Hi Schnappi,

Thanks for the suggestions!

Regarding the charger: I am using a professional Victron power supply, which is usually very 'clean', but I will definitely try a session without the charger attached for a few minutes to see if the keys/s rate improves. It's a good point about potential EMI interference.

As for the laptops: I will make sure to disable Windows Defender and Firewall completely during the next test to ensure no background process is throttling the connection to the D5T5 servers.

I've tried at various times during the week, but the result has been a consistent 29 keys/s. I’ll report back if the 'no-charger' or 'no-firewall' approach makes any difference in the latency.

Thanks for thinking along!

Is the car in Ignition II (turn the ignition knob to start for 5+ seconds)?

Hi everyone,

I’m sharing my experience regarding a CEM PIN decode on my 2016 XC90 T8 (SPA), as I’ve hit a technical brick wall that makes the process impossible.

My VDASH session is consistently stuck at a rate of only 29 keys per second, resulting in an estimated time of 160 hours (nearly a full week).

Technical Setup & Troubleshooting performed:

Connection: Switched from Wi-Fi to a dedicated 200 Mbit wired Ethernet (UTP) line.
Network Latency: A manual ping test to d5t5.com shows a stable 40ms. However, VDASH fluctuates wildly, showing a "red" latency of 1752ms at times, and staying around 143ms at best.
Hardware: Tested with two different laptops and various USB ports to rule out driver/OS bottlenecks.
Power: Car is connected to a 20A Victron charger, maintaining 13.9V.
Interface: Using the Super J2534 cable.

The Issue:
Despite the healthy 40ms internet connection, there is a massive bottleneck in the communication between VDASH and the servers. A 160-hour decoding time is not feasible for a home setup.

I opened a support ticket 9 days ago but have received zero response.

Is anyone else experiencing this 29 keys/s limit on the SPA platform? Is this a server-side throttle, or is there a specific J2534 driver setting that needs to be tweaked to get closer to the expected 100+ keys/s?

I would appreciate any insights from the developers or the community, as the current performance makes the PIN recovery feature unusable.