I recently encountered similar issues with an XTS5000. The programming was the easy part. I have a hotspot and 1 P25 repeater (that is a multi-mode based off PiStar) to use. I noticed the following.
1) Even when not in use, the first call takes multiple tries to get in.
2) BER is all over the place, from perfect to not great, even with a strong signal (-43dbm)
3) There is often ridiculous packet loss coming back into pistar (80%+), even with excellent internet connections (hardwired 400Mbps)
4) Often I notice it can take up to 3 seconds for pistar to accept the call
5) none of these happen with DMR, D-Star or YSF on the same hardware
Someone mentioned it could be related to frequency stability. I've exhausted the tools I have available. Utilizing a dummy load and a power meter/frequency counter (uncalibrated) I determined output power is expected on both high and low, and frequency is stable to 1kHz (the accuracy of my equipment). I can't tell if it's exactly on frequency, because its not calibrated. I also used the dummy load and my TinySA, and the P25 matches the wave form from everything I see about calibrating P25 equipment. Obviously its not a true alignment or measurement, but it does show that I either need to find someone to stick it on good equipment for an alignment or deal with it.
It could be two different things. Yours could be old & drifting too, or ....... PiStar does support P25, but its the just glitchy. I'm thinking its a bit of both but have no way to verify.