favor TLC. Our roundups of the best 2TB NVMe SSDs and the best PCIe 5.0 SSDs flag sustained performance for exactly this reason.
Common misconceptions
The biggest myth is that SLC caching means your drive contains actual SLC memory. It doesn’t. The controller just runs regular TLC or QLC cells in a faster one-bit mode temporarily. Another common mix-up is blaming DRAM cache for write slowdowns. DRAM is a separate buffer that stores the drive’s mapping tables, not your file data, so a DRAM-less drive and an SLC cache cliff are two different problems. People also assume the cache is fixed in size. On most drives it’s dynamic, so it shrinks as the drive fills, which is why a full drive feels slower than an empty one even though nothing broke.
Frequently asked
Does SLC caching wear out my SSD faster?
Not meaningfully. Writing to the cache and folding data back does add some internal write activity, but controllers account for this in their endurance ratings. Modern drives are rated for hundreds of terabytes written, far more than a typical user reaches in years of normal use.
Why does my SSD slow down during big transfers?
You’ve filled the SLC cache. Once the fast buffer is full, writes fall back to the drive’s native TLC or QLC speed, which is slower. Let the drive idle for a few minutes and the controller re-folds the cached data, restoring burst speed for your next transfer.
Can I turn SLC caching off?
On consumer drives, no. It’s managed by the controller firmware and isn’t user-configurable. Some enterprise drives skip it entirely in favor of consistent sustained speed, but for desktop and gaming use the cache is a net win you can’t and wouldn’t want to disable.
Does a bigger SSD have a bigger cache?
Usually, yes. Since dynamic caching borrows from free space, a larger drive with more empty capacity can run a bigger SLC region. That’s one reason a 2TB drive often sustains large writes better than a 500GB model, even within the same product line.
Is TLC or QLC better for caching?
TLC handles it better. QLC’s native write speed is much lower, so when the cache runs dry the drop is steeper, sometimes under 100 MB/s. TLC drives fall back to a more livable native speed and recover their cache faster, which is why heavy writers should favor TLC. For price-sensitive builds, check our best SSD deals to weigh the tradeoff.
Write Your Review
No reviews yet. Be the first to share your experience!