There is one technical thing to keep in mind about hardsyncing digital oscillators in general. There is a relation between the samplerate at which the waveform is calculated and something called 'jitter', a kind of distortion that sounds like the name implies. Jitter is in fact a very small amount of asymmetry in the waveform caused by the fact that, when using hardsync, the starting point of a synced waveform can be no more precise than the samplerate allows. That wouldn't be so bad but for the fact that the amount of asymmetry changes dynamically in a rate depending on the difference in frequency between the samplerate and the syncing wave's frequency. This means that hardsyncing a digital oscillator can introduce a very small spike in the waveform. This occurs when mixing the synced waveform with the syncing waveform, e.g. when crossfading between the waveforms or ringmodulating them. Due to the fixed samplerate of 96kHz the zero crossing point, which is the syncing point, will be quantized to the samplerate on the synced oscillators sync input. The resulting small spike is the difference between the anti-aliased syncing waveform that actually crossses through zero "in between" two samples and the quantized sync moment of the synced waveform. Notably the pulse waveform can suffer from this phenomenon. Normally the spike is filtered away in the filter section, but at high cutoff and/or resonance settings of the filter the spike can excite the filter in an unwanted way.
How the pulse waveform on the Modular suffers from jitter depends largely on the polarity of the syncing waveform.

Here is a testpatch of a master oscillator hardsynced to a slave oscillator.

With the module POLARITY the output of both oscillators can be inverted by selecting buttons 3 or 4. Test out combinations of two sawtooth waveforms, two square waveforms with and without PWidth on the master oscillator and PWidth on the master and a sawtooth on the slave oscillator. Do the same tests with the slave oscillator set to a ratio of 2:1, so one octave higher.
Notice the difference in sound between mixing inverted and non-inverted waveforms for both sawtooth and squarewave settings. Inverting sawtooth waveforms will raise the mix waveform by an octave. However non-inverted square waveforms will sound much thinner than inverted square waveforms.
If the master oscillator has PWidth then the slave oscillator will follow the PWM effect as a siren-like vibrato if the outputs are not inverted. If the outputs are inverted the slave will not follow the PWM effect and it has a stable frequency. Note that on a traditional analog oscillator the sawtooth will not follow the PWM effect and always has a stable frequency.

The conclusion of this test is that it is a good idea to add the possibility of inverting the output of the oscillator that hardsyncs the other oscillator. That other oscillator can have this option as well, to be able to correct for the differences when using sawtooth or square waveforms. This option is only needed when hardsync or a 'suboctave divider' is applied. The suboctave divider will be constructed later in this workshop.