U-Blox M8T vs. Datagnss HD9310

Recently the price for the Allystar HD9310 based dual-frequency GNSS receiver dropped significantly. Compared to the well-known single-frequency U-Blox M8T receiver, the price difference is less than 50 $. Both of them are not available on the european market.
In order to rule out an antenna-specific influence on the measurement results, both receivers should share one antenna uplink. Unfortunately, the professional splitter equipment Tallysman TW-150 (32-0150DCL-0) is very expensive and also not available to end customers of the european market. At the recommendation of my brother, I used a Technisat Sat Splitter which performed very well at a comparatively low price.
After 3 hours of stabilization no statistically significant differences could be observed.
The frequency coverage of the used high quality antenna Tallysman TW2712 is limited to L1/G1/B1/E1, so in order to explore the full potential of the L2 capability of the Datagnss receiver I should be using a triple-band antenna as recommend by Dr. Bertges.
But at the moment the price tag of a TW3972 or a TW7972 is just below the height of a navigation satellite.

Left: U-Blox M8T // Right: Datagnss HD9310 both connected to a single Tallysman TW2712
Short-Term performance: Position Performance within one hour in sub-centimeter range!
Long-Time Performance: More than 8 hours of measurement data; position precision better than 10cm.
3D Skyview Plot, created with my fork of RaspAP. I will update my GitHub repository shortly, to make this feature publically available.