The initial issue was to determine whether the MAX232 was properly functional. That can be done with a voltmeter, and further, seems it hasn't been done.

The question of whether a logic analyzer can interpret RS232 signal levels isn't relevant at all in answering that question, and it's not adequate in determining whether there's something wrong, either with the PC or with the expectations.

If the V+ and V- outputs are at nominally +/- 10 volts, respectively, and the device produces valid RS232 levels when its logic inputs are driven to +5 or Gnd, and produces valid logic levels when its RS232 inputs are driven to valid RS232 levels, e.g. V+ and V-, that answers the question about the MAX232, doesn't it? An oscilloscope will answer any questions regarding timing. Without the 'scope, the performance issue is open, of course, and not every logic analyzer will settle those questions.

RE