{"product_id":"szsjkj-rtx-2080ti-22gb-gddr6-mining-ai-gpu","title":"SZSJKJ GeForce RTX 2080 Ti 22GB GDDR6 Mining \u0026 AI GPU — Turing 352-bit","description":"\u003cmeta charset=\"UTF-8\"\u003e\n\u003cmeta name=\"viewport\" content=\"width=device-width, initial-scale=1.0\"\u003e\n\u003clink href=\"https:\/\/fonts.googleapis.com\/css2?family=Exo+2:wght@400;600;700;800;900\u0026amp;family=Rajdhani:wght@400;500;600;700\u0026amp;display=swap\" rel=\"stylesheet\"\u003e\n\u003cstyle\u003e\n  *, *::before, *::after { box-sizing: border-box; margin: 0; padding: 0; }\n  :root {\n    --navy-deep: #0A1628; --navy-mid: #0D1F3C; --navy-card: #0F2544;\n    --navy-border: #1E3A5F; --gold: #F5A32A; --gold-dim: #C87D1A;\n    --cyan: #00C8FF; --cyan-dim: #0099CC; --white: #FFFFFF;\n    --silver: #C8D8E8; --muted: #6B8CAE; --text: #D8E8F5;\n  }\n  body { font-family: 'Rajdhani', sans-serif; background: var(--navy-deep); color: var(--text); line-height: 1.6; }\n  .mh-section { padding: 36px; border-bottom: 1px solid var(--navy-border); }\n  .mh-section:last-child { border-bottom: none; }\n  .mh-overview p { font-size: 15px; color: var(--text); line-height: 1.7; margin-bottom: 14px; }\n  .mh-overview p:last-child { margin-bottom: 0; }\n  .mh-highlight { color: var(--gold); font-weight: 600; }\n  .mh-cyan { color: var(--cyan); font-weight: 600; }\n  .mh-spec-table { width: 100%; border-collapse: collapse; font-size: 13.5px; }\n  .mh-spec-table thead tr { background: #071830; border-bottom: 2px solid var(--gold); }\n  .mh-spec-table thead th { padding: 12px 16px; text-align: left; font-size: 10px; font-weight: 700; letter-spacing: 0.16em; text-transform: uppercase; color: var(--muted); }\n  .mh-spec-table tbody tr { border-bottom: 1px solid #132338; transition: background 0.15s; }\n  .mh-spec-table tbody tr:hover { background: #0D2040; }\n  .mh-spec-table td { padding: 12px 16px; vertical-align: top; }\n  .mh-spec-table td:first-child { color: var(--cyan); font-weight: 600; font-size: 12px; letter-spacing: 0.06em; white-space: nowrap; width: 38%; }\n  .mh-spec-table td:last-child { color: var(--text); font-weight: 500; }\n  .mh-spec-group td { padding: 8px 16px 4px; font-size: 10px; font-weight: 700; letter-spacing: 0.14em; text-transform: uppercase; color: var(--gold); background: #071830; border-bottom: 1px solid #132338; }\n  .mh-acl { font-size: 11px; color: var(--muted); line-height: 1.6; border-top: 1px solid var(--navy-border); padding-top: 16px; margin-top: 4px; }\n  @media (max-width: 640px) { .mh-section { padding: 24px 20px; } }\n\u003c\/style\u003e\n\n\n\u003c!-- OVERVIEW --\u003e\n\u003cdiv class=\"mh-section mh-overview\"\u003e\n\u003cp\u003eThe \u003cspan class=\"mh-highlight\"\u003eSZSJKJ GeForce RTX 2080 Ti 22GB\u003c\/span\u003e is a remanufactured NVIDIA Turing card with extended memory. The unit is built around the \u003cspan class=\"mh-cyan\"\u003eTU102 die with 4352 CUDA cores, 544 tensor cores, and 68 RT cores\u003c\/span\u003e, paired with \u003cspan class=\"mh-highlight\"\u003e22 GB of GDDR6 memory\u003c\/span\u003e across the full 352-bit memory bus — double the 11 GB found on stock retail RTX 2080 Ti cards. These extended-memory variants are produced by Chinese rebuilders using salvaged TU102 dies (often from server-pull or refurbished sources) combined with higher-density GDDR6 modules and a custom PCB designed for the doubled capacity.\u003c\/p\u003e\n\u003cp\u003eImportant transparency note: \u003cspan class=\"mh-highlight\"\u003ethis is not a factory NVIDIA product\u003c\/span\u003e. It is a third-party rebuild of the RTX 2080 Ti reference design with modified memory. The card identifies as an RTX 2080 Ti to Windows and uses the standard NVIDIA GeForce driver stack with no modification required. Warranty is provided by the rebuilder\/seller rather than NVIDIA. Performance in gaming and most mining algorithms is comparable to a stock 2080 Ti — the value of the modification is in the \u003cspan class=\"mh-cyan\"\u003e22 GB VRAM headroom for AI\/ML inference, large-DAG mining algorithms, and creator workloads with heavy memory demands\u003c\/span\u003e.