Making QDatDroid feel native on Dragonwing: the work behind Zebra RFID SDK 2.0.5.275

When the RFID reader is the phone

For a decade, mobile RAIN RFID has meant a handheld plus a Bluetooth sled — a separate radio you pair, charge, and carry. That is changing. Qualcomm's Dragonwing Q-6690, launched in September 2025 and a CES 2026 Innovation Award honoree, is the first enterprise mobile processor with a UHF (RAIN) RFID reader built directly into the silicon. At NRF 2026 (9 January 2026) Zebra announced three devices built on it: the TC501 mobile computer, its extra-ruggedized twin the TC701, and the ET401 — the first RFID-enabled enterprise tablet. They are orderable now and reaching mass availability through 2026.

That matters because it removes the sled: one rugged device, no pairing, no second battery, cheaper hardware. But it also means the RFID radio is no longer the Zebra sled the SDK has spoken to for years — it is an on-SoC Qualcomm radio with its own rules. Getting QDatDroid to feel exactly as effortless on a sledless TC701 as it does on an RFD40 sled took real engineering at Meerv Inc. This is the story of that work.

The upgrade: Zebra RFID SDK 2.0.5.275

QDatDroid is built on Zebra's RFID SDK. Supporting Dragonwing meant moving to Zebra RFID SDK 2.0.5.275 (released by Zebra on 22 April 2026) — the first SDK to add the TC501 / TC701 / ET401 and the new RFD4051 sled. The bump from the previous 2.0.5.238 was not a drop-in: we diffed Zebra's own reference app between the two SDK versions and pulled four priority changes into QDatDroid.

The hard part: the radio doesn't behave like a sled

Adopting the SDK was only the start. The real effort was making QDatDroid's ease of use survive the move — every convenience operators rely on had to work identically on a radio that behaves very differently from a sled. Bringing the app up on a TC701, we documented seven distinct limitations of the on-SoC Qualcomm radio and built a workaround for each.

What it took: a six-day sprint, six builds

This was concentrated work. From 9 to 14 June 2026, QDatDroid went through six build increments — versions 2.1.7 through 2.3.2 (internal build numbers 82 to 88) — across roughly three dozen commits, each a revision validated on real TC701 hardware: integrated-reader detection and connection, the emulated locate, a factory-reset region wizard, the SDK-adoption work above, and finally the 2.3.2 locate-and-settings polish. The result, QDatDroid 2.3.2, was regression-tested against the Zebra TC701 and its integrated Qualcomm RFID engine — with the Bluetooth sleds (RFD40 / RFD4031) confirmed unaffected.

Why it matters

The point of all of it is that none of it shows. Pick up a sledless TC701, TC501, or ET401 and QDatDroid does exactly what it does on a sled: one-tap instant demo, Visual Tag Locate, temperature off CoolTag and Axzon sensor tags, every scan GPS-and-time stamped and streamed live. The on-SoC radio trades range for form factor — roughly 2 m and 200+ tags per second versus a sled's ~12 m and 1,300 — which is exactly the right trade for item-level and cold-chain work at the shelf, the bench, or the receiving door.

As the TC501, TC701, and ET401 reach mass availability through 2026, customers can collapse “handheld + sled” into a single rugged device — and, on the Dragonwing Q-6690's 6-TOPS AI engine, eventually run shelf-life prediction and excursion detection on the device itself. QDatDroid is ready for that hardware on day one, because the work to get there is already done.

For a private, on-premise deployment on Dragonwing hardware inside your own perimeter, book a demo.