It's impossible to determine if there's anything wrong at all with the JTAG signals due to the low time resolution. You also don't tell if you experience an actual communication problem. The waveforms seem to show valid digital signal levels.
According to my experience, the cause for different behaviour of USB blaster versions is more in the driver impedance than the cable. Insufficient TCK signal quality, particularly ringing edges causing double clocking are observed with some boards. Often the problem can be fixed by adding a small capacitor (e.g. 15 - 22 pF) between TCK and ground near the FPGA.
P.S.: I overlooked the reported error. I don't think however that it can be primarily associated with the signal level variations.