\u003c\/p\u003e\n\u003cp\u003eFor GPU mining, approximate hashrates with tuned clocks: \u003cspan class=\"mh-highlight\"\u003eEtchash (Ethereum Classic) ~58–60 MH\/s at ~220 W\u003c\/span\u003e, \u003cspan class=\"mh-highlight\"\u003eAutolykos2 (Ergo) ~190–210 MH\/s at ~180 W\u003c\/span\u003e, \u003cspan class=\"mh-highlight\"\u003eKawPow (Ravencoin) ~32–35 MH\/s at ~230 W\u003c\/span\u003e, \u003cspan class=\"mh-highlight\"\u003eOctopus (Conflux) ~78–85 MH\/s at ~240 W\u003c\/span\u003e. The 22 GB frame buffer provides multi-year headroom for DAG growth on every current GPU-mineable algorithm, and opens up workloads that 11 GB cards cannot handle — for example, hosting larger AI inference models alongside mining when the card is idle.\u003c\/p\u003e\n\u003cp\u003eFor Australian operators, the card draws approximately \u003cspan class=\"mh-highlight\"\u003e250–260 W TDP from 2 × 8-pin PCIe connectors\u003c\/span\u003e (some rebuild revisions use 8+6-pin), with a recommended PSU of 650 W+ for a single-card build. The 2080 Ti is a large dual-slot card — typical dimensions are around 270 × 116 × 38 mm in the reference-style cooler. A licensed electrician should be consulted for circuit planning on multi-card rigs, and installations must comply with AS\/NZS 3000:2018. As a refurbished\/remanufactured product sold in Australia, the card is fully covered by Australian Consumer Law guarantees of acceptable quality.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003c!-- SPECIFICATIONS --\u003e\n\u003cdiv class=\"mh-section\"\u003e\n\u003ctable class=\"mh-spec-table\"\u003e\n\u003cthead\u003e\u003ctr\u003e\n\u003cth\u003eSpecification\u003c\/th\u003e\n\u003cth\u003eValue\u003c\/th\u003e\n\u003c\/tr\u003e\u003c\/thead\u003e\n\u003ctbody\u003e\n\u003ctr class=\"mh-spec-group\"\u003e\u003ctd colspan=\"2\"\u003eGPU\u003c\/td\u003e\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eGPU Model\u003c\/td\u003e\n\u003ctd\u003eNVIDIA GeForce RTX 2080 Ti (extended-memory rebuild)\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eGPU Chip\u003c\/td\u003e\n\u003ctd\u003eTU102\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eArchitecture\u003c\/td\u003e\n\u003ctd\u003eTuring\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eProcess Node\u003c\/td\u003e\n\u003ctd\u003eTSMC 12 nm FinFET\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eCUDA Cores\u003c\/td\u003e\n\u003ctd\u003e4352\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eRT Cores\u003c\/td\u003e\n\u003ctd\u003e68 (1st generation)\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eTensor Cores\u003c\/td\u003e\n\u003ctd\u003e544\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eBase \/ Boost Clock\u003c\/td\u003e\n\u003ctd\u003e~1350 MHz \/ ~1545 MHz (Founders Edition reference; rebuild PCB matches)\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr class=\"mh-spec-group\"\u003e\u003ctd colspan=\"2\"\u003eMEMORY (MODIFIED FROM STOCK)\u003c\/td\u003e\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eMemory Capacity\u003c\/td\u003e\n\u003ctd\u003e22 GB GDDR6 (modded — stock is 11 GB)\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eMemory Bus\u003c\/td\u003e\n\u003ctd\u003e352-bit\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eMemory Speed\u003c\/td\u003e\n\u003ctd\u003e14 Gbps effective\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eMemory Bandwidth\u003c\/td\u003e\n\u003ctd\u003e616 GB\/s\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eMemory Modification\u003c\/td\u003e\n\u003ctd\u003eHigher-density GDDR6 modules on custom PCB by third-party rebuilder\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr class=\"mh-spec-group\"\u003e\u003ctd colspan=\"2\"\u003eMINING (APPROX., TUNED)\u003c\/td\u003e\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eEtchash (ETC)\u003c\/td\u003e\n\u003ctd\u003e~58–60 MH\/s @ ~220 W\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eAutolykos2 (ERG)\u003c\/td\u003e\n\u003ctd\u003e~190–210 MH\/s @ ~180 W\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eKawPow (RVN)\u003c\/td\u003e\n\u003ctd\u003e~32–35 MH\/s @ ~230 W\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eOctopus (CFX)\u003c\/td\u003e\n\u003ctd\u003e~78–85 MH\/s @ ~240 W\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr class=\"mh-spec-group\"\u003e\u003ctd colspan=\"2\"\u003eAI \/ COMPUTE\u003c\/td\u003e\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eUse Case\u003c\/td\u003e\n\u003ctd\u003eLocal LLM inference, Stable Diffusion, CUDA compute workloads\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eVRAM Advantage\u003c\/td\u003e\n\u003ctd\u003e22 GB supports models that cannot fit on 11–12 GB cards\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eCUDA Compute