#678 breaks [hw tune] high frequency voltage readings on some platforms

[0x2b3bfa0]

Since the merge of #678, the high frequency antenna voltage readings from hw tune are far below the values reported by older firmware versions. The device used for testing purposes was Elechouse Proxmark3 Easy, but this might be happening on all the non-RDV4 devices.

---

hw tune high frequency readings before 050aa18b13084b0d57ffcee352bf7a30b828b8f5

# LF antenna: 26.12 V @   125.00 kHz
# LF antenna: 31.49 V @   134.00 kHz
# LF optimal: 33.55 V @   130.43 kHz
# HF antenna: 36.26 V @    13.56 MHz

hw tune high frequency readings after 050aa18b13084b0d57ffcee352bf7a30b828b8f5

# LF antenna: 26.26 V @   125.00 kHz
# LF antenna: 31.62 V @   134.00 kHz
# LF optimal: 33.55 V @   130.43 kHz
# HF antenna:  6.24 V @    13.56 MHz
# Your HF antenna is marginal.

[pwpiwi]

Did you rebuild the client after flashing the firmware?

[0x2b3bfa0]

Did you rebuild the client after flashing the firmware?

Indeed! The testing procedure used to chase the offending commit has been semi-automated with the following script:

while read commit; do
    git checkout "$commit"
    make clean && make all
    client/flasher "$PORT" -b bootrom/obj/bootrom.elf \
                              armsrc/obj/fullimage.elf
    client/proxmark3 "$PORT" -c 'hw tune'
done < commits

Note: $PORT represents the path of the character device (`/dev/ttyACM), and the iteration over the lines in the file namedcommits` is an oversimplification of a hand-operated binary search algorithm.*

[pwpiwi]

Sorry, but I cannot confirm the issue. I compared the most current HEAD (5bc3841) with the release before 050aa18 (which is 2758d83) on three different PM3s (Original, RDV2, RDV3 Easy). HF voltage results look OK and consistent:

PM3          |   5bc3841    |   2758d83 
-------------+--------------------------
Original     |    22.05V    |    22.12V
RDV2         |    32.26V    |    32.47V
RDV3 Easy    |    30.91V    |    30.91V

[0x2b3bfa0]

I've bought my PM3 Easy (blue circuit board) on AliExpress, and chances are that it was a knockoff. I've sent it back to the seller for a full refund, just in case it was a hardware issue, but I'll repeat the tests after getting a new device.

What a strange issue!