{"id":248476,"date":"2025-11-07T02:10:14","date_gmt":"2025-11-07T02:10:14","guid":{"rendered":"https:\/\/www.newsbeep.com\/uk\/248476\/"},"modified":"2025-11-07T02:10:14","modified_gmt":"2025-11-07T02:10:14","slug":"as-expected-nvidias-3999-mini-ai-supercomputer-is-terrible-for-gaming-dgx-spark-struggles-to-hit-50-fps-at-1080p-on-medium-settings-in-cyberpunk-2077","status":"publish","type":"post","link":"https:\/\/www.newsbeep.com\/uk\/248476\/","title":{"rendered":"As expected, Nvidia’s $3,999 mini AI supercomputer is terrible for gaming \u2014 DGX Spark struggles to hit 50 FPS at 1080p on medium settings in Cyberpunk 2077"},"content":{"rendered":"

The DGX Spark<\/a> is an excellent, compact system tailored for artificial intelligence developers and related professionals; however, its gaming performance falls significantly short of the best graphics cards<\/a>. A Reddit<\/a> user tested the DGX Spark’s gaming capabilities, and the results were disappointing despite the device’s graphical specifications being comparable to those of a GeForce RTX 5070<\/a>.<\/p>\n

The DGX Spark, equipped with the GB10 Blackwell Superchip<\/a>, flaunts remarkable specifications that would make anyone drool. Nvidia has fused a 20-core Arm processor with integrated Blackwell graphics and 128GB of LPDDR5X unified memory into the GB10 SoC. The result is a Blackwell iGPU with 6,144 CUDA cores with access to an aggregate bandwidth of approximately 273 GB\/s. It runs with a 1,665 MHz base clock but features a boost clock that escalates to 2,525 MHz.<\/p>\n

<\/p>\n

On paper, the DGX Spark’s gaming performance is somewhat in the same alley as the GeForce RTX 5070, which utilizes the GB205 silicon with 6,144 CUDA cores and offers a memory bandwidth of up to 672 GB\/s. As a desktop discrete graphics card, the GeForce RTX 5070 has a higher base clock (2,325 MHz) than the GB10.<\/p>\n

You may like<\/p>\n

Cyberpunk 2077 runs on DGX Spark<\/a> from r\/nvidia<\/a><\/p>\n

The issue with the GB10 lies in its architecture based on the Arm platform and its operation on the Nvidia DGX OS, a customized Linux distribution derived from Ubuntu 24.04 LTS. Consequently, it requires an additional layer of emulation to run x86 games. Although Linux-based emulation software has advanced significantly, it has yet to unlock the hardware’s performance fully. As of the most recent information, Box64 (utilized in the experiment by the Reddit user) and Box86 are capable of achieving approximately 80% of native performance.<\/p>\n

The DGX Spark achieved a frame rate of 50 FPS while running Cyberpunk 2077<\/a> at a 1080p (1920×1080) resolution, with image fidelity adjusted to medium settings. It is possible that DLSS could have assisted the DGX Spark in attaining 60 FPS or higher; however, the Redditor noted that this option was unexpectedly unavailable. The author remarked that the stability was “actually pretty good,” although occasional game crashes occurred. Nevertheless, the game remained playable.<\/p>\n

It is unlikely that anyone would allocate $3,999 solely for a DGX Spark for gaming purposes. However, AI developers can also be gamers, making gaming not entirely uncommon. Although the DGX Spark could satisfy a casual gaming itch, it is not suitable for serious gaming endeavors. For those seeking a device capable of both artificial intelligence applications and gaming, one might consider AMD’s Strix Halo<\/a>, which competes directly with the DGX Spark<\/a> within this compact form factor.<\/p>\n

Update (November 6):\u00a0Corrected Blackwell iGPU bandwidth.<\/p>\n

Get Tom’s Hardware’s best news and in-depth reviews, straight to your inbox.<\/p>\n

<\/p>\n

\n

\"Google\n<\/p>\n

<\/a><\/p>\n

Follow Tom’s Hardware on Google News<\/a>, or add us as a preferred source<\/a>, to get our latest news, analysis, & reviews in your feeds.<\/p>\n","protected":false},"excerpt":{"rendered":"The DGX Spark is an excellent, compact system tailored for artificial intelligence developers and related professionals; however, its…\n","protected":false},"author":2,"featured_media":248477,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[20],"tags":[554,733,4308,86,56,54,55],"class_list":{"0":"post-248476","1":"post","2":"type-post","3":"status-publish","4":"format-standard","5":"has-post-thumbnail","7":"category-artificial-intelligence","8":"tag-ai","9":"tag-artificial-intelligence","10":"tag-artificialintelligence","11":"tag-technology","12":"tag-uk","13":"tag-united-kingdom","14":"tag-unitedkingdom"},"_links":{"self":[{"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/posts\/248476","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/comments?post=248476"}],"version-history":[{"count":0,"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/posts\/248476\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/media\/248477"}],"wp:attachment":[{"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/media?parent=248476"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/categories?post=248476"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/tags?post=248476"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}