Capability\u003c\/td\u003e\n\u003ctd\u003e7.5 (Turing)\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr class=\"mh-spec-group\"\u003e\u003ctd colspan=\"2\"\u003ePOWER\u003c\/td\u003e\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eTDP\u003c\/td\u003e\n\u003ctd\u003e~250–260 W\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003ePower Connectors\u003c\/td\u003e\n\u003ctd\u003e2 × 8-pin PCIe (some rebuilds: 8+6-pin)\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eRecommended PSU\u003c\/td\u003e\n\u003ctd\u003e650 W+ ATX (single card)\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr class=\"mh-spec-group\"\u003e\u003ctd colspan=\"2\"\u003eDISPLAY OUTPUTS\u003c\/td\u003e\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eHDMI\u003c\/td\u003e\n\u003ctd\u003e1 × HDMI 2.0b\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eDisplayPort\u003c\/td\u003e\n\u003ctd\u003e3 × DisplayPort 1.4a\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eUSB-C\u003c\/td\u003e\n\u003ctd\u003e1 × USB-C (VirtualLink, FE reference layout)\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr class=\"mh-spec-group\"\u003e\u003ctd colspan=\"2\"\u003eINTERFACE \u0026amp; COOLING\u003c\/td\u003e\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eBus Interface\u003c\/td\u003e\n\u003ctd\u003ePCIe 3.0 x16\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eCooler\u003c\/td\u003e\n\u003ctd\u003eDual axial fan, vapour chamber heatsink (varies by rebuild)\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eSlot Width\u003c\/td\u003e\n\u003ctd\u003e2-slot\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr class=\"mh-spec-group\"\u003e\u003ctd colspan=\"2\"\u003ePHYSICAL\u003c\/td\u003e\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eForm Factor\u003c\/td\u003e\n\u003ctd\u003eATX, full-length dual-slot\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eApprox. Dimensions\u003c\/td\u003e\n\u003ctd\u003e~270 × 116 × 38 mm\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eApprox. Weight\u003c\/td\u003e\n\u003ctd\u003e~1.3 kg\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr class=\"mh-spec-group\"\u003e\u003ctd colspan=\"2\"\u003eSOFTWARE SUPPORT\u003c\/td\u003e\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eAPIs\u003c\/td\u003e\n\u003ctd\u003eDirectX 12 Ultimate, Vulkan, OpenGL 4.6, CUDA 7.5, OptiX\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eDriver\u003c\/td\u003e\n\u003ctd\u003eNVIDIA GeForce Game Ready \/ Studio (Windows), NVIDIA proprietary (Linux)\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eCard Recognition\u003c\/td\u003e\n\u003ctd\u003eIdentifies as RTX 2080 Ti — no driver modification required\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr class=\"mh-spec-group\"\u003e\u003ctd colspan=\"2\"\u003eIN THE BOX\u003c\/td\u003e\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eIncluded\u003c\/td\u003e\n\u003ctd\u003eSZSJKJ RTX 2080 Ti 22GB graphics card\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eNote\u003c\/td\u003e\n\u003ctd\u003eRefurbished \/ remanufactured product — warranty provided by rebuilder, ACL applies\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003c\/tbody\u003e\n\u003c\/table\u003e\n\u003cp class=\"mh-acl\"\u003e\u003cstrong\u003eAustralian Consumer Law Notice:\u003c\/strong\u003e All products sold by MinerHub come with guarantees that cannot be excluded under the Australian Consumer Law. You are entitled to a replacement or refund for a major failure and compensation for any other reasonably foreseeable loss or damage. You are also entitled to have goods repaired or replaced if they fail to be of acceptable quality. Manufacturer warranty terms apply in addition to your statutory rights. All mining hashrate, power, and gaming performance figures are approximate values dependent on driver version, BIOS, tuning, ambient temperature, and host system. MinerHub does not guarantee mining returns or profitability.\u003c\/p\u003e\n\u003c\/div\u003e","brand":"SZSJKJ","offers":[{"title":"Default Title","offer_id":49177296601340,"sku":"SZSJKJ-RTX2080TI-22G","price":949.99,"currency_code":"AUD","in_stock":true}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0805\/8067\/3788\/files\/RTX2080Ti-22G-Desktop-Computer-Gaming-Graphics-Card.webp?v=1781062647","url":"https:\/\/minerhub.com.au\/products\/szsjkj-rtx-2080ti-22gb-gddr6-mining-ai-gpu","provider":"MinerHub","version":"1.0","type":"link"}