# Microsoft Stepping into Spark with New Box

**URL:** <https://forums.developer.nvidia.com/t/microsoft-stepping-into-spark-with-new-box/372033>\
**Category:** DGX Spark / GB10\
**Created:** [June 2, 2026, 5:13pm UTC](https://forums.developer.nvidia.com/t/microsoft-stepping-into-spark-with-new-box/372033 "2026-06-02T17:13:14Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![robert287](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@robert287](https://forums.developer.nvidia.com/u/robert287)\
**Post date:** [June 2, 2026, 5:13pm UTC](https://forums.developer.nvidia.com/t/microsoft-stepping-into-spark-with-new-box/372033/1 "2026-06-02T17:13:14Z")

</div>

[https://venturebeat.com/infrastructure/microsoft-debuts-surface-rtx-spark-dev-box-to-run-large-ai-models-without-cloud-costs](https://venturebeat.com/infrastructure/microsoft-debuts-surface-rtx-spark-dev-box-to-run-large-ai-models-without-cloud-costs)

---

<div class="post-metadata">

**Author:** ![eggman](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@eggman](https://forums.developer.nvidia.com/u/eggman)\
**Post date:** [June 2, 2026, 5:27pm UTC](https://forums.developer.nvidia.com/t/microsoft-stepping-into-spark-with-new-box/372033/2 "2026-06-02T17:27:17Z")

</div>

Artificial segmentation? That’s the only thing I can think of. They’ve literally doubled/quadruped the surface area for 1/2 the power? Somehow also with less ports.

Either that or somehow without the Mellanox port they’ve reduced idle further from 25-35W? But that seems unlikely imo.

Or maybe it’s the other way around, the Spark guzzles power for minimal performance benefit (like 600W RTX PRO 6000 Workstations vs Max-Q) - I guess one possibility is power limiting Spark to ~65-70W which is the average I’ve seen for LLM usage.