I am not an expert on the DisplayPort, but from what I read about it I understood that there are different speeds each DP version supports (called HBR1, HBR2, HBR3, etc.). DP 1.2 runs at HBR2. This is what 5k+, 8k, and 8k+ uses and what is sufficient for them.
8k-X requires HBR3 and DP 1.4 (technically it was DP 1.3 which introduced HBR3 speed). We know from some leaked info that the DP bridge Pimax used in 5k+ and 8k does not support HBR3. If they use the same bridge in 8k+ (technically there is no need to change it) then there is no difference and 8k+ should behave the same.
In other words, extension cables which worked on 8k or 5k+ should work exactly the same, the wireless (if it ever worked) should work the same etc.
Now, what about 8k-X, which has to run (in its native mode) at HBR3 possibly with DSC as well? First consequence is you need a better cable, (HBR2 ~ 5 Gpbs, HBR3 ~ 8 Gbps), most likely shorter, with better shielding and better material overall.
This is the reason why Pimax cannot confirm any working extension (I am not sure they tried the fiber, but I would be very careful about claiming the fiber can solve it without testing it either, because the speed specs for the fiber bridges might not be up to the HBR3 specs either).
You may (and rightly so) assume that in the upscaler mode the bandwidth 8k-X requires is exactly the same as the one of all three previous headsets (basically QHD res), but it does not necessarily imply that the implemented DP transmission runs at the same configuration (i.e. HBR2 speed). I can imagine that there is handshake on DP interface happening, which tries to run at the highest possible speed (HBR3) and if it fails there is no fall-back solution implemented to try the slower speed as well.
I do not think that the control over which bitrate the DP uses is exposed via the normal OS API, but probably only through some proprietary graphics card driver interface and Pimax needs to implement it. Which probably they did not.
The other option could be to instruct 8k-X to “downgrade” its DP interface to HBR2 speed only, to force the host to switch as well, but this again looks like an additional feature which needs to be implemented.
So what you consider to be 10 minutes test could be several man-days work.