Yes, and that report indicates a null result. But they also indicate that the driving frequency was way mistuned for the cavity. They were driving it with a magnetron from a microwave oven, which output a frequency far from optimal. This indicates a low-budget operation, and not one in a lab with lots of microwave gear.

NASA, at least, has a microwave source with the right frequency. But they report "researchers were now working on a new integrated analytical tool to help separate EmDrive thrust pulse waveform contributions from the thermal expansion interference". That indicates this thing is still way too close to the noise threshold.

Back when cold fusion was taken seriously, I went to a Stanford talk where a physicist described their attempts to replicate the experiment. At first, he said, they had the apparatus surrounded with radiation detectors and alarms, in case it produced a dangerous burst of neutrons. After a while, they realized that wasn't going to happen. They discovered that the effect being measured was about twice background radiation. Then they discovered that people moving around the apparatus affected neutron readings by more than the measured amount. (Humans are mostly water, and thus reflect neutrons.) Finally they moved the experiment to a "neutron cube" built from lead bricks, where the background radiation was very low. The measured neutron readings went way down. That's what it's like when a phenomenon is near the noise threshold.

The German team did it for a BBC documentary on "junk science" - they were looking for a null result, so while I take claims that the tech works with a pinch of salt, the same applies here.

Everybody acts to their own interests.

Ah yes, the typical "let's prove this doesn't work" so do it half-assed and then go "see, it doesn't"

It is possible to "disprove" a lot of known science with those kind of experiments

My understanding was that the report didn't show a null report:

The device produced positive thrusts in the positive direction and negative thrusts in the negative direction of about 20 micronewtons in a hard vacuum, consistent with the low Q factor.

Besides being tested horizontally in both directions on the torsion pendulum, the cavity was also set upwards as a "null" configuration. However, this vertical test intended to be the experimental control showed an anomalous thrust of hundreds of micronewtons that could be caused by a magnetic interaction with the power feeding lines going to and from liquid metal contacts in the setup.

(from Wikipedia)

Am I misreading this somehow?

> That's what it's like when a phenomenon is near the noise threshold.

That's not really noise level, that's just unknown systematic errors.

he just means here are all kinds of piddly effects that will throw it off. If you switch the thing on and you get a huge effect size with high signal/noise you don't need to worry about stuff like who is standing near the tank other than to worry that they're being irradiated with fast neutrons.

If you're that low-budget why would you not just size the cavity to match a commercial microwave oven magnetron?